Understanding the Factors

Influencing Children’s Participation in

Brand Communities

Margurite Louise Hook

BBus (First Class Hons), BCom

A thesis presented in fulfilment of the requirements

for the Degree of Doctor of Philosophy

in Management

University of Newcastle

June, 2018

This research was supported by an Australian Government Research

Training Program (RTP) Scholarship

Declarations

Statement of Originality

I hereby certify that to the best of my knowledge and belief this thesis is my own work

and contains no material previously published or written by another person except where

due references and acknowledgements are made. It contains no material which has been

previously submitted by me for the award of any other degree or diploma in any university

or other tertiary institution. I give consent to the final version of my thesis being made

available worldwide when deposited in the University’s Digital Repository, subject to the

provisions of the Copyright Act 1968.

Signed Dated 29/06/2018

(Margurite L Hook)

Thesis by Publication

I hereby certify that this thesis is in the form of a series of *papers. I have included as part

of the thesis a written statement from each co-author, endorsed in writing by the Faculty

Assistant Dean (Research Training), attesting to my contribution to any jointly authored

papers. (*refer to clause 39.2 of the Rules Governing Research Higher Degrees for

acceptable papers).

Signed Dated 29/06/2018

(Margurite L Hook)

i
Acknowledgements

The past three years have been an exciting journey, one that has ended too soon. My PhD

experiences were shaped by the amazing people I was lucky to encounter and have

helping me through.

Firstly, to my supportive and hard-working supervisors Stacey Baxter and Alicia

Kulczynski. I will be forever grateful to you both for helping me achieve my PhD, and

gain a publication record so early in my career. Even though at times your feedback and

suggestions weighed heavily on me, you always encouraged me and pushed me to achieve

work that I could be proud of (a high bar, as you both know!). I hope the conclusion of

my PhD does not mean the end of our collaboration, and that we can work together in the

future as well.

Second, to my fiancé Nathan. Thank you, Nate, for always supporting me, even though I

may be in my own PhD world most of the time. You have helped me keep on track, and

importantly told me when to take a much-needed break. Hopefully now this is over I will

be less muddled and thinking about less than a million things all the time…hopefully.

Third, to my Mum and Dad. I cannot express how much your support has helped me

through the past few years, even with Dad’s constant ‘when does your PhD finish?’ and

‘how many papers have you published now?’ You have always instilled in me to be the

best I can be and without that, I doubt I would have even contemplated attempting a PhD,

let alone achieving one.

Lastly, to all my friends and colleagues at the University of Newcastle, there are too many

of you to list by name. Thank you for all the advice, support and friendly smiles I have

received over the years.

ii
1. Table of Contents

1. Table of Contents .........................................................................................................iii

2. List of Publications Included as Part of the Thesis ....................................................... 1

2.1 Published Work, in Order of Appearance ............................................................... 1

2.2 Unpublished Work, in Order of Appearance ........................................................... 1

2.2 Copyright Permission Statement ............................................................................. 2

3. List of Additional Publications ..................................................................................... 3

4. Abstract ......................................................................................................................... 4

5. Thesis overview ............................................................................................................ 6

5.1 Introduction ............................................................................................................. 7

5.2 Research Aim ........................................................................................................ 11

5.3 Research Questions ............................................................................................... 11

5.4 Method................................................................................................................... 12

5.4.1 Ethical Considerations ................................................................................... 15

5.5 Discussion ............................................................................................................. 16

5.6 Implications ........................................................................................................... 20

5.7 Significance and Contribution ............................................................................... 21

5.8 Structure of Thesis ................................................................................................. 23

6. Published Papers ......................................................................................................... 24

6.1 Criteria for Journal Selection................................................................................. 25

6.1.1 Journal of Brand Management ....................................................................... 25

iii
 6.1.2 Journal of Marketing Behavior ...................................................................... 26

6.1.3 International Journal of Consumer Studies .................................................... 26

6.1.4 Journal of Brand Management ....................................................................... 27

6.2 Paper One: Literature Review ............................................................................... 29

6.2.1 Statement of Contribution of Others .............................................................. 29

6.2.2 Third Party Copyright Acknowledgement ..................................................... 30

6.2.3 Overview and Contribution of Paper ............................................................. 31

6.2.4 Full Paper ....................................................................................................... 34

6.3 Paper Two: Replication and Extension ................................................................. 71

6.3.1 Statement of Contribution of Others .............................................................. 71

6.3.2 Third Party Copyright Acknowledgement ..................................................... 72

6.3.3 Overview and Contribution of Paper ............................................................. 73

6.3.4 Full Paper ....................................................................................................... 76

6.4 Paper Three: New Model ...................................................................................... 89

6.4.1 Statement of Contribution of Others .............................................................. 89

6.4.2 Third Party Copyright Acknowledgement ..................................................... 90

6.4.3 Overview and Contribution of Paper ............................................................. 91

6.4.4 Full Paper ....................................................................................................... 94

6.5 Paper Four: Similarity Attraction ........................................................................ 120

6.5.1 Statement of Contribution of Others ............................................................ 120

6.5.2 Overview and Contribution of Paper ........................................................... 121

iv
 6.5.3 Full Paper ..................................................................................................... 124

7. Additional Papers ...................................................................................................... 151

7.1 Conference Paper 1.............................................................................................. 152

7.1.1 Full Paper ..................................................................................................... 153

7.2 Conference Paper 2.............................................................................................. 167

7.2.1 Full Paper ..................................................................................................... 168

8. References ................................................................................................................. 175

9. Appendices ................................................................................................................ 182

9.1 Human Ethics Approval ...................................................................................... 182

9.1.1 Approval for Studies 2 and 3 (Papers Two and Three) ................................ 182

9.1.2 Approval for Studies 4, 5 and 6 (Paper Four) .............................................. 183

9.2 Participant Information Sheets ............................................................................ 184

9.2.1 Paper Two and Paper Three: Participant Information Sheet ........................ 184

9.2.2 Paper Four: Participant Information Sheet ................................................... 186

9.3 Questionnaires Employed.................................................................................... 188

9.3.1 Paper Two: Replication and Extension ........................................................ 188

9.3.2 Paper Three: New Model ............................................................................. 200

9.3.3 Paper Four: Similarity Attraction ................................................................. 204

9.4 Third Party Copyright Permission Communications .......................................... 215

9.4.1 Paper One ..................................................................................................... 215

9.4.2 Paper Two .................................................................................................... 218

v
 9.4.3 Paper Three .................................................................................................. 219

9.5 Reference Lists for Papers ................................................................................... 223

9.5.1 Paper One Reference List ............................................................................ 223

9.5.2 Paper Two Reference List ............................................................................ 230

9.5.3 Paper Three Reference List .......................................................................... 231

9.5.4 Paper Four Reference List............................................................................ 235

9.5.5 Conference Paper 1 Reference List .............................................................. 239

9.5.5 Conference Paper 2 Reference List .............................................................. 241

9.6 ABDC Ranking Explanation ............................................................................... 243

9.7 H-index Explanation............................................................................................ 244

vi
2. List of Publications Included as Part of the Thesis

2.1 Published Work, in Order of Appearance

1. Hook, M., Baxter S.M. & Kulczynski A. (forthcoming) Antecedents and

Consequences of Participation in Brand Communities: A Systematic Literature

Review, Journal of Brand Management, doi: 10.1057/s41262-017-0079-8

This paper is provided in Section 6.2 of the thesis.

2. Hook, M., Baxter S.M. & Kulczynski A. (2017). Antecedents and Consequences

of Children’s Brand Community Participation: A Replication and Extension

Study, Journal of Marketing Behavior, 3(1), 63-72, doi: 10.561/107.00000042

This paper is provided in Section 6.3 of the thesis.

3. Hook M., Baxter S.M. & Kulczynski A. (2016). Children's Participation in Brand-

Based Social Networks: Examining the Role of Evaluative Social Identity, Self-

Esteem and Anticipated Emotions on Commitment and Desire to Recommend,

International Journal of Consumer Studies, 40(5), 552-561, doi:

10.1111/ijcs.12300

This paper is provided in Section 6.4 of the thesis.

2.2 Unpublished Work, in Order of Appearance

1. Title: “I'm Like You, You’re Like me, We Make a Great Brand Community!’

Similarity-Attraction and Children’s Brand Community Participation”. Submitted

to the Journal of Brand Management June 2018

This paper is provided in Section 6.5 of the thesis.

1
2.2 Copyright Permission Statement

I warrant that I have obtained, where necessary, permission from the copyright owners to

use any third-party copyright material reproduced in the thesis, or to use any of my own

published work (journal articles) in which the copyright is held by another party (e.g.

publisher, co-author).

2
3. List of Additional Publications

The following additional publications are relevant to the thesis and were prepared by the

candidate. They are provided in Section 7 of this thesis.

Additional Publication 1:

Hook M., Baxter S.M. & Kulczynski A., (2016). Understanding the Complex Interplay

Between Evaluative Social Identity, Negative Anticipated Emotions and Self-Esteem on

a Child’s Commitment to Brand Communities, 45th European Marketing Academy

Conference 2016 – ‘Marketing in the Age of Data’, Oslo, Norway.

Additional Publication 2:

Hook M., Baxter S.M. & Kulczynski A., (2017). ‘You’re like me’. Children’s brand

community participation, Australian and New Zealand Marketing Academy Conference

2017 – ‘Marketing for Impact’, Melbourne, Australia.

Note: The candidate also presented a work-in-progress version of this thesis as part of the

2016 Australian and New Zealand Marketing Academy Doctoral Colloquium in

Christchurch, New Zealand.

3
4. Abstract

Child-oriented brand communities are increasingly being utilised by marketers, to

encourage children’s engagement and participation with brands. While brand

communities have been extensively researched for adult participants, children are

cognitively and socio-emotionally different to adults, prompting the need for targeted

research in the area. This thesis by publication, comprising of four core publications and

two supplementary papers, addresses the need to understand the factors influencing

children’s brand community participation. Through undertaking a mixture of secondary,

descriptive and causal research, this thesis advances the knowledge of child brand

community participants, aiding academics and practitioners alike.

Firstly, an analysis of the extant literature on brand communities was undertaken.

Through consolidating the literature, it was revealed that several areas require academic

attention, including that of child-oriented brand communities. Secondly, the replication

of a seminal adult-oriented brand community model was performed, and extended to a

child context. Results indicated that children’s participation in brand communities was

different to that of adults. Thirdly, a new model to predict children’s participation was

developed. Findings revealed that subjective group dynamics and personal self-esteem

influenced children’s brand community participation. Children with low personal self-

esteem were seen to have stronger recommendation intentions and were more committed

members, than children with high personal self-esteem. Lastly, causal research was

undertaken to produce a model that predicts children’s participation. The three-part

experimental study showed that children had a stronger desire to participate when the

community members were characteristically like themselves, and loyal to the community.

4
This thesis contributes to the brand community literature, presenting much needed

information on child participants. The models developed and interdisciplinary theories

employed, contribute to the brand community literature, providing new avenues through

which to investigate brand community participation. Advice is provided to practitioners

regarding the unique characteristics of child participants and how to effectively manage

and promote child-oriented brand communities.

5
5. Thesis overview

This section provides an overview, explaining how the four core papers and two

supplementary papers are linked to the overall research aim of the thesis. It begins with

an introduction on the thesis topic, brand communities, which is followed by a summary

of the thesis’s aim, and the research questions. The overall research method is described

briefly, as well as the associated ethical considerations that were addressed. The

remainder of this section outlines the papers of the thesis, detailing how they relate to one

another, the implications of each paper, and finally the significance and contributions of

the papers.

Section 5 concludes with a description of the structure of the thesis.

6
5.1 Introduction

Brand communities are a popular marketing tool, encouraging consumers to build

relationships with a brand and with other consumers. Defined as ‘a specialized, non-

geographically bound community, based on a set of social relationships among admirers

of a brand’ (Muniz Jr & O’Guinn, 2001, p 412), these communities facilitate the

development of brand relationships and ultimately influence factors such as: brand

loyalty, word-of-mouth, brand trust and purchase intentions (Algesheimer, Dholakia, &

Herrmann, 2005; Laroche, Habibi, Richard, & Sankaranarayanan, 2012; Woisetschläger,

Hartleb, & Blut, 2008).

Marketers have been employing brand communities as branding tools for many years

with great success. One example, is the ‘Harley Owners Group’ Harley-Davidson brand

community, which helped launch the brands’ value to over $7.8 billion (Fournier & Lee,

2009). Other successful brand community examples include: Camp Jeep (McAlexander,

Schouten, & Koenig, 2002), the My Nutella Community (Cova & Pace, 2006) and

Sephora (Thumm, 2015; Ungerleider, 2014). In addition, GoPro has recently amassed a

thriving community with around 6,000 videos uploaded everyday by loyal fans and users

of their products (Foote, 2017).

Many brands are focussing their sights on encouraging child consumers such as: Lego (N.

Lee, 2017; Lego, 2016), Moshi Monsters (Mind Candy, 2016), and Mattel’s Barbie

(Mattel, 2016), to bolster excitement and engagement with the brand. For example, the

‘Lego Life’ brand community is specifically designed for children aged under 13, for use

on modern devices such as tablets and smart phones, to connect members together (N.

Lee, 2017). The app for ‘Lego Life’ had over 1 million android device downloads in its

first year alone (LEGO System A/S, 2017).

7
The importance of child consumers cannot be underestimated. Children’s spending and

influencing power has reached over $200 billion per year in the US, with spending

focussing on big brand names such as Disney, Mattel and Nike (Collins & Mitchell, 2015).

Marketing efforts are often directed at child consumers since they represent not one, but

three market avenues: (1) consumers, (2) influencers and a (3) future market (Götze,

2002; McNeal, 1998).

Children form a relationship with brands, which becomes part of their self identity,

meaning that to them, the brand has a sense of importance in their everyday life (Ji, 2002).

Due to the significance of brands for children, the desire to use only brand-specific

products is very strong (Reveladvertising, 2017). Marketing is often directed towards

child consumers, since brands are of high importance to them (Collins & Mitchell, 2015).

Despite child-oriented brand communities being employed by marketing practitioners,

there is scant attention by academics.

The study of adult-oriented brand communities has led to hundreds of published papers.

Since Muniz & O’Guinn’s (2001) well-accepted definition of a brand community, many

academics have been active in examining brand communities, and behaviour in these

communities. Academics have studied several contexts, applied multiple approaches,

examined various brand community forms (e.g. online and offline), and produced

numerous unique models that utilise a range of constructs. The large number of studies

and models produced during the last few decades have provided worthwhile findings;

however, it is suggested that the vast array of information may be a source of confusion

to academics and practitioners, due to the amount of information and lack of consistency

across studies.

8
No summary or review of the literature had been undertaken, prior to the commencement

of this thesis. The first paper of this thesis addressed the need to consolidate the brand

community literature. Paper One entitled: “Antecedents and Consequences of

Participation in Brand Communities: A Literature Review” (hereafter referred to as

‘Paper One: Literature Review’), addressed the first research question of the thesis:

What are the antecedents and consequences of brand community participation?

Paper One: Literature Review is a systematic literature review, undertaken to provide an

in-depth analysis of the current knowledge of brand community participation. From the

findings of Paper One: Literature Review, research questions were formulated for future

research into the area of child-oriented brand communities, as well as other brand

community areas. The subsequent papers of this thesis provide a foundation for

investigating the proposed child-oriented brand community research questions.

One of the most cited adult-oriented brand community participation models, as identified

in Paper One: Literature Review, was developed and tested by Bagozzi and Dholakia

(2006). This led to an investigation examining whether this seminal adult-oriented model

could be applied to the context of children. Paper Two addressed the research question:

Do the antecedents and consequences of brand community participation differ

between adult- and child- participants?

To address this research question, a replication and extension study was undertaken,

retesting Bagozzi and Dholakia’s (2006) model, utilising a child sample. Paper Two is

entitled: “Antecedents and Consequences of Children’s Brand Community Participation:

A Replication and Extension Study” (hereafter referred to as ‘Paper Two: Replication

and Extension’). Comparisons were made between the adult participant findings in the

9
original study (Bagozzi & Dholakia, 2006), and the child participants of Paper Two:

Replication and Extension’s study.

Having found in Paper Two: Replication and Extension that the model used to explain

adult-oriented brand community participation did not fully predict the child participant

behaviour, Papers Three and Four sought to begin research on the final research question

of the thesis:

What antecedents of brand community participation are unique to child

participants?

Paper Three, entitled: “Children’s Participation in Brand-Based Social Networks:

Examining the Role of Evaluative Social Identity, Self-Esteem and Anticipated Emotions

on Commitment and Desire to Recommend” (hereafter referred to as ‘Paper Three: New

Model’), created the first model to investigate the context of child-oriented brand

community participation. Drawing from the theories identified in Paper Two: Replication

and Extension and developmental psychology, Paper Three: New Model introduced the

construct of personal self-esteem and the theory of Subjective Group Dynamics (SGD;

Abrams & Rutland, 2008) into the field of brand communities.

Due to the significance of SGD on children’s group interactions, and the findings of Paper

Three: New Model showing that the theory explains child-oriented brand communities,

SGD theory was chosen to be explored further. Paper Four continued the investigation

on the impact of SGD on the drivers of children’s brand community participation. A

review of the previous studies exploring SGD amongst children revealed that similarity

attraction, with regards to shared characteristics, can impact children’s relationships

(Berger, 2008; Haselager, Hartup, Lieshout, & Riksen‐Walraven, 1998). Drawing from

this knowledge, Paper Four investigated the role of similarity attraction and SGD on

10
children’s brand community participation through undertaking three experimental

studies. Paper Four is entitled: “‘I'm Like You, You’re Like me, We Make a Great Brand

Community!’ Similarity-Attraction and Children’s Brand Community Participation”

(hereafter referred to as ‘Paper Four: Similarity Attraction’).

Two conference papers were also developed based on the overall research question. These

papers were prepared and presented at separate international conferences by the

candidate. Conference Paper One is entitled: “Understanding the Complex Interplay

Between Evaluative Social Identity, Negative Anticipated Emotions and Self-Esteem on

a Child’s Commitment to Brand Communities” and Conference Paper Two is entitled:

“‘You’re like me’. Children’s Brand Community Participation”. These are included in

Section 7 at the end of the thesis as additional papers.

5.2 Research Aim

The primary aim of the research was to:

Understand the factors influencing children’s participation in brand

communities

5.3 Research Questions

i. What are the antecedents and consequences of brand community participation?

ii. Do the antecedents and consequences of brand community participation differ

between adult- and child- participants?

11
 iii. What antecedents of brand community participation are unique to child

participants?

5.4 Method

This section provides an overview of the research methods. Further explanation of the

research method utilised for each paper is covered in the respective publication. Over the

four papers, six studies were undertaken using a variety of research approaches

(secondary, descriptive, and causal). The use of multiple research approaches was

undertaken to broaden the dimensions and scope of the research (Mertens, 2014; Morse,

2016). The core of this thesis was quantitative, consisting of descriptive and causal

methods, with a supplementary secondary research component, so that multiple research

purposes were achieved.

The first study (Paper One), consisted of secondary data collection on brand community

literature. A systematic review of over 1,900 articles was undertaken, with 41 examined

and synthesised in detail. The systematic literature review method was chosen, as opposed

to a narrative review. Although narrative reviews are often undertaken in social sciences,

they are criticized for being singular accounts, with reviewed articles chosen at the

authors’ discretion, therefore subject to bias (Fink, 2001; Hart, 1998; Tranfield, Denyer,

& Smart, 2003). Systematic reviews, on the other hand, are favoured by many science

disciplines due to the rigorous process undertaken to select articles for review (Mulrow,

1994). The systematic review method was chosen to ensure all relevant articles were

analysed, not just the most cited or most ‘important’. The five remaining studies all

entailed the collection of primary data.

Study 2 (Paper Two), employed a descriptive cross-sectional research design, comprising

of a quantitative questionnaire (questionnaires for all studies are provided in Appendix

12
9.3), using two samples (branded community and non-branded community) of Australian

children aged 6-14 years. Participants were contacted via an external research panel

company (“ResearchNow”). The use of external research panels to source participant

samples is a popular method for behavioural researchers (Evans & Mathur, 2005),

providing reliable data that is both representative of the general population (Roulin,

2015), and yields insightful results (Fulgoni, 2014).

Participants completed the survey, for Study 2 (Paper Two), online. Online survey

collection has a number of significant advantages over offline collection that benefited

all empirical studies. Specifically, the online collection method allowed greater flexibility

for participants (Evans & Mathur, 2005) in terms of the time chosen to complete the

respective survey. This meant that the survey could be undertaken after school hours,

such as during the weekend. The online method also allowed for both parental consent

and child assent to be collected simultaneously, fulfilling the ethics requirements of the

research. Lastly, by conducting the research online, a large number of relevant

participants were able to be sourced in a relatively short time, providing a more diverse

and representative sample than face-to-face data collection (Evans & Mathur, 2005).

The two samples of Study 2 (Paper Two) were compared against each other, and against

the original paper findings (Bagozzi & Dholakia, 2006). Structural Equation Modelling

(SEM), using Analysis of a Moment Structures (AMOS), was employed to analyse the

data. This analysis method was chosen to replicate the approach adopted in the original

study. The use of SEM and AMOS in the field of marketing is common practice and

largely accepted by academics (e.g. Hair, Gabriel, & Patel, 2014; Steenkamp &

Baumgartner, 2000).

13
Study 3 (Paper Three), again undertook a descriptive cross-sectional research design,

using a sample of Australian children (6-14 years old) to participate in a questionnaire.

The sample was accessed via an external research panel (“ResearchNow”) and directed

to complete an online questionnaire, as per Study 2 (Paper Two). The data was analysed,

and a model produced, through the application of the PROCESS macro in SPSS.

PROCESS is a computational tool used to test mediation, moderation, as well as

moderated mediation (Hayes, 2012). Study 3 (Paper Three) investigated these forms of

relationships (mediation, moderation and moderated mediation) and therefore required a

data analysis tool that could analyse for these effects. This form of data analysis has

become commonly accepted, with many studies employing the method (Hayes, Montoya,

& Rockwood, 2017).

Studies 4, 5 and 6 were combined to form Paper Four. Due to the purpose of examining

cause-and-effect relationships (causal research), the research for Paper Four employed

an experimental research design. Experimental research design is highly regarded,

associated with robustness and producing trustworthy causal findings (Bryman & Bell,

2011). Research methods employing experimental designs are a prominent research

approach in psychology and behavioural research (Christensen, Johnson, Turner, &

Christensen, 2011).

Three separate studies were undertaken. For each study (4, 5 and 6) a separate sample of

Australian children (6-17 years old), sourced via an external research panel

(“ResearchNow”), were used. The experiment was conducted online, in an environment

that allowed variables to be controlled for, to ensure manipulations were unlikely to be

affected by extraneous factors (Bryman & Bell, 2011). All data analysis for Paper Four

was undertaken though implementing the PROCESS macro in SPSS. The PROCESS tool

14
was again chosen due to its ability to analyse mediation, moderation and moderated

mediation (Hayes, 2012).

The candidate undertook all aspects (across all methods: secondary, descriptive and

causal) of the data collection, questionnaire development, data analysis and synthesis, for

the six studies, across the four papers.

5.4.1 Ethical Considerations

Ethics approval was obtained for those empirical studies that involved human participants

(see Appendix 9.1 for copies of the approvals given by the University of Newcastle

Human Research Ethics Committee (HREC)). Due to the collection of only secondary

data for Paper One, no ethics approval was required.

The primary ethical considerations, for all empirical studies, related to the participation

of child respondents. When using a sample of children, guidelines advise parental or

guardian consent needs to be acquired (Hill, 2005). In line with the University of

Newcastle HREC guidelines (University of Newcastle, 2014), potential child respondents

for the empirical papers (Papers Two, Three and Four), and their guardians, were

provided with an information sheet explaining the general purpose of the research, and

information on how confidentiality and anonymity would be maintained (copies of

information sheets are provided in Appendix 9.2). Guardian consent and child assent to

the research was required, prior to participation of the child. To ensure anonymity, no

identifying information was gathered, with only basic demographic details collected (age,

gender and school year).

15
5.5 Discussion

The popularity of brand communities amongst practitioners has spurred an increase in

research investigating brand communities, however, only for adult participants. Given the

importance of children as consumers and growing number of child-oriented brand

communities, examining this area is argued to yield fruitful information. Whilst adult-

orientated research may guide the understanding of brand community participation, it is

argued differences in socio-emotional (Cicchetti & Cohen, 2006) and cognitive skills

(Piaget, 1972) could impact a child’s brand community participation. Due to these

differences, the behaviours evidenced in adult-oriented brand communities may not

convey the behaviours of child participants.

Socio-emotional development plays a key role in children’s behaviour within group

contexts (Betts & Stiller, 2014). Social skills are developed over time, a result of life

experiences, interpretations of events, and the way experiences are valued (Voinea &

Damian, 2014). With relationships being the centre of a brand community (Muniz Jr &

O’Guinn, 2001; Ouwersloot & Odekerken‐Schröder, 2008), children’s brand community

participation is likely to be influenced by these less-developed social skills. Given this,

an examination of developmental psychology literature was suggested to yield useful

insights for understanding child-oriented brand communities. Developmental psychology

studies have extensively explored children’s social relationships and the influence of

children’s less developed social-emotional skills is evident (e.g. Abrams, Rutland, &

Cameron, 2003; Bigler & Jones, 1997).

As well as socio-emotional skills, a child’s cognitive development impacts how they

interact in group contexts (Abrams et al., 2003). Children’s perceptions of groups and

social judgements take time to develop, thus impacting interactions in group settings

16
(such as brand communities), as evidenced in developmental psychology literature (e.g.

Abrams et al., 2014, 2003; Engelmann, Herrmann, Rapp, & Tomasello, 2016). Cognitive

development is not a simple process, and develops over time (Wohlwill, 1962).

Consequently, children do not have the same cognitive ability as that of adults.

Drawing from the socio-emotional and cognitive ability differences between adults and

children, the four papers and two conference papers of this thesis have provided a valuable

step towards filling the gap in the brand community literature, regarding child-oriented

brand community participation. Paper One: Literature Review identified the antecedents

and consequences of brand community participation, compiling and organising them into

logical categories, allowing for comparisons to be made across studies. Analysis of the

extant literature revealed that some brand community areas are highly researched, and

others need to be examined further, as they have received little to no attention, yet are

important. Considering these findings, research questions were proposed to guide future

research in the field.

Paper Two: Replication and Extension found that the seminal adult-oriented model

produced by Bagozzi & Dholakia (2006), did not fully replicate for the child context. It

was proposed that this may be due to developmental differences in cognitive (Piaget,

1972) and socio-emotional skills (Cicchetti & Cohen, 2006) between adults and children.

Paper Three: New Model examined the effect of evaluative social identity on brand

community commitment and recommendations, including the mediating roles of positive

and negative anticipated emotions, and the moderating influence of personal self-esteem.

These variables were drawn from developmental psychology and the model examined in

Paper Two: Replication and Extension (Bagozzi & Dholakia, 2006; Harter, 1993). Key

findings of Paper Three: New Model related to the moderating role of personal self-

17
esteem and the influence of subjective group dynamics. Low personal self-esteem

children showed stronger relationships between evaluative social identity and both

commitment and recommendations than children with high personal self-esteem. In

addition, the mediating effect of positive and negative anticipated emotions felt by child

members was due to the influence of in-group and out-group perceptions, as suggested

by subjective group dynamics theory (Abrams & Rutland, 2008).

Paper Four: Similarity Attraction examined the effect of member similarity on

participation desire, including the mediating role of respect towards the brand community

and the moderating influence of member deviance. These variables were drawn from

Subjective group dynamics theory (Abrams & Rutland, 2008), similarity attraction theory

(Aboud & Mendelson, 1996; Haselager et al., 1998; Hunter, Fox, & Jones, 2016) and

Paper One: Literature Review’s findings (Bagozzi & Dholakia, 2006; Dholakia, Bagozzi,

& Pearo, 2004; Kim, Choi, Qualls, & Han, 2008). Through undertaking three studies in

Paper Four: Similarity Attraction, children who perceived members of a brand

community to be characteristically similar to themselves were shown to have higher

participation desire. Once one member was seen as disloyal, however, this similarity

attraction effect attenuated. This moderation effect was suggested to be due to the impact

of subjective group dynamics on child-oriented brand communities.

Figure 1 provides a summary of how the key thesis topics are integrated, and which papers

they correspond with. Two avenues of research were identified as relevant to child-

oriented brand community participation, namely: findings from adult-oriented brand

communities and developmental psychology. Through undertaking each separate study,

these streams of research were examined and analysed, with links established amongst

the streams.

18
Figure 1 - Integration of Key Concepts

19
5.6 Implications

Children have the opportunity to participate in some form of brand community every day,

whether it is in a school playground, online via a website, or through an app on a smart device.

This thesis has uncovered several important insights for the field of marketing related to child-

oriented brand communities, with a particular focus on participation.

Firstly, marketers are provided with a concise summary of information regarding brand

community participation, and how the form of community (online, offline or social-media-

based) impacts participation. The results imply that the most common drivers and

consequences of participation differ depending upon which form the brand community takes.

Therefore, it is suggested that marketing strategies to encourage brand community participation

should be adapted, based on the form of brand community employed.

Second, brand communities with child participants were found to be uniquely different to those

with adult members. Due to this, it is proposed that marketers should not use the same strategies

for both distinct target audiences (adults and children).

Third, the thesis highlights that children with low personal self-esteem are at risk of being

20
manipulated by brands when they participate in their brand communities. Children and

guardians should be educated on the potential misuse of brands, to help protect vulnerable child

participants in brand communities. Marketers and academics also need to be made aware of

this issue to help ensure children’s wellbeing.

Fourth, the importance of developmental psychology in understanding child-oriented brand

community participation is shown. Subjective group dynamics theory was applied to explain

20
the influence of member deviance on respect towards the community, and subsequently

participation desire. There is evidence to suggest that communities can be harmed by deviant

behaviour, and deviant members. Marketers are advised to monitor their brand communities

for deviant members, due to the negative effects they present on attracting new participants and

members.

Lastly, this thesis shows that unique antecedents are evident for child-oriented brand

communities. Similarity attraction was shown to influence a child’s desire to participate in a

brand community. This suggests that elements such as member similarity should be promoted

when seeking child participation in a brand community.

5.7 Significance and Contribution

This thesis makes a significant contribution to the field of marketing, in particular brand

communities, through its six studies, across four separate papers. Although research into brand

communities is extensive (e.g. Dholakia, Bagozzi, & Pearo, 2004; Hung, 2014; Sicilia &

Palazón, 2008), this thesis adds to the literature through investigating the under-researched area

of child-oriented brand communities.

By summarising and compiling the extant brand community literature, the current knowledge

of brand community participation is synthesised. From an analysis of the antecedents and

consequences of participation, it is shown that academics still need to address some areas, even

though a substantial amount of research has been undertaken, as research questions are yet to

be answered. The understanding of how these antecedents and consequences differ across

brand community forms is highlighted, expanding upon academics’ knowledge in this area.

21
The research of this thesis responds to calls for future research on child brand community

participation (Flurry, Swimberghe, & Parker, 2014). Children are known to participate in these

brand communities, and brands are developing communities specifically for children. Through

undertaking the replication and extension study, this thesis demonstrates that child-oriented

communities are different to adult-oriented communities, and therefore, adult findings are not

applicable to children. This extends the understanding of brand community participation, and

offers areas for future research.

The impact of personal self-esteem is presented to the area of brand communities to show how

some children are more effected by the relationship between evaluative social identity and

anticipated emotions. Anticipated emotions are introduced as a mediator on children’s brand

community relationships between evaluative social identity and the constructs: community

commitment and community recommendations. By doing so, the understanding of anticipated

emotions and social identity, with regards to brand communities, is also extended.

The theory of subjective group dynamics is introduced to the area of brand communities to

explain children’s participation. By showing that in-group norms and deviant behaviour

influences child-oriented brand communities, the use of subjective group dynamics theory is

extended. Specifically, member deviance is identified as a moderator on the relationship

between member similarity and respect towards the community in the context of children.

Through introducing, and demonstrating, the effect of similarity attraction in child-oriented

brand communities; the thesis expands the understanding of the drivers of children’s brand

community participation. Similarity attraction has been widely explored and evidenced in

influencing children’s relationship formation in only the field of psychology. Findings of this

thesis indicate that children will desire participation when they perceive members are

22
characteristically similar. This demonstrates another context where similarity attraction occurs

amongst children.

5.8 Structure of Thesis

Having provided an introduction in Section 5, Section 6 presents the papers of the thesis.

Section 6 begins with an explanation on the choice of journals for each of the four papers

(Section 6.1). Following this, the papers are presented, in the order given above in Section 5.

Each paper is prefaced by an introduction, and details how the paper contributes to the thesis,

and more broadly the contribution made to the field of marketing. Then the full papers (as

published or submitted) are provided.

Paper One: Literature Review is presented in Section 6.2 and provides the published review

on the literature. The literature reviewed is used as the basis for the remaining papers of the

thesis. Paper Two: Replication and Extension is then delivered in Section 6.3, Paper Three:

New Model in Section 6.4, and finally Paper Four: Similarity Attraction in Section 6.5.

Section 7 then provides the additional papers (Conference Paper One and Conference Paper

Two).

23
6. Published Papers

This section presents the papers that constitute this thesis. Three of the papers have been

published in peer-reviewed journals and one is under review by a peer-reviewed journal. It

begins with an explanation of the journal selection that occurred for each of the four papers.

Following this, the papers are presented in order. Before each respective paper there is a

statement of the contribution that the candidate and the co-authors made to the paper. In

addition, an introduction on each paper is presented prior to the full publication being given.

After the four papers are given, Section 7 provides the additional papers that were submitted

as conference papers.

24
6.1 Criteria for Journal Selection

This section gives an explanation as to why the respective journals were chosen as a publication

outlet for the individual papers that constitute this thesis. Several criteria were assessed prior

to approaching the journal including: aims of the journal, Australian Business Deans Council

(ABDC) ranking (for more on this ranking guideline see Appendix 9.6), h-index (for more on

this index see Appendix 9.7) and prior publications by the journal. All four journals chosen are

discussed separately below in terms of this criteria, and any other relevant factors.

6.1.1 Journal of Brand Management

Paper One: Literature Review was published in the Journal of Brand Management (JBM).

JBM’s aim is to be the leading journal on brand management and strategy, focussing on

discussing theory and practice of specific topics including that of online brand communities

(Springer, 2018). This journal is an A-ranked journal, as defined ABDC ranking standards,

signifying it is in the top 15-25% of journals in the field of Marketing. The strong standing of

the journal is also shown through its h-index score of 28 (Scimago Lab, 2018b). The aim and

ranking of JBM is a good fit with the purpose of Paper One: Literature Review, being a

reference for academics and practitioners on brand community participation.

Since Paper One: Literature Review is a literature review, another key criterion during journal

selection was the prior publication of literature review papers. Some journals in the marketing

field only publish empirical papers and therefore would have been inappropriate to pursue in

this case. JBM has previously published literature review papers on other topics, and comprises

empirical papers on brand communities, some of which were incorporated into the review done

by Paper One: Literature Review, strengthening the suitability of the paper for the journal.

25
6.1.2 Journal of Marketing Behavior

Paper Two: Replication and Extension is unique since it is a replication of a previously

published article. The original paper (Bagozzi & Dholakia, 2006) was published in the

International Journal of Research in Marketing (IJRM), an A*-rank journal (ABDC ranking),

the highest ranking bestowed by ABDC. IJRM welcomes academics to replicate and extend

upon its previously published studies, with a dedicated ‘replication corner’ that publishes

replication papers in the recently established partner journal, the Journal of Marketing Behavior

(JMB). JMB, while only formed in 2016, has been given a B-rank by ABDC. Although JMB

is a B-ranked journal, all editors are from the parent, A*-ranked journal, IJRM. This signifies

that the high standard of the parent journal (IJRM) remains for JMB.

Another benefit of the journal is that both JMB and IJRM are associated with the European

Marketing Academy (EMAC), a professional society for marketing academics and

practitioners (EMAC, 2018). This association provides an extensive network through which

the papers can be promoted and viewed.

To date Paper Two: Replication and Extension is the only brand community context paper

published by JMB.

6.1.3 International Journal of Consumer Studies

Paper Three: New Model was published in the International Journal of Consumer Studies

(IJCS), in a special issue release entitled “Children as Consumers”. This journal was chosen

due to its high-ranking standard, and Paper Three: New Model was a good fit for the special

issue topic: “Children as Consumers”.

26
IJCS is an A-ranked journal (ABDC), making it amongst the top journals in the marketing field,

receiving a strong impact factor of 1.51 in 2016 (Clarivate Analytics, 2018b). Topics examined

by IJCS include: consumer protection, consumer behaviour, consumer ecosystem and, family

and household studies (John Wiley & Sons, 2018). Prior to accepting Paper Three: New Model,

IJCS had previously published empirical papers on brand communities, using similar research

methodologies. This further highlighted that Paper Three: New Model would be appropriate

for, and in line with IJCS.

Paper Three: New Model has already been cited by other publications, showing the paper is

having an impact in the field. Specifically, the following papers have referenced Paper Three:

New Model:

1. Berge, S. (2017). Food co-operatives sustainably managing common pool resources as

hyper-communities as outlined by consumer culture theory, International Journal of

Consumer Studies, 41(5), 509-517, doi: 10.1111/ijcs.12359.

2. Lopex, A. and Rodriguez, R. (). Children and their brands: How young consumers relate to

brands, Journal of Consumer Marketing, https://doi.org/10.1108/JCM-06-2016-1842.

