bs_bs_banner

International Journal of Consumer Studies ISSN 1470-6423

Consumer decision-making styles and post purchase

behaviour of poor for FMCGs
Arvind Kumar, Anupama Vohra and Hamendra Kumar Dangi
Faculty of Management Studies, University of Delhi, New Delhi, Delhi, India

Keywords Abstract
Consumer behaviour, consumer decision-
making styles, FMCGs, post purchase More than half of the world’s population is poor. Certainly, their purchasing power
behaviour, the poor. cannot be compared with that of the riches, but it is their collective purchase potential
which makes them a substantial market. The fast moving consumer goods (FMCGs), on
Correspondence the other hand, are the fourth largest industry in the world. FMCGs are relatively low
Arvind Kumar, Faculty of Management cost products and the poor, by necessity, spend a significant amount of their income on
Studies, University of Delhi, New Delhi, Delhi, FMCGs. So, by virtue of being a large consumer base for FMCGs, the poor are a
India. promising market for the FMCG marketers. But little is known about their buying
E-mail: [email protected] behaviour for FMCGs as only a few studies have been conducted on them in this regard.
This study aims to explore their shopping orientations towards price, quality and brand
doi: 10.1111/ijcs.12320 for FMCGs in the context of one of the largest developing countries like India by
exploring their consumer decision-making styles (CDMSs), especially their price
consciousness, quality consciousness, brand consciousness and brand loyalty. Besides it,
it also explores their post purchase behaviour and testifies its mechanism with the above
mentioned CDMSs in the present context through structured equation modelling.

Introduction 2007; Shukla, 2010), TV sets (Banerjee and Duflo, 2007;

Shukla, 2010), mobile phones (Sarin and Jain, 2009).
More than half of world’s population is poor and live on a rela-
It suggests that poor are a potential market for marketers and
tively low-income. Obviously, the purchasing power of these
understanding their purchase pattern can create new avenues of
people cannot be compared with that of the riches, but it is the
growth for the marketers. In return, poor shall get the prod-
collective purchasing power of the poor by virtue of which
uct(s) they need. But to do so, one has to ‘listen’ to the poor
they represent a substantial purchasing power. The aggregate
purchasing power of poor is estimated to be approximately US patiently and open mindedly (Smith, 2005, p. 155) and this
$5 trillion a year (Hammond et al., 2007) and qualifies to be a could be done through consumer behaviour research. But when
significant market. Recent scholarly work also justify that poor one looks towards the preexisting consumer behaviour research
are a significant market for the marketers and deserve attention of poor, there is nothing much that one can explore from. This
from the profit oriented business firms. scarceness becomes more severe when one looks for inductive
Prahalad and Hart (2002) state that business firms can make research, especially in the context of developing world where
profit even from the poor people. Prahalad (2005), in his best- most of the world poor population lives in.
selling book ‘The Fortune at the Bottom of the Pyramid’ detail This study bridges the gap existing in the pertinent literature
all the examples – ITC (e-choupal), Hindustan Liver Ltd. by evaluating the shopping orientations of poor towards price,
(Annapurna salt, candy and detergent), Cemex (housing), Casas quality and brand for fast moving consumer goods (FMCGs) in
Bahia, Jaipur foot, the Aravind eye care system, ICICI bank the context of one of the largest developing countries like India
and Voxiva – wherein the firms have earned substantial profits by exploring their consumer decision-making styles (CDMSs),
even from the people who live on mere US $2 per day. Kirch- especially their price consciousness, quality consciousness,
georg and Winn (2006), Hammond et al. (2007) and Silver- brand consciousness and brand loyalty. It also explores their
thorne (2007) also propound that poor are a substantial market post purchase behaviour and evidences the association between
for marketers. However, a few researchers (for example Kar- their post purchase behaviour and aforementioned CDMSs
nani, 2007; Jaiswal, 2008) argue that the poor can be a market through structural equation modelling.
only for basic necessities. But, poor are also found to be the As far as organization is concerned, this study consists of
buyers of other than the basic necessities like alcohol (Baklien five sections. First section provides a brief introduction of the
and Samarasinghe, 2004), costlier school education for their study while the second section reviews the pertinent literature.
kids (Tooley and Dixon, 2006), radio sets (Banerjee and Duflo, The third section attempts to the research methodology adapted

International Journal of Consumer Studies 00 (2016) 00–00 1

C 2016 John Wiley & Sons Ltd
V
The poor, FMCGs, consumer behaviour A. Kumar et al.

in the study whereas the fourth section analyses the data. The significant influence on CDMSs (Mafini and Dhurup, 2014;
last section concludes the findings. South Africa: generation Y consumers).
As explained above, the CDMSs have been explored across
Literature review different cultures, ages and generations but the decision-making
styles of economically poor consumers are still remain unex-
Consumer decision-making styles plored. In this study, the researchers evaluate their CDMSs,
CDMSs, also known as shopping styles (Tai, 2005; Hou and especially their price consciousness, quality consciousness,
Lin, 2006; Park et al., 2010), have been studied through vari- brand consciousness and brand loyalty.
Why not all the CDMSs?
ous approaches like the consumer typology approach (Darden
Though each CDMS, suggested by Sproles and Kendall
and Ashton, 1974; Moschis, 1976), the psychographic/lifestyle
(1986), has its own importance, the research pertaining to
approach (Wells, 1975; Lastovicka, 1982) and the characteristic
CDMSs has not always been conducted while taking all of
approach (Sproles, 1985; Sproles and Kendall, 1986). But
these eight altogether, that is, research has also been conducted
among all these, the characteristic approach based CDMSs, as
on a few of them as per the objectives and interests of the
proposed by Sproles (1985) and Sproles and Kendall (1986),
researchers (for example Sinha and Batra, 1999: price con-
are arguably the best explanatory measure (Lysonski et al.,
sciousness; Batra and Sinha, 2000: price consciousness; Alford
1996; Lysonski and Durvasula, 2013).
Initially, Sproles (1985) introduced a 50-item inventory to and Biswas, 2002: price consciousness; Lee, 2008: price con-
sciousness and brand consciousness; Rzem and Debabi, 2012:
assess the predilections through which the consumers make
price consciousness; Ye et al., 2012: brand consciousness and
their shopping choices. These 50 items were grouped into nine
brand loyalty, etc.). In this study as well, the authors were
traits and the principal component analysis was carried out for
interested in exploring the shopping orientations of poor
the purpose of confirmation. Although the factor analysis con-
towards price, quality and brand for FMCGs, so they selected
firmed only six of the traits, the unconfirmed ones were also
the concerned CDMSs i.e. price consciousness, quality con-
considered similar to the confirmed ones. Later, Sproles and
sciousness, brand consciousness and brand loyalty on the basis
Kendall (1986) transformed the original 50-item inventory into
of their practical relevance to the research at hand and the liter-
a 40-item consumer style inventory (CSI). This time, the factor
ature review on these four is described as follows:
analysis of the transformed 40-item confirmed eight traits: per-
fectionism/high-quality consciousness, brand consciousness,
Price consciousness
brand loyalty, price consciousness, impulsiveness, recreational
shopping consciousness, novelty-fashion consciousness and The term ‘price consciousness’ was first used in 1961 when
confusion by over-choice. Gabor and Granger (1961, p. 172) defined it as the extent to
To check its applicability, Hafstrom et al. (1992) compared which the buyers are conscious of a product’s price during the
the CDMSs of American and Korean consumers and revealed purchase. Later, Monroe and Petroshius (1981, p. 44) articu-
that CSI instrument possesses the ingredients of construct valid- lated it as the degree to which the buyers are unwilling to pay
ity and can be used worldwide. Further, Durvasula et al. (1993) more for a product and may even refrain from buying it if they
evaluated the cross-cultural applicability of the CSI scale in the find the price to be greater than what is acceptable to them.
context of New Zealand and found that the scale could be used Subsequently, the term ‘price consciousness’ was continu-
for cross-cultural evaluations. Since then, the CSI instrument ously used in the marketing literature to represent buyer
has been used worldwide to explore the CDMSs of the consum- response to the price but its definitions have been a little incon-
ers. For instance: Lysonski et al. (1996) compared the CDMSs sistent. Sproles and Sproles (1990) defined it as the shopping
of the undergraduate students living in Greece, US, New Zea- predilection of buyers towards the buying of products at ‘sale
land and India; Leo et al. (2005) compared the CDMSs of Sin- prices and lower prices in general’ while the American Market-
gaporean and Australians; Wesley et al. (2006) explored the ing Association (1995) defined it as the degree to which the
CDMSs in relation to the shopping mall behaviour in US; Kav- customers are ‘sensitive to differences in price between alterna-
kani et al. (2011) evaluated the CDMSs of Iranian consumers tive choices’. Though, the term ‘price consciousness’ has been
for clothes; Anic et al. (2012) explored the CDMSs of young defined a little bit differently by the researchers but the sole
consumers in Bosnia and Herzegovina; Lysonski and Durvasula concept has been unanimous and it has been being used as a
(2013) compared the CDMSs of Indian young adults through perfect measure of buyers’ response to prices.
the longitudinal analysis of data from 1994 to 2009. As far as poor are concerned, most of the price related
A few studies of CDMSs were also conducted to address the research on poor has been centripetal to the two themes – (i)
association between CDMSs and demographics like gender, do the poor pay more? and (ii) why do the poor pay more?
and age etc. Mitchell and Walsh (2004: Germany) compared Alcaly and Klevorick (1971: US) observed that price of each
the CDMSs of male and female customers and found that male of 32 product categories studied was higher at small independ-
shoppers are somewhat less perfectionists, slightly less fashion ent stores than it was at chain stores and poor, by doing most
conscious and less confused by over-choice. Later, the differ- of their shopping at small independent stores, pay more for
ence in CDMSs of male and female shoppers was also their food purchases. MacDonald and Nelson (1991), in their
observed in the generation Y context (Hanzaee and Aghasibeig, study on 10 large metropolitan areas of US, revealed that, on
2008: Iran; Mokhlis and Salleh, 2009: Malaysia). As far as age an average, the poor pay more because the prices in a poverty
is concerned, it has been revealed that age does not have any area store were about 2% higher than prices in other area

2 International Journal of Consumer Studies 00 (2016) 00–00

C 2016 John Wiley & Sons Ltd
V
A. Kumar et al. The poor, FMCGs, consumer behaviour

stores. They also found that most of the urban poor reside in As far as poor are concerned, Goldman (1976), in his study
central city areas and the products are 4.2% costlier than subur- on Jewish people of Israel, found that quality is the most
ban prices. On the same lines, Chung and Myers (1999) found important reason for poor for buying furniture at a particular
that the poor in Minnesota, US pay more for grocery products. store. It shows that poor are quality concerned and are quality
They found that the main reasons behind this were – (i) the conscious consumers. Contrary to it, Leibtag and Kaufman
presence of less numbers of chain stores in the poor areas and (2003) revealed that poor households in US economize their
(ii) higher prices by poor area stores, i.e. convenience or small food purchases by purchasing lower quality food products. In
grocery stores. On the other hand, Hayes (2000), in their study China also, the low-income households, as compared to their
on New York, US, found that the overall food basket costs 2% high-income counterparts, demand relatively low-quality food
lesser in poor areas than the upper-income areas but cereal, products (Gale and Huang, 2007; Huang and Gale, 2009).
orange juice, apples and bananas are significantly costlier in It suggests that only a very few studies have been carried out
poor areas. on the poor regarding their quality preferences and moreover,
The poor are aware of the fact that the products in their the findings have also not been unanimous. So, to generalize
impoverished region are costlier than in affluent localities. But, the findings on quality consciousness of the poor, the hypothe-
because of mobility and transportation barriers, they are left sis to be tested in the context of India is
with no choice but to shop in the area where they live in H2: the poor are not quality conscious in their non-durable
(Andreasen, 1993). So, as compared to their non-poor counter- purchases.
parts, the poor pay more for most of their purchases. In spite of
this fact, fruits and vegetables were affordable to the poor in Brand consciousness
the quantities they normally buy (Dibsdall et al., 2003).
Brand consciousness is the mental disposition of a buyer due to
A few studies have also been conducted on the poor about
which s(h)e purchases the product(s) of well-known or national
their strategies in coping with the prices. Leibtag and Kaufman
brand(s) rather than the store or distributors’ brand(s) (Bruner
(2003), in their study in US, revealed that the poor purchase
II, n.d.-a). Brand consciousness may be defined as the ‘consum-
less expensive fruits and vegetables. Later, Gbadamosi (2009),
er’s orientation towards the purchase of expensive and well-
in his study on the low income people of metropolitan Greater
known brands’ (Leo et al., 2005).
Manchester, England, revealed that low-income consumers are
As far as the poor are concerned, Jones and Zufryden
responsive to groceries’ sales promotions and do compare pro-
(1982), in their research on Californian households’ purchase
motional offers in their neighbouring grocery stores in order to
behaviour, revealed that in each price category, brand penetra-
economize on grocery purchases. It suggests that the poor are
tion in low income households is more than the higher income
price conscious. A few researches like Batra and Sinha (2000)
households. They also ascertained a positive relationship
also supports this notion wherein it has been proved that the
between the presence of children in poor households and the
price consciousness and income are negatively related i.e. poor
number of brand purchases, that is, a family with children pur-
are supposed to be more price conscious.
chases more numbers of brands than a childless family pur-
Though the research, as discussed above, infers that the poor
chases. Leibtag and Kaufman (2003) revealed that low-income
are price conscious, the degree of their price consciousness is
households in US cope with brand and affordability issues by
yet to be known. Apart from it, the price consciousness of the
buying private-label brands of fixed-weight cheese, and ready
poor is yet to be statistically proved as the nature of the
to eat cereal. Elliott and Leonard (2006), in their research on
research on the poor on the concept has largely been qualitative
Britain’s poor households, found that, in order to satisfy their
and has not at all been quantitative. Further, the price con-
kids’ demands regarding the branded shoes, the parents even
sciousness of the poor is yet to be explored in the context of
spend more on branded shoes as they have little knowledge
developing world, where the bulk of the poor population found
about less known brands.
in. So, to these ends, the hypothesis framed in the context of a
On the same lines, Hamilton and Catterall (2007, 2008), in
developing country like India is
their exploratory research on Britain’s poor households, found
H1: the poor are price conscious in their non-durable
that the needs and wants of the children was the focal point
purchases.
around which the whole household’s consumption is structured.
Hamilton and Catterall (2007) found that mothers in poor
Quality consciousness
households try to minimize poverty’s negative impact on their
Generally, not all consumers purchase products on the bases of children, and their way of doing it is by purchasing branded
their prices i.e. some make their purchases after evaluating the products for their young ones. Hamilton (2009), in her research
quality of the products. The more cautiously they search for on UK’s low-income households revealed that children are
quality products, the more quality conscious they are (Radder highly motivated by branded clothing and they do not shop at
et al., 2006). So, quality consciousness is the degree to which discount stores as it could malign their reputation among their
the customers make efforts in purchasing ‘products perceived peers. This shows that the youngsters are not only brand con-
to be of the highest quality’ (Bruner II, n.d.-b). Sproles and scious but also image conscious. Boston Consulting Group
Sproles (1990) define quality consciousness as an extent up to (2012, p. 27), in its in-depth analysis of the consumption pat-
which a buyer cautiously searches for the best quality in the tern of Indian customers, found that 66% of Indian poor buy
products. any from a basket of preferred brands in food and groceries.

International Journal of Consumer Studies 00 (2016) 00–00 3

C 2016 John Wiley & Sons Ltd
V
The poor, FMCGs, consumer behaviour A. Kumar et al.