6.1.4 Journal of Brand Management

Paper Four: Similarity Attraction has been submitted to the Journal of Brand Management

(JBM) and is currently under review. As stated above JBM is an A-ranked journal (ABDC

ranking) with a high reputation in the field of marketing. The journal has been prevalent in

publishing papers on the area of brand communities since the early 2000s (e.g. Boyd, Clarke,

& Spekman, 2014; Hatch & Schultz, 2010; Hickman & Ward, 2013; Lee, Lee, Taylor, & Lee,

2011; Quinn & Devasagayam, 2005; Wang, Butt, & Wei, 2011). Many of these brand

community papers have been successful with strong citation rates associated e.g. over 380
27
(Hatch & Schultz, 2010), 44 (Lee et al., 2011) and 25 (Wang et al., 2011). JBM has also

previously published studies on similarity attraction, in the context of social media (Wallace,

Buil, & de Chernatony, 2014). Similarity-attraction is a core theory in Paper Four, further

supporting the alignment of JBM for this paper.

28
6.2 Paper One: Literature Review

Full Citation:

Hook, M., Baxter S.M. & Kulczynski A. (forthcoming) Antecedents and Consequences of

Participation in Brand Communities: A Systematic Literature Review, Journal of Brand

Management, doi: 10.1057/s41262-017-0079-8

6.2.1 Statement of Contribution of Others

By signing below, I confirm that Margurite Hook was the sole contributor to the paper entitled

“Antecedents and consequences of participation in brand communities: a literature review”.

The co-authors (Stacey Baxter and Alicia Kulczynski) only provided guidance for the paper,

with limited intellectual input.

X
Associate Professor Stacey Baxter
Co-author Faculty Assistant Dean Research Training

X
Dr Alicia Kulczynski
Co-author

29
6.2.2 Third Party Copyright Acknowledgement

Reprinted by permission from Springer Nature: Springer, Journal of Brand Management,

Hook, M., Baxter S.M. & Kulczynski A., (forthcoming), advance online publication, doi:

10.1057/s41262-017-0079-8.

See Appendix 9.4.1 for a copy of the communications received from the rights holder.

30
6.2.3 Overview and Contribution of Paper

Brand community research in the marketing field has been undertaken for some time, with

studies dating back to the early 2000s (e.g. Bagozzi & Dholakia, 2002; McWilliam, 2000;

Muniz Jr & O’Guinn, 2001). Due to this there is no shortage of articles examining a key issue

in the area, namely brand community participation. To date hundreds of articles have examined

brand communities, proposing models that highlight antecedents and consequences of

participation in these communities. The vast amount of literature on the area makes

comparisons and the identification of under-researched areas difficult. Paper One: Literature

Review sought to address this issue. The aim of Paper One: Literature Review was to identify

and compare the antecedents and consequences of brand community participation that have

been examined in academic journals. This aim was achieved through undertaking a systematic

literature review.

A systematic literature review is an explicit and transparent review of published material, using

a reproducible method, with set exclusion and inclusion criteria (Pluye, Hong, Bush, & Vedel,

2016; Tranfield et al., 2003). A strict process and method of collecting the articles is undertaken

to ensure the results can be replicated, if desired.

A three-stage collection process was undertaken, with over 1,900 articles analysed. An

inclusion and exclusion criteria process was followed, with articles that were not in line with

the aim and scope of the research excluded from further review. All inclusion and exclusion

criteria were documented to ensure the process could be reproduced in the future. The final

article sample consisted of 41 articles which were then examined further. For each article, key

details were collected, such as: type of brand community, research method, model developed,

31
and antecedents and consequences examined. This data was then summarised and compiled

into meaningful groups so that comparisons could be made.

The main findings centred around the antecedents and consequences studied across three forms

of brand communities (offline, online and social-media-based). An analysis of these

antecedents and consequences showed that they could be placed into categories. Five categories

for antecedents (self-related, social-related, information-related, entertainment-related and

technology-related) and three categories for consequences (brand-related, brand community-

related, and social-related). These categories uncovered interesting insights when comparing

across forms of communities. For example, antecedents relating to the self (aspects to do with

the individual member themselves) were the most prevalent for all brand community forms,

except online brand communities. With regards to consequences, brand-related (e.g. brand

loyalty) were the most common overall, however, brand-community related consequences (e.g.

community commitment and community loyalty) was the highest for online brand

communities.

In addition, areas that have received less attention were brought to focus, such as the lack of

child-oriented brand community research, a finding pivotal for the current thesis and

subsequent papers. Child-oriented brand communities were not the only under-researched area

revealed. Research questions were proposed to address those areas that had been under-

researched, drawn from the review findings.

The revelation that child-oriented brand communities have received very little attention sets

the premise for the remainder of the current thesis. Paper One: Literature Review provides the

literature base for the subsequent papers and proposed two research questions on child-oriented

32
brand communities. These research questions are addressed in Paper Two: Replication and

Extension, Paper Three: New Model and Paper Four: Similarity Attraction.

Paper One: Literature Review presents a much-needed summary and review of the extant

literature on brand communities. The information given provides a depository of information

on brand communities for both academics and practitioners. Academics are also made aware

of the numerous areas that require more attention, through the research questions developed.

The usefulness and contribution Paper One: Literature Review makes to the field was summed

up by the following comment made by a reviewer from the Journal of Brand Management:

“The aim and intention of the article "Antecedents and Consequences of Participation

in Brand Communities: A Systematic Literature Review" is very relevant and could

provide a fruitful next step in the continuous development of the body of knowledge on

brand communities, which, as the author also states, is growing rapidly. The article

could also provide a very useful overview of the excising body of knowledge for business

managers, as in a range of practice fields there is also a widespread and growing belief

in the brand community's potential in brand building and relationship management.

Therefore, this article could become a relevant and beneficial contribution to both

scholars within this field and to a board range of practitioners.”

33
6.2.4 Full Paper

Antecedents and Consequences of Participation in

Brand Communities: A Literature Review

34
Abstract

With hundreds of articles dedicated to investigating brand communities, there is now a need

to consolidate the literature. This review addresses the need to reconcile the findings of brand

community participation literature through undertaking a literature review. Over 1900

articles were examined, 41 in detail. Findings reveal that three forms of brand community

participation have been studied: offline, online, and social-media-based, each uncovering the

antecedents and consequences of brand community participation. Antecedents were grouped

into five categories (self-related, social-related, information-related, entertainment-related

and technology-related) and consequences into three categories (brand-related, brand

community-related, and social-related). From the review, several future research directions

are uncovered, including 16 specific research questions. By scrutinising the vast literature on

brand community participation, and presenting multiple avenues for future research, this

review presents findings useful for academics and practitioners alike.

Keywords

Brand community participation, Literature review, Online brand communities, Offline brand

communities, Social-media-based brand communities

35
Introduction

The brands Nutella, Jeep, Lego and Apple may at first appear to have little in common;

however, all use a brand community (or multiple) as part of their marketing strategy (Cova

and Pace 2006; Lego 2016; McAlexander et al. 2002; Muñiz and Schau 2005). Numerous

success stories show that a brand community can transform a brand. For example, Harley-

Davidson used a brand community-centered marketing strategy to rescue their declining

brand, which contributed to a brand value boasting $7.8 billion (Filipe Lages and

Montgomery 2004). The French cosmetics brand Sephora demonstrated that brand

community success is not only for motor vehicle brands, with one million viewers every

month participating in the brand community (Thumm 2015), and members of the brand

community spending 2.5 times more than non-members (Ungerleider 2014).

Claimed by some as ‘the holy grail of brand loyalty’ (McAlexander et al. 2002, p 38), brand

communities can provide great value for a brand. With the potential to offer brand

differentiation and a sustainable competitive advantage (Thompson and Sinha 2008), brand

communities present the opportunity to develop and foster long-term relationships with

customers by providing a platform through which loyal customers can participate in activities

together (Carlson et al. 2008; Hur et al. 2011; Muniz et al. 2001; Stokburger-Sauer 2010).

By definition, a brand community brings together brand devotees to participate in shared

rituals and traditions (Muniz et al. 2001).

Since brand communities provide an avenue for sustaining relationships with customers, a

large amount of research has been dedicated to investigating the characteristics of brand

communities and member participation in these brand communities (e.g. Cova and Pace

2006; Schau and Muniz 2006; Sierra et al. 2016). In addition, the rapid emergence of brand

36
communities developed by organisations, and communities developed by passionate brand

advocates (for example see Cova and Pace 2006), has seen a rise in academic research into

the area of brand communities (e.g. Annett-Hitchcock and Xu 2015; Baldus et al. 2015;

Pahnila and Väyrynen 2015; Sierra et al. 2016; Syrjälä 2016). To date, several hundred

articles have been published in the field of brand communities, with the number rapidly

increasing in recent years (e.g. Sierra et al. 2016; Syrjälä 2016; Zheng et al. 2015). In

particular, a key focus has been the investigation of antecedents and consequences of brand

community participation. A review of the literature reveals that researchers are yet to

consolidate this extensive body of knowledge. As a result, this paper seeks to encapsulate

brand community participation by isolating the antecedents and consequences of

participation identified in the extant literature.

The aim of the present study is to identify and compare the antecedents and consequences of

brand community participation that have been examined in academic journals through an

extensive literature review. From an academic point of view, consolidating the literature will

be helpful in identifying areas of further significant research. In addition, the future research

directions (including specific research questions) will help guide future studies in the field.

From a practitioner’s perspective, a comprehensive overview of the current knowledge on

antecedents and consequences of brand community participation will help inform the

creation and management of brand communities, and provide guidance on harnessing the full

potential of brand communities for brands. The following sections define brand communities

and discuss the method, findings, and future research directions arising from this

comprehensive review.

37
Brand community definition

A brand community is defined as ‘a specialised, non-geographically bound community,

based on a set of social relationships among admirers of a brand’ (Muniz et al. 2001, p 412).

This definition is widely acknowledged and accepted in the brand community literature (e.g.

Carlson et al. 2008; Tsai et al. 2012; Zhou et al. 2012). Another term used to explain these

groups of brand devotees is ‘consumer tribes’ (Canniford 2011). Consumer tribes are groups

of consumers that centre around a specific interest, idea, or behaviour; and sometimes include

brands (Cova and Cova 2002). However, a brand is not an essential component of a consumer

tribe, unlike brand communities where the brand takes centre focus (Canniford 2011).

Although there is a difference between consumer tribes and brand communities, the literature

crosses over substantially with many brand community studies integrating consumer tribe

literature (e.g. Kuo and Feng 2013; Luo et al. 2015; Muniz et al. 2001). Due to this, consumer

tribe literature incorporating a brand was included in the current review. For the purposes of

this review, the term ‘brand community’ will be used to refer to both brand communities and

consumer tribes (that focus on a brand).

All brand communities are said to display three characteristics: consciousness of kind, shared

rituals and traditions, and a sense of moral responsibility (Muniz et al. 2001). Consciousness

of kind refers to the connection an individual feels towards the brand and community

members, and the level of legitimacy they associate with this connection (Muniz et al. 2001).

Shared rituals and traditions represent the shared consumption experiences by brand

community members, the history, and the stories that are told in the community (Muniz et al.

2001). Lastly, moral responsibility refers to the duty that community members feel to stay in

the group, retain members, and introduce new members (Muniz et al. 2001). A range of terms

38
 have been applied to refer to different brand communities, these three characteristics,

however, remain consistent (Casaló, Flavián, and Guinaliu, 2008; Madupu and Cooley,

2010a; Zhou and Amin, 2014). These brand community characteristics by Muniz et al. (2001)

even provide three antecedents and consequences, depending upon the perspective viewed.

Madupu and Cooley (2010b) viewed these characteristics as consequences, and contrastingly

Zhou and Amin (2014) saw these as antecedents. This finding highlights the somewhat

confusing nature of the antecedents and consequences of brand community participation, as

many can have a dual role, that is, they may be viewed as either an antecedent or a

consequence depending on the perspective taken.

Method

A systematic literature review entails an explicit and transparent review of published

material, using a reproducible method, with set exclusion and inclusion criteria (Pluye et al.

2016; Tranfield et al. 2003). In order to ensure that the method was reproducible for this

review, the following steps were undertaken. First, guidelines were established regarding the

scope and boundaries of the study. Second, a plan was made as to where the literature would

be sourced. Third, selection criteria were established, with specific inclusion and exclusion

criteria. For the last step, the final sample of literature was synthesised and the results were

examined.

Scope of study

Due to the popularity and success of brand communities (Belk and Tumbat 2005; Cova and

Pace 2006; Sicilia and Palazón 2008), as well as the value that brand communities provide

for brands (e.g. Hur et al. 2011; Thompson and Sinha 2008); many studies have been

39
 undertaken to identify the determinants of brand community participation (antecedents;

Filipe Lages and Montgomery 2004), and what occurs as a result of participation

(consequences). With regards to these antecedents, the terms ‘antecedents’ and ‘drivers’ are

used synonymously in the brand community literature (e.g. Carlson et al. 2008; Hung 2014)

to explain those variables that influence the dependent variable of study. In the current

context the dependent variable is most commonly ‘brand community participation’.

Search of articles

Searches were conducted in multiple journal databases to identify articles that included the

term ‘brand community/ ies’ in their abstract, title or keywords. In addition, due to the high

level of similarity between ‘consumer tribes’ and ‘brand community, the term ‘consumer

tribes’ was also used as a search term. Some have used the term ‘tribe’ or ‘consumer tribe’

synonymously with ‘brand communities’ (Kozinets 1999), whereas others suggest key

differences exist (Canniford 2011). Due to use of the term ‘consumer tribes’ appearing in,

and the incorporation of consumer tribe literature in many brand community studies (e.g.

Kuo and Feng 2013; Luo et al. 2015; Wang et al. 2015); the term ‘consumer tribe/s’ was also

employed in the article search. For the remainder of this paper the term ‘brand community’

will be used to refer to both those termed ‘brand community’ and ‘consumer tribe’ by the

original author.

Research into brand communities has been undertaken in various journals and disciplines.

For example marketing (e.g. Bagozzi and Dholakia 2006; Bruhn et al. 2014); computer

science (e.g. Habibi et al. 2014; Kang et al. 2007); management (e.g. Baldus et al. 2015;

Carlson et al. 2008; Zaglia 2013), and psychology (e.g. Lin 2008; Stokburger-Sauer 2010).

Care was taken to ensure articles in a variety of fields and disciplines were included by using

40
 a range of databases, specifically: Business Source Complete, ABI/Inform, Academic Source

Complete, JSTOR, Proquest, Science Direct and WARC. Consistent with other systematic

reviews in the area of marketing and management (Crawford and Gregory 2015; Snyder et

al. 2016; Witell et al. 2015); only academic journals were studied. No books or other

literature were included as not all these resources are readily available. Additional insights

may be found upon examining other literature sources. This is a limitation of the current

review.

Selection of articles

As the aim of this research is to identify all antecedents and consequences of brand

community participation, inclusion and exclusion criteria were put in place during the

literature searches, however, to ensure all relevant articles were included this criteria was

broad. To be included in the first sample the following criteria had to be met: (1) the article

was peer-reviewed, (2) published in English, (3) full text was available to download and (4)

published in a scholarly journal (see Fig. 1).

After the initial search of articles was undertaken, a second process of analysis was

undertaken to determine the final sample of articles to be included in the review. Although

many of the articles initially mentioned the term ‘brand community/ies’, in their abstract,

only 178 focussed specifically on defining, explaining or analysing brand communities.

These 178 articles were further analysed, and those that did not examine antecedents or

consequences of brand community participation were excluded from the sample (n = 137).

The final sample examined consisted of 41 articles.

41
Coding and analysis

Information was extracted and compiled from each individual article chosen for the final

sample. The information collected included, but was not limited to, the following: year of

publication, type of article (quantitative, qualitative or conceptual), research design, brand

community form of focus, geographical context, brand community participation definition

used, antecedents studied, consequences examined, and major findings. This information was

compiled in a spreadsheet to create a comprehensive summary of all information used for

this study.

Figure 1 - Selection of Articles Process

ABI/Inform, Academic Inclusion criteria:

Source Complete, Business (1) Peer reviewed
Search terms “brand
Source Complete, Science (2) Published in English
community”, “brand
Direct, WARC, Proquest and (3) Full text available
communities” and“consumer
JSTOR databases: (4) Scholarly journals
tribes”
1925 articles

Excluded articles which did not focus

primarily on brand communities or
consumer tribes
(n = 1771)

Chosen for further

analysis:

178 articles
Excluded articles which did not focus
primarily on antecedents and
consequences of participation and
those which focussed on brand
community engagement
(n = 137)
Final sample:

41 articles

42
Analysis and results

The analysis and results have been categorised into five subsections. Specifically, these

subsections discuss the (1) publication activity, (2) research design, (3) brand community

forms, (4) brand community participation definitions, and (5) antecedents and consequences

of brand community participation.

Publication activity

This subsection discusses the publication activity with regards to the 41 articles chosen for

review. The final sample of articles emerged from over a decade of research (2006–2016, see

Fig. 2, note the articles were complied in mid-2016).

Figure 2 - Year Distribution of Articles

Whilst the literature on brand communities began earlier with Muniz et al. (2001) seminal

article introducing and defining brand communities, studies started investigating brand

community participation more specifically in 2006 (Bagozzi and Dholakia 2006; Shang et

al. 2006). Brand community participation received low and uneven attention until 2010.

43
 Since 2010, there has been a steady publication of articles on brand community participation,

with 24 of the 41 articles (59%) published since 2010.

Research design

This subsection provides an overview of the research methodology of the brand community

literature analysed in this review. From the 41 articles, only two (2) were conceptual

(Madupu and Cooley 2010a; Zhou and Amin 2014), with the rest taking an empirical

approach (n = 39). Of these, four took a qualitative approach (Enginkaya and Yılmaz 2014;

Goulding et al. 2013; Mitchell and Imrie 2011; Morandin et al. 2013), three used mixed

methods (Taute and Sierra 2014; Royo-Vela and Casamassima 2011; Tsai et al. 2012), and

32 applied quantitative techniques only (e.g. Bagozzi and Dholakia 2006; Manthiou et al.

2014; Shang et al. 2006). Of those articles that took a quantitative or mixed method approach,

all studies were cross-sectional in nature. The high number of empirical articles highlights

researchers’ preference for empirical evidence in brand community research, making greater

theoretical and more generalisable findings.

High-involvement products were the most commonly studied, with a number examining

technology brands, e.g. Apple, Samsung and Sony (Habibi et al. 2016; Shang et al. 2006;

Wang et al. 2013, 2015), and car brands, e.g. Harley-Davidson, Ford and Mazda (Bagozzi

and Dholakia 2006; Marzocchi et al. 2013; Morandin et al. 2013; Zhou et al. 2013). At the

other end of the scale, fast food brand communities were also found in the final sample of

articles (e.g. Habibi et al. 2016; Manthiou et al. 2014).

The majority of articles used a sample of respondents from Asian countries (n = 19), with

China being the most common (e.g. Zhou and Amin 2014; Zhou et al. 2012), seven used an

44
 American sample (e.g. Bagozzi and Dholakia 2006; Carlson et al. 2008; Habibi et al. 2016;

Manthiou et al. 2014), and five were based in European countries (e.g. Casaló et al. 2008;

Sánchez-Franco et al. 2012). Interestingly, only one study employed a sample from the

Pacific-region countries, specifically New Zealand (Mitchell and Imrie 2011), and only one

from Africa (Mzoughi et al. 2010). Five studies investigated a range of countries in their

sample (e.g. Dholakia et al. 2004; Morandin et al. 2013; Royo-Vela and Casamassima 2011),

and the remaining three did not specify where their sample was geographically based (Habibi

et al. 2016; Madupu and Cooley 2010a; Woisetschläger et al. 2008).

Observation of the sample demographics used throughout the entire final sample found that

brand community participation has only been examined for adult community members.

Brands and product categories that arguably appeal to children, in addition to adults, were

studied such as Nike (Jung et al. 2014), football teams (Woisetschläger et al. 2008) and theme

parks (Carlson et al. 2008). However, relationships for child-participants were not explored.

Although two of the studies employed a student sample (Manthiou et al. 2014; Sánchez-

Francoet al. 2012), the youngest age for these participants was 18.

Brand community forms

When observing the 41 articles studied, four forms of brand communities were found: offline,

online, virtual, and social-media-based. An offline brand community constitutes the in-

person face-to-face meetings of community members united around a focal brand, with

infrequent interaction, and a high level of involvement required from the brand itself (e.g.

‘Camp Jeep’ see McAlexander et al. 2002). Online brand communities, on the other hand,

have no geographical limitations and are instead located in an online, or virtual environment

where members share information about a common brand (Jang et al. 2008; Lee et al. 2011;

45
Madupu and Cooley 2010b). Participation in online brand communities occurs in ways not

possible for offline brand communities, with members able to participate via instant photo

and video sharing at a global scale, as well as through discussions among members without

talking face-to-face (Zaglia 2013). These online communities have also been termed virtual

brand communities. A virtual brand community is a social group originating on the internet

where information exchange occurs around one focal brand (Casaló et al. 2008). The

definitions of both online and virtual brand communities emphasise the importance of

exchanging information about a focal brand (Jang et al. 2008; Lee et al. 2011; Wang et al.

2012). As the two terms refer to the same overarching concept, for this review only the term

‘online brand community’ will be used to refer to both online and virtual brand communities.

Lastly, social-media-based brand communities are formed on social media platforms such as

‘Facebook’ (Habibi et al. 2016) and ‘Weibo’ (Luo et al. 2015). Social media platforms are

capable of hosting multiple branded communities simultaneously, unlike online brand

communities where only one brand is the focus (Shang et al. 2006). For example, ‘Facebook’

hosts millions of brand communities (De Vries et al. 2012). Across the three brand

community forms, studies have been undertaken to investigate why individuals participate

and how their participation impacts their behaviours.

Online brand communities (n = 19, e.g. Casaló et al. 2008; Chen and Ku 2013; Hur et al.

2011) were the most discussed in the sample, with less attention given to offline (n = 13, e.g.

Bagozzi and Dholakia 2006; Carlson et al. 2008; Tsai et al. 2012), and social-media-based

brand communities (n = 9, e.g. Sánchez-Franco et al. 2012; Sung et al. 2010; Zhang et al.

2015). Although social-media-based brand communities had only nine articles in the final

sample, there appears to be a current trend towards research on this form of community. All

46
 the social-media-based brand community articles in the final sample were recently published

(2010-present), compared with the other brand community forms (2006-present). Upon

further examination of the publishing dates of the final sample, interest in online brand

community research (online and social media) appears to be increasing, with 16 of the sample

articles published since 2012.

Brand community participation definition

Various terms were used throughout the 41 articles to explain brand community participation.

Although there was some variation in terms used, key features were evidenced. Firstly, social

intention forms the basis of participation and therefore has been used to measure brand

community participation (Bagozzi and Dholakia 2006; Mzoughi et al. 2010; Zhou et al.

2013). In other words, the intention to participate provides evidence of participation

occurring (e.g. Bagozzi and Dholakia 2006; Mzoughi et al. 2010). The notion that intention

can be used as an indicator of participation is important as in some brand communities, such

as those using online mediums, member participation can be hard to observe (Shang et al.

2006). Participation in online brand communities need not be visible, as participation can

involve ‘lurking’ or browsing the brand community without visible interactions occurring

(Madupu and Cooley 2010a; Shang et al. 2006). Participation has also been measured based

on observable behaviours, such as involvement in activities (Sánchez-Franco et al. 2012; Tsai

et al. 2012; Wang et al. 2012), and providing help to other members (Casaló et al. 2008). By

providing help to others (Casaló et al. 2008), and actively involving in the brand community

(Sánchez-Franco et al. 2012; Tsai et al. 2012; Wang et al. 2012), members are not only

participating, but it is suggested they are also committing to the community. Although the

term ‘brand community commitment’ is consistently studied in isolation from brand

47
community participation, there appears to be a conceptual and empirical overlap across these

two concepts. Specifically, brand community commitment refers to a members desire to

sustain relationships formed within the brand community (Zhou et al. 2012). This is achieved

by revisiting the community and exchanging information among members (Munnukka et al.

2015; Zhou et al. 2012). These elements all involve continued brand community

participation, and therefore it is argued that brand community participation and brand

community commitment both are indicators of participation.

Evidence that brand community participation and brand community commitment have the

same core components can be seen when examining the specific measures used in prior

research (e.g. Bagozzi and Dholakia 2006; Mzoughi et al. 2010; Tsai et al. 2012). Item

similarity is seen when examining the constructs used to measure brand community

participation and brand community commitment. Brand community participation items

include statements such as ‘I intend to participate in activities…’ (Bagozzi and Dholakia

2006; Mzoughi et al. 2010) and ‘I actively participate in brand community activities’ (Tsai

et al. 2012). In comparison, brand community commitment items include statements related

to, or explicitly involving participation such as ‘I will exchange information and opinions

with brand community members’ (Hur et al. 2011; Jang et al. 2008) and ‘I am motivated to

participate actively’ (Hur et al. 2011; Munnukka et al. 2015). The overlapping nature of brand

community participation and commitment constructs requires researchers to clearly

distinguish the two constructs in order to establish construct validity. Based on these findings,

the terms brand community participation and brand community commitment will be treated

synonymously for this review, and will be hereafter only referred to as ‘brand community

participation’.

48
Antecedents and consequences

This final subsection discusses the antecedents and consequences of brand community

participation identified in the 41 articles studied. A large range of antecedents and

consequences were examined in detail. After examining all the articles, categories were

developed with relation to the focal area of each respective antecedent and consequence of

brand community participation studied. Five categories of antecedents were found, and three

categories of consequences, these are discussed below. Table 1 presents a summary of all the

review findings, in descending order of frequency for each element. The results are grouped

by form of brand community (offline, online and social-media-based).

Antecedents

Five categories of antecedents were found and were termed, in order of overall prominence

among the articles: self-related (n = 44), social-related (n = 34), information-related (n = 24),

entertainment-related (n = 8) and technology-related (n = 3). These categories were decided

upon based on the main focus of the construct in question, and as interpreted by the original

author/s. These categories are discussed separately; however, they are, by nature, all

interconnected. From simply looking at the prominence of the categories of antecedents

found, it can be seen that self-related, social-related, and information-related have had the

biggest impact and influence in brand community literature; however, each category of

antecedents presents useful insights into brand community participation.

49
Table 1 - Summary of Findings

Offline Brand Community (n=13) Online Brand Community (n=19) Social-Media-Based Brand Community (n=9)
Antecedents
Frequency Frequency Frequency
Self-related 20 Information-related 17 Self-related 8
Social-related 12 Social-related 16 Social-related 6
Information-related 2 Self-related 16 Information-related 5
Entertainment-related 1 Entertainment-related 5 Entertainment-related 2
Technically-based 0 Technically-based 2 Technically-based 1
Consequences
Brand-related 8 Brand Community-related 17 Brand-related 6
Brand Community-related 4 Brand-related 11 Brand Community-related 2
Social-related 4 Social-related 6 Social-related 2
Geographic Context studied
Americas 4 Asia 11 Asia 5
Asia 3 Multiple 3 Europe 2
Europe 2 Americas 2 Americas 1
Multiple 2 Not Specified 2 Not Specified 1
Africa 1 Europe 1 Africa 0
Australia/Pacific 1 Australia/Pacific 0 Australia/Pacific 0
Not Specified 0 Africa 0 Multiple 0
Research Method
Quantitative 8 Quantitative 16 Quantitative 8
Qualitative 3 Conceptual 2 Qualitative 1
Mixed 2 Mixed 1 Mixed 0
Conceptual 0 Qualitative 0 Conceptual 0
Sample Demographics
Adult 8 Adult 17 Adult 6
Student 0 Student 0 Student 3
Children 0 Children 0 Children 0

50
 Interestingly, all but one of these categories (technology-related) take the perspective

of the individual looking to participate in the community (consumer perspective). This

shows that the main focus of the literature to date has been to study antecedents from

the perspective of the consumer, rather than other viewpoints, such as the brand itself.

One paper, however, uniquely addressed this brand-based perspective (Veloutsou and

Moutinho 2009), investigating the impact of brand-related antecedents such as ‘brand

reputation’ and ‘social visibility of the brand’. Since this was the only paper found to

take this perspective, more research should be done in this area, in particular looking at

these brand-related antecedents for different forms of brand communities (online,

offline, and social-media-based).

Self-related antecedents

The most commonly examined antecedent category was the self-related antecedents,

referring to those aspects that are to do with the individuals themselves. These

antecedents looked at either how the individual (the consumer looking to participate in

the community) perceives they relate to the brand community, or the personal benefits

they will gain from participation. The importance of identity, and the formation of an
101
individual’s social identity within the brand community context was found throughout

the brand community literature sample (e.g. Bagozzi and Dholakia 2006; Morandin et

al. 2013; Mzoughi et al. 2010) and was the most common self-related antecedent of

brand community participation found. Social identity refers to when an individual sees

themselves as part of the group (brand community) and feels an emotional significance

by being part of that group (Tajfel 1978). Social identity is strongly connected to the

social-related category of antecedents; however, since identification is primarily about

how the individual sees themselves in relation to the community group, this was seen

as a self-related antecedent rather than a social-related antecedent.

51
 Zhang et al.’s (2015) study identified a unique self-related antecedent to participation,

in the context of social-media-based brand communities: ‘information technology

habit’, referring to an individual’s use of certain information technology based on prior

behaviours. When an individual uses a social media platform (e.g. ‘Facebook’) on a

daily basis, and has done so for some time, this habit will influence their participation

in a social-media-based brand community. The individual would be more likely to

participate in a social-media-based brand community on ‘Facebook’ if there is already

an established behaviour of using the ‘Facebook’ platform (Zhang et al. 2015).

Other self-related antecedents found included attitude (Bagozzi and Dholakia 2006),

and self-related motives for participating in the brand community, such as self-

discovery (Dholakia et al. 2004; Madupu and Cooley 2010b, 2010b), and rewards (Jang

et al. 2008; Sung et al. 2010; Zhou and Amin 2014). These rewards refer to incentives

offered to consumers to encourage participation, such as coupons and special offers

(Sung et al. 2010), termed by some as ‘opportunity seeking’ (Enginkaya and Yılmaz

2014), and present an interesting dilemma. Although rewards and incentives were found

to have a strong positive impact on participation, when looking at the overall effect on

participation outcomes, incentive seeking displayed the weakest relationship to loyalty

(Sung et al. 2010). This suggesting the inclusion of incentives to members is not enough

for an effective brand community, and other factors, perhaps social-related antecedents,

are also needed.

Social-related antecedents

The second most common category of antecedents examined was social-related. The

main focus of these antecedents is on the interpersonal relationships formed within a

brand community, termed ‘social benefits’ that a member will desire from the brand

52
 community (Jung et al. 2014; Kuo and Feng 2013), or ‘social needs’ (Wang et al. 2012).

The ability to form relationships and connect with other individuals who are devoted to

the brand is a key antecedent to brand community participation. Related to this notion

is the culture that is shared between members (Zhou and Amin 2014), and the support

given by community members (Sánchez-Franco et al. 2012); with these also acting as

antecedents to participation.

The impact of trust on brand community participation was first identified in the online

brand community sample, the need to establish a trust in the community and its current

members (Casaló et al. 2008). With the brand community residing in the online

environment, the issue of trust arises as it is harder for the member to establish trust,

due to a lack of face-to-face interactions (Shang et al. 2006). Many studies in the online

contexts (online and social-media-based) have acknowledged that consumers need to

establish trust prior to participation (Casaló et al. 2008; Chen and Ku 2013; Hur et al.

2011; Shang et al. 2006; Tsai et al. 2012); however, no offline brand community studies

have noted this.

The approval of others is also highlighted as an antecedent, for all forms of brand

communities, through the term ‘subjective norms’ (Bagozzi and Dholakia 2006;

Mzoughi et al. 2010). If important others of the individual, such as friends and family,

approve of the brand community and its members there is a stronger likelihood of

participation occurring. This is drawn from the theory of reasoned action that has been

applied in the brand community literature (e.g. Mzoughi et al. 2010).

Information-related antecedents

The third category of antecedents was concerned with the information members could

gain from the brand community (information-related). Members seek a brand

53
 community with the expectation of receiving information about the products or services

of the brand in return (e.g. Jung et al. 2014; Kuo and Feng 2013; Madupu and Cooley

2010a, 2010b). Studies identified simply the ‘informational benefit’ (Jung et al. 2014)

or ‘information need’ (Madupu and Cooley 2010a; Wang et al. 2012) that individuals

desire from the community. A related aspect identified was the quality of information

given (Chen and Ku 2013; Jang et al. 2008; Zhang et al. 2015; Zhou and Amin 2014)

whereby a higher perceived quality leads to greater participation.

Interestingly, the information-related antecedents appear to be less important for offline

brand communities, with only two studies identifying information-related antecedents.

In contrast, a large number of information-related antecedents were discovered for

online brand communities, with this being the largest category of antecedent for this

form of brand community. Social-media-based brand communities also had a fairly

high number, taking into account the sample size, compared to offline. This could

signify that participants in offline community contexts are seeking more social and self-

related benefits rather than informational and that participants in online community

contexts demand more informational benefits from the community. This finding

highlights that consumers desire different things, depending on the form of community

in question.

Entertainment-related antecedents

Entertainment-related antecedents were also found, across all types of brand

communities. Individuals expect a level of entertainment, and entertainment-related

benefits, from the brand community (Madupu and Cooley 2010a). Although these

antecedents were not as commonly found as the previous three categories (self-, social-

and information-related), these antecedents provide interesting insights into brand

54
 community participation.

Termed simply entertainment value (Dholakia et al. 2004) or hedonic benefits (Kuo and

Feng 2013), the level of enjoyment or fun that a member can have in the brand

community has been evidenced as an antecedent to participation across all community

forms. These entertainment-related motives, however, are more prevalent in online

brand communities, rather than offline or social-media-based. This could either imply

that entertainment is not a key antecedent or that more research on entertainment-

related antecedents is needed in these contexts (offline and social-media-based). Since

the information-related category presents a much larger percentage of the antecedents

across all types of communities, there is evidence to suggest that the need for consumers

to gain information rather than entertainment is more important.

Technology-related antecedents

A category unique to the online-oriented brand communities (online and social-media-

based) was the technology-related category. This category, in contrast to the other four,

is concerned with the design and features of the community itself, rather than the

individual members or benefits the brand community provides. These technology-

related features of a brand community, even though they may be uncontrollable by the

brand itself (e.g. social media platforms for social-media-based brand communities),

can act as antecedents to brand community participation. In addition, the quality of the

system used to run the brand community can also be an antecedent to participation (Jang

et al. 2008; Zhou and Amin 2014). This category is the least frequently used throughout

the sample, implying that these technology-related features are not as important as the

other categories of antecedents. However, these features can still influence brand

community participation and should not be ignored.

55
Consequences

In relation to the consequences of brand community participation, three categories were

discovered. These were, in order of prominence: brand-related (n = 25), brand

community-related (n = 23) and social-related (n = 12). Although each category is

distinct, they are all interconnected. Most studies viewed consequences in terms of

positive implications towards the brand (i.e. consequences from the perspective of the

company). The only study in the current sample found to take a different view was

Wang et al. (2013), who studied consequences from the view of the customer, or

members of the brand community. Interestingly, these consequences (cognitive, social-

integrative, personal-integrative and affective) can fit into the antecedent categories

developed by this study (cognitive—information-related, social-integrative—social-

related, personal-integrative—self-related, affective—entertainment-related). This was

the only study that explored this context of consequences, emphasising more research

is needed on member-related consequences.

Brand-related consequences

Brand-related consequences were the most prominent consequences throughout the

sample analysed, and more especially for offline and social-media-based brand

communities. A popular perspective many brand community studies have taken is

concerned with the influence participation has on the brand as a whole (brand-related

consequences).

With a very high interest in the marketing field generally, it is unsurprising that the

subject of brand loyalty is of much interest in brand community literature and was the

most studied brand-related consequence (e.g. Habibi et al. 2016; Jang et al. 2008; Luo

et al. 2015; Madupu and Cooley 2010a; Munnukka et al. 2015; Scarpi 2010). Brand

56
 community participation was consistently found to positively influence brand loyalty

(e.g. Munnukka et al. 2015; Scarpi 2010). This area has even been extended to

investigate ‘oppositional brand loyalty’, that suggests brand community members have

such a high loyalty to the brand that they will strongly oppose competing brands

(Madupu and Cooley 2010a). Purchase and repurchase intentions were other brand-

related consequences found, with participation having a strong positive effect on both

(e.g. Ho 2015; Lee et al. 2011; Munnukka et al. 2015).

Brand Community-related consequences

With the highest frequency of consequences for online brand communities, brand

community-related consequences have received a lot of attention in the literature. These

consequences are concerned with the influence that member participation has on the

brand community itself, rather than the brand more generally. Interestingly, all brand

community-related consequences were seen from the perspective of positive influences,

with a clear avoidance of potential negative effects that could occur. This category of

consequences included factors such as commitment to the community (Casaló et al.

2008; Hedlund 2014; Kuo and Feng 2013; Munnukka et al. 2015; Royo-Vela and

Casamassima 2011; Zhou and Amin 2014), integrating into the community (Sánchez-

Franco et al. 2012), and a loyalty to the brand community (Chen and Ku 2013;

Woisetschläger et al. 2008). These brand community-related consequences were found

for all forms of brand communities; however, significantly more attention has been

given in this category to online brand communities. This implies that a key outcome of

online brand communities is to create a loyalty to the community itself, as well as the

brand, whereas for offline and social-media-based brand communities the emphasis is

more directed towards brand-related consequences, such as brand loyalty and purchase

intentions. However, this is only one interpretation and more research into this finding
57
 is needed.

Social-related consequences

The last category of consequences found was social-related, referring to the actions of

brand community members to talk to others about the brand and the brand community,

after participating in the community themselves. These social-related consequences

focused on the notion of word-of-mouth and recommendation to others, terms used

synonymously in the brand community literature (e.g. Hedlund 2014; Hur et al. 2011).

Unique terms employed by Scarpi (2010), ‘brand evangelism’ and ‘community

evangelism’, also refer to these concepts. These social-related consequences, by nature,

directly link to the other two categories, as they are concerned with informing others

about both the brand and the brand community.