Price The most important determinant of satisfaction is consumption

Consciousness
process, because dissatisfaction may be caused even by a good
Quality product if the consumer does not know how to consume it
Consciousness Consumer Decision Post purchase
Making Styles behavior
properly (Blackwell et al., 2001, p. 80).
Brand
Consciousness
It is hypothesized that customers’ satisfaction has a positive
association with their repurchase intentions (Oliver, 1980; Fang
Brand
Loyalty
et al., 2011). But repurchase intention is not limited only to the
brand. Researchers (Bansal and Taylor, 1999; Tsai and Huang,
Figure 1 Hypothesized model of post purchase behaviour and 2007) have also theorized that satisfaction of customers also
consumer decision-making styles. determine their patronization of the purchase point. In addition
to these, ‘recommendation’ is also one of the crucial facets of
post purchase behaviour and the phrase which states that satis-
The aforementioned studies show that poor purchase brands.
fied customers tell three people and unsatisfied customers tell
It infers that they are brand conscious. But these inferential
3 000 (Blackshaw, 2008) justifies the same. So, post purchase
claims are yet to be verified statistically as poor are yet to be
behaviour aspect can be subdivided into three dimensions –
evaluated on the brand consciousness measure/scale. To do so,
repurchase intentions regarding the product (brand), repurchase
the hypothesis to be tested in the Indian context is
intentions regarding the purchase point and recommendation to
H3: the poor are brand conscious in their non-durable
the others.
purchases.
In the case of the poor, Tate (1961) was the first who
explored the post purchase behaviour of the poor. He revealed
Brand loyalty
that low-income households patronage the purchase point and
Brand loyalty is the degree to which a customer expresses loy- are one-shop shoppers while middle and upper income level
alty to specific brand(s) in the concerned product category households do not patronage and are disloyal to the purchase
(Bruner II, n.d.-a). It is that characteristic of the consumers that points. Later, Viswanathan (2007) also found that the poor
make them to repetitively purchase the same brands which are patronize the purchase point and one of the main reasons
their favourite (Sproles and Sproles, 1990). Brand loyalty is behind the patronizing act was bonding, that is the poor try to
defined as a ‘deeply held commitment to rebuy or repatronize a have a bonding with neighbouring retailers so that they can
preferred product/service consistently in the future, thereby avail credit facility in times of hardship. Though, patronizing
causing repetitive same-brand or same brand-set purchasing, the purchase point is one of the dimensions of post purchase
despite situational influences and the marketing efforts having behaviour, post purchase behaviour of the poor has largely
the potential to cause switching behaviour’ (Oliver, 1999). been explored on this point only. To bridge this literature gap
As far as the brand loyalty of poor is concerned, a very few and generalize the findings on the aspect of store’s store, fol-
studies are existing on poor regarding this aspect. Gbadamosi lowing hypotheses may be formed to test in the context of a
(2009) was the first to explore the brand loyalty of poor regard- developing country like India is
ing grocery purchases. In his study on UK’s poor, he revealed H5: If satisfied with a FMCG brand, poor repurchase it and
that poor people are not brand loyal in their grocery purchases. revisit the same store.
Later, Boston Consulting Firm (2012, p. 27) also explored the H6: Poor recommend FMCG brand(s) to others.
brand loyalty aspect of poor in the Indian context and revealed
that 95% of the poor are not brand loyal for their food and gro- CDMSs and post purchase behaviour
cery purchases. Though, these studies conclude that poor are
The CDMSs significantly influence the entire consumer behav-
not brand loyal for their food and grocery purchases, the fact is
iour (Chen et al., 2012; Hunjra et al., 2012), and post purchase
that the studies are very few in number and are of exploratory
behaviour, being the constituent part of consumer behaviour, is
nature. To test the statistically significance of the brand loyalty
also expected to be influenced by the CDMSs. But till the date,
of the poor in non-durables and generalize these findings, the
there is a paucity of literature regarding the association
hypothesis framed in Indian context is as follows
between the post purchase behaviour and the purchase
H4: the poor are not brand loyal in their non-durable
decision-making style. Further, price consciousness, quality
purchases.
consciousness, brand consciousness and brand loyalty, being
the constituents of the CDMSs construct, also need be verified
Post purchase behaviour
as the constructive measures of the ‘CDMSs’ in the present
Post purchase behaviour is that part of consumer behaviour context.
whereat the consumers actually consume the products/services So, to ascertain the abovementioned associations in the con-
and evaluate their performance (Assael, 2004, p. 44). Schiffman text of the poor, the researcher proposed a hypothetical model.
and Kanuk (2004, p. 571) state that there can be three possible The model exhibits that CDMSs is a second order construct
evaluations, viz. a neutral feeling, satisfaction and dissatisfac- comprising of price consciousness, brand consciousness, quality
tion. A neutral feeling is experienced if the actual performance consciousness and brand loyalty and positively influences the
matches the expectations while satisfaction is experienced if post purchase behaviour (refer to Fig. 1).
performance exceeds the expectations. On the other hand, dis- The hypothesized model led to the formulation of following
satisfaction is caused if performance is below the expectations. research hypotheses:

4 International Journal of Consumer Studies 00 (2016) 00–00

C 2016 John Wiley & Sons Ltd
V
A. Kumar et al. The poor, FMCGs, consumer behaviour

Table 1 Cronbach’s alpha values

Construct Bathing soap Cooking oil Tea Tooth-paste Washing soap

Price consciousness 0.800 0.797 0.800 0.802 0.797

Quality consciousness 0.741 0.706 0.748 0.741 0.738
Brand consciousness 0.709 0.729 0.715 0.704 0.735
Brand loyalty 0.884 0.928 0.893 0.883 0.887
Recommendations to others 0.921 0.919 0.918 0.921 0.921

Source: Primary data.

H7: ‘CDMSs’ is specified as a second order construct of reckoned to be the most robust measure of CDMSs (Lysonski
price consciousness, brand consciousness, quality et al., 1996; Lysonski and Durvasula, 2013). In this study,
consciousness and brand loyalty. CDMSs constructs were adapted from Sproles and Sproles
H8: ‘CDMSs’ positively influences the post purchase (1986) and Sproles and Sproles (1990). Price consciousness
behaviour. construct was comprised of five items while remaining three
CDMSs were comprised of four items each (refer to appendi-
Research methodology ces). Further, all the items were measured on a 5-point Likert
scale (1 5 strongly disagree and 5 5 strongly agree).
Research design The post purchase behaviour aspect was comprised of two
constructs – repurchase intention and recommendations to
A two stage research design – an exploratory research design
others. There were two items in each of these two constructs
at the initial stage, and a descriptive research design at the sub-
(refer to appendices) and all the items were measured on 5-
sequent stage – was used in this study. To begin with, an
point Likert scale (1 5 strongly disagree and 5 5 strongly
exploratory research design was used through extensive litera-
agree). The research instrument used in the research was
ture survey, in-depth interviews and focus group discussions to
adapted from Fang et al. (2011), Tsai and Huang (2007) and
understand and define the research problem. It also provided
Gupta and Stewart (1996) and may be found in the
insights to the researcher regarding the purchase pattern of the
Appendices.
poor, which further facilitated the identification of the variables
on which the consumer behaviour of poor was largely depend- Reliability assessment
ent on. Subsequently, a descriptive research design (single
cross-sectional) was used to describe the consumer behaviour The extent to which a scale yields consistent results on
of poor on the concerned variables. repeated administrations is known as the reliability of a scale.
The Cronbach’s alpha method is the most widely used method
Data collection method for reliability estimation (Salkind, 2010, p. 160; Bindak, 2013).
The thumb rule for a scale to be internal consistent is to have
To have insights about the data collection method, the existing value of more than 0.7 of Cronbach’s alpha (Nunnally, 1978).
literature was exhaustively reviewed (for example Greenberg In this study, the Cronbach’s alpha values of all five measures
and Dervin, 1970; Goldman, 1976; Gorn and Goldberg, 1977; were above 0.7 (refer to Table 1). So, each measure was found
Robinson et al., 2000; Hamilton and Catterall, 2007; Gbada- to be reliable.
mosi, 2009; Hamilton, 2009; Hamilton, 2011; Bacha and Neto,
2014). The review of research methodology espoused in exist- Validity assessment
ing literature showed that data on the consumer behaviour of
poor have been mostly collected through in-home interviews Validity of an instrument is defined as the degree to which it
(for example Greenberg and Dervin, 1970; Goldman, 1976; accurately measures the characteristics it is made to measure
(Aaker and Day, 1980, p. 227). Validity of a measure is subdi-
Gorn and Goldberg, 1977; Robinson et al., 2000; Hamilton and
vided into two parts – content validity and construct validity.
Catterall, 2007; Gbadamosi, 2009; Hamilton, 2011). So, on the
Content validity, also known as face validity (Hair et al.,
basis of these insights, data for the present study was collected
2006a,b, p. 355), is the evaluation of how adequately the mea-
through in-home interview method as well and a structured
sure assesses the domain of the construct (Churchill et al.,
questionnaire was used as the research instrument.
2009, p. 248). Content validity cannot be measured mathemati-
cally. It largely depends on how accurately the researcher
Research instrument
understands the domain of the measure and then how efficiently
The research instrument was developed in accordance with the s(h)e defines it. In the present research, an exhaustive literature
proposed research framework and was composed to collect review was carried out to understand the domain of the instru-
information on two broad aspects – CDMSs and the post pur- ment. The opinions of experts were also sought in expressing
chase behaviour. the domain in the form of statements/items/variables. These
When the CDMSs are talked about, the work of Sproles and experts were a group of three professors and practitioners
Kendall (1986) and Sproles and Sproles (1990) are considered working in the fields of Economics and Business Statistics. All
to be the most seminal and the CSI proposed by them is of them were from a well-known Indian B-school and were

International Journal of Consumer Studies 00 (2016) 00–00 5

C 2016 John Wiley & Sons Ltd
V
The poor, FMCGs, consumer behaviour A. Kumar et al.

Table 2 AVE scores of the constructs

Construct Bathing soap Cooking oil Tea Tooth-paste Washing soap

Price consciousness 0.733 0.716 0.689 0.712 0.738

Quality consciousness 0.748 0.732 0.733 0.727 0.736
Brand consciousness 0.718 0.773 0.688 0.663 0.767
Brand loyalty 0.827 0.812 0.820 0.821 0.822
Recommendations to others 0.879 0.865 0.876 0.844 0.874

Source: Primary data.

part of the expert panel formed to provide assistance to this member, who usually takes the purchase decisions for selected
research. They not only proposed the CSI inventory as a categories, was taken as the sample element.
research instrument for this research, but also suggested how it
could be better used in the context of Indian poor by its Hindi Why Delhi is taken as the area of study?
language transformation.
Construct validity, on the other hand, is mainly measured by con- In India, Delhi was selected as the area for this study because,
vergent validity and discriminant validity. Aaker and Day (1980, p. apart from being the most populous area in India, it is the sec-
228) claim that construct validity of a measure cannot be reckoned ond most populous city in the world. As far as poor population
until it’s convergent and discriminant validities are assured. Conver- is concerned, it is home to 1.696 million poor people (refer to
gent validity is the extent to which an item positively correlates with Table 3) and records related to the poor are easily accessible.
other items of a construct (Churchill, 1979, p. 70). It is measured by Further, being the capital city of India, it has become a minia-
average variance extracted (AVE) score. An instrument is said to be ture abode for the poor of different cultures, traditions and lan-
convergent valid if its AVE score is more than 0.5 (Fornell and guages of country due to migration from all corners of India.
Larcker, 1981). In the present research, AVE score of each construct
for each of the selected product categories was more than 0.5. It Why BPL Family is taken as the sample unit?
assured the convergent validity of the research instrument (refer to
Table 2). Poor are those set of people whose income level falls short of
As far as discriminant validity is concerned, it is the extent predetermined poverty line (Bourguignon, 2006, p. 80). In
to which a construct significantly differs from other constructs India, per capita poverty lines are calculated separately for
that are actually different to it (Hair et al., 2006a,b, p. 354). It urban and rural areas at national and state levels (Department
is measured by making the comparisons of the inter-construct of Planning, 2009, p. 238). Subsequently, the people whose
correlation coefficients and respective square root values of family income is less than the weighted poverty line (no. of
AVEs (Bove et al., 2009). An instrument is said to be discrimi- family members multiplied by the per capita poverty line of the
nant valid if the square roots of AVE of all the constructs are area the family lives in) of their living area are declared as
greater than their respective inter-construct correlation coeffi- poor. For example – the per capita poverty line in Delhi in a
cients (Hair et al., 2006a,b, p. 778). In present research, square year is 1145  per month, then a family of five is declared poor
roots of AVEs of all the constructs for each of the selected if its monthly income was less than 5725 , and a family of 10
product category were greater than their respective inter- members is declared poor if the aggregate monthly income of
construct correlation coefficients. Hence, the discriminant valid- all its family members is less than 11 450.
ity of the instrument was proved. Poor in India are further categorized into two subclasses –
BPL and poorest of poor – wherein the income of BPL people
Area of the study & sample unit is more than that of the other. Apart from the income, the pop-
To explore the consumer behaviour of poor in Indian context, ulation strength of BPL families is also more than to that of
the metropolitan area of Delhi was selected as the area of the poorest of poor families (Department of Food, Civil Supplies &
study and the latest below poverty line (BPL) family list of Consumer Affairs, Himachal Pradesh, 2012). In Delhi too, the
Delhi, available at the webpage of Department of Food & Sup- BPL people are not only financially better off than poorest of
plies, Government of Delhi, was taken as the sampling frame. poor but they also represent a larger part of overall poor popu-
BPL family was taken as the sample unit and the BPL family lation (refer to Table 4).

Table 3 Monthly per capita poverty line of Delhi, India (2011–2012)

Rural Delhi Urban Delhi

Monthly per capita Monthly poverty line Monthly per capita Monthly poverty line Total number of BPL
Year poverty line for a five member family poverty line for a five member family people in Delhi

2011–2012 1145 5725 1134 5670 1.696 million

Source: Planning commission, 2013.

6 International Journal of Consumer Studies 00 (2016) 00–00

C 2016 John Wiley & Sons Ltd
V
A. Kumar et al. The poor, FMCGs, consumer behaviour

Table 4 Categorization of poor population in Delhi, India rule approach. In the case of confidence interval approach
S. No. Category Population strength
(mean) (Black, 2008, p. 277; Malhotra and Dash, 2009, p.
371), the population standard deviation was unknown and it
1 Below poverty line (BPL) 2 44 262 families was calculated through a pilot survey of 30 poor families. The
2 Poorest of poor 1 29 685 families highest item standard deviation was 1.601 (refer to Table 5).
Source: Department of Food & Supplies, Government of Delhi, Delhi Thus, the sample size with D 5 0.05, z 5 1.96 and r 5 1.601
(2012). was calculated to be 196.
Thumb rule approach states that the sample size must be at
least four or five times of the number of variables in a study
So, taking the above mentioned factors into account, the peo- (Basilevsky, 1994). The total number of variables in present
ple belonging to BPL families were taken as the target popula- study was 17 and the minimum sample size was calculated to
tion for the present study and a BPL family was considered as be 85 (5 17 3 5).
the sample unit. So, by considering the confidence interval approach (mean)
and thumb rule approach along with the insights from the sam-
Product categories selected ple sizes of the existing studies as the approximate estimates
for the sample size, 360 was selected as the sample size for
The focus group discussions, besides being insightful of the this study.
purchase pattern of the poor, also assisted the researcher in list-
ing the packaged goods that poor were mostly the consumers Sampling procedure
of. Subsequently, a sample of five product categories was
selected in such a way that the product categories of FMCGs To have a representative sample, the sampling procedure was
like food & beverages, household care and personal care could carried out in two parts – short listing of constituencies and
get an overall representation. The final sample was comprised proportionate sampling. Both the parts of the sampling proce-
of five product categories – bathing soap, cooking oil, tea, tooth dure are explained as follows:
Delhi consists of 70 assembly constituencies (Election Com-
paste and washing soap. The usage amount of these product
mission of India, 2014). These constituencies are also referred
categories was also statistically substantial in the poor house-
as circles. For a better public distribution system (PDS), Gov-
holds (National Council of Applied Economic Research, 2005,
ernment of Delhi has established a separate food & supply
p. 143).
office in each of these 70 constituencies.
Taking the constraints of time and resources into account, 12
Sample plan
constituencies were surveyed in this study. The selection proce-
To choose the appropriate size, the approached used in this dure of these constituencies is explained as follows. In the first
study were: confidence interval approach (mean) and thumb step, the BPL family list of Delhi was sorted to find the

Table 5 Item standard deviation

Construct Item Bathing soap Cooking oil Tea Tooth-paste Washing soap

Price consciousness PC1 1.311 1.348 1.278 1.311 1.251

PC2 1.095 1.124 1.167 1.085 1.106
PC3 1.324 1.357 1.322 1.324 1.324
PC4 1.450 1.487 1.431 1.450 1.450
PC5 1.413 1.601 1.502 1.413 1.423
Quality consciousness QC1 0.907 0.547 0.898 0.907 0.907
QC2 1.119 1.251 1.196 1.119 1.119
QC3 0.850 0.868 0.850 0.850 0.850
QC4 0.999 1.055 0.999 0.999 0.999
Brand consciousness BC1 0.730 0.803 0.73 0.730 0.730
BC2 0.691 1.114 0.973 0.691 0.691
BC3 0.907 1.326 1.033 0.907 0.907
BC4 1.251 1.251 1.251 1.251 1.251
Brand loyalty BL1 0.547 0.535 0.547 0.547 0.547
BL2 0.834 0.994 0.834 0.834 0.834
BL3 1.135 1.135 1.135 1.135 1.135
BL4 1.029 1.083 1.015 1.029 1.029
Post purchase behaviour PPB1 0.551 0.626 0.551 0.551 0.551
PPB2 1.332 1.373 1.373 1.332 1.332
PPB3 1.351 1.426 1.426 1.388 1.388
PPB4 1.392 1.392 1.392 1.392 1.392

Source: Primary data.

International Journal of Consumer Studies 00 (2016) 00–00 7

C 2016 John Wiley & Sons Ltd
V
The poor, FMCGs, consumer behaviour A. Kumar et al.

Table 6 Proportionate sampling procedure

Sr. Geographical No. of BPL No. of BPL

No. District area* (Km2) families Constituency families Sample size

1 North West Delhi 440.31 28 363 Narela 4453 45

Rithala 463 05
Bawana 7957 80
2 North Delhi 59.16 31 957 Timarpur 760 10
Adarsh Nagar 9388 95
Shakur Basti 1194 15
3 Central Delhi 24.68 8536 Chandni Chowk 857 10
4 West Delhi 129.52 14 808 Rajouri Garden 1698 20
5 South West Delhi 420.54 9337 Dwarka 748 10
6 South Delhi 249.85 11 493 Deoli 2183 25
7 East Delhi 63.76 9994 Vishwas Nagar 1603 20
8 North East Delhi 60.29 16 830 Seelampur 2441 25
Total 139 567 139 567 360

Source: Compiled by the researcher, *forest survey of India (http://www.fsi.nic.in/sfr2003/delhi.pdf).