Although these social-related consequences appear to have a positive impact on the

brand, there is the possibility that social activities of members can be negative in nature,

rather than positive (Luo et al. 2015). Word-of-mouth is difficult to control, and there

is the high possibility that negative, as well as positive, word-of-mouth can occur as a

consequence of brand community participation, especially in the context of social-

media-based brand communities (Luo et al. 2015). Interestingly, the majority of studies

avoided this issue with attention focussed on the positive implications word-of-mouth

can have for a brand and its brand community (e.g. Hedlund 2014; Munnukka et al.

2015; Woisetschläger et al. 2008).

Hur et al. (2011) focused on a somewhat negative social-related consequence,

‘constructive complaints’. These ‘constructive complaints’ refer to members

complaining about the brand as a consequence of participation (a seemingly negative

perspective), however, in a form that brands can then use to improve the product or

58
 service in question (turning the negative reaction into a positive outcome). So, although

in some forms this is a negative consequence of participation, this was viewed from a

positive perspective for the brand, rather than negative (Hur et al. 2011).

Dual nature variables

An important note to make is that this review has discussed the antecedents and

consequences in terms of how the original authors viewed them; however, there are

multiple ways to view each respective antecedent and consequence. For example, Kuo

and Feng (2013) investigated certain benefits that a brand community can provide for

its members. These benefits are identified by Kuo and Feng (2013) as antecedents in

the context of the study, that is, consumers are more likely to participate in a brand

community if they perceive that the brand community would provide certain benefits

to them. However, these could also be viewed as a consequence of participation, i.e. an

individual will gain benefits from participation in the community (Wang et al. 2013).

This is just one example of a dual nature variable (can be both an antecedent and a

consequence), and there are many variables in the brand community literature that could

fit into this criterion. ‘Community integration’, studied as a consequence (Sánchez-

Franco et al. 2012), is similar to such concepts as identification with the community,

commonly seen as an antecedent, not a consequence (Carlson et al. 2008; Madupu and

Cooley 2010a; Marzocchi et al. 2013).

This presents an issue of much confusion when examining the brand community

literature, as variables can be both antecedents and consequences, depending upon the

perspective taken. These dual nature variables present a challenge to both practitioners

and academics alike, and warrant further investigation.

59
Future research directions

The findings of this review highlight multiple avenues for future research. The

following sections discuss suggested future research in the area of brand community

participation, with specific research questions proposed.

Arising from the overlap found between the terms ‘brand community participation’ and

‘brand community commitment’, there is a need to better define these constructs. As

identified from this review, an examination of the measures used for each term reveals

that the two terms have been measured similarly (e.g. Bagozzi and Dholakia 2006;

Mzoughi et al. 2010; Tsai et al. 2012). The similarity in terms raises the issue of

construct validity, which needs to be addressed. Future research needs to analyse the

terms ‘brand community participation’ and ‘brand community commitment’ together,

rather than in isolation as evidenced by the articles studied in this review. To address

this need, the following research question is proposed:

RQ1 What are the conceptual and empirical differences between ‘brand

community participation’ and ‘brand community commitment’?

As identified in the results, none of the final sample employed child-age participants.

Children, as young as five, are participating in brand communities when talking with

peers about brands at school (Chaplin and John 2005). Some brands are also employing

online brand communities targeted specifically for children including: Lego (Lego

2016), Moshi Monsters (Mind Candy 2016), and Mattel’s Barbie (Mattel 2016).

Children participate in brand communities, based on exploratory studies (Flurry et al.

2014); however, their behaviour and the impact of the factors that influence their

participation remain unknown, with no research investigating the antecedents and

consequences, as evidenced by this review. Studies have investigated child-brand

60
relationships (e.g. Chaplin and John 2005; Chaplin and Lowrey 2010; Ji 2002, 2008);

however, these also did not specifically examine the area of brand communities, or

antecedents and consequences of brand community participation.

Whilst adult-orientated research may guide the understanding of brand community

participation, it is argued that differences in socio-emotional (Cicchetti and Cohen

2006), and cognitive skills (Piaget 1972) could impact a child’s brand community

participation. Children participate in brand communities and have a substantial impact

in the marketplace (Gorn and Florsheim 1985), however, have cognitive and socio-

emotional differences to that of adults and therefore should be studied in a brand

community context. To address this need, three separate, yet related, research questions

are proposed as a starting point to understand child-participants in brand communities:

RQ2 What are the antecedents to children participating in brand communities?

RQ3 What are the consequences of children participating in brand communities?

RQ4 Do the antecedents and consequences of brand community participation

differ between adult- and child- participants?

The findings of this review show that a majority of brand community studies focussed

on an Asian context, with limited attention on other geographical contexts, especially

Pacific regions (such as Australia and New Zealand), and African regions. Due to this,

a need arises for future research to investigate whether or not these Asian context

findings can be applied to other geographical contexts. One study has already provided

evidence that these contexts are more dissimilar than similar (Madupu and Cooley

2010b). This study compared the contexts of India and America, finding that cultural

differences significantly affected antecedents and consequences of brand community

participation (Madupu and Cooley 2010b). However, this is the only study to be
61
undertaken that compares geographical contexts found by the current review, signifying

more research can, and should be done in this area. To address this, the following

research question is proposed:

RQ5 Do the antecedents and consequences of brand community participation

differ across cultural contexts?

The majority of studies examined by this review focused on antecedents taken from the

perspective of the customer, that is, what does the customer want or desire from the

brand community. Although this is important for both practitioners and academics to

understand, there are other perspectives that should be taken into account. Veloutsou

and Moutinho (2009) were the only authors to look at the antecedents of brand

community participation from a brand-related perspective, in an offline community

only. Due to the limited research conducted thus far on perspectives other than the

customer, it is suggested that future studies examine this further, through the following

research question:

RQ6 What roles do other perspectives (e.g. brand, company, lurker customer,

active customer) play on antecedents to brand community participation, across

all forms of brand communities?

Even though the self-related antecedents of incentives and rewards were highlighted in

this review, there is evidence to suggest these antecedents do not have as strong of an

influence on brand community participation as other antecedents identified. This

finding warrants an investigation into whether or not incentives and rewards alone are

enough to influence participation, or, as this review suggests, other self-related and

social-related antecedents are needed for participation to occur. To address this, the

following research question is advised:

62
 RQ7 How effective are incentives and rewards (self-related antecedents) as

antecedents to brand community participation?

The finding of this review suggests that the categories of antecedents were not found to

be consistently prominent across all forms of brand communities, highlighting that

consumers may desire different benefits and features depending upon the form of brand

community they are interested in participating in. A study comparing the desires of

consumers across all three forms of brand communities (offline, online, and social-

media-based) would shed light on this issue. These findings would especially help

practitioners when developing a brand community. If certain benefits are more

important for online than offline for example, such as information-related as this review

suggests, developers should emphasise and feature these information-related benefits.

In addition, the entertainment-related antecedents were found to be more prominent in

online brand communities, which could suggest these benefits are also highly valued

by members. However, due to the small amount of antecedents found in the

entertainment-related antecedents for offline and social-media-based contexts, more

research should be done before confirming this proposition. Future studies should

examine entertainment-related antecedents in offline and social-media-based brand

communities. To aid future studies in on this issue, two research questions are

suggested:

RQ8 How do the desires of brand community members differ between forms of

brand communities (offline, online, and social-media-based)?

RQ9 How do entertainment-related antecedents influence brand community

participation in offline and social-media-based brand communities?

Based on the findings of the self-related antecedent category, trust was only identified
63
as an antecedent for online brand community participation. Due to the fact the brand

community is located online and not face-to-face, it is suggested trust is difficult to

establish in the online environment (Casalo´ et al 2008). In an offline brand community,

members meet face-to-face and can even interact directly with the brand (Marzocchi et

al. 2013). Trust is built on interactions between individuals and can be developed more

quickly when interactions are face-to-face, rather than online (Gefen and Straub 2004).

When the community is online, individuals have less information about other members

(Casalo´ et al. 2008) and, therefore, it is suggested trust needs to be established prior to

particpation, based on prior brand experiences outside of the brand community context.

In offline brand communities, it is proposed that trust is assumed, or quick to develop

due to face-to-face interactions (McAlexander et al. 2002), and therefore not an

antecedent. Future research will need to investigate the antecedent of trust in terms of

offline brand communities to confirm or disprove these propositions. The following

research question is offered for future studies:

RQ10 How does trust impact offline, online, and social-media-based brand

community participation, respectively?

A category of antecedents discovered by this review was technology-related

antecedents. These were aspects such as the design and layout of online and social-

media-based brand communities. These technology-related antecedents were only

found for online and social-media-based communities, which is understandable as only

these forms are dependent upon technology to operate. However, what is more

interesting is that no design-related antecedents were found for offline brand

communities. There are arguably design-related features that could act as antecedents

for offline brand communities, for example the format of gatherings, and the systems

in place to communicate with members during and outside of meetings. Future research
64
should explore what design-related antecedents there are for offline brand communities

and how much they ultimately influence participation, with the following research

question provided as a starting point:

RQ11 How do design-related antecedents influence offline brand community

participation?

The majority of consequences found by this review related to the brand (brand-, brand

community-, and social-related). Only one study was found to look at consequences

from the perspective of the members of the brand community (Wang et al. 2013). Even

the antecedents found by the current study suggest there are a number of member-

related consequences that could occur from participation. Factors like social identity,

entertainment, and information benefits could all arguably be viewed from a member’s

perspective as consequences. For example, a social identity can be formed due to

participating in the brand community, and benefits can be gained from participation.

These were reinforced by the member-related consequences found by Wang (2013) as

they could be grouped into the five antecedent categories developed by this review. To

aid future studies on this issue, the following research question is proposed:

RQ12 What are the member-related consequences of brand community

participation, across all forms of brand communities (offline, online, and social-

media-based)?

Based on this review, there is a suggestion that to achieve brand community-related

consequences, online brand communities are the most appropriate form to take, and for

brand-related consequences the best choice is a social-media-based or offline brand

community. However, due to the lack of studies comparing different forms of brand

communities it is hard to determine whether or not this assumption, that certain

65
consequences are more likely depending upon the type of community, is correct.

Drawing from this, the following research question is proposed for future studies:

RQ13 How does the form of community (offline, online, or social-media-based)

influence the consequences of brand community participation?

All articles studied were cross-sectional in nature, with none employing a longitudinal

research design. This highlights that the long-term consequences of brand community

participation are yet to be accurately captured. A comparative study, with a longitudinal

research design, would be helpful to practitioners planning to develop a brand

community, yet unsure as to which form (offline, online, or social-media-based)

provides stronger positive brand-related consequences in the long-term. Future research

should investigate the following research question:

RQ14 What are the long-term consequences of brand community participation?

As evidenced by this review, there has been very little attention given to examine

whether brand communities can have negative consequences, as well as positive. Of all

the articles studied, none looked at the negative consequences of brand community

participation. Although this is understandable, as practitioners are arguably more

interested in positive consequences, rather than negative, research into this area could

provide very valuable information. An exploratory study should explore the extent to

which brand community participation can lead to negative consequences for a brand;

for example, when word-of-mouth is used in a negative manner, which has already been

evidenced in offline brand communities (Hickman and Ward 2007; Luo et al. 2015;

Phillips-Melancon and Dalakas 2014). It is likely that there are more negative

consequences than just social-related and hence more research should be done in this

area, specifically addressing the following:

66
 RQ15 What are the negative consequences (for all brand community stakeholders,

e.g. community participants, the company) of brand community participation,

across all forms of brand communities?

One issue highlighted by this review was dual nature variables, which are those

variables that can act as both antecedents to participation and consequences from

participation, depending upon the perspective taken. These variables present an issue

of confusion currently in the literature, as each author usually takes only a one-sided

view of the variable, without acknowledgment of the other views that can be taken.

This makes a comparison of findings across brand community studies very difficult.

The development of a new term to describe these dual nature variables could help aid

this confusion and current difficulty, helping practitioners and academics alike. Future

research should look into these dual nature variables and provide better definitions for

these variables in terms of brand communities, with the following research question

suggested as a starting point for future studies:

RQ16 What role do dual nature variables play in brand communities?

Limitations

Although a thorough literature review has been undertaken, the nature of a literature

review presents an overarching limitation to the current, and all, literature reviews. A

review entails an examination and consolidation of only the findings that have been

published. Hence if a specific issue is less discussed in the reviewed literature, it is

difficult to establish whether this issue is less relevant or, alternatively, whether there

has been a lack of empirical focus. For example, in the current review some antecedents

and consequences were less prevalent than others (e.g. entertainment-related

67
 antecedents). There is not enough evidence to conclude that these entertainment-related

antecedents are less important for brand community participation, or whether there is a

lack of research on entertainment-related antecedents. All reviews cannot conclusively

determine whether the findings uncovered are due to an empirical or researcher-based

explanation.

Conclusion

A total of 1925 articles were analysed as part of a literature review on brand community

participation. The examined final sample consisted of 41 articles examining three

forms: offline, online, and social-media-based brand communities. Multiple insights

were found with regards to the antecedents and consequences of brand community

participation. Firstly, five categories of antecedents were developed, based on an

analysis of all 41 articles, these categories were: self-related, social-related,

information-related, entertainment-related and technology-related. Secondly, three

categories of consequences were developed, which were: brand-related, brand

community-related, and social-related. From an analysis of these categories of

antecedents and consequences, future research avenues were highlighted, through the

formation of research questions. These research questions can act as a guide for future

studies in the field of brand communities.

This review has consolidated the findings of brand community participation literature,

presenting findings useful for both academics and practitioners. The findings of this

study will be helpful to academics as antecedents and consequences of brand

community participation have been summarised and many future research directions

have been identified to aid studies in the field. The findings will also be useful to

practitioners when developing brand communities as the findings provide a better

68
 understanding of why consumers participate in brand communities and what this

participation can result in for brands.

Compliance with ethical standards

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of

interest.

69
References

Refer to Appendix 9.5.1 for Paper 1 Reference List

70
6.3 Paper Two: Replication and Extension

Full Citation:

Hook, M., Baxter S.M. & Kulczynski A. (2017). Antecedents and Consequences of

Children’s Brand Community Participation: A Replication and Extension Study, Journal

of Marketing Behavior, 3(1), 63-72, doi: 10.561/107.00000042

6.3.1 Statement of Contribution of Others

By signing below, I confirm that Margurite Hook was the sole contributor to the paper

entitled “Antecedents and Consequences of Children’s Brand Community Participation:

A Replication and Extension Study”. The co-authors (Stacey Baxter and Alicia

Kulczynski) only provided guidance for the paper, with limited intellectual input.

X
Associate Professor Stacey Baxter
Co-author Faculty Assistant Dean Research Training

X
Dr Alicia Kulczynski
Co-author

71
6.3.2 Third Party Copyright Acknowledgement

Permission to copy and communicate this work entitled: “Antecedents and Consequences

of Children’s Brand Community Participation: A Replication and Extension Study” has

been granted by Zac Rolnik of Now Publishers.

See Appendix 9.4.2 for a copy of the communications received from the rights holder.

72
6.3.3 Overview and Contribution of Paper

Brand community literature is dominated by an abundance of studies examining adult

participants. Paper One: Literature Review highlighted this by finding that no empirically

tested models had been developed or examined for child brand community participants.

Of the hundreds of articles examining brand communities, there have been a few seminal

studies that have arguably received more attention than others. One of these seminal

papers is the 2006 work of Bagozzi & Dholakia entitled: “Antecedents and purchase

consequences of customer participation in small group brand communities”.

Bagozzi & Dholakia, 2006 has received over 1,190 citations (as of May 2018), making it

an extremely influential paper in the brand community field. Given the popularity and

influence of Bagozzi & Dholakia, 2006, Paper Two: Replication and Extension sought to

empirically test whether this model applied for children. The replication was undertaken

to begin addressing Paper One: Literature Review’s research question of: Do the

antecedents and consequences of brand community participation differ between adult-

and child- participants?

The model developed by Bagozzi & Dholakia, 2006 focusses on the theory of planned

behaviour model (Ajzen, 1991), adding the components of social identity (Ellemers,

Kortekaas, & Ouwerkerk, 1999), anticipated emotions (Bagozzi & Pieters, 1998) and

desire (Bagozzi, 1992). The study showed that a mixture of social and psychological

processes are undertaken for consumers to have a desire to, and subsequently, participate

in a brand community. Given the differences in cognitive (Piaget, 1972) and socio-

emotional (Cicchetti & Cohen, 2006) skills that exist between adults and children, Paper

Two: Replication and Extension argued that the seminal Bagozzi & Dholakia, 2006 model

would not fully replicate for the child context.

73
Through collecting data from two samples of adult participants (Harley-Davidson brand

community members and a non-branded motorcycle riding group) this model was

validated in Bagozzi & Dholakia, 2006. Paper Two: Replication and Extension utilised

two samples (self-identified members of either: Minecraft brand community or a non-

branded computer game community) of Australian children aged 6-14, replicating the

process employed Bagozzi & Dholakia, 2006. The results showed that the model

replicated fairly well, however, not identically to the adult sample. Specifically,

relationships that were significant for adults were found to be insignificant for children,

and other relationships that were insignificant in the original study were found to be

significant for children.

The findings of Paper Two: Replication and Extension highlight the significance of this

thesis’s research, providing empirical evidence that adult models cannot be directly

applied to the child context. More broadly Paper Two: Replication and Extension has

impact for both practitioners and academics. For practitioners the findings imply that

methods used to attract adult participants may be inappropriate to use for child

participants, signifying alternative methods should be developed and applied.

Academics are informed that adult-based models do not accurately capture children’s

behaviour in brand communities. This signifies more studies should be undertaken to

develop child-context models. Paper Two: Replication and Extension sets the scene for

the final two papers of the thesis. Drawing from the finding that adult-based models do

not fully explain child-participants, new models specifically for children need to be

developed. Both Paper Three: New Model and Paper Four: Similarity Attraction present

new child-oriented models to spur research into the field.

74
The thoroughness of the research undertaken in Paper Two: Replication and Extension

was praised by the Journal of Marketing Behavior editor through the following comment:

“Congratulations on a solid piece of research”

75
6.3.4 Full Paper

Antecedents and Consequences of Children’s

Brand Community Participation: A Replication

and Extension Study

76
Abstract

Brand communities are a popular tool brands use to develop relationships with customers.

Bagozzi and Dholakia’s (2006) seminal article provides one model to explain

participation in these brand communities. This research replicates and extends this model

to the demographic of children. Results show that most relationships reflected those

observed in the original study, however, some distinct differences were found. Findings

highlight that adult-orientated brand community models may not be suitable to explain all

child-members’ attitudes and behaviors in brand communities.

Keywords:

Brand communities; branding and brand equity; children; social identity; theory of

planned behavior

77
Introduction

Since the beginning of the new millennium there have been several hundred articles

published regarding brand communities. With over 1,000 citations, Bagozzi and Dholakia

(2006) would be classified a seminal article within brand community literature. A brand

community is defined as ‘a specialized, non-geographically bound community, based on

a set of social relationships among admirers of a brand’ (Muniz and O’Guinn 2001, p.

412). These branded communities have become a popular marketing resource due to the

valuable role they play in brand and product promotion, as well as customer relationship

management (Muniz and O’Guinn 2001). Brand community popularity has prompted the

development of many conceptual models, applying numerous theories, to explain brand

community participation. One model is that presented by Bagozzi and Dholakia (2006),

which introduced the Theory of Planned Behavior in combination with other elements

into the brand community context.

Brand community research is dominated by adult-oriented studies. However, children as

young as 5 also engage in these communities (Flurry et al. 2014). Little is known,

however, about children’s behavior, and factors of influence in this area. Whilst adult-

orientated research may guide our understanding of brand community participation,

differences in socio-emotional (Cicchetti and Cohen 2006) and cognitive skills (Piaget

1972) could impact a child’s brand community participation. To the best of the authors

knowledge, only two papers have investigated child or youth brand community

participants: Sicilia and Palazón (2008) and Flurry et al. (2014). These two studies show

that children participate in brand communities, however, they do not provide any

empirical evidence as to the motives of children’s participation. Due to the popularity and

influence of Bagozzi and Dholakia’s (2006) model in the field of brand community

78
research, an investigation into whether this model applies to the demographic of children

will yield insightful results for both academics and practitioners.

Method

This study replicates Bagozzi and Dholakia’s (2006) model, extending it to the

demographic of child brand community participants (Australian children aged 6–14 years

old). The product category of computer games was chosen in place of motorcycles for

this study. Replicating Bagozzi and Dholakia (2006), data were collected from two

independent groups: (1) those who self-identified as being a Minecraft brand community

member; and (2) those who self-identified as being a computer game community member.

The survey was conducted online with parental consent, and child assent obtained prior

to participation. A total of 761 child participants completed the survey, 372 in the brand

community group and 389 in the non-branded community group. The age of the

participants was approximately evenly distributed within each group (MBC = 9.98 years;

MnBC = 10.71 years). For the brand community, the majority of participants were male

(63.4%; 36.6% female) and for the non-branded community, the majority were female

(63.2%; 36.8% male). Table 1 shows a comparison between the current and original

samples. Differences exist between the ages of participants, year of data collection,

gender distribution of participants, focal brand community and data collection method.

79
Table 1 - Method Comparison
Current Study Bagozzi & Dholakia
Year of Data Collection 2016 2006
Brand Studied Minecraft Harley-Davidson
Participant Population
Brand Community 372 154
Non-Branded Community 389 298
Total 761 452
Participant Characteristics –
Brand Community
Age 6 – 14 (mean = 9.98) 23 – 73 (mean = 47.5)
Gender 63.4% male (36.6% female) 74% male (26% female)
Participant Characteristics –
Non-Branded Community
Age 6 – 14 (mean = 10.71) 20 – 67 (mean = 43.2)
Gender 63.2% female (36.8% male) 83.6% male (16.4% female)
Data Collection
Brand Community Online Mail
Non-Branded Community Online Online
Data Analysis Confirmatory Factor Analysis Confirmatory Factor Analysis
and Structural Equation and Structural Equation
Modeling Modeling

All constructs from the original article were employed. However, in some instances,

minor alterations were made to reflect the language ability of participants through the use

of synonyms (for example, ‘depressed’ was altered to state ‘sad’). Additionally, all scales

were changed from 7-point Likert scales to 5-point Likert scales as these are more suitable

for child participants (Borgers and Hox 2001).

Table 2 summarizes the means, standard deviations and reliability scores of the 13

constructs for both groups of data collected. The reliability scores for the majority of

measures were above 0.70, except for perceived behavioral control, and cognitive social

identity in the non-branded community. The original article also had low reliability for

the perceived behavioral control measure. Unlike the original article the measures of

group behavior and brand-related behavior resulted in high reliability scores in this study.

80
Table 2 - Means, Standard Deviations, and Reliabilities of Scales

Current Study Bagozzi & Dholakia (2006)

Non-Branded Non-Branded
Brand Community Brand Community
Scale Community Community
M SD α M SD α M SD α M SD α
Attitudes 5.59 0.74 0.86 4.84 0.71 0.82 5.31 1.15 .94 5.59 1.05 .88
Subjective Norms 5.16 0.96 0.84 4.57 0.83 0.79 5.76 1.50 .87 5.77 1.45 .80
Positive Anticipated
Emotions 5.80 0.71 0.92 5.35 0.84 0.93 4.59 1.59 .95 4.97 1.39 .91
Negative Anticipated
Emotions 2.86 0.96 0.95 2.46 0.94 0.94 1.82 1.08 .95 2.24 1.19 .90
Desires 4.35 1.05 0.87 3.52 1.01 0.85 5.19 1.43 .93 6.07 1.03 .85
Cognitive Social
Identity 4.92 0.94 0.83 4.36 0.89 0.68 4.14 1.71 .90 4.21 1.62 .88
Affective Social
Identity 5.56 0.83 0.83 4.93 0.98 0.85 4.61 1.64 .91 4.88 1.57 .87
Evaluative Social .94
Identity 4.39 0.95 0.73 3.60 0.92 0.69 4.30 1.89 .96 4.74 1.92
Perceived
Behavioural Control 5.32 0.77 0.35 3.13 0.73 0.36 4.71 1.64 .57 4.86 1.48 .62
Brand Identification 2.91 1.18 - - - - 4.79 2.04 - - - -
Social Intention 4.51 1.12 0.86 3.61 1.13 0.82 3.86 .93 .90 4.32 .90 .82
Group Behaviour 3.28 0.92 0.78 2.64 0.90 0.85 1.02 .73 .74 .95 .60 .51
Brand-Related
Behaviour 2.91 0.86 0.81 1.54 0.32 0.71 2.15 .66 .62 2.01 .63 .55

Note: All 5-point scale means from the current study were converted to 7-point means for comparison.

Results and Discussion

Based on the original analysis method, Confirmatory Factor Analysis (CFA) and

Structural Equation Modelling (SEM) were performed to produce a model fitting both

groups of data (brand and non-branded communities), to identify the significant

relationships between variables of interest. Both models did not fit as well as the original

study: brand community: χ2(471) = 1078.78, p ≈ .00, RMSEA = .059, NNFI = .92, CFI =

.92; non-branded community: χ2(440) = 997.48 p ≈ .00, RMSEA = .057, NNFI = .92 CFI

= .93. In addition, as shown in Figures 1 (brand community) and 2 (non-branded

community), the paths and significance did not directly replicate those reported by

Bagozzi and Dholakia (2006). Interestingly, some of the paths that were insignificant are

significant in the current study and vice versa, differences were also observed in respect
81
to relationship strength and direction. In particular, while Bagozzi and Dholakia found a

positive relationship between perceived behavioral control and desire, a negative

relationship was observed in this study. The change in relationship direction may be

driven by the nature of the sample. Specifically, compared to adults, children possess less

control over their behavior (Baumrind 1978), resulting in the negative effect.

Figure 1 - Findings for Structural Equation Model: Brand Community (n = 372)

Another interesting finding is that both attitudes and subjective norms were not

significantly associated with desire (both brand and non-branded communities), unlike

the Bagozzi and Dholakia’s (2006) findings. This suggests that these components of the

theory of planned behavior may not be suitable for a child context, in particular for child

participants inbrandcommunities. In contrast, therelationshipbetween group behavior and

82
 brand behavior (for the brand community) changed from non-significant to significant in

the current study suggesting that group behavior has a more significant influence on brand

behavior in the context of children compared to adults. These findings reinforce prior

research that demonstrated group influence is of particular importance for children (e.g.

Dishion and Tipsord 2011; Hawkins and Coney 1974).

Figure 2 - Findings for Structural Equation Model: Non-Branded Community (n = 389)

Considering the model overall, five coefficients transferred from insignificant to

significant, five coefficients switched from significant to insignificant and sixteen

coefficients remained the same in terms of significance. This suggests that the results

replicate fairly well in the new context of children.

83
Tests of mediation were also performed and compared to Bagozzi and Dholakia’s (2006)

findings (see Table 3). Unlike the original findings, social identity, attitudes, positive

anticipated emotions and subjective norms were found to have a significant direct effect

on social intentions, and were not fully mediated by desire. For the non-branded

community, similar results were found with social identity, attitudes and negative

anticipated emotions all having a significant direct effect on social intentions.

Table 3 - Summary of Direct Effects (χ² difference and p-values)

Current Study Bagozzi & Dholakia (2006)
Brand Non-Branded Brand Non-Branded
Community Community Community Community
Social Identity – social intentions χ² = 8.43, χ² = 5.55, χ² = .66, χ² = .37,
p < .001 p < .001 p > .30 p > .56
Attitude – social intentions χ² = 8.19, χ² = 8.47, χ² = .13, χ² = 5.29,
p < .001 p < .001 p > .70 p < .05
Positive anticipated emotions – χ² = 2.12, χ² = 1.25, χ² = .16, χ² = 1.50,
intentions p = .02 p = .13 p > .68 p > .21
Negative anticipated emotions – χ²= .04, χ²= 6.92 χ² = .55, χ² = .55,
social intentions p = .81 p = .24 p > .46 p > .48
Subjective Norm – social χ² = 16.90, χ² = 18.37, χ² = .48, χ² = .2.68,
intentions p < .01 p = .34 p > .49 p > .10
Social identity – group behavior χ² = .53, χ² = .55, χ² = 3.38, p > χ² = 9.07,
p = .32 p = .39 .07 p < .001
Attitude – group behavior χ² = 1.41, χ² = 3.83, χ² = .48, χ² = .99,
p = .16 p < .03 p > .49 p > .35
Positive anticipated emotions – χ² = 1.75, χ² = .26, χ² = .23, χ² = 4.54,
group behavior p = .12 p = .60 p > .66 p < .05
Negative anticipated emotions – χ² = .56, χ² = .73, χ² = .16, χ² = 1.38,
group behavior p = .45 p = .37 p > .68 p > .24
Subjective Norm – group χ² = .68, χ² = .04, χ² = .67, χ² = .81,
behavior p = .39 p = .83 p > .30 p > .40
Perceived Behavioral Control – χ² = 1.49, χ² = .04, χ² = .34, χ² = 2.00,
group behavior p = .21 p = .83 p > .57 p > .17

Note: These results are the direct effects after accounting for the indirect effects of desire and social intention.

In addition, for the non-branded community positive anticipated emotions had a

significant direct effect on group behavior, similar to the findings of Bagozzi and

Dholakia (2006). Interestingly, the fit levels were similar across all four cases (the two

from the current study and the two from Bagozzi and Dholakia 2006) also highlighting

that the results replicated fairly well.

84
Employing methods used by Bagozzi and Dholakia (2006), tests were also conducted to

assess the difference in the correlations between the two models (brand community versus

non-branded community). To do this a simultaneous CFA for the latent variables constant

across the groups was performed to show whether there was a significant difference

between the correlation coefficients. The 12-factor model did not fit as well as the original

study: χ2(880) = 2016.08, p ≈ .00, RMSEA = .041, NNFI = .92, CFI = .93. Unlike the

original study, however, the test of equality of factor loadings showed that the hypothesis

of invariance could be rejected: χ2(20) = 20.116, p ≈ .00.

Table 4 - Summary of Significant Differences Between Models

Brand Non-Branded
Community Community
Attitude – Desire .12 .13
Positive Anticipated Emotions – Desire .27 .31
Negative Anticipated Emotions – Desire .30 .30
Subjective Norm – Desire .11 .10
Perceived Behavioral Control – Desire -.27*** -.16
Social Identity – Desire .17*** .07
Perceived Behavioral Control – Intentions .65** .59
Desire – Intentions .85 .78
Intentions – Group Behavior .65 .68
Social Identity – Brand Behavior .07* .20
Group Behavior – Brand Behavior .06*** .15

*p < .05, **p < .01, ***p < .001

This result shows that the addition of the ‘brand’ into the community significantly

modifies the structureofthemodel. Further analysis showsthat five pathswere significantly

different between the groups, based on χ2 difference tests (see Table 4).

Conclusion, Limitations and Avenues for Future Research

This paper has replicated Bagozzi and Dholakia’s (2006) seminal work, using different

participants, namely children aged 6–14 years old. Bagozzi and Dholakia’s (2006) model

replicates well for this new context. However, there are some unique differences for

85
children’s brand community participation, in overall model fit, path significance and

moderating effects. Unlike Bagozzi and Dholakia (2006), results of this study show a

significant difference between the brand community and non-branded community

models. This finding is particularly interesting as it may suggest the incorporation of the

‘brand’ into the community has a stronger influence on children compared to adults.

In addition, the mediating role of desire was found to be vastly different for child

members. The results show that once a child forms positive attitudes, positive anticipated

emotions or a social identity with the community, a direct positive effect on social

intentions is observed, with desire not necessary for the relationship to occur. This may

be due to children not being able to distinguish between desire, attitudes and emotions

(Schult 2002), therefore desire adds no predictive element to the formation of this

relationship. This is further supported by the lack of significant relationship between

attitudes and desire, for both the branded and non-branded communities, in contrast to the

findings of Bagozzi and Dholakia (2006). It is therefore suggested that future research be

conducted to understand the role of desire for child brand community members.

Since Bagozzi and Dholakia’s (2006) model does not fully explain children’s brand

community participation, additional future research is warranted. Specifically, future

research should look at developing a new model that better explains children’s brand

community participation. Researchers should consider the inclusion of variables that are

uniquely relevant to child-aged brand community participants (such as self-esteem).

86
Appendix – Correlation Matrices

Minecraft Brand Community

PBC SN NAE PAE ATT SIDE DES SINT BID GBEH ESI ASI CSI
PBC 1
SN .66 1
NAE -.42 -.16 1
PAE .59 .46 -.12 1
ATT .52 .50 -.08 .80 1
SIDE .00 .00 .00 .00 .00 1
DES -.30 -.07 .37 .11 .12 .71 1
SINT .04 .13 .20 .27 .26 .61 .77 1
BID .00 .00 .00 .00 .00 .61 .44 .37 1
GBEH .10 .13 .09 .22 .20 .39 .47 .64 .24 1
ESI .00 .00 .00 .00 .00 .86 .62 .53 .53 .34 1
ASI .00 .00 .00 .00 .00 .70 .50 .43 .43 .27 .60 1
CSI .00 .00 .00 .00 .00 .71 .51 .44 .44 .28 .62 .50 1
BBEH .04 .05 .04 .09 .08 .26 .26 .32 .26 .43 .22 .18 .18

Non-Brand Community

PBC SN NAE PAE ATT SIDE DES SINT GBEH ESI ASI CSI BB
PBC 1
SN -.61 1
NAE .15 -.17 1
PAE -.46 .44 .04 1
ATT -.45 .46 .18 .76 1
SIDE .00 .00 .00 .00 .00 1
DES .15 .01 .42 .29 .30 .52 1
SINT -.28 .24 .26 .40 .40 .40 .71 1
GBEH -.18 .16 .18 .27 .27 .28 .49 .69 1
ESI .00 .00 .00 .00 .00 .88 .46 .35 .24 1
ASI .00 .00 .00 .00 .00 .78 .41 .31 .22 .68 1
CSI .00 .00 .00 .00 .00 .81 .43 .33 .23 .71 .63 1
BBEH <-.01 <.01 <.01 .01 .01 .15 .09 .08 .07 .13 .12 .12

Note: PBC = Perceived Behavioral Control, SN = Subjective Norms, NAE = Negative Anticipated
Emotions, PAE = Positive Anticipated Emotions, ATT = Attitude, SIDE = Social Identity, DES = Desire,
SINT = Social Intentions, GBEH = Group Behavior, ESI = Evaluative Social Identity, ASI = Affective
Social Identity, CSI = Cognitive Social Identity, BBEH = Brand Behavior

87
References

Refer to Appendix 9.5.2 for Paper 2 Reference List

88
6.4 Paper Three: New Model

Full Citation:

Hook M., Baxter S.M. & Kulczynski A. (2016). Children's Participation in Brand-Based

Social Networks: Examining the Role of Evaluative Social Identity, Self-Esteem and

Anticipated Emotions on Commitment and Desire to Recommend, International Journal

of Consumer Studies, 40(5), 552-561, doi: 10.1111/ijcs.12300

6.4.1 Statement of Contribution of Others

By signing below, I confirm that Margurite Hook was the sole contributor to the paper

entitled “Children’s Participation in Brand-Based Social Networks: Examining the Role

of Evaluative Social Identity, Self-Esteem and Anticipated Emotions on Commitment and

Desire to Recommend”. The co-authors (Stacey Baxter and Alicia Kulczynski) only

provided guidance for the paper, with limited intellectual input.

X
Associate Professor Stacey Baxter
Co-author Faculty Assistant Dean Research Training

X
Dr Alicia Kulczynski
Co-author

89
6.4.2 Third Party Copyright Acknowledgement

Permission to copy and communicate this work entitled: “Children's Participation in

Brand-Based Social Networks: Examining the Role of Evaluative Social Identity, Self-

Esteem and Anticipated Emotions on Commitment and Desire to Recommend” has been

granted by John Wiley and Sons.

See Appendix 9.4.3 for a copy of the communications received from the rights holder.

90
6.4.3 Overview and Contribution of Paper

Having established that child- and adult-participants do not behave identically, Paper

Three: New Model sought to produce the first child-oriented brand community model.

Paper Three: New Model incorporated constructs from the model used in Bagozzi &

Dholakia, 2006 (studied extensively in Paper Two: Replication and Extension),

constructs identified in Paper One: Literature Review, and introduced variables from

psychology, to develop and test a new child-oriented model. The aim of Paper Three:

New Model was to examine the influence of evaluative social identity on brand-based

social network1 commitment and network recommendations, specifically for children.

Drawing from the Bagozzi & Dholakia, 2006 model examined in Paper Two: Replication

and Extension, the theories of anticipated emotions and evaluative social identity were

again tested. The dependent variables employed for the study were commitment and

recommendations. These variables were identified in Paper One: Literature Review as

consequences of participation, with evidence that suggested children too show these

consequences in a brand community context (Buckingham, 1993; Cicchetti & Cohen,

2006).

Evaluative social identity centres around the preference of the in-group (brand

community), as opposed to the out-group (those not in the brand community) (Ellemers

et al., 1999). This in-group, out-group notion has also been studied, outside of the brand

community field, in developmental psychology, using the term Subjective Group

1
For this paper the term ‘brand-based social networks’ was used instead of “brand
communities” to align with the chosen journal’s (International Journal of Consumer
Studies) focus.

91
Dynamics (SGD) (Abrams & Rutland, 2008). Paper Three: New Model introduced SGD

theory to the brand community field to help explain children’s participation.

The underlying proposition of SGD theory is that the value of the in-group (in this case

the brand community) is sustained when members endorse the norms of the group

(Marques, Abrams, Paez, & Martinez-Taboada, 1998). The value of the group is

threatened when members endorse the norms of opposing groups or go against their in-

group norms (Marques et al., 1998). SGD theory suggests child members of a group have

a bias towards the in-group, desiring to sustain the value of the group (Abrams & Rutland,

2008). Due to this bias, the members have a strong desire to uphold the norms of the

group.

Paper Three: New Model proposed that a group norm in a brand community context is

loyalty, shown through regular participation. The desire to uphold these norms results in

positive anticipated emotions when the child can participate, and negative anticipated

emotions when the child cannot participate. These anticipated emotions consequently

lead to community commitment and recommending the community to others.

Children with low personal self-esteem flourish when supported by a group (Boulton &

Smith, 1994) and therefore benefit more from commitment and recommendations. They

also have more to gain from upholding the in-group norms. In addition, due to the stronger

sense of disapproval they feel (Harter, 1993) children with low self-esteem are the most

likely to strive to uphold the norms. Due to this, the moderating role of personal self-

esteem was also introduced in Paper Three: New Model.

The proposed model was empirically tested using a sample of Australian children (6-14

years old) that were self-identified members of a Minecraft brand community. The

collected data supported the model proposed, with significant relationships found for all

92
hypotheses. Findings revealed that children with low personal self-esteem were the most

likely to experience negative emotions if they could not participate, and positive emotions

when they could participate. These emotions (both positive and negative) then lead to a

commitment to the community and recommending the community to others. The

underlying cause of these intentions (community commitment and recommendations)

was due to the desire to uphold the in-group norms and fear of being ostracised from the

community if the norms were disobeyed, in line with SGD theory (Abrams, Rutland, &

Cameron, 2003).

The findings of Paper Three: New Model present useful insights for academics and

practitioners. The first model for a child-oriented brand community was produced and

empirically tested, with new mediating and moderating relationships. A new theory

through which children’s brand community participation can be examined, namely SGD

theory, was provided. Practitioners were advised that children with low personal self-

esteem may be more vulnerable and therefore care should be taken to avoid societal

backlash. Society in general was also made aware that targeting certain groups of children

could occur.

The significance and relevance of Paper Three: New Model was indicated by one

reviewer from International Journal of Consumer Studies through the following

comment:

This article is certainly relevant and of interest for the field of marketing;

particularly in relation to increasing children's online brand engagement.

93
6.4.4 Full Paper

Children’s participation in brand-based social

networks: Examining the role of evaluative social

identity, self-esteem and anticipated emotions on

commitment and desire to recommend

94
Abstract

Social networks involving the social interactions and personal relationships of brand

devotees (brand-based social networks) are valuable company and marketing resources,

playing a major role in brand and product promotion, and facilitating word-of-mouth.

This research sought to examine the influence of evaluative social identity on brand-based

social network commitment and network recommendations, specifically for children. The

sample for the study comprised 394 Australian children, 6–14 years of age, who

participate in an informal offline social network for the brand ‘Minecraft’. This research

introduces anticipated emotions (positive and negative) as the mechanisms underlying the

influence of evaluative social identity on brand-based social network commitment and

network recommendations. Specifically, when children are unable to participate in the

brand network, they will experience negative emotions. When children are allowed to

participate positive emotions are experienced. These emotions, both positive and

negative, are found to enhance children’s commitment to the brand-based social network

and also their desire to refer the network to non-members. Further, this research provides

evidence that the relationship between evaluative social identity, and both network

commitment and network recommendations, is only observed for children with low

personal self-esteem. This research provides unique insight into the under-researched area

of children and brand-based social networks, and introduces new moderating and

mediating effects on established relationships, with findings useful for both academics

and practitioners.

Keywords

Brand-based social networks, children, self-esteem, negative anticipated emotions,

evaluative social identity.

95
Introduction

The importance of children as consumers cannot be underestimated. Children’s spending

and influencing power has reached over $200 billion per year in the US, with children

spending money on big brand names such as Nike, Mattel and Disney (Collins and

Mitchell, 2015). The purchasing of branded products arises from the increasing

engagement of children with brands, with one popular engagement medium being social

networks. Whilst ‘social network’ is today synonymous with online social networking

sites such as Facebook (Rennie and Morrison, 2013), the term refers to a set of social

relationships (Li and Zhang, 2015), not bound by context (Subrahmanyam et al., 2008).

One type of social network utilized by marketers is brand-based social networks (also

termed brand communities). Brand-based social networks are ‘a specialized, non-

geographically bound community, based on a set of social relationships among admirers

of a brand’ (Muniz and O’Guinn 2001, p. 412). Brand networks are incorporated into

marketing strategies to encourage engagement with brands (Habibi et al., 2014). These

networks can be of any size, online or offline, and informal or formal in nature, as long

as the focus of the network is that of a brand (Muniz and O’Guinn, 2001). A brand

network is unique in the sense that the group is united by a focal brand as opposed to

another interest such as a topic, religion or hobby.

Brand network engagement has been shown to lead to commitment, whereby members

continually maintain a relationship with the brand and network (Kuo and Feng, 2013),

and recommend the network to others (Algesheimer et al., 2005). Network commitment

is displayed through loyalty and caring about the future of the network by members (Chan

and Li, 2010), whereas network recommendations are shown when a member

recommends the network to other non-members (Algesheimer et al., 2005). A review of

literature reveals researchers have focused on understanding adult brand network

96
commitment and network recommendations (e.g., Harley Davidson and Jeep;

McAlexander et al., 2002), with limited interest in understanding outcomes for children’s

network engagement (Flurry et al., 2014). The emergence of child-orientated brand-based

social networks (e.g. Mattel, My Lego Network, Flurry et al., 2015), however, shows

brands understand the value of establishing a forum where child and adolescent brand-

users can engage. Since children are known to participate in these brand-based social

networks it is important to understand how these networks influence child participants.

This research will aid in understanding the implications associated with children’s

participation in brand-based social networks, at a marketing and also societal level.

With social identity playing a crucial role in prior adult-orientated brand network studies

(Dholakia et al., 2004; Bagozzi and Dholakia, 2006; Casaló et al., 2008), and subjective

group dynamics used to understand children’s group behaviour (Abrams et al., 2003) this

research aims to build on these theories in order to understand the effect of children’s

evaluative social identity on offline brand-based social network commitment and network

recommendations. Whilst previous adult-oriented studies have found a positive

relationship between social identity and network commitment (e.g. Casaló et al., 2008),

they have not considered the role of negative and positive anticipated emotions in

explaining the relationship, or how personal self-esteem might impact this relationship.

Evaluative social identity has also been widely cited as a key influencer of an adult’s

desire to engage with brand networks (e.g. Dholakia et al., 2004). Psychologists have

demonstrated a relationship between evaluative social identity and collective self-esteem

defined as self-esteem derived from group involvement (De Cremer and Oosterwegel,

1999). This research, however, adopts an alternative perspective, investigating the

moderating role of personal self-esteem, which is an individual’s sociability and values

(Luhtanen and Crocker, 1992). In addition, the mediating effect of negative anticipated

97
emotions on the relationship between evaluative social identity and network commitment

and recommendations is explored. With children playing a critical role in the market

place, as consumers, influencers and the future market (Gorn and Florsheim, 1985),

coupled with a child’s need to engage and be part of a group (Harter, 1999); it is argued

that understanding the combined role of evaluative social identity, personal self-esteem

and emotional anticipation will not only provide a unique contribution to marketing

literature, but will also be of interest to brand managers and society as a whole, seeking

to better understand child brand-based social network members and the implications of

child participation.

The remainder of this article is organized as follows. Firstly, a discussion of the

theoretical framework of the research, followed by an outline of the method and design

approach. Next, the results are presented, followed by the discussion, limitations and

proposed future research.

Literature review

Social identity

Social identity theory is widely recognized as a key aspect of brand network participation

(Algesheimer et al., 2005; Bagozzi and Dholakia, 2006). An individual forms a social

identity when they see themselves as part of a group (or groups), and understands the

emotional significance of being in that group (Tajfel, 1978). An individual’s social

identity comprises of three components: cognitive, affective and evaluative (Ellemers et

al., 1999). Evaluative social identity, the component of social identity considered in this

research, refers to the comparison of in-group and out-group choices (Bagozzi and

Dholakia, 2006). This aspect of social identity centres on the value the member places on

being part of one group instead of another, or how the individual evaluates their in-group

98
membership. Evaluative social identity has strong ties to collective self-esteem, unlike

the other two components of social identity, that is, cognitive and affective (Ellemers et

al., 1999).

Bennett and Sani (2004) demonstrate that from approximately five years of age, children

have the cognitive ability to self-categorise themselves as being part of a group (or

groups), and acknowledge their identities within these groups. From approximately seven

years of age children experience evaluative social identity beginning with children

differentiating between in-group and out-group, and is experienced when a child

positively evaluates their in-group membership (Bennett and Sani, 2004).

The relationship between evaluative social identity and network commitment has been

well established in adult-oriented research (e.g. Dholakia et al., 2004; Bagozzi and

Dholakia, 2006). Network commitment refers to an individual’s desire and willingness to

develop and maintain a relationship with the brand-based social network (Garbarino and

Johnson, 1999). Once members engage and develop relationships, the desire to continue

these relationships strengthens, based on their need to sustain their identity within the

network.

Research in developmental psychology has established that children begin to develop

loyal and stable relationships from around the age of six years (Cicchetti and Cohen,

2006). During this stage children have a mature appreciation of friendship and the feelings

of others (Cicchetti and Cohen, 2006). Since children can develop these loyal

relationships, it is suggested that children will feel a sense of commitment to a brand-

based social network in the same way that they commit to other types of relationships.

99
Whilst researchers are yet to examine the effect of evaluative social identity on children’s

brand-based social network commitment, drawing from developmental psychology

literature, it is hypothesized that:

H1: There is a positive relationship between evaluative social identity and

network commitment.

Members of brand-based social networks are described to display three characteristics:

consciousness of kind, rituals and traditions, and a sense of moral responsibility (Muniz

and O’Guinn, 2001). Moral responsibility refers to the responsibility network members

feel towards telling others about their network (Muniz and O’Guinn, 2001), that is,

network recommendations. Literature has examined how engagement in brand-based

networks leads to network recommendations (Algesheimer et al., 2005) for adults,

however, no study has tested the direct relationship between evaluative social identity and

network recommendations, for either adult or child networks. Since network

recommendations are vital to brand network growth and sustainability (Carlson et al.,

2008), it is important for both practitioners and academics to understand the direct

antecedents of network recommendations.

Psychologists have found that children discuss topics with friends, including brands and

products (Buckingham, 1993). This shows that children make recommendations to others

and are capable of feeling a sense of moral responsibility in relation to the brand network.

It is argued that when a child feels high evaluative social identity with a brand network

they will establish a sense of moral responsibility to inform others about the group. They

feel important to the group and hence feel a need to recommend the group to others in an

attempt to expand the group. If a child has a low level of evaluative social identity, then

the brand network is not important to them and hence will not form a sense of moral

responsibility. This leads to the following hypothesis:

100
 H2: There is a positive relationship between evaluative social identity and
network recommendations.

Subjective group dynamics

The subjective group dynamics model can explain how children interact within social

groups (Abrams et al., 2003). Whilst subjective group dynamics is yet to be examined in

regards to brand-based social networks, it is suggested that the theory provides a useful

framework for explaining brand network behaviour. Subjective group dynamics theory

holds that members have a bias towards in-group members and strive to uphold the norms

of the in-group (Abrams et al., 2003). Favouritism for in-group members and rituals is

derived from a desire to increase an individual’s social identity (Tajfel and Turner, 1979).

It is posited that a child who feels important to the brand-based social network (i.e., part

of the ‘in-group’) will also feel a sense of evaluative social identity, resulting in a need to

uphold in-group norms.

Group loyalty is a prescriptive norm within subjective group dynamics theory

(Valkenburg et al., 2006; Abrams and Rutland, 2008), and maintaining contact is

suggested to be a way to show loyalty to the group (brand-based social network). One

key characteristic of brand network members is moral responsibility to recommend the

network to others (Muniz and O’Guinn, 2001), and therefore represents another in-group

norm. Children actively talk about brands with peers (Elliott and Leonard, 2004) and are

therefore capable of directly or indirectly making brand network recommendations. By

following the in-group norm of maintaining contact, the child will feel positive emotions

(e.g. happy, glad, excited), in the same way children will also want to recommend the

network to others, to uphold the moral responsibility of being a member.

101
It is posited that when children feel positive emotions from participation in the brand

network, they will want to continue a relationship with the network and hence establish a

commitment to the network. In addition, when children associate positive emotions with

a brand, it is likely that they will talk about the brand to others (Elliott and Leonard, 2004).

It is also argued that positive anticipated emotions, from following in-group norms, will

have a mediating effect on the relationships between evaluative social identity and

network commitment, and evaluative social identity and network recommendations. That

is, positive anticipated emotions will explain the relationship between evaluative social

identity and network commitment and recommendations. These arguments lead to the

following hypotheses:

H3: Positive anticipated emotions mediate the relationship between evaluative

social identity and (a) network commitment and (b) network recommendations.

According to subjective group dynamics theory, a deviant member is a threat to the in-

group and the norms that they uphold (Abrams and Rutland, 2008). When a child feels a

group is important to them, their group social identity has a high influence on the feelings

a child will experience (Cassidy, 2009). It is suggested that if a child, with high evaluative

social identity (i.e., part of the in-group), is temporarily impeded from participating in the

brand network (e.g., parent limiting contact), the in-group norm of maintaining contact

will not be upheld. When the child returns to the group, they may anticipate that they will

be treated differently, because of their inability to uphold the in-group norm. A deviant

member is seen as a threat to the reputation and values of the group (Abrams et al., 2008)

and as such is less accepted by members than normative members (Abrams et al., 2007).

In these instances it is suggested that a child, with high evaluative social identity, who is

impeded from participating with the brand-based social network will experience negative

emotions (e.g., sadness, anger) since they cannot engage with the brand network, and may

102
be seen as a deviant member. These negative anticipated emotions arising from an

obstruction to participate in the brand-based social network will mediate (explain) the

relationship between evaluative social identity and the consequences of engagement

(network commitment and network recommendations).

When children experience evaluative social identity they form strong group favouritism

(Bennett and Sani, 2004). The development of in-group favouritism is strong and lasting,

occurring in children as young as five (Aboud, 2003). Regardless of whether or not the

child has deviated against the norms, they themselves will still feel an evaluative social

identity with the brand-based social network. This continued feeling of evaluative social

identity with the brand network conveys continued favouritism with the in-group.

Because of this, members will still uphold their commitment and recommend the network

even if negative anticipated emotions occur. The strong favouritism with the in-group

will not change by fear of potential removal, and therefore will not change a child’s

commitment to the brand network. It is also argued that the network commitment will

continue since failing to do so would constitute a further possible reason for others to

ostracize them from the network, since this could be considered as continued deviance of

the in-group norms. According to subjective group dynamics theory deviant members

may be kept in the in-group if they begin to follow the rules after breaking them, however,

continuing to not follow the rules will violate the group membership (Abrams and

Rutland, 2008).

This leads to the following hypotheses:

H4: Negative anticipated emotions mediate the relationship between evaluative

social identity and (a) network commitment and (b) network recommendations.

103
Moderating role of personal self-esteem

A child’s social identity is strongly influenced by their self-esteem (Corenblum, 2014),

which constitutes collective and personal aspects (Luhtanen and Crocker, 1992). The

collective aspect refers to an individual’s self-esteem within a group context, and the

personal aspect refers to an individual’s values and sociability as a whole (Luhtanen and

Crocker, 1992). Personal self-esteem is not limited to a single group influence and is most

volatile around the ages of 6 through to 11 (Trzesniewski et al., 2003). The theory of

social identity posits an increase in evaluative social identity is positively related to

collective self-esteem (Corenblum, 2014), however, since personal self-esteem

encompasses underlying sociability and values (Luhtanen and Crocker, 1992) it is

suggested personal self-esteem will play a moderating role on evaluative social identity

relationships. That is, the level of personal self-esteem will change the relationship

between evaluative social identity and network recommendations and commitment.

Children with low personal self-esteem flourish when supported by a peer or group, as

this increases their sociability (Boulton and Smith, 1994), suggesting children with low

self-esteem will benefit from brand networks more than children with high self-esteem.

Personal self-esteem is, therefore, posited to moderate the relationship between

evaluative social identity and both network commitment and network recommendations.

The supporting benefits the child with low personal self-esteem receives from the network

will drive commitment, so they can continue to receive support. In other words, children

with low personal self-esteem will feel supported in the brand network and hence will

want to keep participating in the network to receive more of this support. Children with

high personal self-esteem, however, will not require this support. This is because of their

high level of personal self-esteem, and hence network commitment will not be as high as

those children with low personal self-esteem. A larger group, or brand network, provides
104
more support (Pendley et al., 2002), suggesting children with low personal self-esteem

will make recommendations to increase members and the size of their support group

(brand-based social network). In addition, a child with low personal self-esteem will want

to recommend the network so that others can see they have a sense of moral responsibility

and are therefore a normative member (Muniz and O’Guinn, 2001).

Maintaining a connection to the in-group is more beneficial to children with low self-

esteem (Hogg and Abrams, 1990). The connection to the network helps the child raise

their personal self-esteem, perhaps unconsciously, and without the network the child’s

personal self-esteem could fall further. These theories imply that children with low

personal self-esteem will benefit more from both network commitment and network

recommendations than that of children with high personal self-esteem. Those children

with high personal self-esteem will not rely on the brand network as much for support as

they already have high sociability (Luhtanen and Crocker, 1992). These arguments lead

to the following hypotheses:

H5: Personal self-esteem moderates the relationship between evaluative social

identity and (a) network commitment and (b) network recommendations.

The relationship between evaluative social identity and negative anticipated emotions is

only expected to occur when a child has low personal self-esteem, as individuals with low

personal self-esteem cognitively process negative implications more than positive

implications, because of the ultimate consequences that could occur (Baumeister et al.,

2001). For example, for a child participating in a brand-based social network, the

consequence felt from following the in-group norms is simply a continuation to be

accepted, however, when in-group norms are not followed the consequence felt is higher,

as this could mean being ostracised from the group.

105
Social and group interactions have been widely evidenced as impacting a child’s personal

self-esteem (Boulton and Smith, 1994; McGee et al., 2006). It is therefore argued that

personal self-esteem will impact the relationship between evaluative social identity and

network commitment and recommendations, and negative anticipated emotions and

network commitment and recommendations. According to psychologists, children with

low personal self-esteem feel the consequence of approval, or disapproval, of others more

strongly than children with high self-esteem (Harter, 1993). Therefore the consequences

of not following in-group norms is suggested to be more strongly felt for children with

low personal self-esteem, when the group feels important to them (i.e. high evaluative

social identity). A child with high evaluative social identity, yet low personal self-esteem

will experience strong negative emotions when brand-based social network contact is

temporarily impeded. The child will be concerned with brand network rejection, since

they are not upholding an in-group norm. Because of this it is hypothesized that:

H6: Personal self-esteem moderates the relationship between evaluative social

identity and negative anticipated emotions.

Method

Sample and data collection

Three hundred and ninety four randomly selected Australian children (6–14 age, Mage =

9.94), who self-identified as participating in an informal offline brand-based social

network, completed an online questionnaire. The ages of 6–14 were chosen, in line with

the United Nations definition of a child (United Nations, 2013). Parental consent, and

child assent, was obtained prior to participation. The research focused on informal offline

social networks formed by child admirers of the brand, Minecraft, where networks are

formed in social settings such as the school playground and after-school groups, of an

106
undefined size. Minecraft is a contemporary brand with significant youth engagement.

There are over 100 million registered Minecraft users (Mojang, 2015), with 20% of these

players under the age of 15 (Minecraft Seeds, 2015). Because of this large following and

popularity among children, the brand-based social network was considered appropriate.

To establish the child participants were active in a Minecraft brand-based social network

filter questions were asked at the beginning of the questionnaire. To qualify for this

research participants needed to both play Minecraft and talk to others offline about

Minecraft.

Measurement

All measures were drawn from existing literature, with minor alterations made in some

instances to reflect the language ability of participants (e.g., ‘depressed’ was altered to

state ‘sad’), see Table 1 for a summary of all measures. Specifically, the scale items for

evaluative social identity, negative anticipated emotions and positive anticipated

emotions were adapted from Bagozzi and Dholakia (2006); network commitment from

Chan and Li (2010); network recommendations from Algesheimer et al. (2005); and the

items for personal self-esteem were those which were employed by Harter (1982). In

addition, a measure for perceived behavioural control was adapted from Bagozzi and

Dholakia (2006) and included in the models as a covariate. The covariate was included in

the analysis to account for any bias relating to the ease with which the child could engage

with (have access or be permitted to engage with) the brand-based social network. These

data were collected so that the measure of negative anticipated emotions and positive

anticipated emotions was reliable, since these directly measure the impact when network

contact was impeded or allowed.

107
Table 5 - Summary of Measurement Items
Construct Measure Items Cronbach
Source Alpha
Evaluative Social Bagozzi and The friends I talk about Minecraft to, need me .706
Identity Dholakia I am the leader of the friends I talk about
(2006) Minecraft with
Negative Bagozzi and If I can’t talk to my friends about Minecraft I .944
Anticipated Dholakia will feel:
Emotions (2006) Not sad – sad
Not at all angry – angry
Like I haven’t done something bad – like I
have done something bad
Not afraid – afraid
Not annoyed – annoyed
Not worried – worried
Not guilty – guilty
Not uncomfortable – uncomfortable
Not nervous – nervous
Positive Anticipated Bagozzi and If I talk to my friends about Minecraft I will .903
Emotions Dholakia feel:
(2006) Not excited – excited
Not happy – happy
Not great – great
Not proud – proud
Not brave – brave
Not comfortable – comfortable
Community Chan and Li I would feel sad if Minecraft wasn’t around .898
Commitment (2010) anymore
I care about the future of Minecraft
If I didn’t play Minecraft for a few days, I
would try and play as soon as I could
Community Algesheimer et I always want to tell people that Minecraft is .872
Recommendations al. (2005) awesome
If my friends or family are looking for a game
to play, I would tell them to play Minecraft
Personal self-esteem Harter (1982) Some kids feel they are a good person/ other .701
kids do not feel they are a good person
Some kids are happy the way they are/ other
kids are not happy the way they are
Some kids feel good/ other kids do not feel
good
Some kids are sure they are doing the right
thing/ other kids are sure they are not doing the
right thing
Some kids want to stay the same/ other kids do
not want to stay the same
Some kids do things fine/ other kids do not do
things fine
Perceived Bagozzi and To talk with my friends about Minecraft is -
Behavioural Control Dholakia (Easy – Hard)
(Covariate) (2006)

108
Data analysis

For analysis the PROCESS macro bootstrapping procedure (Hayes, 2013) in SPSS was

applied to test all hypotheses. Two separate models were employed to firstly test the

mediating effects of negative anticipated emotions and positive anticipated emotions

(Model 4, n = 10,000) and secondly to test the moderating effect of personal self-esteem

(Model 8, n = 10,000). The primary reason for choosing the PROCESS method was so

that multiple mediating effects could be tested simultaneously while also testing

moderated mediation. In addition, both indirect and direct effects of the interactions can

be found using the PROCESS method (Hayes, 2013).

Results

First, Model 4 (Hayes, 2013) was employed to test the effect of evaluative social identity

on network commitment and network recommendations as well as the mediating role of

anticipated emotions (both positive and negative). A significant positive relationship was

found between evaluative social identity and network commitment, in support of H1 (p

= <.001). This result shows that as evaluative social identity increased, so too did network

commitment. A higher level of evaluative social identity led to a higher level of network

commitment, and a lower evaluative social identity led to a lower level of network

commitment. As expected a significant positive relationship was also found between

evaluative social identity and network recommendations (p = < .001), supporting H2.

That is, as evaluative social identity increased, network recommendations also increased,

and as evaluative social identity decreased, so too did network recommendations.

To test for a mediating effect (for both positive anticipated emotions and negative

anticipated emotions) the PROCESS macro bootstrapping procedure (n = 10,000) was

employed (Preacher et al., 2007). The PROCESS macro procedure calculates two

109
regression models: the first tests the effect of the independent variable (evaluative social

identity) on the mediator (positive anticipated emotions); the second model tested the

effect of the mediator on the dependent variables: network commitment and network

recommendations. The results from this model showed a significant indirect effect of

evaluative social identity on network commitment and network recommendations (that

is, via positive anticipated emotions) for both dependent variables (network commitment:

β = .556, 95% CI = .438 to .674; network recommendations: β = .631, 95% CI = .501 to

.761), supporting H3a and H3b.

The same procedure was applied by estimating the model again using network

commitment and network recommendations as the dependent variables, and this time with

negative anticipated emotions as the mediating variable. The first model tested the effect

of the independent variable (evaluative social identity) on the mediator (negative

anticipated emotions); the second model tested the effect of the mediator on the dependent

variables: network commitment and network recommendations. Similar to positive

anticipated emotions, significant results were observed when the mediating role of

negative anticipated emotions was considered (network commitment: β = .303, 95% CI

= .218 to .388; network recommendations: β = .246, 95% CI = .152 to .340) supporting

H4a and H4b.

Results indicate that both positive and negative anticipated emotions explain the

relationship between evaluative social identity and network commitment and

recommendations. The results are presented in Figs. 1 and 2.

110
Figure 1 – Research Model of Mediating Effects on Network Commitment (β)

Figure 2 – Research Model of Mediating Effects on Network Recommendations (β)

Next, to test the moderating role of personal self-esteem, Model 8 was estimated (Hayes,

2013). The impact of personal self-esteem on the relationship between evaluative social

identity and both dependent variables (network commitment and network

recommendations) was found to be significant (network commitment: p = .007, network

recommendations: p = .048), supporting H5a and H5b. The relationship between

evaluative social identity and network commitment and network recommendations,

111
respectively, was stronger for children with low personal self-esteem (β = 0.444, p <

.001; β = 0.363, p < .001) than children with high self-esteem (β = 0.211, p < .001;

β = 0.169, p = .014). Personal self-esteem was also found to moderate the effect of

evaluative social identity on negative anticipated emotions (p = .022), where the

relationship between evaluative social identity and negative anticipated emotions was

only significant for children with low personal self-esteem. As predicted, personal self-

esteem had no impact on the relationship between evaluative social identity and positive

anticipated emotions (p = .263).

When testing for moderated-mediation (H6), the key indicator is the indirect effect

of the interaction on the dependent variable through the mediator (Preacher et al.,

2007). To determine the moderated-mediation effect, a bootstrapping approach was

employed using the PROCESS macro (Preacher et al., 2007) that estimated the

significance of two paths: evaluative social identity x personal self-esteem

 negative anticipated emotions  network commitment, and evaluative social

identity x personal self-esteem  negative anticipated emotions  network

recommendations. This is the Model 8 approach recommended by Preacher et al.

(2007). In the context of the models estimated, moderated-mediation is demonstrated

when the 95% confidence interval for the parameter does not include zero (Preacher

et al., 2007). Results show that the 95% boot-strapped confidence interval for the

indirect effect of the interaction (evaluative social identity x personal self-esteem)

did not include zero across the two dependent variables: network commitment (β =

0.071, 95% CI = 0.003 to 0.155) and network recommendations (β = 0.057, 95% CI =

0.002 to 0.132), supporting the existence of a significant moderated indirect effect. That

is, negative anticipated emotions explain the effect of the interaction between evaluative

112
social identity and personal self-esteem on network commitment and network

recommendations. The detailed results are presented in Fig. 3.

Figure 3 – Research Model of Mediating Effects on Network Commitment (β)

Discussion

The results of this study present numerous insights into children’s interactions with

offline brand-based social networks, highlighting important societal and managerial

implications. Firstly, results show that children, who had high evaluative social identity,

felt positive emotions when they could interact with the brand-based social network,

leading to both network commitment and network recommendations. These results are in

line with subjective group dynamics theory, showing children who are complying with

the in-group norms, such as maintaining group contact (Valkenburg et al., 2006) feel

positive emotions from following the in-group norms. Children are then motivated to

commit to the brand-based social network and also recommend the network to others.

113
Secondly, children who had a strong sense of evaluative social identity felt negative

emotions when they believed they would be hindered from connecting to the brand-based

social network. Although these negative emotions were evident, the consequences of

commitment and recommendations were also still evident. This is in line with subjective

group dynamics theory whereby children form a strong favouritism towards the in-group,

and still feel group favouritism even after going against the in-group norms (Abrams et

al., 2003). These findings highlight an issue of concern relating to brand-based social

networks, and how they have the ability to affect the emotions of child-age participants

to make them feel sad, angry and worried when they might not be able to participate.

Parents should be aware of this to ensure the well-being of their children when

participating in brand networks.

Thirdly, results show that personal self-esteem moderates the relationship between

evaluative social identity and network commitment, and evaluative social identity and

network recommendations. When a child with low personal self-esteem establishes an

evaluative social identity with a brand-based social network they will be more committed

to the network, however, when a child has high personal self-esteem, the effect of

evaluative social identity on network commitment and network recommendations

weakens. These results reflect self-esteem literature arguments whereby children with

low personal self-esteem cope better when they feel group support (Boulton and Smith,

1994). In addition, the need of a child with low personal self-esteem to continue

participating in the network in order to increase their self-esteem is also seen through this

result. These findings highlight concerns that children with low personal self-esteem may

be targeted by brands as an avenue for commitment to a brand network. Parents aware of

their child having low personal self-esteem should monitor their child’s interactions in

these brand networks to make sure brands do not take advantage of them.

114
Similarly, findings show that children with low personal self-esteem and high evaluative

social identity with the brand-based social network are more likely to make

recommendations to others about the brand network. This highlights that children with

low personal self-esteem are pressured to uphold the in-group norm of group

recommendations to others, a form of moral responsibility for the brand network.

Children with high personal self-esteem were not likely to make network

recommendations, highlighting that children with high personal self-esteem were less

susceptible to peer pressure and did not feel that in-group norms had to be followed. This

is in line with predictions based on previous literature (Boulton and Smith, 1994) and

shows that children with low personal self-esteem were more affected by negative

emotions when prevented from participating in the brand-based social network. The

results could imply a stronger desire to be part of the brand-based social network for

children with low personal self-esteem, compared with the children with high personal

self-esteem. In addition, only negative anticipated emotions, significantly affected

commitment and recommendations, supporting previous theory that negative emotions

have a stronger impact on those with low personal self-esteem than positive emotions

(Leary et al., 1995). Again, these findings highlight a concern that children with low

personal self-esteem can be targeted by brands, in this case to spread recommendations

to others to join the brand-based social network. Of perhaps even more concern is the

possibility that these recommendations will lead to even more children with low personal

self-esteem becoming vulnerable to these brand networks.

Lastly, negative anticipated emotions were found to act as the driving mechanism on

network commitment and network recommendations, in line with subjective group

dynamics theory (Abrams et al., 2003). This emphasizes that children’s commitment and

recommendations are mediated by anticipated negative emotions (fear of not upholding

115
in-group norms) when temporarily impeded from network participation. Children are

committed to the group out of fear that if in-group norms are neglected they will be

ostracized from the in-group (brand-based social network).

Limitations and directions for future research

A number of limitations are recognized highlighting avenues of future research. The

current study focused on one brand: ‘Minecraft’. Whilst it is anticipated that results

obtained in this study would be replicated when examining alternate brand-based social

networks, future research should confirm this assumption. This research only focused on

an informal, offline brand-based social network. The conceptual framework produced

should be tested for an online brand-based social network, to compare findings between

offline and online brand-based social networks. The characteristics of online brand-based

social networks, such as the lack of face-to-face interactions, could change the level of

support children with low personal self-esteem feel. The online environment allows users

to become almost anonymous (or assume another identity), and can dramatically change

how users interact. Because of this, it is argued that the results of this study would be

different when applied to an online brand-based social network.

In addition, this research focused on one component of social identity only, that is,

evaluative social identity. Future research may look to extend the conceptual framework

presented in this study to also include the cognitive and affective (Ellemers et al., 1999)

components, to produce a holistic view of the issue. Self-esteem has links to children’s

well-being, children with low self-esteem often being victims of bullying (e.g.

Valkenburg et al., 2006). Because of the high tendency for a child to be victimized when

they experience low self-esteem, it is speculated that children with low self-esteem would

feel bullied if norms were not upheld. In addition, it is suggested that children with low

116
self-esteem may be bullied into obeying the norms, instead of following the norms

because of network loyalty. The present study did not include bullying in the model,

however, future research should explore this area.

Finally, as this study is one of the first to test children’s interactions in brand-based social

networks, more research should look into this area. Observations of children’s behaviour

will uncover insights into child participation in comparison with adults, and provide

valuable knowledge into how to design networks for children. A comparison study

between adult and children’s participation in a brand-based social network would

highlight interesting differences key to designing networks suitable for child interaction.

Conclusion

In conclusion, results revealed that evaluative social identity has an effect on offline

brand-based social network commitment and network recommendations for children aged

6–14 years. Specifically, the relationship between evaluative social identity and two

consequences (network commitment and network recommendations) can be explained by

anticipated emotions when participating with or being prevented from participating in the

brand-based social network. That is, those children with evaluative social identity were

more likely to experience positive emotions when they could engage, and negative

emotions when impeded from participating in the brand-based social network, leading to

stronger network commitment and network recommendations. Further, the relationship

for negative anticipated emotions is dependent upon whether a child has low or high

personal self-esteem, with the relationship not observed for children with high personal

self-esteem.

With limited research undertaken to examine children’s brand-based social network

commitment and network recommendations, this research provides a unique contribution

117
to social network, social identity and subjective group dynamics literature. Findings of

this research would be of interest to academics, and marketing practitioners, specifically

those involved with social networks aimed at children. Parents too would take interest in

the findings, as the issues raised should be monitored to ensure children are safely

interacting in brand-based social networks.

118
References

Refer to Appendix 9.5.3 for Paper 3 Reference List

119
6.5 Paper Four: Similarity Attraction

Title: “I'm Like You, You’re Like me, We Make a Great Brand Community!’ Similarity-

Attraction and Children’s Brand Community Participation”

Submitted to the Journal of Brand Management June 2018

6.5.1 Statement of Contribution of Others

By signing below, I confirm that Margurite Hook was the sole contributor to the paper

entitled “’I'm Like You, You’re Like me, We Make a Great Brand Community!’

Similarity-Attraction and Children’s Brand Community Participation”. The co-authors

(Stacey Baxter and Alicia Kulczynski) only provided guidance for the paper, with limited

intellectual input.

X
Associate Professor Stacey Baxter
Co-author Faculty Assistant Dean Research Training

X
Dr Alicia Kulczynski
Co-author

120
6.5.2 Overview and Contribution of Paper

‘Birds of a feather flock together’, a famous saying that typifies the psychological concept

of similarity attraction. Paper Four: Similarity Attraction contributes to the thesis by

providing a comprehensive three-part study investigating the role of similarity attraction

and continuing the investigation on Subjective Group Dynamics (SGD), begun in Paper

Three: New Model. Drawing from studies in psychology and marketing, three

experimental studies were undertaken with the aim of understanding the impact of

similarity between child brand community members (sharing of characteristics such as:

age, background and opinions) and SGD on their brand community participation desire.

Similarity attraction has been evidenced in a wide range of areas, from adult romantic

relationships (Montoya & Horton, 2004; Singh, Tay, & Sankaran, 2016) to interpersonal

relationships amongst children as young as three (Fawcett & Markson, 2010). The desire

to establish relationships with others is often due to a common or shared possession of a

characteristic, especially in the context of children (Aboud & Mendelson, 1996;

Haselager et al., 1998; Hunter et al., 2016).

Given that the core of a brand community is the relationships amongst individuals (Muniz

Jr & O’Guinn, 2001) and children have been widely shown to participate with similar

others (e.g. Aboud & Mendelson, 1996; Haselager et al., 1998; Hunter et al., 2016), the

theory of similarity attraction was anticipated to play a role in child-oriented brand

community relationships. Children that perceived a brand community to have

characteristically similar members to themselves, were posited to have an increased

respect for the community and therefore a higher desire to participate.

Building upon the study of Paper Three: New Model, SGD theory was also examined.

Specifically, the role deviant members have on the reputation of the group. Deviant

121
members threaten the reputation of the entire group, not just their own (Abrams et al.,

2003). Given this, a moderating role of member deviance was proposed on the model.

Three experimental studies were undertaken, each employing separate samples of

Australian children aged 6-17 years. Studies 1 and 2 were between subjects designs, with

participants allocated to either one of two experimental conditions: low member

similarity or high member similarity. Study 3 was a 2 (member similarity) x 2 (member

deviance) factorial design, with participants randomly allocated to one of four

experimental conditions: low member similarity and deviant member, low member

similarity and normative member, high member similarity and deviant member, or high

member similarity and normative member.

The findings of the three studies showed that children were more likely to desire

participation in the brand community when the current members were perceived as

similar to themselves. This relationship was explained by the respect felt towards the

community. However, these relationships were only seen when there was no deviant

member present. A deviant member caused a significantly lower desire to participate,

even when the members were perceived as similar to the child, due to a lower level of

respect towards the community.

Paper Four: Similarity Attraction’s findings contribute significantly to the current thesis

and to the field. For the current thesis, Paper Four: Similarity Attraction provides a model

through which to explain children’s brand community participation, a significant

contribution to the aim of this thesis (understanding the factors influencing children’s

participation in brand communities). For academics, the theory of similarity attraction

was introduced into the brand community field, along with the concept of respect towards

122
a brand community. Both of these concepts can be integrated into future studies to

uncover new insights in the field.

Drawing from the results of Paper Four: Similarity Attraction, practitioners are advised

to promote children’s brand communities in a manner that emphasises the similarity of

current members to new potential members (e.g. ‘come play with kids just like you’). The

role of deviant members is also brought to attention, an important issue for both

academics and practitioners to be aware of.

Paper Four: Similarity Attraction is yet to be published, however, is currently under

review at the Journal of Brand Management.

123
6.5.3 Full Paper

‘I'm Like You, You’re Like me, We Make a Great

Brand Community!’

Similarity-Attraction and Children’s Brand

Community Participation.