number of BPL families in each of the 70 constituencies. In the the FMCG purchase decisions are made by the females. The
second step, all the constituencies were listed in an ascending majority of the respondents are in the age group of 21–40 years
order on the basis of the number of BPL families. The third (50.5%). The data collected in this study shows that most of
step was for the categorization, where first 24 constituencies the purchase decision makers are illiterate. As far as family
were grouped under low BPL strength category, next 23 con- size is concerned, most of the surveyed families are large sized
stituencies were put under moderately populated category and (6–10 members).
the last 23 constituencies were grouped under highly populated
category. In the third step, all the constituencies were well Consumer decision-making styles
mixed up within their respective categories and four constituen- The overall mean scores of CDMSs for the selected product
cies were randomly chosen from each of these three categories. categories were as follows – quality consciousness: 3.254;
After selecting the constituencies, a proportionate sample brand consciousness: 3.456; brand loyalty: 3.650; and price
size was allotted to each constituency (refer to Table 6) and a consciousness: 3.341 (refer to Table 8). To check the statistical
random sample of the allotted strength was taken out from significance of the scores, one-sample t-test was used and it
each of the shortlisted constituencies. The proportionate sam- was found that the scores were statistically significant
pling procedure is described as follows – first, the strength of (P < .001) in each of these CDMSs. Thus, H1 (which stated
the constituency was made a four digit number i.e. if it was a that the poor are price conscious in their non-durable pur-
three digit number; a zero was prefixed to it (refer to Table 6). chases), H2 (which proposed that the poor are not quality con-
Second, the first two digits numbers were taken and adjusted to scious in their non-durable purchases) and H3 (which phrased
make them the multiples of 5 to reach at the sample size for that the poor are brand conscious in their non-durable pur-
the concerned constituency. For example – the BPL strength of chases) are failed to be rejected. Further, H4, which assumed
the Narela circle was 4453. Here, first two digits number was that the poor are not brand loyal in their non-durable purchases,
44. It was adjusted to 45 to reach at the sample size for Narela is rejected.
(Figure 2). But, Black (2008, p. 293) cautions that statistical significance
of results may lead to deceptive interpretations if their substan-
Data analysis tiality is not ascertained. In this study, the magnitude of overall
mean scores of all the CDMSs, except brand loyalty, is in
The analysis of the collected data was descriptive as well as
between 3 and 3.5 i.e. it is marginally deviated from the neutral
inferential. In the case of descriptive analysis, mean was used
point of 3 (refer to Table 8). It evinces that even though the
the measure of central tendency while standard deviation was
results of all the CDMSs are statistically significant, they are
used as the measure of dispersion. On the other hand, the infer-
not substantive. So, it may be interpreted that (i) poor are mod-
ential analysis was carried out through t-test (one sample), and
erately brand loyal; (ii) poor are marginally favourable to price
chi square test. The statistical tools used to carry out the data
V R V R V R consciousness, quality consciousness and brand consciousness.
analysis were IBMV R SPSS 16.0 and IBM SPSS Amos 20.0.
The findings regarding brand loyalty, brand consciousness
and quality consciousness may be so because of the factors like
Results and analysis nature of selected product categories and purchase size. It may
be that the poor consider brand as a cue to quality and by
The demographic profile of the respondents
reducing their purchase sizes, which is relatively easier to do in
The demographic profile of the respondents is presented in the the selected product categories, they are able to afford the
Table 7. In a little more than half of the poor families (56.1%), brands. By doing this repetitively, they might have shortlisted a

8 International Journal of Consumer Studies 00 (2016) 00–00

C 2016 John Wiley & Sons Ltd
V
A. Kumar et al. The poor, FMCGs, consumer behaviour

Figure 2 Surveyed area. [Colour

figure can be viewed at
wileyonlinelibrary.com]

few brands they are most gratified with and may have become
Table 7 The demographic profile of the respondents (n 5 360)
loyal to them.
As far as price consciousness is concerned, the poor should Demographics Categories Frequency %
be substantially price conscious, but the findings suggest that
Age Up to 20 years 51 14.2
they are marginally price conscious. It might be so because of
21–40 years 182 50.5
price awareness and brand loyalty. Price consciousness measure 41–60 years 105 29.2
was consisted of three types of attitudes (cognitive, affective 61–80 years 22 06.1
and conative) towards low price of FMCGs. All these three ele- Gender Male 158 43.9
ments were well associated with price awareness. But product Female 202 56.1
categories selected in this study are those which are used in Education Illiterate 117 32.5
most of the poor households in Delhi. In addition to it, pur- 1st-primary 54 15.0
chase frequency of most of these products is mostly once or 6th to high school 114 31.7
twice a week and poor are largely aware about the prices of 11th to senior secondary 46 12.8
selected product categories. As a result, the price awareness Undergraduate to post graduate 29 07.7
has belittled the degree of price consciousness and has pushed Occupation Housewife 147 40.8
it towards the neutral point. Further, brand loyalty also appears Labour (daily wages 1 regular) 50 13.9
Student 45 12.5
to have influenced the price consciousness. It seems that poor
Shopkeeper 34 9.4
have shortlisted a few brands they are most satisfied with and
Private job (other than labour) 28 7.8
have become loyal to them. As a result, they do not do much
Sewing work 12 3.3
price-quantity comparisons of various brands across different Housemaid 15 4.2
shops and their price consciousness score is belittled. Hawker 18 5.0
Others 11 3.1
Post purchase behaviour Family size Small size family (1–2) 15 4.2
The mean scores of repurchase intentions regarding both the Medium size family (3–5) 152 42.2
Large size family(6–10) 163 45.3
aspects – brand and purchase store – were more than 3 but the
Very large size family (11–20) 30 8.3
mean score of recommendations aspect was less than 3 (refer
to Table 9). When checked for the statistical significance, all Source: primary data.

International Journal of Consumer Studies 00 (2016) 00–00 9

C 2016 John Wiley & Sons Ltd
V
The poor, FMCGs, consumer behaviour A. Kumar et al.

Table 8 Consumer decision-making styles

One-sample t-test**

Consumer decision-making style N Overall mean* Standard deviation T P

Quality consciousness 360 3.254 0.7772 06.198 .000

Brand consciousness 360 3.456 0.6910 12.528 .000
Brand loyalty 360 3.650 0.8561 14.409 .000
Price consciousness 360 3.341 0.7984 08.114 .000

Source: Primary data. *Overall mean score of the selected product categories on 5-point Likert scale (1 5 strongly disagree and 5 5 strongly agree).
**Test value 5 3.

Table 9 Post purchase behaviour

t-test**

Facet of post purchase behaviour N Overall mean* Standard deviation T P

Repurchase intention1 360 3.8317 0.6463 24.517 .000

Repurchase intention2 360 3.5001 1.2786 6.859 .000
Recommendations to others 360 2.8797 1.0487 22.176 .030

Source: Primary data. 1Repurchasing the FMCG brand one is satisfied with, 2revisiting the store one bought the satisfied brand from. *Overall mean
of selected product categories on a 5-point Likert scale (1 5 strongly disagree and 5 5 strongly agree). **Test value 5 3.

the scores were statistically significant (P < .05) and H5, which PC2 respectively (refer to Table 10). Third, in the quality con-
posited that if satisfied with a FMCG brand, poor repurchase it sciousness construct, QC4 with an estimate of 0.870 was the
and revisit the store where they bought it from, was supported. most substantially contributing variable, QC3 (0.820) was the
H6, which stated that poor recommend the FMCG brand to second most contributing variable while QC1 (50.545) was the
others, was rejected (refer to Table 9). But when checked for
the substantiality, the results were found to be substantive only
for repurchase intention. So, it may be concluded that poor are
meagerly unfavourable on recommendations aspect but they
repurchase the brand they are satisfied with and revisit the store
they bought the brand from.

CDMSs and post purchase behaviour

The analysis regarding the association between CDMSs and
post purchase behaviour is done by the means of the analysis
of three models, i.e. confirmatory factor analysis (CFA) model,
measurement model and structural model, and the orderly anal-
ysis of each of these three models is described as follows.

Confirmatory factor analysis (CFA) model

Confirmatory factor analysis was carried out to confirm CDMSs
as a reflective measure of price consciousness, quality con-
R
sciousness, brand consciousness and brand loyalty. IBMV
R
SPSSV Amos 20.0 was used to run the confirmatory factor
analysis (refer to Fig. 3).