124
Abstract

‘Birds of a feather flock together’, a famous saying that typifies the psychological concept

of similarity-attraction. Moving beyond a common interest in a brand, this research

employs similarity-attraction theory to understand the impact of similarity between child

brand community members (sharing of characteristics such as: age, background and

opinions) on their brand community participation desire. Australian children (n = 466)

aged 6 - 17 participated in one of three experimental studies to assess the impact of

member similarity, respect, and member deviance on brand community participation

desire. Results suggest that greater member similarity enhances children’s desire to

participate in a brand community, with this effect explained by an increased respect felt

towards the community. In addition, when a community member is deviant (disloyal to

the community); respect, and subsequently participation desire, declines. By introducing

the theory of similarity-attraction to child-oriented brand communities; the research

contributes to the sparse literature on child-oriented brand communities, with results

highlighting that marketers should emphasize member similarity when promoting brand

communities aimed at children.

Keywords:

Brand Communities, Children, Similarity-attraction, Respect, Group Dynamics

125
1. Introduction

Brand communities aimed at a child audience are a very popular and valuable marketing

resource, playing a major role in brand and product promotion, and facilitating word-of-

mouth. For example, in 2016, Lego launched their brand community ‘Lego Life’ (Lee,

2017). This brand community received instant popularity, with over one million

downloads of the app on android devices alone, within its first year of operation (LEGO

System A/S, 2017). ‘Lego Life’ is specifically targeted at children under the age of 13,

and encourages its members to share their experiences of the ‘Lego’ brand with others

(Lee, 2017). Although many child-orientated brand communities are employed to engage

children with brands; only a small number of academics have attempted to understand

and explain child-orientated brand community behavior, and the implications for

marketers and brands (Flurry, Swimberghe, and Parker, 2014; Hook, Baxter, and

Kulczynski, 2016).

Brand communities are a specialized community that rely on the social relationships

established between admirers of a brand (Muniz and O’Guinn, 2001). The relationships

in a brand community are formed due to a common interest towards a focal brand (Cova

and Pace, 2006). Whilst a common interest in the focal brand is important, and is at the

center of every brand community (Cova and Pace, 2006); there is evidence to suggest a

greater similarity between a group of individuals is especially important for successful

interactions among children. Developmental psychology explains that the more

similarities there are between children, the greater the likelihood a relationship will form

(Aboud and Mendelson, 1996). Of even more importance, relationships will last, for an

extended period, when there is a higher level of shared characteristics between children

(Aboud and Mendelson, 1996). Drawing from these findings, it is suggested that a greater

similarity between child brand community members will lead to the formation of
126
enduring relationships within a brand community. Investigating similarity between child

brand community members (beyond the similarity of a shared interest in the brand) will

yield useful results for practitioners looking to create long term relationships with

children through employing a brand community.

The attraction between similar individuals has been termed by psychologists as

‘similarity-attraction’ (Byrne, 1971). The majority of psychology researchers agree that

a similarity between individuals is more likely to foster strong social relationships, than

a dissimilarity between individuals (e.g. Montoya and Horton, 2004; Singh, Chen, and

Wegener, 2014). Research to date on similarity-attraction has focused primarily on the

formation of, and interaction within interpersonal relationships. For instance, the

formation of social relationships (e.g. interactions within a school classroom) among

children has been found to be influenced by similarity-attraction (Hunter, Fox, and Jones,

2016). When children perceive others to be similar to themselves, they will form

relationships with them (e.g. Haselager, Hartup, Lieshout, and Riksen-Walraven, 1998;

Hunter et al, 2016). Although there is a collection of research on similarity-attraction in

the psychology field, there has been little attention from other fields, including marketing.

This research introduces similarity-attraction theory to explain children’s brand

community participation.

When looking at interpersonal relationships in marketing, the concepts of trust (e.g.

Becerra and Badrinarayanan, 2013; Delgado-Ballester, 2004; Morgan and Hunt, 1994),

loyalty (e.g. Marzocchi, Morandin, and Bergami, 2013; McAlexander, Kim, and Roberts,

2003) and even love (e.g. Albert, Merunka, and Valette-Florence, 2008; Noel and

Merunka, 2013) have been examined extensively in relation to brands. The current study,

however, is introducing respect as an alternative construct that explains participation

desire. Drawing from previous studies in psychology (e.g. Gueguen, Martin, and Meineri,
127
2011; Montoya and Horton, 2004; Singh et al, 2014), similarity between members is

proposed to increase the level of respect felt towards the brand community, and this

respect is what leads to a desire to participate. In addition, building upon findings that

children harshly evaluate disloyal group members (Abrams, Rutland, Ferrell, and

Pelletier, 2008), it is proposed that a disloyal member will influence the respect members

feel towards the community, and subsequently influence their brand community

participation desire.

To examine the effect of similarity-attraction, respect towards the brand community, and

member deviance on children’s desire to participate in a brand community, three

experimental studies were undertaken. Study 1 establishes that a high level of brand

community member similarity results in a higher desire to participate in the community

when compared to a low level of brand community member similarity. In Study 2, it is

demonstrated that the indirect effect of member similarity on desire to participate is

explained through a perceived respect for the community. The bounds of the effect of

respect are empirically revealed in Study 3. Specifically, the results provide evidence to

suggest that the effectiveness of member similarity on respect is attenuated when there is

a deviant member (disloyal to the community) in the community, subsequently diluting

the desire to participate in the brand community.

2. Similarity-Attraction and Child Brand Community Participants

Similarity-attraction theory (termed by some as the ‘homophily hypothesis’; Kandel,

1978) suggests individuals interact with those that share similar characteristics to

themselves (Gueguen et al, 2011). There are multiple theories as to why this attraction

occurs (Nangle, Erdley, Zeff, Stanchfield, and Gold, 2004), with one theory termed the

‘effectance motive’ describing the need for a predictable, certain, and meaningful

128
interpretation of the world (Byrne and Clore, 1967; White, 1959). Through interacting

with others who support and validate an individual’s views (due to being

characteristically similar); this ‘effectance motive’ is strengthened and nurtured (Byrne

and Clore, 1967). Conversely, when an individual interacts with others that have

dissimilar views (differing characteristics to their own) there is little support and

validation (Byrne and Clore, 1967).

Similarity-attraction has been evidenced across a wide range of contexts, from adult

romantic-oriented relationships (e.g. Montoya and Horton, 2004; Singh and Ho, 2000),

through to the formation of interpersonal relationships in children as young as three years

old (Fawcett and Markson, 2010). Focusing specifically on child-oriented studies,

similarity between a range of characteristics, both physical and non-physical have been

shown to influence relationship formation. For example, physical characteristics such as

hair color (Fawcett and Markson, 2010) and age (Hartup, 1989); as well as non-physical

characteristics like opinions (Epstein, 1989), and type of humor (Hunter et al, 2016). The

more of these characteristics children have in common, the greater the likelihood a

relationship will be formed and, the more likely the relationship will last (Aboud and

Mendelson, 1996).

To date, the majority of studies on brand community participation have focused on adult

participants, with very few dedicated to a child context (Flurry et al, 2014). Initial

evidence suggests that findings from adult-oriented studies are not directly applicable to

children (Hook, Baxter, and Kulczynski, 2017). Specifically, relationships evidenced as

being significant in adult participants, for example between attitudes and desire, were not

found for children (Hook et al, 2017). For example, findings indicated that desire was not

necessary for social interaction and related community behavior to occur, the formation

of positive attitudes lead directly to social interaction, unlike adults (Hook et al, 2017).
129
In addition, relationships between group behavior and brand behavior, found insignificant

for adults were instead found to be significant in the child context (Hook et al, 2017).

These results signify that more needs to be done to understand this niche area.

Little is known about the drivers of children’s participation in brand communities, nor

how children form relationships in branded communities. Similarity-attraction theory

informs us that children are more likely to form stronger, and longer lasting relationships

when they share similar characteristics (Haselager et al, 1998; Hunter et al, 2016).

Drawing from this, it is proposed that similarity-attraction theory can be used to explain

children’s participation in brand communities, since these communities constitute groups

of interpersonal relationships. Specifically, when a child perceives the brand community

members to be similar to themselves (high member similarity), in terms of range of

characteristics (age, background, culture and opinions), they will have a higher desire to

participate in the community. Drawing from similarity-attraction theory, the following

hypothesis is posited:

Hypothesis 1: High (low) member similarity will result in a greater (lower) desire

to participate in the brand community.

3. Similarity-Attraction and Respect

Respect, or admiration, towards an individual is known to influence similarity-attraction

relationships (Lydon, Jamieson, and Zanna, 1988), and individuals assess the extent to

which they respect others (termed ‘cognitive quality’) prior to commencing a relationship

(Montoya and Horton, 2004). The characteristics an individual holds are ranked as being

correct and superior (Montoya and Horton, 2004), and therefore when others are

characteristically similar (portray similar characteristics as themselves); they will be

highly admired, or respected, in turn leading to a greater desire to establish a relationship

130
(Singh et al, 2014). On the other hand, if an individual has little to no characteristics in

common, there is no level of admiration, or respect, and in turn little, or no desire to

establish a relationship (Singh et al, 2014).

The respect an individual feels can be directed not only towards one specific individual,

but also towards a group of individuals (Ellemers, 2012; Huo and Binning, 2008). Just as

one can respect an individual due to the attitudes and opinions they hold, an individual

can also respect a collective (Ellemers, Doosje, and Spears, 2004), such as a group, team

or community. Based upon previous findings (Ellemers, 2012; Ellemers et al, 2004; Huo

and Binning, 2008), in the current context of brand communities, it is proposed that there

are two levels of respect. Firstly, the respect felt towards each individual member

(individual respect), and secondly the respect felt towards the collective brand community

as a whole (collective respect), inclusive of all members. Since a brand community is a

‘set of social relationships’ (Muniz and O’Guinn, 2001, p. 412), it is the interactions

between multiple members, rather than one on one interactions that are of more

importance, and therefore, collective respect is more relevant to the current study of brand

communities.

In reference to adult interpersonal relationships, collective respect, sometimes called

‘intragroup respect’ (Stürmer, Simon, and Loewy, 2008) or ‘respect within group’ (Tyler,

Degoey, and Smith, 1996), is very important within group contexts, and is suggested to

influence group-related behavior, such as participation (e.g. Branscombe, Spears,

Ellemers, and Doosje, 2002; Simon and Stürmer, 2003; Tyler et al, 1996). When an

individual is respected by the group, and in turn respects the group as a whole, it is argued

that there will be a higher likelihood of group-based participation. In other words, it is

expected that a positive relationship between collective respect and participation, will

occur.
131
Previous child-oriented studies (Dunham, Newheiser, Hoosain, Merrill, and Olson, 2014;

Haselager et al, 1998) also suggest that collective respect is relevant to child interpersonal

relationships, with utilization of terms such as ‘social acceptance’ and ‘social rejection’,

relating to those the child ‘liked most’ and ‘liked least’. When a child perceives another

to be characteristically similar, they are more likely to be accepted socially (‘liked most’),

and contrarily when there is little similarity they are socially rejected (‘liked least’)

(Haselager et al, 1998). Although ‘respect towards a group’ has not specifically been

investigated before, ‘liking’ (social acceptance or social rejection) is argued to be a

simplified portrayal of respect.

Drawing from these previous findings, it is proposed that respect towards the brand

community can be used to explain the relationship between brand community member

similarity and desire to participate in the brand community, in the context of child

participants. Specifically, respect is suggested to mediate the relationship between

member similarity and participation desire, whereby an increase in respect towards the

community will result in an increase in participation desire. Respect towards the

community will increase when the child perceives the brand community members’ to be

characteristically similar, rather than dissimilar, to themselves. This leads to the following

hypothesis:

Hypothesis 2: Respect mediates the relationship between member similarity and

participation desire.

4. Member Deviance

Subjective group dynamics theory states that children strive to sustain a positive

social identity within a group (Abrams, Rutland, Cameron, and Marques, 2003).

Sustaining a positive social identity is impacted by whether group norms (rules within the

132
group) are upheld (Abrams et al, 2008). In brand communities, loyalty to the community

is one such norm (Hook et al, 2016), whereby members of the brand community are

expected to stay loyal, i.e. stay a member of the community. If a member of the brand

community becomes disloyal (for example, wants to join a competing brand community),

this member is considered a deviant member (Hook et al, 2016). Disloyalty can be

determined as a result of a combination of factors over time, or arise from a single instance

(Abrams et al, 2003). Research has suggested that a single-instance preference for a

competing brand community can be seen as a source of disloyalty (Hook et al, 2016). A

deviant member is a threat to the group and its norms (Abrams and Rutland, 2008) and as

such are less accepted by members, both at an individual level and collectively (Abrams

et al, 2007). Normative members are more accepted, individually and collectively, since

they are not going against the norms (Abrams et al, 2007). Previous child-oriented studies

in psychology have investigated subjective group dynamics and similarity attraction in

isolation, however, the current study draws these concepts together in an alternative

context of brand communities.

The threat deviant members pose to the group (Abrams et al, 2008) is suggested to

negatively influence the respect felt towards the brand community. If a child is looking

to join the brand community but sees that not all members are loyal to the group (in other

words there is a deviant member present), it is likely they will have a lower respect

towards the brand community as a whole. A misalignment of values and attitudes will

occur, since there is a current member being disloyal, which in turn will result in a loss

of respect for the community. This loss of respect will lead to a lower desire to participate

in the community. Conversely, if all current members are seen as loyal to the brand

community, there is no reason to disrespect the community.

133
Drawing from subjective group dynamics theory stating that deviant members are judged

harshly and impact group interactions (Abrams, Rutland, Pelletier, and Ferrell, 2009), it

is argued that member deviance will moderate the relationship between member

similarity and respect. That is, an individual will lose respect for the brand community

when they witness deviant behavior by a member, irrespective of how characteristically

similar they are to the group members. These arguments lead to the following hypothesis:

Hypothesis 3: Member deviance moderates the relationship between member

similarity and respect.

Drawing from prior literature in developmental psychology, social psychology and

marketing, Figure 1 presents a conceptual model for the impact of similarity-attraction

within the context of child-oriented brand communities.

Figure 2 - Conceptual Model and Hypotheses

Respect towards
the community
Member
Deviance

Member Desire to
Similarity participate

5. STUDY 1

Study 1 aims to demonstrate the effect of brand community member similarity on desire

to participate in the brand community. The study proposes that when child-age

participants perceive that brand community members are similar to themselves (in terms

134
of age, background, culture and opinions) they will have a higher desire to participate in

the brand community.

5.1 Participants and Procedures

A sample of 142 child-members of the Australian general public aged between six and

17 years (71 male, 71 female; MAge = 11.53, SDAge = 3.32), were recruited through an

Australian research panel company to participate in an online experiment. This method is

consistent with the approach undertaken by Baxter, Ilicic, Kulczynski, & Lowrey (2017),

Baxter & Lowrey (2014) and Hook et al (2016). A between subjects design was

employed, with participants randomly allocated to one of the two experimental

conditions: low member similarity (n = 72) or high member similarity (n = 70). This

sample exceeded that required for statistical power of .80, with an a priori alpha level of

.05 and estimated medium effect size (f =.25) (that is, n > 128; G*Power).

First, participants were asked to rate their attitude towards the brand ‘Tip Top’ on three

seven-point scales drawn from Mitchell and Olsen (1981) (dislike/like, bad/good,

uninteresting/interesting; Cronbach α = .913). ‘Tip Top’ is a well-known Australian food

brand. Participants were then presented with the following scenario: Imagine that you are

thinking of joining the 'Tip Top' community. There are lots of members in the ‘Tip Top’

community. Participants were also told in the scenario, dependent upon the condition, that

the members of the community: are the same (different) age as you, have the same

(different) culture as you, have the same (different) opinions as you and have the same

(different) background as you. The high similarity condition used the word same, the low

similarity condition used the word different. The manipulation of member similarity was

adapted from prior similarity-attraction studies (Balmer, Powell, Karaosmanoğlu, Banu

Elmadağ Baş, & Zhang, 2011; Hung, 2014). Participants were then asked to report the

135
degree to which they agreed with three statements (seven-point scales) that focused on

their desire to participate in the ‘Tip Top’ brand community (e.g. “I want to interact

together with the ‘Tip Top’ community” does not describe me/describes me very much;

Cronbach α = .954) drawn from Dholakia, Bagozzi, and Pearo (2004). Next, participants

were asked to report the degree to which they were similar to the members of the brand

community to assess the perceived level of member similarity (four statements using

seven-point scales; not at all/a lot) (e.g. “The members of the ‘Tip Top’ community are

similar to me in terms of age” Cronbach α = .951). Simple demographic information was

also obtained.

5.2 Manipulation and Confound Checks

An independent samples t-test was performed to check if the manipulation of member

similarity was successful. There was a significant difference in perceived member

similarity across the two experimental conditions, with those participants in the high

member similarity condition perceiving a higher similarity than those in the low member

similarity condition (p < .001, MHighSim = 4.80, SDHighSim= 1.46; MLowSim = 3.36, SDLowSim

= 1.51). This result signifies the manipulation of member similarity was successful. In

addition, another independent samples t-test was performed to ensure no confounding

effects occurred with regards to pre-formed attitudes towards the brand ‘Tip Top’. Results

of the t-test showed that there was no significant difference across participants with

regards to attitude towards the brand ‘Tip Top’ (p = .109 MHighSim = 4.84, SDHighSim =

1.65; MLowSim = 4.42, SDLowSim = 1.46).

5.3 Results

Hypothesis one proposed that member similarity would influence a child’s desire to

participate in a brand community. Specifically, it was expected that when a child-

136
participant perceived the members of the brand community to be similar to themselves

(high member similarity), there would be a higher desire to participate when compared to

brand community members who were not similar (low member similarity). To test this

hypothesis an ANOVA model was estimated, with desire to participate in the brand

community as the dependent variable. Results confirmed that member similarity had a

significant effect on brand community participation desire (F(1,140) = 6.65, p = .011, η2

= .045). Results showed that those who were informed that brand community members

were very similar to themselves (high member similarity) reported a significantly greater

participation desire when compared to those who were informed the members were not

similar to themselves (low member similarity) (MHighSim = 4.09, SDHighSim = 1.79; MLowSim

= 3.34, SDLowSim = 1.67), supporting Hypothesis one.

Given the wide age range of participants used in the current study, an additional test was

performed to check that age did not impact the findings. Two age cohorts were created

(6-11 years old and 12-17 years old), similar to groups used by others when investigating

children (Abrams et al, 2007, 2003). Using these two age cohorts, the PROCESS model

1 was employed (n = 10,000, Preacher et al, 2007), to confirm that the age of participants

did not moderate the observed similarity effect, using desire to participate as the

dependent variable. A significant moderating effect was not observed (β = .686, p = .239,

R2∆ = .010), demonstrating that the effect of member similarity is not conditional on age.

6. STUDY 2

Study 2 aims to demonstrate that the effect of member similarity on desire to participate

is explained through a perceived respect for the members of the community. This second

study posits that when child-participants perceive the members of the brand community

to be characteristically similar to themselves, they will have a higher level of respect for

137
the brand community. The level of respect will, in turn, positively influence desire to

participate in the brand community.

6.1 Participants and Procedures

A sample of 143 child-members of the Australian general public aged between six and

17 years (83 male, 60 female; MAge = 11.94, SDAge = 3.54), were recruited through an

Australian research panel company to participate in an online experiment. A between

subjects design was employed, with participants randomly allocated to one of the two

experimental conditions: low member similarity (n = 71) or high member similarity (n =

72). This sample exceeded that required for a statistical power of .80, with an a priori

alpha level of .05 and estimated medium effect size (f =.25) (that is, n > 128; G*Power).

First participants were asked to rate their attitude towards the brand, ‘Kleenex’, using the

measures from Study 1 (Cronbach α = .913). Then participants were presented with the

same scenario and member similarity situation (random allocation to one member

similarity condition) as per Study 1, however, using the brand ‘Kleenex’ where

applicable. An alternative brand was chosen for each separate study to strengthen the

generalizability of the findings across brands. ‘Kleenex’ is a well-known tissue brand in

Australia. As per Study 1, participants desire to participate in the community was then

measured (Cronbach α = .954). Next, participants were asked to report the degree to

which they agreed with four statements (seven-point scales; strongly disagree/strongly

agree) that measured their perceived respect for the brand community members (e.g.

“The members of the ‘Kleenex’ community are probably good at everything they do”

Cronbach α = .942) adapted from Singh, Chen, and Wegener (2014). Participants were

also asked to report the degree to which they were similar to brand community members

138
as per Study 1, however, replacing the brand ‘Tip Top’ with the brand ‘Kleenex’

(Cronbach α = .951). Finally, basic demographic information was obtained.

6.2 Manipulation and Confound Checks

An independent samples t-test was performed to check that the manipulation of member

similarity was successful. There was a significant difference between the two

manipulation conditions, with those participants in the high similarity condition

perceiving a higher similarity than those in the low similarity condition (p < .001, MHighSim

= 5.08, SDHighSim = 1.25; MLowSim = 3.60, SDLowSim =1.39). This result signifies that the

manipulation of member similarity was successful.

In addition, replicating the process applied in Study 1, an independent samples t-test was

performed to ensure no confounding effects occurred with regards to pre-formed attitudes

towards the brand ‘Kleenex’. Results of the t-test showed that there was no significant

difference across participants with regards to attitude towards the brand ‘Kleenex’ (p =

.524, MHighSim = 5.15, SDHighSim = 1.37; MLowSim = 5.01, SDLowSim = 1.32).

6.3 Results

Hypothesis two argued that the relationship evidenced in Study 1, between member

similarity and desire to participate, could be explained by the perceived level of respect

felt towards the brand community. To test this hypothesis, the PROCESS macro

bootstrapping procedure (n = 10,000, Model 4) was employed (Preacher et al, 2007).

Results demonstrated that respect mediated the relationship between member similarity

and participation desire (β = .404, 95% CI = .066 to .742, p = .020), see Table 1 for

summarized regression results. These results support Hypothesis two, showing that

respect towards the community is the driver for desire to participate in the brand

community. A high level of similarity between members, increased the respect felt
139
towards the brand community. This increase in respect, lead to an increase in desire to

participate in the brand community. Conversely, when there was a low level of similarity

between members, the level of respect decreased, as to did desire to participate.

Table 6 - Regression Results (Study 2)

Beta t p
Dependent variable: Respect towards the community
Member Similarity (X) .404 2.362 .020
Dependent variable: Desire to Participate
Respect towards the community (M) .867 7.422 <.001
Member Similarity (X) .336 1.390 .167

To confirm that the age of participants did not moderate the observed respect effect, the

procedure employed in Study 1 was again utilized for Study 2 with participants grouped

into the same two age cohorts (6-11 and 12-17 years old). Specifically, the PROCESS

model 1 was employed (n = 10,000, Preacher et al, 2007), with respect as the dependent

variable. A significant moderating effect was not observed (β = -.151, p = .667, R2∆ =

.001), demonstrating that the effect of respect is not conditional on age.

7. STUDY 3

Study 3 aims to empirically test the bounds of the effect of respect, by examining the

moderating effect of member deviance on the relationship between member similarity

and respect towards the community. This concluding study proposes that member

deviance will impact the relationships evidenced in Study 2. Specifically, when a deviant

member is introduced into the brand community, respect felt towards the community is

hypothesized to decrease, when there is a high member similarity.

140
7.1 Participants and Procedures

A sample of 181 members of the Australian general public aged between six and 17 years

(83 male, 98 female; MAge = 11.85, SDAge = 4.00), were recruited through an Australian

research panel company to participate in an online experiment. A 2 x 2 (member

similarity: high member similarity vs low member similarity; member deviance: deviant

member vs normative member) factorial design was employed, with participants

randomly allocated to one of the four experimental conditions: low member similarity

and deviant member (n = 43), low member similarity and normative member (n = 49),

high member similarity and deviant member (n = 48), or high member similarity and

normative member (n = 41). This sample exceeded that required for a statistical power of

.80, with an a priori alpha level of .05 and estimated medium effect size (f =.25) (that is,

n > 180; G*Power).

As per the previous studies (Study 1 and Study 2), the attitude towards the brand, ‘Kraft’,

was measured (Cronbach α = .884), and the same scenario and member similarity

situation (random allocation to one member similarity condition) were presented, using

the brand ’Kraft’ where applicable. ‘Kraft’ is a well-known Australian food brand. A

manipulation for member deviance was also introduced. Specifically, participants were

given one of the following scenarios, dependent upon the condition they were randomly

allocated (deviant or normative member). As it only takes one member to impact the

dynamics of a child-oriented group (Abrams, Rutland, Palmer, and Purewal 2014), a

single deviant member was chosen for the ‘member deviance’ condition. For the ‘deviant

member’ condition participants were told: “Alex is one member of the ‘Kraft’ community

but thinks that another community is better and has heaps of fun things to do there. Alex

wants to leave the ‘Kraft’ community and join another community”. For the ‘normative

member’ condition participants were told: “Alex is one member of the ‘Kraft’ community
141
and thinks that this community is good and has heaps of fun things to do. Alex wants to

stay a member of the ‘Kraft’ community”. The scenarios for member deviance were

adopted and adapted from Abrams et al (2014). The name ‘Alex’ was chosen as this is a

non-gender specific name, and therefore less likely to produce gender-biased results

(Abrams, Rutland, Palmer, Pelletier, et al, 2014). Following the scenario all participants

were presented with the measures for desire to participate (Cronbach α = .944) and

respect towards the community (Cronbach α = .928) as per Study 1 and Study 2, using

the brand ‘Kraft’ where applicable. Participants were asked to report the degree to which

they agreed with four statements to determine if they had correctly interpreted the

manipulation of member similarity as per the previous studies (Cronbach α = .951). In

addition, participants were also asked to report the degree to which they agreed with one

statement (seven-point scale; very unloyal/loyal) to determine if they had correctly

interpreted the manipulation of member deviance (“To what extent do you think Alex is

being loyal to the ‘Kraft’ community”). Finally, basic demographic information was

obtained.

7.2 Manipulation and Confound Checks

An independent samples t-test was performed to check the manipulation of member

similarity was successful. There was a significant difference between the two member

similarity manipulation conditions, with those participants in the high similarity condition

perceiving a higher similarity than those in the low similarity condition (p < .001, MHighSim

= 4.83, SDHighSim = 1.21; MLowSim = 3.74, SDLowSim = 1.39). This result signifies the

manipulation of member similarity was successful and occurred as desired.

An independent samples t-test was performed to check the manipulation of member

deviance was successful. There was a significant difference between the two member

142
deviance manipulation conditions, with those participants in the member deviance

condition perceiving ‘Alex’ as being more disloyal than those in the normative member

condition (p < .001, MNorm = 5.28, SDNorm = 1.30; MDev = 4.00, SDDev = 1.60). This result

shows the manipulation of member deviance was successful.

To ensure no confounding effects occurred with regards to pre-formed attitudes towards

the brand ‘Kraft’, a between-subjects ANOVA was performed. Results of the ANOVA

showed that there was no significant difference across participants, with regards to

attitude towards the brand ‘Kraft’ (F(1,177) = 2.32, p = .077, η2 = .038). This result

signifies that preexisting attitudes towards the brand ‘Kraft’ did not impact the results of

the study (MHighSimDev = 4.90, SDHighSimDev = 1.46; MHighSimNorm = 5.43, SDHighSimNorm =

1.14; MLowSimDev = 4.83, SDLowSimDev = 1.43; MLowSimNorm = 4.77, SDLowSimNorm = 1.13).

7.3 Results

Hypothesis three proposed that member deviance would moderate the relationship

between member similarity and respect, as shown in Study 2. When a deviant member

was introduced to the brand community, the effect of member similarity on respect

towards the brand community was expected to attenuate. To test Hypothesis three, the

PROCESS macro bootstrapping procedure (n = 10,000, Model 7) was used (Preacher et

al, 2007). Significant results were found across the models supporting the propositions

made.

Firstly, the interaction between the two manipulated variables (member similarity x

member deviance) significantly predicted respect towards the community (β = .659, p =

.032). This respect, consequently, predicted desire to participate in the brand community

(β = .938, p = <.001). As expected, a moderated mediation effect was observed (β = .618,

95% CI = .064 to 1.025). Specifically, results showed that the mediating effect of respect

143
was found to attenuate when member deviance was present (β = .019, 95% CI = -.409 to

.449). In the deviant member and high member similarity condition, there was a

significantly lower level of respect when compared to the normative member and high

member similarity condition (p = .003, MHighSimDev = 4.28, SDHighSimDev = 1.12;

MHighSimNorm = 5.04, SDHighSimNorm = 1.10). For the low member similarity conditions,

member deviance had no significant influence on respect (p = .965, MLowSimDev = 4.26,

SDLowSimDev = 1.11; MLowSimNorm = 4.34, SDLowSimNorm = 0.76). These results all support

Hypothesis three. See Table 2 for a summary of the regression results and Figure 2 for a

graphical representation.

Table 7 - Regression Results (Study 3)

Beta t p
Dependent variable: Respect towards the community
Member Similarity (X) .348 2.277 .024
Member Deviance (W) .426 2.784 .006
XxW .659 2.156 .032

Dependent variable: Desire to Participate

Respect towards the community (M) .938 11.118 <.001
Member Similarity (X) .210 1.173 .243

To confirm that the age of participants did not moderate the observed member deviance

effect, a similar procedure to Studies 1 and 2 was applied with participants grouped again

into two age cohorts (6-11 and 12-17 years old). Specifically, the PROCESS model 3 was

employed (n = 10,000, Preacher et al, 2007), with respect as the dependent variable. A

significant moderating effect was not observed (β = .631, p = .298, R2∆ = .006),

demonstrating that the effect of member deviance is not conditional on age.

144
 Figure 3 - Effect of Similarity and Deviance on Respect

8. General Discussion and Conclusions

Despite children actively engaging in brand communities, limited research has been

conducted to understand this phenomenon. Drawing from previous research in

psychology that has evidenced the effect of similarity-attraction in relationships amongst

children (e.g. Haselager et al, 1998; Rubin, Lynch, Coplan, Rose-Krasnor, and Booth,

1994), this research introduced similarity-attraction as a factor influencing children’s

brand community participation desire. The results from the three studies suggest that a

common brand interest is not the only factor of similarity influencing participation desire;

with children shown to have a stronger desire to participate in a brand community when

they believe the members are similar to themselves (e.g. age and opinions). These results

reflect those obtained by psychologists who have demonstrated that a higher similarity

between children is more likely to promote relationship formation, and enduring

relationships (Aboud and Mendelson, 1996).

145
The effect of member similarity on participation desire was found to occur due to an

increase in the level of respect felt towards the brand community. That is, when a child

perceives community members as similar to themselves, they have a greater respect for

the collective community. These results are consistent with previous findings in

psychology that demonstrate respect highly influences similarity-attraction relationships

(Lydon et al, 1988; Singh et al, 2014). In addition, the current results provide further

evidence that respect is also important for child relationship formation, building upon

previous studies in this niche area (Haselager et al, 1998) and drawing attention to respect

in the field of marketing.

However, the mediating effect of respect is not without boundaries. The results from the

last study showed that when a deviant member was introduced into the brand community,

there was a significant decrease in the level of respect and ultimately the desire to

participate. Respect decreased, when the community members were similar, emphasizing

the impact of member deviance on respect towards the brand community. Specifically,

when a member of a brand community was disloyal, respect for the group decreased, even

though the community members were similar. It is proposed that this is due to a

misalignment between the values and attitudes held by the individual looking to join the

community and the community itself, since one member was going against the group.

Subjective group dynamics theory explains this result, by stating that deviant members

threaten the reputation of the group, and are harshly evaluated (Abrams et al, 2009). It is

suggested that the deviant member introduced into the current study threatened the

reputation and values of the group. In the conditions when there was high similarity

between the members, the reputation of the group was damaged because there was a

deviant member present. This caused the child to negatively evaluate the group, leading

to a lower perceived respect and subsequently lower participation desire.

146
The results of the three studies present interesting findings, with implications for

academics and practitioners alike. Firstly, the current research adds to the sparse literature

on children’s involvement with brand communities. The theory of similarity-attraction is

introduced to the area of brand communities, providing new evidence for child-oriented

brand community participation. For practitioners, results suggest that marketers should

emphasize member similarity when promoting brand communities to children (e.g.,

“come talk to kids just like you!”). Accentuating to children that they will be engaging

with children, who are similar to them, is likely to have a positive impact on participation

and increase the number of members in the community. Further, results indicate that

member deviance has a negative effect on participation desire. It is suggested that

marketers should monitor member interactions with the goal of identifying and

eliminating deviant behavior. Marketers should also portray that members are loyal to the

brand community, in promotion material targeting new members, in an attempt to

minimize any negative influence of deviance that could occur. Future research should be

undertaken to determine how deviant members can be identified, and how to manage

deviant behavior to aid practitioners utilizing brand communities.

While the current research has been conducted in a method that prioritizes reliability and

validity, the research is not without limitations. One main limitation is the use of a

fictitious brand community. A fictitious brand community was used, and was necessary,

to ensure the effect of member similarity could be controlled, however, due to this,

additional factors may influence the results in a real-world context. Factors such as the

number of members in the community may impact the relationships found. When there

is a large brand community (many members), there is a high likelihood that there will be

a greater range of ages, personalities and backgrounds, this may change the results

evidenced in the current study. In addition, when the community is online, a number
147
factors unique to the online environment come into play, such as avatars (Sicilia and

Palazón, 2008), that could change the results. Children in these communities may in fact

want to take on a different identity, and in essence be completely unique, not similar, to

others. Another limitation was the use of characteristics to determine member similarity,

not actual similarity. Although characteristics have been used in previous research to

examine similarity (Balmer et al., 2011; Hung, 2014), other approaches to examining

similarity between members could yield useful results. Instead of informing participants

they are similar or dissimilar to each other, future research should look at incorporating a

measure of similarity into the study.

With regards to the member deviance manipulation, the current study did not specify the

type of member or popularity of the deviant member, however, this could have interesting

implications. For example, if a popular group member or ‘leader’ is the disloyal member

the impact on respect felt towards the group may be more significant than a general

member of the brand community. Future research should explore whether the level of

popularity or status of the deviant member impacts the effects found in the current study.

Again, in relation to the variables employed, the only dependent variable used in this

study was brand community participation desire. Interesting insights may be gained from

an examination of alternative dependent variables. Future research should investigate

whether member-similarity impacts other factors in child-oriented brand communities,

for example, actual behavior in the brand community, rather than just desire.

Lastly, the current research was cross-sectional in nature, signifying no long-term

consequences could be generalized from the results. A longitudinal study, examining the

long-term effects of a brand community with characteristically similar child-members

148
could present interesting insights on the consequences of participation, such as

community loyalty, word-of-mouth recommendations and even brand loyalty.

Conflict of Interest

On behalf of all authors, the corresponding author states that there is no conflict of

interest.

149
References

Refer to Appendix 9.5.4 for Paper 4 Reference List

150
7. Additional Papers

The following section presents the additional papers that were prepared and presented by

the candidate at national and international conferences. Conference Paper 1, presented in

2016 by the candidate, is given first, Conference Paper 2 is then provided. This second

conference paper was delivered in 2017, by the candidate.

To conclude the thesis, the references and appendix are provided in Section 8 and Section

9, respectively.

151
7.1 Conference Paper 1

The following paper (Conference Paper 1) is entitled: “Understanding the complex

interplay between evaluative social identity, negative anticipated emotions and self-

esteem on a child’s commitment to brand communities”. This paper was prepared and

presented on May 26th, 2016, by the candidate, at the European Marketing Academy

Conference (EMAC), with guidance provided by the co-authors (Stacey Baxter and Alicia

Kulczynski). The conference was held in Oslo, Norway, with the theme: “Marketing in

the Age of Data”. Conference Paper 1 is closely tied to the study undertaken in Paper

Three: New Model.

152
7.1.1 Full Paper

Understanding the complex interplay between

evaluative social identity, negative anticipated

emotions and self-esteem on a child’s commitment

to brand communities

153
Abstract:

Applying social identity and subjective group dynamics theory, this research sought to

empirically examine the influence of evaluative social identity, negative anticipated

emotions and self-esteem on children’s brand community commitment. Results obtained

from on online survey of 394 Australian children aged 6 to 14 years revealed that

evaluative social identity has a conditional indirect effect on brand community

commitment. Specifically, for children with low self-esteem, the effect of evaluative

social identity on community commitment was found to be mediated by negative

anticipated emotions; however, this mediated effect was not observed for children with

high self-esteem. This paper provides a unique insight into the under-researched area of

children and brand communities, with findings useful for both academics and

practitioners.

Keywords: Communities, Children, Self-Esteem

Track: Product and Brand Management

Authors: Margurite Hook, Dr Stacey Baxter and Dr Alicia Kulczynski

154
1. Introduction

Brand communities, namely “a specialized, non-geographically bound community, based

on a set of social relationships among admirers of a brand” (Muniz Jr & O’Guinn, 2001,

p 41), have been shown to be a useful tool for marketers, facilitating the development of

consumer-brand relationships (Ouwersloot & Odekerken‐Schröder, 2008). Both online

and offline brand communities are incorporated into marketing strategies to encourage

engagement with brands and products (Bagozzi & Dholakia, 2006). Engagement with a

brand community has been shown to lead to community commitment, whereby members

continually maintain a relationship with the brand (Ouwersloot & Odekerken-Schröder,

2008), and it is these community members that are most valuable for a brand (Garbarino

& Johnson, 1999). A review of literature reveals that researchers to date have focused on

understanding adult brand community commitment (e.g., Harley Davidson and Jeep;

McAlexander, Schouten, & Koenig, 2002), with limited work undertaken to understand

children’s commitment to brand communities (Flurry, Swimberghe, & Parker, 2014). The

recent emergence of child-orientated brand communities (e.g., Mattel, My Lego Network,

Flurry et al., 2015), however, provides evidence that brands understand the potential value

of establishing a forum for which child and adolescent brand-users can engage.

Adopting a social identity and subjective group dynamics framework, this research seeks

to understand children’s brand community commitment. Evaluative social identity (ESI,

i.e., emotional attachment to a community), in particular, has been widely cited as an

important factor influencing an adults desire to engage with a brand community (Bagozzi

& Dholakia, 2006; Dholakia, Bagozzi, & Pearo, 2004). Specifically, ESI was found to be

a determinant of group behaviour in a brand community, and in addition, a stronger

relationship was evident for small group brand communities (Bagozzi & Dholakia, 2006).

Building on existing knowledge, this research seeks to understand the effect of a child’s
155
ESI on brand community commitment. As social interactions among children are known

to be influenced by self-esteem (e.g. Boulton & Smith, 1994) it is suggested that self-

esteem may impact the effect of ESI. Additionally, ESI may lead to negative anticipated

emotions when the child is blocked from the community, which in turn has an influence

on community commitment. It is argued that the effect of ESI on community commitment

will be moderated by an individual’s self-esteem, with negative anticipated emotions

from being blocked from the brand community playing a mediating role.

With children playing a critical role in the market place (consumers, influencers, future

market, Gorn & Florsheim, 1985), coupled with children’s need to engage and be part of

a group (Harter, 1999) it is argued that an understanding of the combined role of ESI,

self-esteem and negative emotional anticipation will not only provide a unique

contribution to marketing literature, but will also be of interest to brand managers, seeking

to understand child brand community members. The following paper will first discuss the

theoretical development of the research, followed by a brief outline of the method and

design approach, the results are then presented with a short discussion following to

conclude the paper.

2. Theoretical development

This research seeks to understand children’s brand community commitment, that is, their

desire and willingness to develop and maintain a relationship with the brand community

(Garbarino & Johnson, 1999). To develop the conceptual framework the theories of social

identity, subjective group dynamics and self-esteem were investigated, each of these is

discussed below.

156
2.1 Social Identity

Social identity theory is widely recognised as a key aspect of brand community

participation (e.g. Bagozzi & Dholakia, 2006) . An individual is said to have achieved

social identity when they see themselves as part of a group (or groups), and understand

the emotional significance of being in the group (Tajfel, 1978). An individual’s social

identity comprises three distinct components: cognitive, affective and evaluative

(Ellemers, Kortekaas, & Ouwerkerk, 1999). Evaluative social identity, the focus of this

research, refers to the negative and positive feelings an individual associates with being

a part of the group (Ellemers, Kortekaas, & Ouwerkerk, 1999). ESI also encompasses an

individual’s comparison of in-group and out-group choices and preferences (Bagozzi &

Dholakia, 2006).

Bennett and Sani (2004) demonstrate that from approximately five years of age, children

have the cognitive ability to not only self-categorise themselves as being part of a group

(or even multiple groups), but also acknowledge that they have identities within these

groups. However, it is not until approximately seven years of age that children experience

ESI (Bennett & Sani, 2004). The emergence of ESI begins with children differentiating

between in-group and out-group and is ultimately identified when a child positively

evaluates their in-group membership (Bennett & Sani, 2004).

The link between ESI and brand community commitment (or continuing participation

behaviour) has been well established in literature (e.g. Bagozzi & Dholakia, 2006;

Dholakia, Bagozzi, & Pearo, 2004), it is therefore hypothesised that a positive

relationship will also be observed for child brand community members (H1).

157
2.2 Subjective group dynamics

The Subjective Group Dynamics (SGD) model can be used to explain how children

interact within social groups (Abrams, Rutland, & Cameron, 2003). Whilst SGD has not

been applied in brand community literature; SGD may provide a useful framework for

explaining brand community behaviour. SGD is based on social identity theory, and holds

that members of groups have a bias towards in-group members and strive to uphold the

norms of the in-group (Abrams et al., 2003). Favouritism for in-group members and

rituals originates from a desire to increase an individual’s social identity (Tajfel & Turner,

1979). It is therefore posited that a child who sees themselves as important to the group

(i.e., part of the ‘in-group’) will feel a sense of ESI, resulting in a need to uphold in-group

norms.

In the context of a children’s brand community, maintaining contact with the brand

community regularly is considered an in-group norm (Valkenburg, Peter, & Schouten,

2006). Therefore if a child, with high ESI (in-group) is temporarily blocked from

participating in the brand community (e.g., parent limiting contact), when the child

returns to the group they may feel that they are treated differently to other in-group

members, or even seen as an out-group member due to an inability to uphold in-group

norms. In these instances it is suggested that the blocked, high ESI community member

will experience negative emotions (e.g., sadness, anger).

When children experience ESI and see themselves as part of the in-group, they form a

strong favouritism towards the group (Bennett & Sani, 2004). It is suggested that when

faced with potential removal from the in-group, a child will remain committed to the

brand community, due to a continued favouritism with the in-group, and they will attempt

to maintain their position in the in-group after being blocked from engagement. It is

158
therefore hypothesised that negative anticipated emotions will mediate the effect of

evaluative social identity on community commitment (H2).

2.3 Moderating role of self-esteem

The theory of social identity posits that an increase in in-group attitudes is positively

related to self-esteem (Corenblum, 2014), however, no research is yet to explore whether

self-esteem moderates social identity relationships. As ESI is directly associated with the

positive and negative feelings experienced when part of a group (community), it is argued

that self-esteem will moderate the effect of ESI on both brand community commitment

and negative anticipated emotions. Social and group interactions have been widely

evidenced at strengthening a child’s self-esteem (e.g. Boulton & Smith, 1994). Self-

esteem has been linked to a child’s well-being, with children possessing low self-esteem

often becoming victims of bullying (Boulton & Smith, 1994; Valkenburg et al., 2006).

Although self-esteem has largely been researched as an outcome of community

involvement, it is argued that self-esteem has a moderating role when considering the

relationship between ESI and negative anticipated emotions.

Drawing from SGD, it is suggested that the effects of not upholding the in-group norms

would be felt more strongly for members with low self-esteem, as they would not have

the confidence that they will remain in the in-group after they have not followed the

norms. Children with low self-esteem lack confidence and rely more on group and peer

support (Boulton & Smith, 1994). As a result, a child with high ESI (e.g., feels they are

needed by the group), yet low self-esteem, will experience strong negative emotions when

contact with the brand community is blocked. Additionally. they will be more concerned

that they will be rejected by the rest of the brand community since they are not upholding

an in-group norm. On the other hand, a child with high ESI and high self-esteem, will not

159
feel concerned with rejection from the brand and therefore will not feel negative

anticipated emotions when blocked from the community. It is therefore hypothesised that

self-esteem moderates the relationship between evaluative social identity and negative

anticipated emotions (H3).

Self-esteem is also suggested to moderate the relationship between evaluative social

identity and community commitment (H4). Specifically, drawing from research that has

shown that children with low self-esteem cope better when they feel they are supported

by a peer or community group (Boulton & Smith, 1994). It is argued that children with

low self-esteem will demonstrate stronger commitment to a brand community when

compared to children with higher levels of self-esteem (who do not require community

support). Figure 1 shows the proposed conceptual model and also a summary of the four

hypotheses.

Figure 1 - Conceptual model of the effects on children’s evaluative social identity in brand

communities

Self-
Esteem
Negative
Anticipated
H2 Emotions
H3
H2

H4

Evaluative Community
Social Identity Commitment
(ESI) H1

160
3. Method

Three hundred and ninety four Australian children (age 6 to 14, Mage = 9.94), who self-

identified as being part of an offline brand community, completed an online

questionnaire. Parental consent, and child assent, was obtained prior to participation. This

research focuses on offline communities formed by child admirers of the brand,

Minecraft, a contemporary brand with significant youth engagement (20% of players are

under the age of 15, and 43% of players are between the ages of 15-21, Minecraft Seeds,

2015).

3.1 Measures

All measures were drawn from existing literature. Measures of social identity (two items,

e.g., “My friends I talk about Minecraft with need me”; α = .706) and negative anticipated

emotions (nine items, e.g., “If I can’t play Minecraft and talk to my friends about

Minecraft I will feel: not sad - sad”; α = .944) were drawn from Bagozzi and Dholakia

(2006). In some instances, items were altered to reflect the language ability of participants

(e.g., ‘depressed’ was altered to state ‘sad’). Three items from Chan & Li (2010) were

used to measure brand community commitment: “I would feel sad if Minecraft wasn’t

around anymore”, “I care about the future of Minecraft” and “If I didn’t play Minecraft

for a few days, I would try and play as soon as I could” (α = .898). Finally, the seven-

item self-esteem measure employed by Harter (1982) was used. Specifically, participants

were asked to indicate what type of child they related to the most, for example, ‘some

kids feel they are a good person’ or ‘other kids do not feel they are a good person’.

Following procedures employed by Harter (1982), responses to self-esteem items were

tallied to create an overall score.

161
4. Results and discussion

To test hypothesis 1 – 4 the PROCESS macro bootstrapping procedure (n = 10,000) was

employed, specifically Model 8 (Hayes, 2013). Table 1 provides a summary of all the

relevant statistical results from the hypothesis tests. The direct effect of ESI on

community commitment was found to be significant, supporting H1. Results show that a

child’s positive evaluation of their identity within a brand community, and their self-

evaluated importance of being in the group are positively related to community

commitment. This result is consistent with the previous findings of adult-oriented brand

community research (e.g., Bagozzi & Dholakia, 2006). The interaction between ESI and

self-esteem was found to have a significant effect on negative anticipated emotions, thus

supporting H3. Specifically, results demonstrate that the effect of ESI on negative

anticipated emotions was only significant for those with lower self-esteem. This is in line

with predictions based on previous literature (Boulton & Smith, 1994) and shows that

children with low self-esteem were more affected by negative emotions when temporarily

blocked from participating in the brand community. The results could imply a stronger

need to be part of the brand community for children with low self-esteem, compared to

children with higher levels of self-esteem.

In support of H4, the interaction between ESI and self-esteem was also found to have a

significant effect on community commitment. When a child with low self-esteem has an

ESI with a brand community they will be more committed to the brand community,

however, when a child has higher self-esteem, the effect of ESI weakens. As

hypothesised, children with low self-esteem have a stronger level of commitment to the

brand community, since they rely on the group more for support. For high self-esteem

children the support of the community network is not as strong, and therefore, the

relationship between ESI and community commitment is not as strong. These results
162
reflect arguments presented in the self-esteem literature whereby low self-esteem children

cope better when they feel they have the support of a group (Boulton & Smith, 1994),

whereas those children with high self-esteem do not need the support of the group.

When testing for moderated-mediation, as is the case for testing H2, the key indicator is

the indirect effect of the interaction on the dependent variable through the mediator.

Results show that the 95% bootstrapped confidence interval for the indirect effect of the

interaction (ESI x self-esteem) did not include zero (effect = .066, 95% CI = .003 to .150),

supporting the existence of a significant indirect effect. These results support H2.

The effect of the interaction between ESI and self-esteem has an indirect effect on

community commitment, via negative anticipated emotions. That is, negative anticipated

emotions acts as the mechanism driving the effect of the interaction on community

commitment. This result is in line with SGD theory, whereby children form strong in-

group favouritism and a commitment to in-group norms. The results emphasise that

children’s commitment is mediated by anticipated negative emotions (fear of not

upholding in-group norms) when they are temporarily blocked from the community.

Children are committed to the group out of fear that if they don’t uphold the in-group

norms (and suffer negative emotions) they will be banished from the in-group.

163
Table 1 – Summary of results

Predictors β t-value p-value Hypothesis

Dependent variable: Negative anticipated emotions
Evaluative .205 4.180 < .001 Supporting H2
Social Identity
(X)

Self-Esteem (W) .093 .954 .340

XxW .233 2.301 .022 Supporting H3
Dependent variable: ‘Minecraft’ Brand Community Commitment
Negative .281 5.875 < .001 Supporting H2
Anticipated
Emotions (M)

Evaluative .483 10.195 < .001 Supporting H1

Social Identity
(X)

Self-Esteem (W) - .016 - .176 .861

XxW - .281 - 2.916 .004 Supporting H4

5. Conclusion

In conclusion, results revealed that ESI has an effect on brand community commitment

for children aged 6-14. Specifically, the relationship between social identity and

community commitment can be explained by negative anticipated emotions when

blocked from the brand community. That is, those children with ESI were more likely to

have negative anticipated emotions leading to stronger brand community commitment.

Further, this relationship is dependent upon whether a child has low or high self-esteem,

with the relationship not observed for children with higher self-esteem.

With limited research undertaken to examine children’s brand community commitment

this research provides a unique contribution to brand community, social identity and

subjective group dynamics literature. It is suggested that the findings of this research

would be of interest to academics in the field of brand communities, and also to marketing
164
practitioners, specifically those that are involved with brand communities aimed at

children. Specifically, this research highlights the importance of self-esteem, which could

further be studied for an adult demographic to compare to the results of the current

findings. Additionally, the research can be used as a starting point to research further the

antecedents and consequences of children participating in brand communities.

165
References

Refer to Appendix 9.5.5 for Conference Paper 1 Reference List

166
7.2 Conference Paper 2

Conference Paper 2 is entitled: ‘‘You’re like me.’ Children’s brand community

participation.” This conference paper was prepared and presented on December 4th, 2017,

by the candidate on at the Australian and New Zealand Marketing Academy Conference

(ANZMAC), with guidance provided by the co-authors (Stacey Baxter and Alicia

Kulczynski). The conference was held in Melbourne, Australia with the theme being:

“Marketing for Impact”. The paper is closely related to the studies undertaken in Paper

Four: Similarity Attraction.

Both reviewers of the conference paper gave positive comments about the research,

stating:

“This is a substantial research paper. Its strength is the rigour in the procedures

followed to produce the results. Based on a sound theoretical foundation, the

researchers have carefully designed their study and produced some new and

insightful outcomes.”

and

“The topic is interesting and perhaps important, given that similarity-attraction

theory, per the authors, has not been looked at in a brand community context.”

167
7.2.1 Full Paper

‘You’re like me.’ Children’s brand community

participation.

168
Abstract:

Similarity-attraction theory is applied to examine the impact of high similarity between

child brand community members (e.g. age, background, opinions) on their desire to

participate in a brand community. Australian children (n = 466) aged 6 - 17 participated

in one of three experimental studies to assess the impacts of member-similarity, respect,

and member deviance on brand community participation desire. Results suggest a

common brand interest is not sufficient for participation alone; with greater member

similarity in respect to an individual’s characteristics required for children to want to

participate. Member similarity had a significant impact on participation desire through

influencing respect towards the collective of members. In addition, when a community

member was deviant (disloyal to the community), respect, and subsequently participation

desire, declined. This study provides useful findings for practitioners, especially those

developing child-oriented brand communities and contributes to the sparse literature on

child-oriented brand communities.

Keywords: Brand Communities, Children, Similarity

169
Introduction

‘Birds of a feather flock together’, is a famous saying that typifies the psychological

concept of similarity-attraction (Byrne, 1971). Similarity-attraction research has focused

on the formation of, and interaction within interpersonal relationships (e.g. Montoya &

Horton, 2004; Singh, Tay, & Sankaran, 2016). Marketers are yet to employ similarity-

attraction to understand the interpersonal relationships formed around brand

communities. Brand communities aimed at children are emerging, and there is an

increasing need to better understand the drivers of participation for child participants

(Flurry, Swimberghe, & Parker, 2014). This research seeks to examine the effect of

similarity-attraction, member respect, and member deviance on children’s desire to

participate in a brand community. Beyond the common interest brand community

members share towards the brand itself; member-similarity, (similarity between the

members themselves (age, background, opinions, and culture)), is necessary for

influencing a child’s desire to participate in a brand community.

Background and Hypotheses

Similarity-attraction theory suggests individuals interact with those that share similar

characteristics to themselves (Gueguen, Martin, & Meineri, 2011). This situation occurs

due to the ‘effectance motive’, which is the need for a predictable, certain, and meaningful

interpretation of the world (Byrne & Clore, 1967; White, 1959). Through interacting with

others who support and validate an individual’s views (due to sharing characteristics),

this ‘effectance motive’ is strengthened and nurtured (Byrne & Clore, 1967). A brand

community is defined as a “specialised, non-geographically bound community, based on

a set of social relationships among admirers of a brand” (Muniz & O’Guinn, 2001, p 412).

Brand communities are, by definition, a set of social relationships, where members have

170
a common interest towards a brand (Cova & Pace, 2006). It is, therefore proposed that

similarity-attraction theory can be used to explain participation. Specifically, when an

individual perceives the brand community members to have similar characteristics as

themselves (high member similarity) they will have a higher brand community

participation desire (H1).

Respect, or admiration, towards an individual is a known influencer to similarity-

attraction relationships (Lydon, Jamieson, & Zanna, 1988). Individuals assess the extent

to which they respect others (termed ‘cognitive quality’) prior to commencing a

relationship (Montoya & Horton, 2004). Individuals rank the characteristics they hold as

being correct and superior (Montoya & Horton, 2004), and therefore when others portray

similar characteristics as themselves; they will be more highly admired, or respected, in

turn leading to a greater desire to establish a relationship (Singh, Chen, & Wegener,

2014). It is therefore hypothesized that respect mediates the relationship between member

similarity and participation desire, whereby an increase in respect will result in an

increase in participation desire (H2).

Subjective group dynamics theory proposes that children strive to sustain a positive social

identity within a group (Abrams, Rutland, Cameron, & Marques, 2003). Sustaining a

positive social identity is impacted by whether group norms are upheld (Abrams, Rutland,

Ferrell, & Pelletier, 2008). In the context of brand communities, loyalty to the community

is one such norm (Hook, Baxter, & Kulczynski, 2016), where a deviant member is a threat

to the group and its norms (Abrams & Rutland, 2008). Drawing from Abrams, Rutland,

Pelletier, & Ferrell (2009), it is argued that member deviance will moderate the

relationship between member similarity and respect (H3). Specifically, when a

characteristically similar member of a brand community presents deviant behaviours,

respect is reduced due to a misalignment of values and attitudes held by the group.

171
Methodology

Australian child participants aged 6 to 17 (252 male, 214 female; MAge = 11.78, SDAge =

3.66) were recruited through a research panel company to participate in one of three

experimental studies (Study 1, n = 142; Study 2, n = 143; Study 3, n = 181). Study 1 and

Study 2 participants were allocated to one of two experimental conditions: low or high

member similarity. A 2 (member similarity: low, high) x 2 (member deviance: absent,

present) factorial design was conducted for Study 3. Member similarity and member

deviance were manipulated using procedures employed by Hung (2014) and Abrams,

Rutland, Palmer, & Purewal (2014) respectively. Each study used a fictitious brand

community, based on a real-world low involvement brand (Study 1 – Tip Top, Study 2 –

Kleenex, Study 3 – Kraft). Measures were adapted from prior literature: brand community

participation desire (Dholakia, Bagozzi, & Pearo, 2004), and respect (Singh et al., 2014).

Results

All manipulations were successful and no confound effects were found. To test H1 (Study

1) an ANOVA model was estimated. Results confirmed that member similarity has a

significant effect on brand community participation desire (p = .011). Specifically, those

who were informed that brand community members were like themselves reported a

greater participation desire (Mhighsimilarity = 4.09; Mlowsimilarity = 3.34), supporting H1. Next,

to examine the mediating role of respect (H2, Study 2) the PROCESS macro

bootstrapping procedure was employed (n = 10,000, Model 4). Results demonstrated that

respect mediated the relationship between member similarity and participation desire (β

= .404, 95% CI = .066 to .742, p = .020), supporting H2. Lastly, Study 3 aimed to examine

the moderating effect of member deviance. To test H3, the PROCESS macro

bootstrapping procedure (n = 10,000, Model 7) was used. Significant results were found

172
across the models; the interaction (member similarity x member deviance) predicted

respect (β = .659, p = .032), and respect predicted participation desire (β = .938, p =

<.001). As expected, a moderated mediation effect was observed; the mediating effect of

respect was found to attenuate when member deviance was present (β = -.497, 95% CI =

-.409 to .449), supporting H3.

Discussion and implications for theory and practice

Despite children engaging in brand communities, limited research has been conducted to

understand this phenomenon. This research introduced similarity-attraction as a factor

influencing children’s brand community participation desire. Results suggest that a

common brand interest is not sufficient; children will have a stronger desire to participate

in a brand community when they believe the members are similar to themselves (e.g. age),

with this effect driven by member respect. Results of this research suggest that marketers

should emphasize member similarity when promoting brand communities to children

(e.g., “come talk to kids just like you!”). Further, results indicate that member deviance

has a negative effect on participation desire. It is suggested that marketers should monitor

member interactions with the goal of identifying and eliminating deviant behavior. Future

research should be undertaken to determine how deviant members can be identified, and

to manage deviant behaviour. In addition, future research should investigate whether

member-similarity impacts other factors in child-oriented brand communities, other than

participation desire.

173
References

Refer to Appendix 9.5.6 for Conference Paper 2 Reference List

174
8. References

ABDC. (2018a). ABDC Members · About us · Australian Business Deans Council.

Retrieved February 28, 2018, from http://www.abdc.edu.au/show_members.php
ABDC. (2018b). Four Categories of Quality · ABDC Journal Quality List · Australian
Business Deans Council. Retrieved February 28, 2018, from
http://www.abdc.edu.au/pages/.html
Aboud, F. E., & Mendelson, M. J. (1996). Determinants of friendship selection and
quality: Developmental perspectives. In The company they keep: Friendship in
childhood and adolescence. (pp. 87–112). New York, NY, US: Cambridge University
Press.
Abrams, D., & Rutland, A. (2008). The development of subjective group dynamics. In
Intergroup attitudes and relations in childhood through adulthood (pp. 47–65). USA:
Oxford University Press.
Abrams, D., Rutland, A., & Cameron, L. (2003). The development of subjective group
dynamics: Children’s judgments of normative and deviant in-group and out-group
individuals. Child Development, 74(6), 1840–1856.
Abrams, D., Rutland, A., Palmer, S. B., Pelletier, J., Ferrell, J., & Lee, S. (2014). The
role of cognitive abilities in children’s inferences about social atypicality and peer
exclusion and inclusion in intergroup contexts. British Journal of Developmental
Psychology, 32(3), 233–247. https://doi.org/10.1111/bjdp.12034
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human
Decision Processes, 50(2), 179–211.
Algesheimer, R., Dholakia, U. M., & Herrmann, A. (2005). The social influence of
brand community: Evidence from european car clubs. Journal of Marketing, 69(3), 19–
34.
Bagozzi, R. (1992). The self-regulation of attitudes, intentions, and behavior. Social
Psychology Quarterly, 178–204.
Bagozzi, R., & Dholakia, U. (2002). Intentional social action in virtual communities.
Journal of Interactive Marketing, 16(2), 2–21.
Bagozzi, R., & Dholakia, U. (2006). Antecedents and purchase consequences of
customer participation in small group brand communities. International Journal of
Research in Marketing, 23(1), 45–61.
Bagozzi, R., & Pieters, R. (1998). Goal-directed emotions. Cognition & Emotion, 12(1),
1–26.
Batista, P. D., Campiteli, M. G., Kinouchi, O., & Martinez, A. S. (2005). Universal
behavior of a research productivity index. ArXiv Preprint Physics/0510142.
Berger, C. (2008). Influence of peer group norms on aggression in early adolescence.
University of Illinois at Urbana-Champaign.

175
Betts, L. R., & Stiller, J. (2014). Centrality in children’s best friend networks: The role
of social behaviour. British Journal of Developmental Psychology, 32(1), 34–49.
https://doi.org/10.1111/bjdp.12020
Bigler, R. S., & Jones, L. C. (1997). Social categorization and the formation of
intergroup attitudes in children. Child Development, 68(3), 530–543.
https://doi.org/10.1111/1467-8624.ep9709050632
Bornmann, L., & Daniel, H. (2007). What do we know about the h index? Journal of
the Association for Information Science and Technology, 58(9), 1381–1385.
Boulton, M. J., & Smith, P. K. (1994). Bully/victim problems in middle school children:
Stability, self-perceived competence, peer perceptions and peer acceptance. British
Journal of Developmental Psychology, 12(3), 315–330.
Boyd, D. E., Clarke, T. B., & Spekman, R. E. (2014). The emergence and impact of
consumer brand empowerment in online social networks: A proposed ontology. Journal
of Brand Management, 21(6), 516–531. https://doi.org/10.1057/bm.2014.20
Bryman, A., & Bell, E. (2011). Business Research Methods (3rd ed.). Oxford University
Press.
Buckingham, D. (1993). Children talking television: The making of television literacy
(Vol. 9). Psychology Press.
Christensen, L. B., Johnson, B., Turner, L. A., & Christensen, L. B. (2011). Research
methods, design, and analysis.
Cicchetti, D., & Cohen, D. J. (2006). Developmental Psychopathology Volume 1.
Hoboken: John Wiley & Sons, Inc. Retrieved from
http://ezproxy.newcastle.edu.au/login?url=http://www.newcastle.eblib.com/EBLWeb/pa
tron?target=patron&extendedid=P 255564 0&
Clarivate Analytics. (2018). InCites International Journal of Consumer Studies.
Retrieved February 5, 2018, from https://jcr-incites-thomsonreuters-
com.ezproxy.newcastle.edu.au/JCRJournalProfileAction.action?pg=JRNLPROF&journ
alImpactFactor=1.51&year=2016&journalTitle=International%20Journal%20of%20Co
nsumer%20Studies&edition=SSCI&journal=INT%20J%20CONSUM%20STUD
Collins, P., & Mitchell, D. M. (2015). The Purchasing Power of Tomorrow: An
Analysis of the Buying Habits of Children Ages 6-15. Retrieved January 28, 2016, from
http://digitalcommons.uri.edu/cgi/viewcontent.cgi?article=1421&context=srhonorsprog
Costas, R., & Bordons, M. (2007). The h-index: Advantages, limitations and its relation
with other bibliometric indicators at the micro level. Journal of Informetrics, 1(3), 193–
203.
Cova, B., & Pace, S. (2006). Brand community of convenience products: new forms of
customer empowerment – the case “my Nutella the community.” European Journal of
Marketing, 40(9/10), 1087–1105. https://doi.org/10.1108/03090560610681023
Dholakia, U. M., Bagozzi, R. P., & Pearo, L. K. (2004). A social influence model of
consumer participation in network- and small-group-based virtual communities.
International Journal of Research in Marketing, 21(3), 241–263.
https://doi.org/10.1016/j.ijresmar.2003.12.004

176
Ellemers, N., Kortekaas, P., & Ouwerkerk, J. W. (1999). Self-categorisation,
commitment to the group and group self-esteem as related but distinct aspects of social
identity. European Journal of Social Psychology, 29(2/3), 371–389.
EMAC. (2018). Home - EMAC. Retrieved February 8, 2018, from http://www.emac-
online.org/r/default.asp?iId=FLFDIE
Engelmann, J. M., Herrmann, E., Rapp, D. J., & Tomasello, M. (2016). Young children
(sometimes) do the right thing even when their peers do not. Cognitive Development,
39, 86–92.
Evans, J. R., & Mathur, A. (2005). The value of online surveys. Internet Research,
15(2), 195–219.
Fawcett, C. A., & Markson, L. (2010). Similarity predicts liking in 3-year-old children.
Journal of Experimental Child Psychology, 105(4), 345–358.
Fink, A. (2001). Conducting research literature reviews: from paper to the Internet.
Sage Publications.
Flurry, L., A., Swimberghe, K., R., & Parker, J., M. (2014). Examining brand
communities among children and adolescents: An exploratory study. Journal of
Consumer Marketing, 31(2), 103–110.
Foote, A. (2017, January 12). #SquadGoals: 21 Outstanding Brand Communities to
Emulate. Retrieved June 11, 2018, from http://landt.co/2017/01/brand-community-
examples/
Fournier, S., & Lee, L. (2009). Getting brand communities right. Harvard Business
Review, 87(4), 105–111.
Fulgoni, G. (2014). Uses and misuses of online-survey panels in digital research:
Digging past the surface. Journal of Advertising Research, 54(2), 133–137.
Götze, E. (2002). The formation of pre-school children’s brand knowledge: An
empirical analysis into the factors of influence. Presented at the Proceedings of the 31
EMAC Annual Conference.
Hair, J. F., Gabriel, M., & Patel, V. (2014). AMOS covariance-based structural equation
modeling (CB-SEM): guidelines on its application as a marketing research tool.
Hart, C. (1998). Doing a literature review: Releasing the social science research
imagination. Sage.
Harter, S. (1993). Causes and Consequences of Low Self-Esteem in Children and
Adolescents. In R. F. Baumeister (Ed.), Self-Esteem: The puzzle of low self-regard (pp.
87–116). Springer US. Retrieved from http://link.springer.com/chapter/10.1007/978-1-
4684-8956-9_5
Harzing, A., & Van Der Wal, R. (2009). A Google Scholar h‐index for journals: An
alternative metric to measure journal impact in economics and business. Journal of the
Association for Information Science and Technology, 60(1), 41–46.
Haselager, G. J., Hartup, W. W., Lieshout, C. F., & Riksen‐Walraven, J. M. A. (1998).
Similarities between friends and nonfriends in middle childhood. Child Development,
69(4), 1198–1208.
177
Hatch, M. J., & Schultz, M. (2010). Toward a theory of brand co-creation with
implications for brand governance. Journal of Brand Management, 17(8), 590–604.
Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable
mediation, moderation, and conditional process modeling.
Hayes, A. F., Montoya, A. K., & Rockwood, N. J. (2017). The analysis of mechanisms
and their contingencies: PROCESS versus structural equation modeling. Australasian
Marketing Journal (AMJ), 25(1), 76–81.
Hickman, T. M., & Ward, J. C. (2013). Implications of brand communities for rival
brands: Negative brand ratings, negative stereotyping of their consumers and negative
word-of-mouth. Journal of Brand Management, 20(6), 501–517.
https://doi.org/10.1057/bm.2012.57
Hill, M. (2005). Ethical considerations in researching children’s experiences.
Researching Children’s Experience, 61–86.
Hirsch, J. E. (2005). An index to quantify an individual’s scientific research output.
Proceedings of the National Academy of Sciences of the United States of America,
102(46), 16569.
Hung, H. (2014). Attachment, identification, and loyalty: Examining mediating
mechanisms across brand and brand community contexts. Journal of Brand
Management, 21(7–8), 594–614.
Hunter, S. C., Fox, C. L., & Jones, S. E. (2016). Humor style similarity and difference
in friendship dyads. Journal of Adolescence, 46, 30–37.
Ji, M. (2002). Children’s relationships with brands: “True love” or “one-night” stand?
Psychology & Marketing, 19(4), 369–387.
John Wiley & Sons. (2018). International Journal of Consumer Studies. Retrieved
February 5, 2018, from
http://onlinelibrary.wiley.com.ezproxy.newcastle.edu.au/journal/10.1111/(ISSN)1470-
6431/homepage/ProductInformation.html
Kim, J. W., Choi, J., Qualls, W., & Han, K. (2008). It takes a marketplace community to
raise brand commitment: The role of online communities. Journal of Marketing
Management, 24(3/4), 409–431.
Laroche, M., Habibi, M. R., Richard, M.-O., & Sankaranarayanan, R. (2012). The
effects of social media based brand communities on brand community markers, value
creation practices, brand trust and brand loyalty. Computers in Human Behavior, 28(5),
1755–1767. https://doi.org/10.1016/j.chb.2012.04.016
Lee, H. J., Lee, D., Taylor, C. R., & Lee, J. (2011). Do online brand communities help
build and maintain relationships with consumers? A network theory approach. Journal
of Brand Management, 19(3), 213–227.
Lee, N. (2017). Lego launches a safe social network for kids to share their creations.
Retrieved August 31, 2017, from https://www.engadget.com/2017/01/31/lego-life/
Lego. (2016). City. Retrieved February 23, 2016, from
https://community.lego.com/t5/CITY/bd-p/1512

178
LEGO System A/S. (2017). LEGO® Life (Version 1.2.8) [Android 4.1 and up]. LEGO
System A/S. Retrieved from
https://play.google.com/store/apps/details?id=com.lego.common.legolife&hl=en
Marques, J., Abrams, D., Paez, D., & Martinez-Taboada, C. (1998). The role of
categorization and in-group norms in judgments of groups and their members. Journal
of Personality and Social Psychology, 75(4), 976.
Mattel. (2016). My World - The Most Pink-tastic World for Girls Online. Retrieved
February 23, 2016, from http://kids.barbie.com/en-us/my-world
McAlexander, J. H., Schouten, J. W., & Koenig, H. F. (2002). Building brand
community. Journal of Marketing, 66(1), 38–54.
McNeal, J. (1998). Taping the three kids’ markets. American Demographics, 20(4), 36–
41.
McWilliam, G. (2000). Building stronger brands through online communities. Sloan
Management Review, 41(3), 43–54.
Mertens, D. M. (2014). Research and evaluation in education and psychology:
Integrating diversity with quantitative, qualitative, and mixed methods. Sage
publications.
Mind Candy. (2016). Moshi Monsters - Sign In. Retrieved February 23, 2016, from
http://www.moshimonsters.com/login/forums
Mingers, J., Macri, F., & Petrovici, D. (2012). Using the h-index to measure the quality
of journals in the field of business and management. Information Processing &
Management, 48(2), 234–241.
Montoya, R. M., & Horton, R. S. (2004). On the importance of cognitive evaluation as a
determinant of interpersonal attraction. Journal of Personality and Social Psychology,
86(5), 696.
Morse, J. M. (2016). Mixed method design: Principles and procedures (Vol. 4).
Routledge.
Mulrow, C. D. (1994). Systematic reviews: rationale for systematic reviews. Bmj,
309(6954), 597–599.
Muniz Jr, A. M., & O’Guinn, T. C. (2001). Brand community. Journal of Consumer
Research, 27(4), 412–432.
Ouwersloot, H., & Odekerken‐Schröder, G. (2008). Who’s who in brand communities –
and why? European Journal of Marketing, 42(5/6), 571–585.
https://doi.org/10.1108/03090560810862516
Piaget, J. (1972). Intellectual evolution from adolescence to adulthood. Human
Development, 15(1), 1–12.
Pluye, P., Hong, Q. N., Bush, P. L., & Vedel, I. (2016). Opening-up the definition of
systematic literature review: the plurality of worldviews, methodologies and methods
for reviews and syntheses. Journal of Clinical Epidemiology, 73, 2–5.
https://doi.org/10.1016/j.jclinepi.2015.08.033

179
Quinn, M., & Devasagayam, R. (2005). Building brand community among ethnic
diaspora in the USA: Strategic implications for marketers. Journal of Brand
Management, 13(2), 101–114.
Reveladvertising. (2017). Generational Marketing: Kids and Their Buying Power.
Retrieved February 26, 2018, from
http://www.reveladvertising.com/blog/2017/4/25/generational-marketing-kids-and-
their-buying-power
Roulin, N. (2015). Don’t throw the baby out with the bathwater: Comparing data quality
of crowdsourcing, online panels, and student samples. Industrial and Organizational
Psychology, 8(2), 190–196.
Scimago Lab. (2018). Journal of Brand Management. Retrieved February 5, 2018, from
http://www.scimagojr.com/journalsearch.php?q=17700156769&tip=sid&clean=0
Sicilia, M., & Palazón, M. (2008). Brand communities on the internet: A case study of
Coca-Cola’s Spanish virtual community. Corporate Communications: An International
Journal, 13(3), 255–270.
Singh, R., Tay, Y. Y., & Sankaran, K. (2016). Causal role of trust in interpersonal
attraction from attitude similarity. Journal of Social and Personal Relationships, 34(5),
717–731. https://doi.org/10.1177/0265407516656826
Springer. (2018). Journal of Brand Management - A Palgrave journal. Retrieved
February 5, 2018, from
http://www.springer.com/business+%26+management/journal/41262
Steenkamp, J.-B. E., & Baumgartner, H. (2000). On the use of structural equation
models for marketing modeling. International Journal of Research in Marketing, 17(2–
3), 195–202.
Thumm, J. (2015, October 12). Sephora Learns the Importance of Online Brand
Communities. Retrieved March 2, 2016, from
http://www.powerretail.com.au/pureplay/sephora-online-brand-communities/
Tranfield, D., Denyer, D., & Smart, P. (2003). Towards a methodology for developing
evidence‐informed management knowledge by means of systematic review. British
Journal of Management, 14(3), 207–222.
Ungerleider, N. (2014, September 22). How Sephora Discovered That Lurkers Are Also
Its “Superfans.” Retrieved March 2, 2016, from
http://www.fastcolabs.com/3035236/how-sephora-discovered-that-lurkers-are-also-its-
superfans
University of Newcastle. (2014). Participant recruitment for human research guidelines
- 001041. Retrieved from https://www.newcastle.edu.au/about-uon/governance-and-
leadership/policy-library/document?RecordNumber=D14_50379
Voinea, M., & Damian, M. (2014). The role of socio-emotional development in early
childhood. Journal Plus Education, 388–392.
Wallace, E., Buil, I., & de Chernatony, L. (2014). Consumer engagement with self-
expressive brands: brand love and WOM outcomes. The Journal of Product and Brand
Management, 23(1), 33–42. https://doi.org/10.1108/JPBM-06-2013-0326

180
Wang, Y. J., Butt, O. J., & Wei, J. (2011). My identity is my membership: A
longitudinal explanation of online brand community members’ behavioral
characteristics. Journal of Brand Management, 19(1), 45–56.
Wohlwill, J. F. (1962). From perception to inference: A dimension of cognitive
development. Monographs of the Society for Research in Child Development, 87–112.
Woisetschläger, D. M., Hartleb, V., & Blut, M. (2008). How to make brand
communities work: Antecedents and consequences of consumer participation. Journal
of Relationship Marketing, 7(3), 237–256. https://doi.org/10.1080/15332660802409605

181
9. Appendices

9.1 Human Ethics Approval

9.1.1 Approval for Studies 2 and 3 (Papers Two and Three)

182
9.1.2 Approval for Studies 4, 5 and 6 (Paper Four)

183
9.2 Participant Information Sheets

9.2.1 Paper Two and Paper Three: Participant Information Sheet

184
185
9.2.2 Paper Four: Participant Information Sheet

186
187
9.3 Questionnaires Employed

9.3.1 Paper Two: Replication and Extension

9.3.1.1 Version 1 (Minecraft Brand Community)

Note: Questionnaire was formatted by Research Now

UNDERSTANDING CHILDREN’S BEHAVIOUR IN BRAND COMMUNITIES
Your child has been invited to participant in a research project that attempts to understand
why children participate in brand communities (a community based on relationships
surrounding a brand) and the marketing-oriented consequences of their involvement.
Whilst it is unlikely that results will be used by marketers, the results of this research will
be published, and as a result, others (including marketers/advertisers/brand managers)
will be free to use the research conclusions. Before consenting to your child’s
participation please discuss the project with your child and confirm the following by
checking the box next to each statement:

 I understand that the project will be conducted as described in the Information

Statement; a copy of which I have saved;
 I understand that due to the anonymous nature of the research, my child’s data
cannot be withdrawn once the completed questionnaire has been submitted;
 I understand that the results of the research will be published;
 If my child cannot read independently I agree to read each question to my child;
 I understand this is not a test, and agree to report my child’s responses accurately;
 I understand that the questionnaire contains forty-six questions and will take
approximately 15 minutes for my child to complete;
 I understand that my child can stop completing the questionnaire at any time prior
to submission, and does not have to give any reason for stopping;
 My child is happy to complete this questionnaire;

 I understand that by checking the above statements, I am providing my informed

consent for my child to participate in this research.
[If all checked move to Q1]

188
1. My:
 Son
 Daughter
will be completing the questionnaire;
2. My son/daughter [linked from Q1] who will be completing this questionnaire is
years of age and is in year at school.