Standardized Regression Weights’ Analysis of CFA Model

After analyzing the standardized regression weights, some of
the critical observations made are as follows. First, the rankings
of the CDMSs (refer to Table 10) constructs was as follows:
brand consciousness: 1st, brand loyalty: 2nd, quality conscious-
ness: 3rd and price consciousness: 4th. Second, in the price
consciousness construct, variables PC3, with a score of 0.731, Figure 3 Conformance of the consumer decision-making styles
has the highest parameter estimate followed by PC1, PC5 and construct. Source: Primary data

10 International Journal of Consumer Studies 00 (2016) 00–00

C 2016 John Wiley & Sons Ltd
V
A. Kumar et al. The poor, FMCGs, consumer behaviour

Table 10 Standardized regression weights of CFA model Table 12 Convergent and discriminant validity of measurement model

Path Estimate Inter construct

Square correlation
Consumer decision-making styles 0.915
Construct AVE root of AVE coefficients
! Brand Consciousness
Consumer decision-making styles 0.628 Price Consciousness 0.503 0.709 20.445, 20.634,
! Brand Loyalty 20.026, 20.337
Consumer decision-making styles 0.601 Quality Consciousness 0.576 0.759 0.416, 20.337,
! Quality consciousness 0.557, 20.002
Consumer decision-making styles 20.677 Brand Consciousness 0.514 0.717 0.550, 20.634,
! Price Consciousness 0.084, 0.557
PC1 Price Consciousness 0.718 Brand Loyalty 0.769 0.877 20.445, 0.550,
PC2 Price Consciousness 0.686 0.136, 0.416
PC3 Price Consciousness 0.731 Post Purchase Behavior 0.924 0.961 0.136, 20.026,
PC5 Price Consciousness 0.703 0.084, 20.002
QC1 Quality consciousness 0.545
Model fit summary of measurement model.
QC3 Quality consciousness 0.820
QC4 Quality consciousness 0.870
BC1 Brand Consciousness 0.654 Measurement model
BC2 Brand Consciousness 0.756 A measurement model, comprising of price consciousness,
BC3 Brand Consciousness 0.735
quality consciousness, brand consciousness, brand loyalty and
BL2 Brand Loyalty 0.931
post purchase behaviour, was drawn. All the constructs were
BL3 Brand Loyalty 0.926
allowed to freely correlate with each other (refer to Fig. 4).
BL4 Brand Loyalty 0.764
Reliability, validity and model fit of the measurement model
Source: Primary data. are explained as follows.

least contributing variable (refer to Table 10). Fourth, in the Reliability of Measurement Model
brand consciousness construct, the order of the variables on the The reliability of the measurement model was tested through the
basis of their parameter estimates was: BC2: 1st, BC3: 2nd, composite reliability method. The reliability scores for price con-
BC1: 3rd (refer to Table 10). Last, BL2, with a standardized sciousness, quality consciousness, brand consciousness, brand loy-
regression weight of 0.931, contributed the most in the explana- alty and post purchase behaviour were 0.752, 0.721, 0.715, 0.819
tion of brand loyalty construct whereas BL4, with a standar- and 0.912 respectively. So, in all the cases the reliability scores
dized regression weight of 0.764, was the least contributing were more than 0.7 and it proved the reliability of the measures.
variable.
Validity of Measurement Model
Model Fit Summary of CFA Model
The AVE of each construct was more than 0.05 (refer to Table
A model fit is well-disposed if CMIN/DF < 3/1 (Kline, 1998). 12) and it proved the convergent validity of the measures. As far
Apart from CMIN/DF score, the overall model-fit can also be
as discriminant validity is concerned, the square root AVE scores
evaluated by evaluating the standardized root mean square resid-
were compared with their corresponding inter construct correla-
ual (SRMR) score along with any of the fit indices out of
tion scores and it was revealed that the former one was greater
RMSEA, CFI and RNI (Hu and Bentler, 1999). Here, a model is
than the later ones. It approved the discriminant validity of all the
said to be good fit if SRMR is less than equal to 0.08, and either
concerned constructs.
CFI/RNI is greater than equal to 0.95 or RMSEA is less than
In the conformance model, the CMIN/DF, RMR, GFI, CFI
equal to 0.06 (Hu and Bentler, 1999). So, a better data fit model
and RMSEA were 2.645/1, 0.052, 0.923, 0.947 and 0.068
is that in which the values of GFI and AGFI are higher and the
respectively (refer to Table 13). It indicates that the concerned
values of SRMR and RMSEA are lower (Lei and Wu, 2007). In
model moderately fits the data.
the conformance model, the CMIN/DF, RMR, GFI, CFI and
RMSEA were 2.597/1, 0.045, 0.937, 0.957 and 0.063 respec-
Structural model
tively (refer to Table 11). It shows that the model significantly
supports the data. The structural model is exhibited in Appendix B and the find-
ings pertaining to this model i.e. parameter and model fit indi-
ces are discussed as follows.

Table 11 Model fit summary of CFA model Table 13 Model fit summary of measurement model

Model CMIN/DF RMR GFI CFI RMSEA Model CMIN/DF RMR GFI CFI RMSEA

Default model 2.597 0.045 0.937 0.957 0.067 Default Model 2.645 0.052 0.923 0.947 0.068

Source: Primary data. Source: Primary data.

International Journal of Consumer Studies 00 (2016) 00–00 11

C 2016 John Wiley & Sons Ltd
V
The poor, FMCGs, consumer behaviour A. Kumar et al.

Table 15 Model fit summary of the structural model

Model CMIN/DF RMR GFI CFI RMSEA

Default Model 2.590 0.054 0.927 0.952 0.067

Source: Primary data.

Model fit summary of the structural model

In the structural model, the CMIN/DF, RMR, GFI, CFI and
RMSEA were 2.590/1, 0.054, 0.927, 0.952 and 0.067 respec-
tively (refer to Table 15). It shows that the structural model
accurately fits the data.
The above discussion points that, H8, which stated that
CDMSs positively influence the post purchase behaviour, is
supported. Besides it, it depicts that, H7, which stated that
CDMSs is specified as a second order construct of the price
consciousness, brand consciousness, quality consciousness and
brand loyalty, is also supported.
It interprets that CDMSs positively influence the post pur-
chase behaviour and it is a second order construct of price con-
sciousness, brand consciousness, quality consciousness and
brand loyalty.

Conclusion
More than half of world’s people, about four billion, in this
21st century are economically poor. Individually they might
spend as meager as ten dollars in an annum (Emmons, 2007),
but collectively they represent a substantial market of US $5
trillion (Hammond et al., 2007; Subrahmanyan and Gomez-
Arias, 2008). On the other hand, FMCGs, with a worth of
Figure 4 Measurement model. Source: Primary data
$570.1 billion, constitute the fourth largest industries of the
world (About-fmcg.com, 2014) and the poor, by necessity,
Parameter Indices of the structural model
spend a significant amount of their income on FMCGs (Karn
The findings regarding parameter indices (refer to Table 14) of et al., 2003; Baklien and Samarasinghe, 2004; Banerjee and
the structural model depict that the magnitudes of all the criti- Duflo, 2007; Boston Consulting Group, 2012). As far as quan-
cal ratios were greater than 1.96 whereas each of the standard tity is concerned, there are a number of FMCGs whose con-
error values was small and precise. In the case of regression sumption amount in the poor households is equivalent to that
weights and squared multiple correlations, it is observed that in their non-poor counterparts’ (National Council of Applied
all the regression weights are significant at .001 levels and all Economic Research, 2005, p. 143). So, poor are a substantial
the squared multiple correlations were more than the recom- market for FMCGs and their purchase behaviour of poor
mended level of 0.300. regarding the concerned products should be explored.

Table 14 Parameter indices of the structural model

Normal Standardized Squared

regression regression Critical Standard multiple
Path weights weights ratio error P value correlations

Consumerdecisionmaking ! Price Consciousness 20.885 20.689 27.504 0.118 *** 0.474

styles
Consumerdecisionmaking ! Quality consciousness 0.949 0.592 7.421 0.128 *** 0.350
styles
Consumerdecisionmaking ! Brand Consciousness 0.975 0.871 7.155 0.136 *** 0.759
styles
Consumerdecisionmaking ! Brand Loyalty 1.000 0.626 – – – 0.392
styles
Consumerdecisionmaking ! Post Purchase 0.703 0.612 9.634 0.178 *** 0.375
styles Behavior

Source: Primary data. ***Significant at .001 level.

12 International Journal of Consumer Studies 00 (2016) 00–00

C 2016 John Wiley & Sons Ltd
V
A. Kumar et al. The poor, FMCGs, consumer behaviour

Though, in the recent years, the attention to explore the con- households were then substituted by other poor households of
sumer behaviour of poor has been increased, it has largely been the same locality. Third, FMCG sector consists of a large num-
restricted to the developed world only (for example Hayes, ber of product categories. Furthermore, the usage of these cate-
2000: New York; Kochuyt, 2004: Belgium; Attanasio and gories also varies across the poor. To overcome the usage
Frayne, 2006: Colombia; Elliott and Leonard, 2006: UK; Ham- limitation, five of the most commonly used product categories
ilton and Catterall, 2006: Ireland; Hamilton and Catterall, 2007: were selected for the study. So, the scope of the study is lim-
Britain; Hamilton and Catterall, 2008: UK; Gbadamosi, 2009: ited to these five categories only.
England; Hamilton, 2009: UK; French et al., 2010: US; Hamil- Thus, further studies may be conducted to explore the con-
ton, 2011: UK; Ethan et al., 2013: New York etc.). As a result, sumer behaviour of the poor in India while overcoming the afore-
the consumer behaviour of poor in the developing world, where mentioned limitations. Apart from this, some of the aspects the
most of the world’s poor reside in, is still largely unexplored. future research may also work upon are explained as follows.
Further, research design espoused in most of the studies on First, there exists scope for further studies covering some more of
poor has been exploratory. The sampling technique employed the FMCG product categories. This will not only allow one to
has largely been non-probability and sample sizes taken for validate the findings of existing research but will also pave the
data collection have been small. So, even by confronting with way to generalize the findings. Second, generally – price, quality,
some of the crucial aspects of consumer behaviour like price, brand and brand loyalty – are the variables that influence the con-
brand and quality, the findings from these studies have never sumer behaviour the most significantly. This phenomenon also
been inductive and inferential. holds merit in the case of economically weaker section people.
This study bridges these literature gaps by evaluating the So, due to the practical relevance of the shopping predilections
consumer behaviour of poor on the aspects of price, quality towards these four variables, this research was confined to the
and brand by assessing their CDMSs for FMCGs, namely price four CDMSs and the remaining CDMSs like novelty-fashion con-
consciousness, quality consciousness, brand consciousness and sciousness, impulsiveness, recreational shopping consciousness
brand loyalty. Second, it adds to the existing consumer behav- and confusion by over-choice were left unexplored. So, there
iour literature by evaluating their post purchase behaviour and exists scope for further studies on poor on the remaining CDMSs.
testifying its association with the specified CDMSs. Third, it is Further, the association between the post purchase behaviour and
based on a descriptive research design. The sampling technique CDMSs was largely unexplored. Though the present research has
employed is cluster (probability) sampling and sample size is tried to bridge this literature gap, yet to validate these findings
also adequately large. So, the findings are inductive, inferential there exist scope for further studies on the concerned aspect.
and statistically tested. Last, it is conducted in the context of Third, the present study explored the consumer behaviour of poor
India, a developing country and an adobe to the largest part of in isolation of the consumer behaviour of the non-poor and the
the world’s poor population, and expands the geographical poorest of poor. Thus, there exists scope for further studies in
extent of the consumer behaviour literature in its own way. which the consumer behaviour of poor could be explored in com-
The findings of this study regarding CDMSs for selected parison to the non-poor and poorest of poor. Last, the present
consumer packaged goods conclude that the poor in Delhi, research was confined to metropolitan area of Delhi and the find-
India are moderately brand loyal. They are marginally price, ings of this research are also confined to the metropolitan poor of
quality and brand conscious consumers. Further, the findings Delhi. So, there exists scope for further research to explore the
pertaining to post purchase behaviour suggest that the poor usu- consumer behaviour of poor in other metropolitan areas of India.
ally do not recommend brand(s) to others. They repurchase the
brands they are satisfied with and patronage the store they References
bought them from. As far as the association between CDMSs Aaker, D.A. & Day, G.S. (1980) Marketing Research, 2nd edn. Wiley,
and post purchase behaviour is concerned, it concludes that New York.
CDMSs is a second order construct of price consciousness, About-fmcg.com. (2014) FMCG industry facts. URL (http://www.about-
brand consciousness, quality consciousness and brand loyalty, fmcg.com/What-is-FMCG/FMCG-industry-facts (retrieved 21 June
and significantly influence the post purchase behaviour. 2012).
Alcaly, R.E. & Klevorick, A.K. (1971) Food prices in relation to income
Limitations and scope for further studies levels in New York City. The Journal of Business, 44, 380–397.
Alford, B.L. & Biswas, A. (2002) The effects of discount level, price con-
Though this study has been theoretically insightful, it has some sciousness and sale proneness on consumers’ price perception and
limitations too. First, in a country like India, the BPL ration behavioral intention. Journal of Business research, 55(9), 775–783.
cards’ issue/reissue process continues throughout the year and American Marketing Association. (1995) Price consciousness. In The
getting a complete and updated BPL list is generally not possi- American Marketing Association’s online dictionary. URL (https://
ble. This study, which has used BPL family list of Delhi as the www.ama.org/resources/Pages/Dictionary.aspx?dLetter5p (retrieved
sampling frame, also faced this very difficulty. So, the assump- 26 March 2014).
Andreasen, A.R. (1993) Revisiting the disadvantaged: old lessons and
tion that some BPL families may be missing in this list cannot
new problems. Journal of Public Policy & Marketing, 12, 270–275.
be denied. Second, the poor in metropolitan cities live in highly Anic, I.D., Rajh, E. & Bevanda, A. (2012) Decision-making styles of
dense shantytowns and the households in these shantytowns young consumers in Bosnia and Herzegovina. Young Consumers:
lack a proper numbering system. This makes it difficult to Insight and Ideas for Responsible Marketers, 13, 86-98.
locate the households. This problem was also faced during this Assael, H. (2004) Consumer Behavior: A Strategic Approach, 1st edn.
study and a few households could not be located. The untraced Houghton Mifflin, Boston.

International Journal of Consumer Studies 00 (2016) 00–00 13

C 2016 John Wiley & Sons Ltd
V
The poor, FMCGs, consumer behaviour A. Kumar et al.

Attanasio, O.P. & Frayne, C. (2006) Do the poor pay more?. In Proceed- Department of Planning. (2009) Economic survey of Delhi 2008-2009.
ings of Eighth BREAD Conference on Development Economics, Cor- http://www.delhi.gov.in/wps/wcm/connect/221e15804e9d2b0ea6-
nell University, Ithaca, New York, USA. b1aeb6d621e583/Poverty1Line1in1Delhi.pdf?MOD5AJPERES
Bacha, M.D.L. & Neto, C.F. (2014) Right here, right now: Brazilian low- Dibsdall, L.A., Lambert, N., Bobbin, R.F. & Frewer, L.J. (2003) Low-
income consumers’ urge to shop. International Journal of Humanities income consumers’ attitudes and behaviour towards access, availabil-
and Social Science, 4, 107–115. ity and motivation to eat fruit and vegetables. Public Health Nutrition,
Baklien, B. & Samarasinghe, D. (2004) Alcohol and Poverty in Sri 6, 159–168. doi:10.1079/PHN2002412
Lanka. Forut, Norway Durvasula, S., Lysonski, S. & Andrews, J.C. (1993) Cross-cultural gener-
Banerjee, A.V. & Duflo, E. (2007) The economic lives of the poor. The alizability of a scale for profiling consumers’ decision-making styles.
Journal of Economic Perspectives, 21, 141–167. Journal of Consumer Affairs, 27, 55–65.
Bansal, H.S. & Taylor, S.F. (1999) The service provider switching model Election Commission of India. (2014) Elections analysis - party wise
(SPSM) a model of consumer switching behavior in the services indus- comparison for all states. URL (http://eci.nic.in/eci_main1/election_
try. Journal of Service Research, 2, 200–218. analysis_ae.aspx (retrieved 21 October 2014).
Basilevsky, A. (1994) Statistical Factor Analysis and Related Methods: Elliott, R. & Leonard, C. (2006) Peer pressure and poverty: explaining
Theory and Applications. Wiley, New York. fashion brands and consumption symbolism among children of the
Batra, R. & Sinha, I. (2000) Consumer-level factors moderating the suc- ‘British poor’. Journal of Consumer Behaviour, 3, 347–359. doi:
cess of private label brands. Journal of Retailing, 76, 175–191. 10.1002/cb.147
Bindak, R. (2013) Relationship between randomness and coefficient Emmons, G. (2007) The business of global poverty. Harvard Business
alpha: a Monte Carlo simulation study. Journal of Data Analysis and School Working Knowledge. Research & Ideas. URL http://hbswk.
Information Processing, 1, 13–17. hbs.edu/pdf/item/5656.pdf
Black, K. (2008) Business Statistics: For Contemporary Decision Mak- Ethan, D., Basch, C.H., Rajan, S., Samuel, L. & Hammond, R.N. (2013)
ing, 4th edn. Wiley, New Delhi. A comparison of the nutritional quality of food products advertised in
Blackshaw, P. (2008) Satisfied Customers Tell Three Friends, Angry Cus- grocery store circulars of high-versus low-income New York City zip
tomers Tell 3,000: Running a Business in Today’s Consumer-Driven codes. International Journal of Environmental Research and Public
World. Crown Business, New York. Health, 11, 537–547.
Blackwell, D.R., Miniard, P.W. & Engel, J.F. (2001) Consumer Behav- Fang, Y.H., Chiu, C.M. & Wang, E.T.G. (2011) Understanding custom-
iour, 9th edn. Harcourt College Publishers, Orlando. ers’ satisfaction and repurchase intentions: an integration of IS model,
Boston Consulting Group. (2012) The tiger roars: an in-depth analysis of trust, and justice. Internet Research, 21, 479–503.
how a billion plus people consume. URL http://www.indiaretailing. Fornell, C. & Larcker, D.F. (1981) Evaluating structural equation models
com/upload/ContentImage/Market_Research_pdf/The_Tiger_Roars. with unobservable variables and measurement error. Journal of Mar-
pdf (retrieved 12 February 2012). keting Research, 18, 39–50.
Bourguignon, F. (2006) From income to endowments: the difficult task of French, S.A., Wall, M. & Mitchell, N.R. (2010) Household income differ-
expanding the income the income poverty paradigm [electronic version]. In ences in food sources. International Journal of Behavioral Nutrition
Poverty and Inequality (ed. by D.B. Grusky & S.M. Ravi Kanbur), chapter and Physical Activity, 7, 77. URL http://link.springer.com/article/10.
4. Stanford University Press, Stanford. URL http://books.google.co.in/ 1186%2F1479-5868-7-77
books?id5Bmx00pzs3lsC&printsec5frontcover#v5onepage&q&f5false Gabor, A. & Granger, C.W.J. (1961) On the price consciousness of con-
Bove, L.L., Pervan, S.J., Beatty, S.E. & Shiu, E. (2009) Service worker sumers. Journal of the Royal Statistical Society Series C (Applied Sta-
role in encouraging customer organizational citizenship behaviors. tistics), 10, 170–188.
Journal of Business Research, 62, 698–705. Gale, F. & Huang K. (2007) Demand for Food Quantity and Quality in
Bruner II, G.C. (n.d.-a) Brand consciousness. In Marketingscales.com. China. Economic Research Report No. 32. USDA, United States
URL (https://www.marketingscales.com/research/brand-consciousness Department of Agriculture, Washington, DC. URL http://agecon-
(retrieved 26 March 2014). search.umn.edu/bitstream/7252/2/er070032.pdf
Bruner II, G.C. (n.d.-b) Quality consciousness. In Marketingscales.com. Gbadamosi, A. (2009) Cognitive dissonance: the implicit explication in
URL (https://www.marketingscales.com/research/quality-conscious- low-income consumers’ shopping behavior for “low-involvement”
ness (retrieved 26 March 2014). grocery products. International Journal of Retail & Distribution Man-
Chen, Y.J., Chen, P.C. & Lin, T.L. (2012) Gender difference analysis agement, 37, 1077–1095.
cross-country in decision-making styles of Taiwanese and American Goldman, A. (1976) Do lower-income consumers have a more restricted
consumers. Journal of International Management Studies, 7, 175–182. shopping scope? Journal of Marketing, 40, 46–54.
Chung, C. & Myers, S.L. (1999) Do the poor pay more for food?: an anal- Gorn, G.J. & Goldberg, M.E. (1977) The impact of television advertising
ysis of grocery store availability and food price disparities. The Jour- on children from low income families. Journal of Consumer Research,
nal of Consumer Affairs, 33, 276–296. 4, 86–88.
Churchill, G.A., Jr. (1979) A paradigm for developing better measures of Greenberg, B.S. & Dervin B. (1970) Use of the Mass Media by the Urban
marketing constructs. Journal of Marketing Research, 16, 64-73. Poor: Findings of Three Research Projects, with an Annotated Bibli-
Churchill, G. A., Jr., Lacobucci, D., & Israel, D. (2009). Marketing ography. Praeger Publishers, New York.
Research: A south Asian perspective (India ed.). Delhi: Cengage Gupta, K. & Stewart, D.W. (1996) Customer satisfaction and customer
Learning India Pvt. Ltd. behavior: the differential role of brand and category expectations.
Darden, W.R. & Ashton, D. (1974) Psychographic profiles of patronage Marketing Letters, 7, 249–263.
preference groups. Journal of Retailing, 50, 99-112. Hafstrom, J.L., Chae, J.S. & Chung, Y.S. (1992) Consumer decision-
Department of Food, Civil Supplies & Consumer Affairs, Himachal making styles: comparison between United States and
Pradesh. (2012) Antodaya Anna Yojna. URL (http://admis.hp.nic.in/ Korean young consumers. Journal of Consumer Affairs, 26,
ehimapurti/schemes.htm#anto (retrieved 4 November 2012). 114–122.
Department of Food & Supplies, Government of Delhi, Delhi. (2012) Hair, J.F., Jr., Black, W.C., Babin, B.J., Anderson, R.E. & Tatham, R.L.
Public distribution system statistics. URL (http://delhi.gov.in/wps/ (2006a) Multivariate Data Analysis, 6th edn. Pearson-Prentice Hall,
wcm/connect/DOIT_Food/food/home (retrieved 21 June 2012). New Jersey.