Please read all parts of the questionnaire to your child. Please remember this is not a test,
and to ensure the accuracy of findings please enter your son/daughter’s [linked from Q1]
response for each question.
3. Do you play Minecraft?
 Yes
 No

4. Do you play Minecraft online with other people?

 Yes
 No
(if yes then continue with this survey version, if no then get directed to “non-
branded community” version survey)
5. I think playing Minecraft and talking to people on Minecraft:

Is Silly 1 2 3 4 5 Is Smart
Will Hurt Me 1 2 3 4 5 Is good for me
Is Bad 1 2 3 4 5 Is Good
Will make bad 1 2 3 4 5 Will make good
things happen things happen

6. If I play Minecraft and talk to people on Minecraft, I will feel:

Not at all
A Lot
Excited 1 2 3 4 5
Happy 1 2 3 4 5
Great 1 2 3 4 5
Proud 1 2 3 4 5
Brave 1 2 3 4 5
Comfortable 1 2 3 4 5

189
7. If I can’t play Minecraft and talk to people on Minecraft, I will feel:

Not at all A Lot

Angry 1 2 3 4 5
Like I have done 1 2 3 4 5
something bad
Afraid 1 2 3 4 5
Annoyed 1 2 3 4 5
Sad 1 2 3 4 5
Worried 1 2 3 4 5
Guilty 1 2 3 4 5
Uncomfortable 1 2 3 4 5
Nervous 1 2 3 4 5

8. Please express how Mum, Dad, your brother/s, your sister/s and your friends feel about
you playing Minecraft and talking to people on Minecraft.
My Mum, Dad, brother/s, sister/s and friends think I:

Should not play Minecraft and talk to

people on Minecraft 1 2 3 4 5 Should play Minecraft and talk to
people on Minecraft
Think its not ok that I play Minecraft Think its ok that I play Minecraft
and talk to people on Minecraft 1 2 3 4 5 and talk to people on Minecraft

9. How much are you allowed to play Minecraft and talk to people on Minecraft?

Never 1 2 3 4 5 Whenever I want

10. To play Minecraft and talk to people on Minecraft is:

Easy 1 2 3 4 5 Hard

11. When playing Minecraft, I act the same as other people I play Minecraft with:

Not at all 1 2 3 4 5 A Lot

190
12. I think I am the same as the people I play Minecraft with:

Not at all 1 2 3 4 5 A Lot

13. How much do you like the people you play Minecraft with?

Not at all 1 2 3 4 5 A Lot

14. I feel I fit in with the people I play Minecraft with:

Not at all 1 2 3 4 5 A Lot

15. The people that I play Minecraft with, need me:

Not at all 1 2 3 4 5 A Lot

16. I am the leader of the people I play Minecraft with:

Not at all 1 2 3 4 5 Very Much

17. I need to play Minecraft and talk to people on Minecraft

Maybe Yes, I do
No, I don’t
1 2 3 4 5

18. How much do you need to play Minecraft and talk to people on Minecraft:

I don’t I need to a lot

need to at 1 2 3 4 5
all

19. I want to play Minecraft and talk to people on Minecraft in the next couple of days:

I don’t I want to very

want to at 1 2 3 4 5 much
all

191
20. I will plan to play Minecraft and talk to people on Minecraft in the next couple of
days:

Not at all 1 2 3 4 5 A Lot

21. The people I play and talk to on Minecraft plan to play and talk in the next couple of
days:

Not at all 1 2 3 4 5 A Lot

22. If Minecraft was a person, do you think that you would be the same type of person?

Not at all 1 2 3 4 5 Very Much

23. How many times do you think you will play Minecraft and talk with your friends
about Minecraft in the next couple of days? Please click on only one box.
 None
 1-3 times
 4-6 times
 7-9 times
 10 times or more

24. How many times do you want to play Minecraft and talk with your friends about
Minecraft in the next couple of days? Please click on only one box.
 None
 1-3 times
 4-6 times
 7-9 times
 10 times or more

25. How many Minecraft toys do you have? Please click on only one box.
 None
 1-3 toys
 4-6 toys
 7-9 toys
 10 toys or more

192
26. How many Minecraft books do you have? Please click on only one box.
 None
 1-3 books
 4-6 books
 7-9 books
 10 books or more
27. Ask an adult how much money they spent on Minecraft toys and books for you this
year. Please click on only one box.
 $0
 $1 - $49
 $50 - $99
 $100 - $149
 $150 or more

28. How often do you look at Minecraft toys and Minecraft books at the shops? Please
click on only one box.

I look at it
I don’t look at
every time in
it in the shops
the shops
1 2 3 4 5

29. I am a:
 Girl
 Boy

30. I am years old

193
9.2.1.2 Version 2 (Non-Branded Community)

Note: Questionnaire was formatted by Research Now

UNDERSTANDING CHILDREN’S BEHAVIOUR IN BRAND COMMUNITIES

Your child has been invited to participant in a research project that attempts to understand
why children participate in brand communities (a community based on relationships
surrounding a brand) and the marketing-oriented consequences of their involvement.
Whilst it is unlikely that results will be used by marketers, the results of this research will
be published, and as a result, others (including marketers/advertisers/brand managers)
will be free to use the research conclusions. Before consenting to your child’s
participation please discuss the project with your child and confirm the following by
checking the box next to each statement:

 I understand that the project will be conducted as described in the Information

Statement; a copy of which I have saved;
 I understand that due to the anonymous nature of the research, my child’s data
cannot be withdrawn once the completed questionnaire has been submitted;
 I understand that the results of the research will be published;
 If my child cannot read independently I agree to read each question to my child;
 I understand this is not a test, and agree to report my child’s responses accurately;
 I understand that the questionnaire contains forty-six questions and will take
approximately 15 minutes for my child to complete;
 I understand that my child can stop completing the questionnaire at any time prior
to submission, and does not have to give any reason for stopping;
 My child is happy to complete this questionnaire;

 I understand that by checking the above statements, I am providing my informed

consent for my child to participate in this research.
[If all checked move to Q1]
1. My:
 Son
 Daughter
will be completing the questionnaire;
2. My son/daughter [linked from Q1] who will be completing this questionnaire is
years of age and is in year at school.
194
Please read all parts of the questionnaire to your child. Please remember this is not a test,
and to ensure the accuracy of findings please enter your son/daughter’s [linked from Q1]
response for each question.

3. Do you play computer games?

 Yes
 No

4. Do you play computer games online with other people?

 Yes
 No
(if no then finish questionnaire)

5. I think playing computer games and talking to my friends about computer games:

Is Silly 1 2 3 4 5 Is Smart
Will Hurt 1 2 3 4 5 Is good for
Me me
Is Bad 1 2 3 4 5 Is Good
Will 1 2 3 4 5 Will make
make bad good
things things
happen happen

6. If I play computer games and talk to my friends about computer games I will feel:

Not at all
A Lot
Excited 1 2 3 4 5
Happy 1 2 3 4 5
Great 1 2 3 4 5
Proud 1 2 3 4 5
Brave 1 2 3 4 5
Comfortable 1 2 3 4 5

195
7. If I can’t play computer games and talk to my friends about computer games I will feel:

Not at
A Lot
all
Angry 1 2 3 4 5
Like I have done 1 2 3 4 5
something bad
Afraid 1 2 3 4 5
Annoyed 1 2 3 4 5
Sad 1 2 3 4 5
Worried 1 2 3 4 5
Guilty 1 2 3 4 5
Uncomfortable 1 2 3 4 5
Nervous 1 2 3 4 5

8. Please express how Mum, Dad, your brother/s, your sister/s and your friends feel about
you playing computer games and talking to your friends about computer games.

My Mum, Dad, brother/s, sister/s and friends think I:

Should not play computer games and Should play computer games and
1 2 3 4 5
talk to my friends about computer talk to my friends about computer
games games

Think its not ok that I play computer Think its ok that I play computer
games and talk to my friends about 1 2 3 4 5 games and talk to my friends
computer games about computer games

9. How much are you allowed to play computer games and talk to your friends about
computer games?

1 2 3 4 5 Whenever I
Never
want

10. To play computer games and talk with my friends about computer games is:

Easy 1 2 3 4 5 Hard

196
11. When playing computer games, I act the same as my friends I talk to computer games
about:

Not at all 1 2 3 4 5 A Lot

12. I think I am the same as my friends I talk to computer games about:

Not at all 1 2 3 4 5 A Lot

13. How much do you like your friends that you talk to computer games about?

Not at all 1 2 3 4 5 A Lot

14. How much do you feel a part of your group of friends that you talk to computer games
about?

Not at all 1 2 3 4 5 A Lot

15. The friends I talk about computer games to, need me:

Not at all 1 2 3 4 5 A Lot

16. I am the leader of the friends I talk about computer games with:

Not at all 1 2 3 4 5 Very Much

17. I need to play computer games and talk to my friends about computer games

Maybe Yes, I do
No, I don’t
1 2 3 4 5

18. How much do you need to play computer games and talk to your friends about
computer games:

I don’t need I need to a lot

1 2 3 4 5
to at all

197
19. I want to play computer games and talk to my friends about computer games in the
next couple of days:

I don’t I want to very

want to at 1 2 3 4 5 much
all

20. I will plan to play computer games and talk to my friends about computer games in
the next couple of days

Not at all 1 2 3 4 5 A Lot

21. My group of friends plan to talk about computer games together in the next couple of
days

Not at all 1 2 3 4 5 A Lot

22. If computer games was a person, do you think that you would be the same type of
person?

Not at all 1 2 3 4 5 Very Much

23. How many times do you think you will play computer games and talk with your
friends about computer games in the next couple of days? Please click on only one box.
 None
 1-3 times
 4-6 times
 7-9 times
 10 times or more

24. How many times do you want to play computer games and talk with your friends
about computer games in the next couple of days? Please click on only one box.
 None
 1-3 times
 4-6 times
 7-9 times
 10 times or more

198
25. How many computer games toys do you have? Please click on only one box.
 None
 1-3 toys
 4-6 toys
 7-9 toys
 10 toys or more

26. How many computer games books do you have? Please click on only one box.
 None
 1-3 books
 4-6 books
 7-9 books
 10 books or more

27. Ask an adult how much money they spent on computer games toys and books for
you this year. Please click on only one box.
 $0
 $1 - $49
 $50 - $99
 $100 - $149
 $150 or more

28. How often do you look at computer games toys and computer games books at the
shops? Please click on only one box.

I look at it
I don’t look at
every time in
it in the shops
the shops
1 2 3 4 5

29. I am a:
 Girl
 Boy

30. I am years old

199
9.3.2 Paper Three: New Model

Note: Questionnaire was formatted by Research Now

UNDERSTANDING CHILDREN’S BEHAVIOUR IN BRAND COMMUNITIES

Your child has been invited to participant in a research project that attempts to understand
why children participate in brand communities (a community based on relationships
surrounding a brand) and the marketing-oriented consequences of their involvement.
Whilst it is unlikely that results will be used by marketers, the results of this research will
be published, and as a result, others (including marketers/advertisers/brand managers)
will be free to use the research conclusions. Before consenting to your child’s
participation please discuss the project with your child and confirm the following by
checking the box next to each statement:

 I understand that the project will be conducted as described in the Information

Statement; a copy of which I have saved;
 I understand that due to the anonymous nature of the research, my child’s data
cannot be withdrawn once the completed questionnaire has been submitted;
 I understand that the results of the research will be published;
 If my child cannot read independently I agree to read each question to my child;
 I understand this is not a test, and agree to report my child’s responses accurately;
 I understand that the questionnaire contains forty-six questions and will take
approximately 15 minutes for my child to complete;
 I understand that my child can stop completing the questionnaire at any time prior
to submission, and does not have to give any reason for stopping;
 My child is happy to complete this questionnaire;
 I understand that by checking the above statements, I am providing my informed
consent for my child to participate in this research.

[If all checked move to Q1]

1. My:
 Son
 Daughter
will be completing the questionnaire;
2. My son/daughter [linked from Q1] who will be completing this questionnaire is
years of age and is in year at school.
200
Please read all parts of the questionnaire to your child. Please remember this is not a test,
and to ensure the accuracy of findings please enter your son/daughter’s [linked from Q1]
response for each question.

3. Do you play Minecraft?

 Yes
 No

4. Do you play Minecraft online with other people?

 Yes
 No

5. The friends I talk about Minecraft to, need me:

Not at all 1 2 3 4 5 A Lot

6. I am the leader of the friends I talk about Minecraft with:

Not at all 1 2 3 4 5 Very Much

7. If I talk to my friends about Minecraft, I will feel:

Not at all
A Lot
Excited 1 2 3 4 5
Happy 1 2 3 4 5
Great 1 2 3 4 5
Proud 1 2 3 4 5
Brave 1 2 3 4 5
Comfortable 1 2 3 4 5

8. If I can’t talk to my friends about Minecraft, I will feel:

Not at all A Lot

Angry 1 2 3 4 5
Like I have done 1 2 3 4 5
something bad
Afraid 1 2 3 4 5
Annoyed 1 2 3 4 5
Sad 1 2 3 4 5
Worried 1 2 3 4 5
Guilty 1 2 3 4 5
Uncomfortable 1 2 3 4 5
Nervous 1 2 3 4 5

201
9. I would feel sad if Minecraft wasn’t around anymore:

Strongly 1 2 3 4 5 Strongly
Disagree Agree

10. If I didn’t play Minecraft for a few days, I would try and play Minecraft as soon as I
could:

Strongly 1 2 3 4 5 Strongly
Disagree Agree

11. I care about the future of Minecraft:

Strongly 1 2 3 4 5 Strongly
Disagree Agree

12. I always want to tell people that Minecraft is awesome

Strongly 1 2 3 4 5 Strongly
Disagree Agree

13. If my friends or family are looking for a game to play, I would tell them to play
Minecraft

Strongly 1 2 3 4 5 Strongly
Disagree Agree

14. To talk with my friends about Minecraft is:

Easy 1 2 3 4 5 Hard

202
15. Click the box that you feel sounds like you, for each different situation. Choose two
boxes per line, one box on the left and one on the right.

Really true Sort of Some kids are but other kids are Really true Sort of true
for me true for me happy the not happy the for me for me
  way they are way they are  
Really true Sort of Some kids are but Other kids do Really true Sort of true
for me true for me feel good not feel good for me for me
   
Really true Sort of Some kids are but Other kids are Really true Sort of true
for me true for me sure they are sure they are for me for me
  doing the not doing the  
right thing right thing
Really true Sort of Some kids but Other kids do Really true Sort of true
for me true for me feel they are a not feel they for me for me
  good person are a good  
person
Really true Sort of Some kids but Other kids do Really true Sort of true
for me true for me want to stay not want to for me for me
  the same stay the same  
Really true Sort of Some kids do but Other kids do Really true Sort of true
for me true for me things fine not do things for me for me
  fine  

16. I am a:
 Girl
 Boy

17. I am years old

203
9.3.3 Paper Four: Similarity Attraction

Note: Questionnaire was formatted by Research Now

STUDY ONE
BRAND COMMUNITY MEMBER SIMILARITY AND THE MODERATING
ROLE OF MEMBER DEVIANCE
Your child has been invited to participant in a research project that attempts to understand
why children participate in brand communities (a community based on relationships
surrounding a brand) and the marketing-oriented consequences of their involvement. The
results of this research will be published, and as a result, others (including
marketers/advertisers/brand managers) will be free to use the research conclusions.
Before consenting to your child’s participation please discuss the project with your child
and confirm the following by checking the box next to each statement:

 I understand that the project will be conducted as described in the Information

Statement; a copy of which I have saved;
 I understand that due to the anonymous nature of the research, my child’s data
cannot be withdrawn once the completed questionnaire has been submitted;
 I understand that the results of the research will be published;
 If my child cannot read independently I agree to read each question to my child;
 I understand this is not a test, and agree to report my child’s responses accurately;
 I understand that the questionnaire contains nine questions and will take
approximately 10 minutes for my child to complete;
 I understand that my child can stop completing the questionnaire at any time prior
to submission, and does not have to give any reason for stopping;
 My child is willing to complete this questionnaire;

 I understand that by checking the above statements, I am providing my informed

consent for my child to participate in this research.
[If all checked move to Q1]
1. My:
 Son
 Daughter
will be completing the questionnaire;

204
2. My son/daughter [linked from Q1] who will be completing this questionnaire is
years of age and is in year at school.

Please read all parts of the questionnaire to your child. Please remember this is not a
test, and to ensure the accuracy of findings please enter your son/daughter’s [linked
from Q1] response for each question.

3. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
My attitude towards the bread brand ‘Tip Top’ is:
Bad 1 2 3 4 5 6 7 Good
Dislike very 1 2 3 4 5 6 7 Like very
much much
Pleasant 1 2 3 4 5 6 7 Unpleasant
Unfavourable 1 2 3 4 5 6 7 Favourable
Poor Quality 1 2 3 4 5 6 7 High Quality

The rest of the questionnaire will relate to the bread brand ‘Tip Top’, please remember
this brand for the rest of the questionnaire.
Imagine you thinking of joining the ‘Tip Top’ community. There are lots of other
members in the ‘Tip Top’ community. The other members of this community are the
same (different) age as you, are the same (different) culture, have the same (different)
opinions and have the same (different) background as you.

4. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
I desire to interact with the ‘Tip Top’ community sometime during the next 2 weeks:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

5. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
My desire for interacting with the ‘Tip Top’ community can be described as:

Very Weak 1 2 3 4 5 6 7 Very Strong

Desire Desire

204
6. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
I want to interact together with the ‘Tip Top’ community during the next 2 weeks:

Does not 1 2 3 4 5 6 7 Describes

describe me very
me at all much

7. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
The other members of the brand community are similar to me in terms of personality:

Not at all 1 2 3 4 5 6 7 A Lot

8. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
The other members of the brand community are similar to me in terms of age:

Not at all 1 2 3 4 5 6 7 A Lot

9. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
The other members of the brand community are similar to me in terms of values:

Not at all 1 2 3 4 5 6 7 A Lot

10. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
The other members of the brand community are similar to me in terms of background:

Not at all 1 2 3 4 5 6 7 A Lot

206
 STUDY TWO

BRAND COMMUNITY MEMBER SIMILARITY AND THE MODERATING

ROLE OF MEMBER DEVIANCE

Your child has been invited to participant in a research project that attempts to understand
why children participate in brand communities (a community based on relationships
surrounding a brand) and the marketing-oriented consequences of their involvement. The
results of this research will be published, and as a result, others (including
marketers/advertisers/brand managers) will be free to use the research conclusions.
Before consenting to your child’s participation please discuss the project with your child
and confirm the following by checking the box next to each statement:

 I understand that the project will be conducted as described in the Information

Statement; a copy of which I have saved;
 I understand that due to the anonymous nature of the research, my child’s data
cannot be withdrawn once the completed questionnaire has been submitted;
 I understand that the results of the research will be published;
 If my child cannot read independently I agree to read each question to my child;
 I understand this is not a test, and agree to report my child’s responses accurately;
 I understand that the questionnaire contains fourteen questions and will take
approximately 15 minutes for my child to complete;
 I understand that my child can stop completing the questionnaire at any time prior
to submission, and does not have to give any reason for stopping;
 My child is willing to complete this questionnaire;

 I understand that by checking the above statements, I am providing my informed

consent for my child to participate in this research.

[If all checked move to Q1]

1. My:
 Son
 Daughter
will be completing the questionnaire;

207
2. My son/daughter [linked from Q1] who will be completing this questionnaire is
years of age and is in year at school.

Please read all parts of the questionnaire to your child. Please remember this is not a
test, and to ensure the accuracy of findings please enter your son/daughter’s [linked
from Q1] response for each question.
3. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
My attitude towards the tissue brand ‘Kleenex’ is:
Bad 1 2 3 4 5 6 7 Good
Dislike very 1 2 3 4 5 6 7 Like very
much much
Pleasant 1 2 3 4 5 6 7 Unpleasant
Unfavourable 1 2 3 4 5 6 7 Favourable
Poor Quality 1 2 3 4 5 6 7 High Quality

The rest of the questionnaire will relate to the tissue brand ‘Kleenex’, please remember
this brand for the rest of the questionnaire.
Imagine you thinking of joining the ‘Kleenex’ community. There are lots of other
members in the ‘Kleenex’ community. The other members of this community are the
same (different) age as you, are the same (different) culture, have the same (different)
opinions and have the same (different) background as you.

4. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
The members of the ‘Kleenex’ community would make good leaders:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

5. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
The members of the ‘Kleenex’ community will achieve all of their goals:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

208
6. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
The members of the ‘Kleenex’ community are probably good at everything they do:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

7. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
The members of the ‘Kleenex’ community will probably be successful in life:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

8. Please show how much you agree with the following statements from 1 to 7, by
clicking on one number.
I desire to interact with the ‘Kleenex’ community sometime during the next 2 weeks:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

9. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
My desire for interacting with the ‘Kleenex’ community can be described as:

Very 1 2 3 4 5 6 7 Very
Weak Strong
Desire Desire

10. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
I want to interact together with the ‘Kleenex’ community during the next 2 weeks:

Does not 1 2 3 4 5 6 7 Describes

describe me very
me at all much

209
11. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
The other members of the brand community are similar to me in terms of personality:

Not at all 1 2 3 4 5 6 7 A Lot

12. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
The other members of the brand community are similar to me in terms of age:

Not at all 1 2 3 4 5 6 7 A Lot

13. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
The other members of the brand community are similar to me in terms of values:

Not at all 1 2 3 4 5 6 7 A Lot

14. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
The other members of the brand community are similar to me in terms of background:

Not at all 1 2 3 4 5 6 7 A Lot

210
 STUDY THREE

BRAND COMMUNITY MEMBER SIMILARITY AND THE MODERATING

ROLE OF MEMBER DEVIANCE

Your child has been invited to participant in a research project that attempts to understand
why children participate in brand communities (a community based on relationships
surrounding a brand) and the marketing-oriented consequences of their involvement. The
results of this research will be published, and as a result, others (including
marketers/advertisers/brand managers) will be free to use the research conclusions.
Before consenting to your child’s participation please discuss the project with your child
and confirm the following by checking the box next to each statement:

 I understand that the project will be conducted as described in the Information

Statement; a copy of which I have saved;
 I understand that due to the anonymous nature of the research, my child’s data
cannot be withdrawn once the completed questionnaire has been submitted;
 I understand that the results of the research will be published;
 If my child cannot read independently I agree to read each question to my child;
 I understand this is not a test, and agree to report my child’s responses accurately;
 I understand that the questionnaire contains fifteen questions and will take
approximately 15 minutes for my child to complete;
 I understand that my child can stop completing the questionnaire at any time prior
to submission, and does not have to give any reason for stopping;
 My child is willing to complete this questionnaire;

 I understand that by checking the above statements, I am providing my informed

consent for my child to participate in this research.

[If all checked move to Q1]

1. My:
 Son
 Daughter
will be completing the questionnaire;

211
2. My son/daughter [linked from Q1] who will be completing this questionnaire is
years of age and is in year at school.

Please read all parts of the questionnaire to your child. Please remember this is not a
test, and to ensure the accuracy of findings please enter your son/daughter’s [linked
from Q1] response for each question.

3. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
My attitude towards the food brand ‘Kraft’ is:

Bad 1 2 3 4 5 6 7 Good
Dislike very 1 2 3 4 5 6 7 Like very
much much
Pleasant 1 2 3 4 5 6 7 Unpleasant
Unfavourable 1 2 3 4 5 6 7 Favourable
Poor Quality 1 2 3 4 5 6 7 High Quality

The rest of the questionnaire will relate to the food brand ‘Kraft’, please remember this
brand for the rest of the questionnaire.
Imagine you thinking of joining the ‘Kraft’ community. There are lots of other members
in the ‘Kraft’ community. The other members of this community are the same (different)
age as you, are the same (different) culture, have the same (different) opinions and have
the same (different) background as you.
Alex is one member of the ‘Kraft’ community that you are a member of, but thinks that
another community is better and has heaps of fun things to do there. They want to leave
the ‘Kraft’ community and join another brand community. (Alex is a member of the
‘Kraft’ community, and thinks that this community is good and has heaps of fun things
to do. They want to stay a member of the ‘Kraft’ community.)

4. Please show how much you agree with the following statements from 1 to 7, by
clicking on one number.
I desire to interact with the ‘Kraft’ community sometime during the next 2 weeks:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

212
5. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
My desire for interacting with the ‘Kraft’ community can be described as:

Very 1 2 3 4 5 6 7 Very
Weak Strong
Desire Desire

6. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
I want to interact together with the ‘Kraft’ community during the next 2 weeks:

Does not 1 2 3 4 5 6 7 Describes

describe me very
me at all much

7. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
The members of the ‘Kraft’ community would make good leaders:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

8. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
The members of the ‘Kraft’ community will achieve all of their goals:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

9. Please show how much you agree with the following statement from 1 to 7, by clicking
on one number.
The members of the ‘Kraft’ community are probably good at everything they do:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

213
10. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
The members of the ‘Kraft’ community will probably be successful in life:

Strongly 1 2 3 4 5 6 7 Strongly
Disagree Agree

11. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
The other members of the brand community are similar to me in terms of personality:

Not at all 1 2 3 4 5 6 7 A Lot

12. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
The other members of the brand community are similar to me in terms of age:

Not at all 1 2 3 4 5 6 7 A Lot

13. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
The other members of the brand community are similar to me in terms of values:

Not at all 1 2 3 4 5 6 7 A Lot

14. Please show how much you agree with the following statement from 1 to 7, by
clicking on one number.
The other members of the brand community are similar to me in terms of background:

Not at all 1 2 3 4 5 6 7 A Lot

214
9.4 Third Party Copyright Permission Communications

9.4.1 Paper One

215
216
217
9.4.2 Paper Two

218
9.4.3 Paper Three

219
220
221
222
9.5 Reference Lists for Papers

9.5.1 Paper One Reference List

Annett-Hitchcock, K., and Y. Xu. 2015. Shopping and virtual communities for
consumers with physical disabilities: Shopping and virtual communities. International
Journal of Consumer Studies 39(2): 136–144. https://doi.org/10.1111/ijcs.12161.
Bagozzi, R.P., and U.M. Dholakia. 2006. Antecedents and purchase consequences of
customer participation in small group brand communities. International Journal of
Research in Marketing 23(1): 45–61.
Baldus, B.J., C. Voorhees, and R. Calantone. 2015. Online brand community
engagement: Scale development and validation. Journal of Business Research 68(5):
978–985.
Belk, R.W., and G. Tumbat. 2005. The cult of Macintosh. Consumption Markets &
Culture 8(3): 205–217.
Bruhn, M., S. Schnebelen, and D. Schäfer. 2014. Antecedents and consequences of the
quality of e-customer-to-customer interactions in B2B brand communities. Industrial
Marketing Management 43(1): 164–176.
Canniford, R. 2011. How to manage consumer tribes. Journal of Strategic Marketing
19(7): 591–606.
Carlson, B.D., T.A. Suter, and T.J. Brown. 2008. Social versus psychological brand
community: The role of psychological sense of brand community. Journal of Business
Research 61(4): 284–291.
Casalò, L.V., C. Flaviaǹ, and M. Guinalìu. 2008. Promoting consumer’s participation in
virtual brand communities: A new paradigm in branding strategy. Journal of Marketing
Communications 14(1): 19–36.
Chaplin, L.N., and D.R. John. 2005. The development of self-brand connections in
children and adolescents. Journal of Consumer Research 32(1): 119–129.
Chaplin, L.N., and T.M. Lowrey. 2010. The development of consumer-based
consumption constellations in children. Journal of Consumer Research 36(5): 757–777.
Chen, C. D., and E.C. Ku. 2013. Bridging indistinct relationships and online loyalty:
evidence from online interest-based communities. Online Information Review 37(5):
731.
Cicchetti, D., and Cohen, D. J. 2006. Developmental Psychopathology Volume 1.
Hoboken: John Wiley & Sons, Inc. Retrieved from
http://ezproxy.newcastle.edu.au/login?url=http://
www.newcastle.eblib.com/EBLWeb/patron?target=patron&ex-tendedid=P 2555640&.
Cova, B., and S. Pace. 2006. Brand community of convenience products: new forms of
customer empowerment—The case ‘‘my Nutella the community’’. European Journal of
Marketing 40(9/ 10): 1087–1105. https://doi.org/10.1108/03090560610681023.

223
Cova, B., and V. Cova. 2002. Tribal marketing: The tribalisation of society and its
impact on the conduct of marketing. European Journal of Marketing 36(5/6): 595–620.
https://doi.org/10.1108/ 03090560210423023.
Crawford, H.J., and G.D. Gregory. 2015. Humorous advertising that travels: A review
and call for research. Journal of Business Research 68(3): 569–577.
https://doi.org/10.1016/j.jbusres.2014.09.005.
De Vries, L., S. Gensler, and P.S. Leeflang. 2012. Popularity of brand posts on brand
fan pages: An investigation of the effects of social media marketing. Journal of
Interactive Marketing 26(2): 83–91.
Dholakia, U.M., R.P. Bagozzi, and L.K. Pearo. 2004. A social influence model of
consumer participation in network- and small-group-based virtual communities.
International Journal of Research in Marketing 21(3): 241–263.
https://doi.org/10.1016/j.ijresmar.2003.12.004.
Enginkaya, E., and Yılmaz, H. 2014. What drives consumers to interact with brands
through social media? A motivation scale development study. In 2nd international
conference on strategic innovative marketing, 148, 219–226. https://doi.org/10.1016/j.
sbspro.2014.07.037.
Filipe Lages, L., and D.B. Montgomery. 2004. Export performance as an antecedent of
export commitment and marketing strategy adaptation: Evidence from small and
medium-sized exporters. European Journal of Marketing 38(9/10): 1186–1214. https://
doi.org/10.1108/03090560410548933.
Flurry, L.A., K.R. Swimberghe, and J.M. Parker. 2014. Examining brand communities
among children and adolescents: An exploratory study. Journal of Consumer Marketing
31(2): 103–110.
Gefen, D., and D.W. Straub. 2004. Consumer trust in B2C e-Commerce and the
importance of social presence: Experiments in e-Products and e-Services. Omega 32(6):
407–424.
Gorn, G.J., and R. Florsheim. 1985. The effects of commercials for adult products on
children. Journal of Consumer Research 11(4): 962–967.
Goulding, C., A. Shankar, and R. Canniford. 2013. Learning to be tribal: Facilitating the
formation of consumer tribes. European Journal of Marketing 47(5/6): 813–832.
https://doi.org/10.1108/ 03090561311306886.
Habibi, M.R., M. Laroche, and M.O. Richard. 2014. The roles of brand community and
community engagement in building brand trust on social media. Computers in Human
Behavior 37: 152–161.
Habibi, M.R., M. Laroche, and M.O. Richard. 2016. Testing an extended model of
consumer behavior in the context of social media-based brand communities. Computers
in Human Behavior 62: 292–302.
Hedlund, D.P. 2014. Creating value through membership and participation in sport fan
consumption communities. European Sport Management Quarterly 14(1): 50–71.

224
Hickman, T., and J. Ward. 2007. The dark side of brand community: Inter-group
stereotyping, trash talk, and Schadenfreude. Advances in Consumer Research 34: 314–
319.
Ho, C.-W. 2015. Identify with community or company? An investigation on the
consumer behavior in Facebook brand community. Telematics and Informatics 32(4):
930–939.
Hung, H. 2014. Attachment, identification, and loyalty: Examining mediating
mechanisms across brand and brand community contexts. Journal of Brand
Management 21(7–8): 594–614.
Hur, W., K. Ahn, and M. Kim. 2011. Building brand loyalty through managing brand
community commitment. Management Decision 49(7): 1194–1213.
https://doi.org/10.1108/00251741111151217.
Jang, H., L. Olfman, I. Ko, K. Joon, and K. Kim. 2008. The influence of on-line brand
community characteristics on community commitment and brand loyalty. International
Journal of Electronic Commerce 12(3): 57–80.
Ji, M. 2002. Children’s relationships with brands: ‘‘True love’’ or ‘‘one-night’’ stand?
Psychology & Marketing 19(4): 369–387.
Ji, M. 2008. Child-brand relations: A conceptual framework. Journal of Marketing
Management 24(5–6): 603–619.
Jung, N.Y., S. Kim, and S. Kim. 2014. Influence of consumer attitude toward online
brand community on revisit intention and brand trust. Journal of Retailing &
Consumer Services 21(4): 581–589.
Kang, I., K.C. Lee, S. Lee, and J. Choi. 2007. Investigation of online community
voluntary behavior using cognitive map. Computers in Human Behavior 23(1): 111–
126.
Kozinets, R.V. 1999. E-tribalized marketing?: The strategic implications of virtual
communities of consumption. European Management Journal 17(3): 252–264.
https://doi.org/10.1016/S0263-2373(99)00004-3.
Kuo, Y. F., and L. H. Feng. 2013. Relationships among community interaction
characteristics, perceived benefits, community commitment, and oppositional brand
loyalty in online brand communities. International Journal of Information Management
33(6): 948–962.
Lee, H.J., D. Lee, C.R. Taylor, and J. Lee. 2011. Do online brand communities help
build and maintain relationships with consumers? A network theory approach. Journal
of Brand Management 19(3): 213–227.
Lego. 2016. City. Retrieved February 23, 2016, from
https://community.lego.com/t5/CITY/bd-p/1512.
Lin, H. F. 2008. Antecedents of virtual community satisfaction and loyalty: An
empirical test of competing theories. CyberPsychology & Behavior 11(2): 138–144.
https://doi.org/10.1089/cpb.2007.0003.

225
Luo, N., M. Zhang, and W. Liu. 2015. The effects of value co-creation practices on
building harmonious brand community and achieving brand loyalty on social media in
China. Computers in Human Behavior 48: 492–499.
Madupu, V., and D.O. Cooley. 2010a. Antecedents and consequences of online brand
community participation: A conceptual framework. Journal of Internet Commerce 9(2):
127–147. https://doi. org/10.1080/15332861.2010.503850.
Madupu, V., and D.O. Cooley. 2010b. Cross-cultural differences in online brand
communities: An exploratory study of Indian and American online brand communities.
Journal of International Consumer Marketing 22(4): 363–375.
Manthiou, A., L.R. Tang, and R. Bosselman. 2014. Reason and reaction: the dual route
of the decision-making process on Facebook fan pages. Electronic Markets 24(4): 297–
308. https:// doi.org/10.1007/s12525-014-0156-8.
Marzocchi, G., G. Morandin, and M. Bergami. 2013. Brand communities: Loyal to the
community or the brand? European Journal of Marketing 47(1/2): 93–114.
https://doi.org/10.1108/ 03090561311285475.
Mattel. 2016. My world—The most pink-tastic world for girls online. Retrieved
February 23, 2016, from http://kids.barbie.com/en-us/ my-world.
McAlexander, J.H., J.W. Schouten, and H.F. Koenig. 2002. Building brand community.
Journal of Marketing 66(1): 38–54.
Mind Candy. 2016. Moshi Monsters - Sign In. Retrieved February 23, 2016, from
http://www.moshimonsters.com/login/forums.
Mitchell, C., and B.C. Imrie. 2011. Consumer tribes: Membership, consumption and
building loyalty. Asia Pacific Journal of Marketing and Logistics 23(1): 39–56.
https://doi.org/10.1108/13555851111099989.
Morandin, G., R.P. Bagozzi, and M. Bergami. 2013. Brand community membership and
the construction of meaning. Scandinavian Journal of Management 29(2): 173–183.
https://doi.org/10.1016/ j.scaman.2013.03.003.
Muniz, J., M. Albert, and T.C. O’Guinn. 2001. Brand community. Journal of Consumer
Research 27(4): 412–432. https://doi.org/ 10.1086/319618.
Muñiz Jr., A.M., and H.J. Schau. 2005. Religiosity in the abandoned apple newton
brand community. Journal of Consumer Research 31(4): 737–747.
Munnukka, J., H. Karjaluoto, and A. Tikkanen. 2015. Are Facebook brand community
members truly loyal to the brand? Computers in Human Behavior 51: 429–439.
Mzoughi, P.N., R.B. Ahmed, and H. Ayed. 2010. Explaining the participation in a
small group brand community: An extended TRA. Journal of Business & Economics
Research 8(8): 17–25.
Pahnila, S., & Väyrynen, K.2015. A cultural comparison of open innovation in online
brand communities. Retrieved from
https://pacis2015.comp.nus.edu.sg/_proceedings/PACIS_2015_submis sion_182.pdf.
Phillips-Melancon, J., and V. Dalakas. 2014. Brand rivalry and consumers’
Schadenfreude: The case of Apple. Services Marketing Quarterly 35(2): 173–186.
226
Piaget, J. 1972. Intellectual evolution from adolescence to adulthood. Human
Development 15(1): 1–12.
Pluye, P., Q.N. Hong, P.L. Bush, and I. Vedel. 2016. Opening-up the definition of
systematic literature review: The plurality of worldviews, methodologies and methods
for reviews and syntheses. Journal of Clinical Epidemiology 73: 2–5. https://
doi.org/10.1016/j.jclinepi.2015.08.033.
Royo-Vela, M., and P. Casamassima. 2011. The influence of belonging to virtual brand
communities on consumers’ affective commitment, satisfaction and word-of-mouth
advertising: The ZARA case. Online Information Review 35(4): 517–542. https://
doi.org/10.1108/14684521111161918.
Sánchez-Franco, M.J., E.M. Buitrago-Esquinas, and R. Yñiguez. 2012. How to intensify
the individual’s feelings of belonging to a social networking site? Management Decision
50(6): 1137–1154. https://doi.org/10.1108/00251741211238373.
Scarpi, D. 2010. Does size matter? An examination of small and large web-based brand
communities. Journal of Interactive Marketing 24(1): 14–21.
https://doi.org/10.1016/j.intmar.2009.10.002.
Schau, H., and A. Muñiz. 2006. A tale of tales: The Apple Newton narratives. Journal
of Strategic Marketing 14(1): 19–33.
Shang, R., Y. Chen, and H. Liao. 2006. The value of participation in virtual consumer
communities on brand loyalty. Internet Research 16(4): 398–418.
https://doi.org/10.1108/10662240610690025.
Sicilia, M., and M. Palazón. 2008. Brand communities on the internet. Corporate
Communications: An International Journal 13(3): 255–270.
https://doi.org/10.1108/13563280810893643.
Sierra, J.J., V.A. Badrinarayanan, and H.A. Taute. 2016. Explaining behavior in brand
communities: A sequential model of attachment, tribalism, and self-esteem. Computers
in Human Behavior 55: 626–632. https://doi.org/10.1016/j.chb.2015.10.009.
Snyder, H., L. Witell, A. Gustafsson, P. Fombelle, and P. Kristensson. 2016. Identifying
categories of service innovation: A review and synthesis of the literature. Journal of
Business Research. https:// doi.org/10.1016/j.jbusres.2016.01.009.
Stokburger-Sauer, N. 2010. Brand community: Drivers and outcomes. Psychology &
Marketing 27(4): 347–368.
Sung, Y., Y. Kim, O. Kwon, and M. Moon. 2010. An explorative study of Korean
consumer participation in virtual brand communities in social network sites. Journal of
Global Marketing 23(5): 430–445. https://doi.org/10.1080/08911762.2010.521115.
Syrjälä, H. 2016. Turning point of transformation: Consumer communities, identity
projects and becoming a serious dog hobbyist. Journal of Business Research 69(1):
177–190. https:// doi.org/10.1016/j.jbusres.2015.07.031.
Tajfel, H. (ed.). 1978. Differentiation between social groups: Studies in the social
psychology of intergroup relations. London: Academic Press.

227
Taute, Ha, and J. Sierra. 2014. Brand tribalism: An anthropological perspective. Journal
of Product & Brand Management 23(1): 2–15. https://doi.org/10.1108/JPBM-06-2013-
0340.
Thompson, S.A., and R.K. Sinha. 2008. Brand communities and new product adoption:
The influence and limits of oppositional loyalty. Journal of Marketing 72(6): 65–80.
https://doi.org/10. 1509/jmkg.72.6.65.
Thumm, J. 2015. Sephora learns the importance of online brand communities. Retrieved
March 2, 2016, from http://www. powerretail.com.au/pureplay/sephora-online-brand-
communities.
Tranfield, D., D. Denyer, and P. Smart. 2003. Towards a methodology for developing
evidence-informed management knowledge by means of systematic review. British
Journal of Management 14(3): 207–222.
Tsai, H. T., H. C. Huang, and Y. L. Chiu. 2012. Brand community participation in
Taiwan: Examining the roles of individual-, group-, and relationship-level antecedents.
Journal of Business Research 65(5): 676–684.
https://doi.org/10.1016/j.jbusres.2011.03.011.
Ungerleider, N. 2014, September 22. How Sephora discovered that Lurkers are also its
‘‘Superfans.’’ Retrieved March 2, 2016, from http://www.fastcolabs.com/3035236/how-
sephora-discovered-that-lurkers-are-also-its-superfans.
Veloutsou, C., and L. Moutinho. 2009. Brand relationships through brand reputation
and brand tribalism. Advances in Brand Management 62(3): 314–322.
https://doi.org/10.1016/j.jbusres.2008.05.010.
Wang, Y., S.F. Chan, and Z. Yang. 2013. Customers’ perceived benefits of interacting
in a virtual brand community in China. Journal of Electronic Commerce Research
14(1): 49–66.
Wang, Y., S.S. Ma, and D. Li. 2015. Customer participation in virtual brand
communities: The self-construal perspective. Information & Management 52(5): 577–
587. https://doi.org/10.1016/j.im.2015.04.003.
Wang, Y., J. Shi, S. Ma, G. Shi, and L. Yan. 2012. Customer interactions in virtual
brand communities: Evidence from China. Journal of Global Information Technology
Management 15(2): 46–69.
Witell, L., H. Snyder, A. Gustafsson, P. Fombelle, and P. Kristensson. 2015. Defining
service innovation: A review and synthesis. Journal of Business Research.
https://doi.org/10.1016/j.jbusres. 2015.12.055.
Woisetschläger, D.M., V. Hartleb, and M. Blut. 2008. How to make brand communities
work: Antecedents and consequences of consumer participation. Journal of
Relationship Marketing 7(3): 237–256. https://doi.org/10.1080/15332660802409605.
Zaglia, M.E. 2013. Brand communities embedded in social networks. Journal of
Business Research 66(2): 216–223. https://doi.org/10. 1016/j.jbusres.2012.07.015.
Zhang, H., K.Z.K. Zhang, M.K.O. Lee, and F. Feng. 2015. Brand loyalty in enterprise
microblogs. Information Technology & People 28(2): 304–326.

228
Zheng, X., C.M.K. Cheung, M.K.O. Lee, and L. Liang. 2015. Building brand loyalty
through user engagement in online brand communities in social networking sites.
Information Technology & People 28(1): 90–106. https://doi.org/10.1108/ITP-08-2013-
0144.
Zhou, Y., and M. Amin. 2014. Factors affecting online community commitment in
China: A conceptual framework. Journal of Technology Management in China 9(1):
24–36. https://doi.org/ 10.1108/JTMC-08-2013-0033.
Zhou, Z., J.P. Wu, Q. Zhang, and S. Xu. 2013. Transforming visitors into members in
online brand communities: Evidence from China. Journal of Business Research 66(12):
2438–2443. https:// doi.org/10.1016/j.jbusres.2013.05.032.
Zhou, Z., Q. Zhang, C. Su, and N. Zhou. 2012. How do brand communities generate
brand relationships? Intermediate mechanisms. Journal of Business Research 65(7):
890–895. https:// doi.org/10.1016/j.jbusres.2011.06.034.

229
9.5.2 Paper Two Reference List

Bagozzi, Richard and Utpal Dholakia (2006), “Antecedents and Purchase Consequences
of Customer Participation in Small Group Brand Communities,” International Journal of
Research in Marketing, 23(1), 45–61.
Baumrind, Diana (1978), “Parental Disciplinary Patterns and Social Competence in
Children,” Youth & Society, 9(3), 239–67.
Borgers, Natacha and Joop Hox (2001), “Item Nonresponse in Questionnaire Research
with Children,” Journal of Official Statistics, 17(2), 321–35.
Cicchetti, Dante and Donald Cohen (2006), Developmental Psychopathology.
Volume 1, Hoboken: John Wiley & Sons, Inc.
Dishion, Thomas and Jessica Tipsord (2011), “Peer Contagion in Child and Adolescent
Social and Emotional Development,” Annual Review of Psychology, 62, 189–214.
Flurry, Laura, Krist Swimberghe, and Janna Parker (2014), “Examining Brand
Communities Among Children and Adolescents: An Exploratory Study,” Journal of
Consumer Marketing, 31(2), 103–10.
Hawkins, Del and Kenneth Coney(1974), “Peer Group Influences on Children’s Product
Preferences,” Academy of Marketing Science. Journal (Pre-1986), 2(2), 322.
Muniz, Albert and Thomas O’Guinn (2001), “Brand Community,” Journal of Consumer
Research, 27(4), 412–32.
Piaget, Jean (1972), “Intellectual Evolution From Adolescence to Adulthood,” Human
Development, 15(1), 1–12.
Schult, Carolyn (2002), “Children’s Understanding of the Distinction between Intentions
and Desires,” Child Development, 6, 1727.
Sicilia, Maria and Mariola Palazón (2008), “Brand Communities on the Internet,”
Corporate Communications: An International Journal, 13(3), 255– 70.

230
9.5.3 Paper Three Reference List

Aboud, F. (2003) The formation of in-group favoritism and out-group prejudice in

young children: are they distinct attitudes? Developmental Psychology, 39, 48.
Abrams D, Rutland A. (2008) The development of subjective group dynamics. In
Intergroup Attitudes and Relations in Childhood through Adulthood, pp. 47–65, Oxford
University, Oxford.
Abrams, D., Rutland, A. & Cameron, L. (2003) The development of subjective group
dynamics: children’s judgments of normative and deviant in-group and out-group
individuals. Child Development, 74, 1840–1856.
Abrams, D., Rutland, A., Cameron, L. & Ferrell, J. (2007) Older but wilier: in-group
accountability and the development of subjective group dynamics. Developmental
Psychology, 43, 134–148.
Abrams, D., Rutland, A., Ferrell, J. & Pelletier, J. (2008) Children’s judgments of
disloyal and immoral peer behavior: subjective group dynamics in minimal intergroup
contexts. Child Development, 79, 444–461.
Algesheimer, R., Dholakia, U. & Herrmann, A. (2005) The social influence of brand
community: evidence from european car clubs. Journal of Marketing, 69, 19–34.
Bagozzi, R. & Dholakia, U. (2006) Antecedents and purchase consequences of
customer participation in small group brand communities. International Journal of
Research in Marketing, 23, 45–61.
Baumeister, R., Bratslavsky, E., Finkenauer, C. & Vohs, K. (2001) Bad is stronger than
good. Review of General Psychology, 5, 323.
Bennett M, Sani F. (2004) Development of the Social Self. Psychology Press, New
York.
Boulton, M. & Smith, P. (1994) Bully/victim problems in middle school children:
stability, self-perceived competence, peer perceptions and peer acceptance. British
Journal of Developmental Psychology, 12, 315–330.
Buckingham D. (1993) Children Talking Television: The Making of Television Literacy
(Vol. 9.) Psychology Press, New York.
Carlson, B., Suter, T. & Brown, T. (2008) Social versus psychological brand
community: the role of psychological sense of brand community. Journal of Business
Research, 61, 284–291.
Casaló, L., Flavián, C. & Guinalíu, M. (2008) Promoting consumer’s participation in
virtual brand communities: a new paradigm in branding strategy. Journal of Marketing
Communications, 14, 19–36.
Cassidy, T. (2009) Bullying and victimisation in school children: the role of social
identity, problem-solving style, and family and school context. Social Psychology of
Education, 12, 63–76.
Chan, K. & Li, S. (2010) Understanding consumer-to-consumer interactions in virtual
communities: the salience of reciprocity. Journal of Business Research, 63, 1033–1040.

231
Cicchetti D, Cohen D.J. (2006) Developmental Psychopathology (Vol 1). John Wiley &
Sons, Inc, Hoboken.
Collins P, Mitchell D.M. (2015) The purchasing power of tomorrow: An analysis of the
buying habits of children ages 6-15. [WWW document] URL
http://digitalcommons.uri.edu/cgi/viewcontent.cgi?article=1421&context=srhonorsprog
(accessed January 28, 2016)
Corenblum, B. (2014) Relationships between racial-ethnic identity, self-esteem and in-
group attitudes among first nation children. Journal of Youth and Adolescence, 43, 387–
404.
De Cremer, D. & Oosterwegel, A. (1999) Collective self-esteem, personal self-esteem,
and collective efficacy in in-group and outgroup evaluations. Current Psychology, 18,
326–339.
Dholakia, U., Bagozzi, R. & Pearo, L. (2004) A social influence model of consumer
participation in network- and small-group-based virtual communities. International
Journal of Research in Marketing, 21, 241–263.
Ellemers, N., Kortekaas, P. & Ouwerkerk, J. (1999) Self-categorisation, commitment to
the group and group self-esteem as related but distinct aspects of social identity. European
Journal of Social Psychology, 29, 371–389.
Elliott, R. & Leonard, C. (2004) Peer pressure and poverty: exploring fashion brands
and consumption symbolism among children of the “british poor”. Journal of Consumer
Behaviour, 3, 347–359.
Flurry L., A., Swimberghe K R., Parker J. M. (2014) Examining brand communities
among children and adolescents: An exploratory study. Journal of Consumer
Marketing, 31, 103–110.
Garbarino, E. & Johnson, M. (1999) The different roles of satisfaction, trust, and
commitment in customer relationships. The Journal of Marketing, 63, 70–87.
Gorn, G. & Florsheim, R. (1985) The effects of commercials for adult products on
children. Journal of Consumer Research, 11, 962–967.
Habibi, M., Laroche, M. & Richard, M. (2014) The roles of brand community and
community engagement in building brand trust on social media. Computers in Human
Behavior, 37, 152–161.
Harter, S. (1982) The perceived competence scale for Children. Child Development, 53,
87.
Harter, S. (1993) Causes and Consequences of Low Self-Esteem in Children and
Adolescents. In Self-Esteem: The Puzzle of Low self-Regard (ed. by R.F. Baumeister),
pp. 87–116. Springer, US.
Harter, S. (1999) The Construction of the Self: A Developmental Perspective. Guilford
Press, New York.
Hayes, A.F. (2013) Introduction to Mediation, Moderation, and Conditional Process
Analysis: A Regression-Based Approach. Guilford Press, New York.

232
Hogg, M. & Abrams, D. (1990) Social Identifications: A Social Psychology of
Intergroup Relations and Group Processes. Routledge, Florence, KY, USA.
Kuo, Y. & Feng, L. (2013) Relationships among community interaction characteristics,
perceived benefits, community commitment, and oppositional brand loyalty in online
brand communities. International Journal of Information Management, 33, 948–962.
Leary, M., Terdal, S., Tambor, E. & Downs, D. (1995) Self-esteem as an interpersonal
monitor: the sociometer hypothesis. Journal of Personality and Social Psychology, 68,
518–530.
Li, T. & Zhang, Y. (2015) Social network types and the health of older adults: exploring
reciprocal associations. Social Science and Medicine, 130, 59–68.
Luhtanen, R. & Crocker, J. (1992) A collective self-esteem scale: Self-evaluation of
one’s social identity. Personality and Social Psychology Bulletin, 18, 302–318.
McAlexander, J., Schouten, J. & Koenig, H. (2002) Building brand community. Journal
of Marketing, 66, 38–54.
McGee, R., Williams, S., Howden-Chapman, P., Martin, J. & Kawachi, I. (2006)
Participation in clubs and groups from childhood to adolescence and its effects on
attachment and self-esteem. Journal of Adolescence, 29, 1–17.
Minecraft Seeds. (2015) Minecraft Player Demographics. [WWW document] URL
http://minecraft-seeds.net/blog/minecraft-player-demographics/(accessed October 4th,
2015)
Mojang (2015) Minecraft. [WWW document] URL https://minecraft.net/ stats (accessed
March 25th, 2015)
Muniz, A. Jr & O’Guinn, T. (2001) Brand community. Journal of Consumer Research,
27, 412–432.
Pendley, J., Kasmen, L., Miller, D., Donze, J., Swenson, C. & Reeves, G. (2002) Peer
and family support in children and adolescents with type 1 diabetes. Journal of
Pediatric Psychology, 27, 429–438.
Preacher, K., Rucker, D. & Hayes, A.F. (2007) Assessing moderated mediation
hypotheses: theory, methods and prescriptions. Multivariate Behavioral Research, 42,
185–227.
Rennie F, Morrison T.M. (2013) E-Learning and Social Networking Handbook:
Resources for Higher Education. Routledge, New York.
Subrahmanyam, K., Reich, S., Waechter, N. & Espinoza, G. (2008) Online and offline
social networks: use of social networking sites by emerging adults. Journal of Applied
Developmental Psychology, 29, 420–433.
Tajfel, H. (Ed.). (1978) Differentiation between Social Groups: Studies in the Social
Psychology of Intergroup Relations. Academic Press, London.
Tajfel, H. & Turner, J. (1979) An integrative theory of intergroup conflict. The Social
Psychology of Intergroup Relations, 33, 74.
Trzesniewski, K., Donnellan, M. & Robins, R. (2003) Stability of self-esteem across the
life span. Journal of Personality and Social Psychology, 84, 205–220.
233
United Nations. (2013) Definition of Youth: United Nations. [WWW document] URL
http://www.un.org/esa/socdev/documents/youth/fact-sheets/youth-definition.pdf
(accessed August 10th, 2015)
Valkenburg, P., Peter, J. & Schouten, A. (2006) Friend networking sites and their
relationship to adolescents’ well-being and social self-esteem. CyberPsychology and
Behavior, 9, 584–590.

234
9.5.4 Paper Four Reference List

Aboud, F. E., & Mendelson, M. J. (1996). Determinants of friendship selection and

quality: Developmental perspectives. In The company they keep: Friendship in
childhood and adolescence. (pp. 87–112). New York, NY, US: Cambridge University
Press.
Abrams, D., & Rutland, A. (2008). The development of subjective group dynamics. In
Intergroup attitudes and relations in childhood through adulthood (pp. 47–65). USA:
Oxford University Press.
Abrams, D., Rutland, A., Cameron, L., & Ferrell, J. (2007). Older but wilier: In-group
accountability and the development of subjective group dynamics. Developmental
Psychology, 43(1), 134–148.
Abrams, D., Rutland, A., Cameron, L., & Marques, J. (2003). The development of
subjective group dynamics: When in-group bias gets specific. British Journal of
Developmental Psychology, 21(2), 155–176.
Abrams, D., Rutland, A., Ferrell, J. M., & Pelletier, J. (2008). Children’s judgments of
disloyal and immoral peer behavior: Subjective group dynamics in minimal intergroup
contexts. Child Development, 79(2), 444–461.
Abrams, D., Rutland, A., Palmer, S. B., Pelletier, J., Ferrell, J., & Lee, S. (2014). The
role of cognitive abilities in children’s inferences about social atypicality and peer
exclusion and inclusion in intergroup contexts. British Journal of Developmental
Psychology, 32(3), 233–247. https://doi.org/10.1111/bjdp.12034
Abrams, D., Rutland, A., Palmer, S. B., & Purewal, K. (2014). Children’s responses to
social atypicality among group members: Advantages of a contextualized social
developmental account. British Journal of Developmental Psychology, 32(3), 257–261.
Abrams, D., Rutland, A., Pelletier, J., & Ferrell, J. M. (2009). Children’s group nous:
Understanding and applying peer exclusion within and between groups. Child
Development, 80(1), 224–243.
Albert, N., Merunka, D., & Valette-Florence, P. (2008). When consumers love their
brands: Exploring the concept and its dimensions. Journal of Business Research,
61(10), 1062–1075. https://doi.org/10.1016/j.jbusres.2007.09.014
Balmer, J. M., Powell, S. M., Karaosmanoğlu, E., Banu Elmadağ Baş, A., & Zhang, J.
(2011). The role of other customer effect in corporate marketing: Its impact on
corporate image and consumer-company identification. European Journal of Marketing,
45(9/10), 1416–1445.
Baxter, S., Ilicic, J., Kulczynski, A., & Lowrey, T. (2017). Using sublexical priming to
enhance brand name phonetic symbolism effects in young children. Marketing Letters,
28(4), 565–577.
Baxter, S., & Lowrey, T. (2014). Examining children’s preference for phonetically
manipulated brand names across two English accent groups. International Journal of
Research in Marketing, 31(1), 122–124.

235
Becerra, E. P., & Badrinarayanan, V. (2013). The influence of brand trust and brand
identification on brand evangelism. The Journal of Product and Brand Management,
22(5/6), 371–383. https://doi.org/10.1108/JPBM-09-2013-0394
Branscombe, N. R., Spears, R., Ellemers, N., & Doosje, B. (2002). Intragroup and
intergroup evaluation effects on group behavior. Personality and Social Psychology
Bulletin, 28(6), 744–753.
Byrne, D. (1971). The attraction paradigm (Vol. 11). Academic Pr.
Byrne, D., & Clore, G. (1967). Effectance arousal and attraction. Journal of Personality
and Social Psychology, 6(4, Pt.2), 1.
Cova, B., & Pace, S. (2006). Brand community of convenience products: new forms of
customer empowerment – the case “my Nutella the community.” European Journal of
Marketing, 40(9/10), 1087–1105. https://doi.org/10.1108/03090560610681023
Delgado-Ballester, E. (2004). Applicability of a brand trust scale across product
categories: A multigroup invariance analysis. European Journal of Marketing, 38(5/6),
573–592.
Dholakia, U. M., Bagozzi, R. P., & Pearo, L. K. (2004). A social influence model of
consumer participation in network- and small-group-based virtual communities.
International Journal of Research in Marketing, 21(3), 241–263.
https://doi.org/10.1016/j.ijresmar.2003.12.004
Dunham, Y., Newheiser, A.-K., Hoosain, L., Merrill, A., & Olson, K. R. (2014). From a
Different Vantage: Intergroup Attitudes Among Children from Low- and Intermediate-
Status Racial Groups. Social Cognition, 32(1), 1–21.
https://doi.org/10.1521/soco.2014.32.1.1
Ellemers, N. (2012). The group self. Science, 336(6083), 848–852.
Ellemers, N., Doosje, B., & Spears, R. (2004). Sources of respect: the effects of being
liked by ingroups and outgroups. European Journal of Social Psychology, 34(2), 155–
172. https://doi.org/10.1002/ejsp.196
Epstein, J. L. (1989). The selection of friends: Changes across the grades and in
different school environments. In T. Berndt & G. Ladd (Eds.), Peer relationships in
child development (pp. 158–187). New York: Wiley.
Fawcett, C. A., & Markson, L. (2010). Similarity predicts liking in 3-year-old children.
Journal of Experimental Child Psychology, 105(4), 345–358.
Flurry, L., A., Swimberghe, K., R., & Parker, J., M. (2014). Examining brand
communities among children and adolescents: An exploratory study. Journal of
Consumer Marketing, 31(2), 103–110.
Gueguen, N., Martin, A., & Meineri, S. (2011). Similarity and Social Interaction: When
Similarity Fosters Implicit Behavior Toward a Stranger. Journal of Social Psychology,
151(6), 671–673.
Hartup, W. W. (1989). Behavioral manifestations of children’s friendships. In T. Berndt
& G. Ladd (Eds.), Peer relationships in child development (pp. 46–70). New York.

236
Haselager, G. J., Hartup, W. W., Lieshout, C. F., & Riksen-Walraven, J. M. A. (1998).
Similarities between friends and nonfriends in middle childhood. Child Development,
69(4), 1198–1208.
Hook, M., Baxter, S., & Kulczynski, A. (2016). Children’s participation in brand-based
social networks: Examining the role of evaluative social identity, self-esteem and
anticipated emotions on commitment and desire to recommend. International Journal of
Consumer Studies, 40(5), 552–561.
Hook, M., Baxter, S., & Kulczynski, A. (2017). Antecedents and Consequences of
Children’s Brand Community Participation: A Replication and Extension Study.
Journal of Marketing Behavior, 3(1), 63–72. https://doi.org/10.1561/107.00000042
Hung, H. (2014). Attachment, identification, and loyalty: Examining mediating
mechanisms across brand and brand community contexts. Journal of Brand
Management, 21(7–8), 594–614.
Hunter, S. C., Fox, C. L., & Jones, S. E. (2016). Humor style similarity and difference
in friendship dyads. Journal of Adolescence, 46, 30–37.
Huo, Y. J., & Binning, K. R. (2008). Why the Psychological Experience of Respect
Matters in Group Life: An Integrative Account. Social and Personality Psychology
Compass, 2(4), 1570–1585. https://doi.org/10.1111/j.1751-9004.2008.00129.x
Kandel, D. B. (1978). Homophily, selection, and socialization in adolescent friendships.
American Journal of Sociology, 84(2), 427–436.
Lee, N. (2017). Lego launches a safe social network for kids to share their creations.
Retrieved August 31, 2017, from https://www.engadget.com/2017/01/31/lego-life/
LEGO System A/S. (2017). LEGO® Life (Version 1.2.8) [Android 4.1 and up]. LEGO
System A/S. Retrieved from
https://play.google.com/store/apps/details?id=com.lego.common.legolife&hl=en
Lydon, J. E., Jamieson, D. W., & Zanna, M. P. (1988). Interpersonal similarity and the
social and intellectual dimensions of first impressions. Social Cognition, 6(4), 269–286.
Marzocchi, G., Morandin, G., & Bergami, M. (2013). Brand communities: loyal to the
community or the brand? European Journal of Marketing, 47(1/2), 93–114.
https://doi.org/10.1108/03090561311285475
McAlexander, J. H., Kim, S. K., & Roberts, S. D. (2003). Loyalty: The Influences of
Satisfaction and Brand Community Integration. Journal of Marketing Theory and
Practice, 11(4), 1–11.
Montoya, R. M., & Horton, R. S. (2004). On the importance of cognitive evaluation as a
determinant of interpersonal attraction. Journal of Personality and Social Psychology,
86(5), 696.
Morgan, R. M., & Hunt, S. D. (1994). The commitment-trust theory of relationship
marketing. Journal of Marketing, 58(3), 20.
Muniz, J., Albert M., & O’Guinn, T. C. (2001). Brand community. Journal of
Consumer Research, 27(4), 412–432. https://doi.org/10.1086/319618

237
Nangle, D. W., Erdley, C. A., Zeff, K. R., Stanchfield, L. L., & Gold, J. A. (2004).
Opposites do not attract: Social status and behavioral-style concordances and
discordances among children and the peers who like or dislike them. Journal of
Abnormal Child Psychology, 32(4), 425–434.
Noel, A., & Merunka, D. (2013). The role of brand love in consumer-brand
relationships. The Journal of Consumer Marketing, 30(3), 258–266.
https://doi.org/10.1108/07363761311328928
Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Assessing moderated mediation
hypotheses: Theory, methods and prescriptions. Multivariate Behavioral Research,
42(1), 185–227.
Rubin, K. H., Lynch, D., Coplan, R., Rose-Krasnor, L., & Booth, C. L. (1994). “Birds
of a feather…”: Behavioral concordances and preferential personal attraction in
children. Child Development, 65(6), 1778–1785.
Sicilia, M., & Palazón, M. (2008). Brand communities on the internet. Corporate
Communications: An International Journal, 13(3), 255–270.
https://doi.org/10.1108/13563280810893643
Simon, B., & Stürmer, S. (2003). Respect for Group Members: Intragroup Determinants
of Collective Identification and Group-Serving Behavior. Personality and Social
Psychology Bulletin, 29(2), 183–193. https://doi.org/10.1177/0146167202239043
Singh, R., Chen, F., & Wegener, D. T. (2014). The Similarity-Attraction Link:
Sequential Versus Parallel Multiple-Mediator Models Involving Inferred Attraction,
Respect, and Positive Affect. Basic & Applied Social Psychology, 36(4), 281–298.
Singh, R., & Ho, S. Y. (2000). Attitudes and attraction: A new test of the attraction,
repulsion and similarity-dissimilarity asymmetry hypotheses. British Journal of Social
Psychology, 39(2), 197–211.
Stürmer, S., Simon, B., & Loewy, M. I. (2008). Intraorganizational Respect and
Organizational Participation: The Mediating Role of Collective Identity. Group
Processes & Intergroup Relations, 11(1), 5–20.
https://doi.org/10.1177/1368430207084842
Tyler, T., Degoey, P., & Smith, H. (1996). Understanding why the justice of group
procedures matters: A test of the psychological dynamics of the group-value model.
Journal of Personality and Social Psychology, 70(5), 913.
White, R. W. (1959). Motivation reconsidered: The concept of competence.
Psychological Review, 66(5), 297–333.

238
9.5.5 Conference Paper 1 Reference List

Abrams, D., Rutland, A., & Cameron, L. (2003). The Development of Subjective Group
Dynamics: Children’s Judgments of Normative and Deviant In-Group and Out-Group
Individuals. Child Development, 74(6), 1840–1856.
Bagozzi, R. P., & Dholakia, U. M. (2006). Antecedents and purchase consequences of
customer participation in small group brand communities. International Journal of
Research in Marketing, 23(1), 45–61.
Bennett, M., & Sani, F. (2004). Development of the Social Self. Psychology Press.
Boulton, M. J., & Smith, P. K. (1994). Bully/victim problems in middle school children:
Stability, self-perceived competence, peer perceptions and peer acceptance. British
Journal of Developmental Psychology, 12(3), 315–330.
Chan, K. W., & Li, S. Y. (2010). Understanding consumer-to-consumer interactions in
virtual communities: The salience of reciprocity. Journal of Business Research, 63(9–
10), 1033–1040.
Corenblum, B. (2014). Relationships Between Racial-Ethnic Identity, Self-esteem and
In-group Attitudes Among First Nation Children. Journal of Youth & Adolescence,
43(3), 387–404.
Dholakia, U. M., Bagozzi, R. P., & Pearo, L. K. (2004). A social influence model of
consumer participation in network- and small-group-based virtual communities.
International Journal of Research in Marketing, 21(3), 241–263.
Ellemers, N., Kortekaas, P., & Ouwerkerk, J. W. (1999). Self-categorisation,
commitment to the group and group self-esteem as related but distinct aspects of social
identity. European Journal of Social Psychology, 29(2/3), 371–389.
Flurry, L., A., Swimberghe, K., R., & Parker, J., M. (2014). Examining brand
communities among children and adolescents: an exploratory study. Journal of
Consumer Marketing, 31(2), 103–110.
Garbarino, E., & Johnson, M. S. (1999). The different roles of satisfaction, trust, and
commitment in customer relationships. The Journal of Marketing, 70–87.
Gorn, G. J., & Florsheim, R. (1985). The Effects of Commercials for Adult Products on
Children. Journal of Consumer Research, 11(4), 962–967.
Harter, S. (1999). The construction of the self: A developmental perspective. Guilford
Press.
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process
analysis: A regression-based approach. Guilford Press.
McAlexander, J. H., Schouten, J. W., & Koenig, H. F. (2002). Building Brand
Community. Journal of Marketing, 66(1), 38–54.
Minecraft Seeds. (2015). Minecraft Player Demographics. Retrieved from
http://minecraft-seeds.net/blog/minecraft-player-demographics/. (Last accessed:
October 4, 2015).

239
Muniz Jr, A. M., & O’Guinn, T. C. (2001). Brand Community. Journal of Consumer
Research, 27(4), 412–432.
Ouwersloot, H., & Odekerken‐Schröder, G. (2008). Who’s who in brand communities –
and why? European Journal of Marketing, 42(5/6), 571–585.
Tajfel, H. (Ed.). (1978). Differentiation between social groups: Studies in the social
psychology of intergroup relations. London: Academic Press.
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. The
Social Psychology of Intergroup Relations, 33(47), 74.
Valkenburg, P. M., Peter, J., & Schouten, A. P. (2006). Friend Networking Sites and
Their Relationship to Adolescents’ Well-Being and Social Self-Esteem.
CyberPsychology & Behavior, 9(5), 584–590.

240
9.5.5 Conference Paper 2 Reference List

Abrams, D., & Rutland, A. (2008). The development of subjective group dynamics. In
Intergroup attitudes and relations in childhood through adulthood (pp. 47–65). USA:
Oxford University Press.
Abrams, D., Rutland, A., Cameron, L., & Marques, J. (2003). The development of
subjective group dynamics: When in-group bias gets specific. British Journal of
Developmental Psychology, 21(2), 155–176.
Abrams, D., Rutland, A., Ferrell, J. M., & Pelletier, J. (2008). Children’s judgments of
disloyal and immoral peer behavior: Subjective group dynamics in minimal intergroup
contexts. Child Development, 79(2), 444–461.
Abrams, D., Rutland, A., Palmer, S. B., & Purewal, K. (2014). Children’s responses to
social atypicality among group members: Advantages of a contextualized social
developmental account. British Journal of Developmental Psychology, 32(3), 257–261.
Abrams, D., Rutland, A., Pelletier, J., & Ferrell, J. M. (2009). Children’s group nous:
Understanding and applying peer exclusion within and between groups. Child
Development, 80(1), 224–243.
Balmer, J. M., Powell, S. M., Karaosmanoğlu, E., Banu Elmadağ Baş, A., & Zhang, J.
(2011). The role of other customer effect in corporate marketing: Its impact on
corporate image and consumer-company identification. European Journal of Marketing,
45(9/10), 1416–1445.
Byrne, & Clore. (1967). Effectance arousal and attraction. American Psychological
Association.
Cova, B., & Pace, S. (2006). Brand community of convenience products: new forms of
customer empowerment – the case “my Nutella the community.” European Journal of
Marketing, 40(9/10), 1087–1105. https://doi.org/10.1108/03090560610681023
Dholakia, U. M., Bagozzi, R. P., & Pearo, L. K. (2004). A social influence model of
consumer participation in network- and small-group-based virtual communities.
International Journal of Research in Marketing, 21(3), 241–263.
https://doi.org/10.1016/j.ijresmar.2003.12.004
Flurry, L., A., Swimberghe, K., R., & Parker, J., M. (2014). Examining brand
communities among children and adolescents: An exploratory study. Journal of
Consumer Marketing, 31(2), 103–110.
Gueguen, N., Martin, A., & Meineri, S. (2011). Similarity and Social Interaction: When
Similarity Fosters Implicit Behavior Toward a Stranger. Journal of Social Psychology,
151(6), 671–673.
Hook, M., Baxter, S., & Kulczynski, A. (2016). Children’s participation in brand‐based
social networks: Examining the role of evaluative social identity, self‐esteem and
anticipated emotions on commitment and desire to recommend. International Journal of
Consumer Studies.
Hung, H. (2014). Attachment, identification, and loyalty: Examining mediating
mechanisms across brand and brand community contexts. Journal of Brand
Management, 21(7–8), 594–614.
241
Lydon, J. E., Jamieson, D. W., & Zanna, M. P. (1988). Interpersonal similarity and the
social and intellectual dimensions of first impressions. Social Cognition, 6(4), 269–286.
Montoya, R. M., & Horton, R. S. (2004). On the importance of cognitive evaluation as a
determinant of interpersonal attraction. Journal of Personality and Social Psychology,
86(5), 696.
Muniz, J., Albert M., & O’Guinn, T. C. (2001). Brand community. Journal of
Consumer Research, 27(4), 412–432. https://doi.org/10.1086/319618
Singh, R., Chen, F., & Wegener, D. T. (2014). The Similarity-Attraction Link:
Sequential Versus Parallel Multiple-Mediator Models Involving Inferred Attraction,
Respect, and Positive Affect. Basic & Applied Social Psychology, 36(4), 281–298.
Singh, R., Tay, Y. Y., & Sankaran, K. (2016). Causal role of trust in interpersonal
attraction from attitude similarity. Journal of Social and Personal Relationships,
265407516656826. https://doi.org/10.1177/0265407516656826
Tidwell, N. D., Eastwick, P. W., & Finkel, E. J. (2013). Perceived, not actual, similarity
predicts initial attraction in a live romantic context: Evidence from the speed-dating
paradigm. Personal Relationships, 20(2), 199–215.
White, R. W. (1959). Motivation reconsidered: The concept of competence.
Psychological Review, 66(5), 297.

242
9.6 ABDC Ranking Explanation

The Australian Business Deans Council (ABDC) represents 39 business schools across

universities in Australia, aiming to make these schools better (ABDC, 2018a). One

method this is achieved is through providing a journal quality ranking to guide academics.

“The ABDC Journal Quality list is divided into four categories of quality, A*, A, B, C as

described below:

• A*: this is the highest quality category, and indicatively represents approximately

the top 5-7% of the journals assigned to the given primary for panel.

• A: this is the second highest quality category, and indicatively represents

approximately the next 15-25% of the journals assigned to the given primary for

panel.

• B: this is the third highest quality category, and indicatively represents

approximately the next 35-40% of the journals assigned to the given primary for

group.

• C: this is the fourth highest quality category, and represents the remaining

recognised quality journals assigned to the given primary for panel.” (ABDC,

2018b, para. 1)

243
9.7 H-index Explanation

The Hirsch index, or h-index as its commonly referred to, is a single number criterion to

measure the scientific input of a journal or an individual (Bornmann & Daniel, 2007).

The h-index is calculated using the following logic: “A scientist has index h if h of his or

her Np papers have at least h citations each and the other (Np − h) papers have fewer than

≤ h citations each” (Hirsch, 2005, p. 16569). This same formula can be applied to a

journal. The higher the h-index, the more impact the journal has had in the field (Mingers,

Macri, & Petrovici, 2012). An h-index of 0 indicates that the individual or journal has, at

best, published articles that have had no impact in the field (Bornmann & Daniel, 2007).

Compared to other indexes, the h-index has been widely praised (e.g. Batista, Campiteli,

Kinouchi, & Martinez, 2005; Bornmann & Daniel, 2007; Costas & Bordons, 2007). A

reason for this is that the h-index has the advantage of measuring research impact of a

researcher’s or journal’s entire contribution over their lifetime (Hirsch, 2005), and it’s

easy to calculate (Harzing & Van Der Wal, 2009). The h-index favours citations across

many papers, rather than a single highly cited paper and many non-cited papers.

244