14 International Journal of Consumer Studies 00 (2016) 00–00

C 2016 John Wiley & Sons Ltd
V
A. Kumar et al. The poor, FMCGs, consumer behaviour

Hair, J.F., Bush, R.P. & Ortinau, D.J. (2006b) Marketing Research: Lei, P.W. & Wu, Q. (2007) An NCME instructional module on introduc-
Within Changing Information Environment, 3rd edn. Tata McGraw- tion to structural equation modeling: issues and practical considera-
Hill, New Delhi. tions. Educational Measurement: Issues and Practice, 26, 33–43. URL
Hamilton, K. (2009) Those left behind: inequality in consumer culture. http://ncme.org/linkservid/47EFEB5A-1320-5CAE-6EC90BFD-
Irish Marketing Review, 20, 40–54. F09AA39E/showMeta/0/
Hamilton, K. (2011) Low-income families and coping through brands: Leibtag, E.S. & Kaufman, P.R. (2003) Exploring Food Purchase Behav-
inclusion or stigma? Sociology, 46, 1–17. doi:10.1177/ ior of Low-Income Households: How Do They Economize?. Agricul-
0038038511416146 ture Information Bulletin No-747-07, United States Department of
Hamilton, K. & Catterall, M. (2006) Consuming love in poor families: Agriculture, Economic Research Service. URL http://www.ers.usda.
children’s influence on consumption decisions. Journal of Marketing gov/publications/aib747/aib74707.pdf
Management, 22, 1031–1052. Leo, C., Bennett, R. & H€artel, C.E. (2005) Cross-cultural differences in
Hamilton, K. & Catterall, M. (2007) Love and consumption in poor families consumer decision-making styles. Cross Cultural Management: An
headed by lone mothers. Advances in Consumer Research, 34, 559–564. International Journal, 12, 32–62.
URL http://www.acrwebsite.org/volumes/v34/500301_101219_v1.pdf Lysonski, S. & Durvasula, S. (2013) Consumer decision making styles in
Hamilton, K. & Catterall, M. (2008) I Can Do It!” Consumer retailing: evolution of mindsets and psychological impacts. Journal of
coping and poverty. Advances in Consumer Research, 35, Consumer Marketing, 30, 75–87.
551–556. URL http://www.acrwebsite.org/volumes/v35/naacr_ Lysonski, S., Durvasula, S. & Zotos, Y. (1996) Consumer decision-
vol35_247.pdf making styles: a multi-country investigation. European Journal of
Hammond, A.L., Kramer, W.J., Katz, R.S., Tran, J.T. & Walker, C. Marketing, 30, 10–21.
(2007) The Next 4 Billion: Market Size and Business Strategy at the MacDonald, J.M. & Nelson, P.E. (1991) Do the poor still pay more?
Base of the Pyramid. World Resources Institute and International Food price variations in large metropolitan areas. Journal of Urban
Finance Corporation, Washington, DC. Economics, 30, 344–359.
Hanzaee, K.H. & Aghasibeig, S. (2008) Generation Y female and male Mafini, C. & Dhurup, M. (2014) Assessing consumer purchasing decision
decision-making styles in Iran: are they different? The International styles: an empirical investigation from South Africa. International
Review of Retail, Distribution and Consumer Research, 18, 521–537. Business & Economics Research Journal (IBER), 13, 679–688.
Hayes, L.R. (2000) Are prices higher for the poor in New York City? Malhotra, N.K. & Dash, S. (2009) Marketing Research: An Applied Ori-
Journal of Consumer Policy, 23, 127–152. entation, 5th edn. Pearson Education, New Delhi.
Hou, S.C. & Lin, Z.H. (2006) Shopping styles of working Taiwanese Mitchell, V.W. & Walsh, G. (2004) Gender differences in German con-
female. In Proceedings of International Conference on Business and sumer decision-making styles. Journal of Consumer Behaviour, 3,
Information, Singapore. 331–346.
Hu, L.T. & Bentler, P. (1999) Cutoff criteria for fit indexes in covariance Mokhlis, S. & Salleh, H.S. (2009) Consumer decision-making styles in
structure analysis: conventional criteria versus new alternatives. Struc- Malaysia: an exploratory study of gender differences. European Jour-
tural Equation Modeling, 6, 1–55. nal of Social Sciences, 10, 574–584.
Huang, K.S. & Gale, F. (2009) Food demand in China: income, quality, and Monroe, K. & Petroshius, S.M. (1981) Buyer’s perceptions of price: an
nutrient effects. China Agricultural Economic Review, 1, 395–409. update of the evidence. In Perspectives in Consumer Behavior (ed. by
Hunjra, A.I., Niazi, G.S.K. & Khan, H. (2012) Relationship between H. Kassarjian & T S Robertson), pp. 43–55. Illinois: Scott Foresman
decision making styles and consumer behavior. Actual Problems of Moschis, G.P. (1976) Shopping orientations and consumer uses of infor-
Economics, 2, 21–26. mation. Journal of Retailing, 52, 61.
Jaiswal, A.K. (2008) The fortune at the bottom or the middle of the pyra- National Council of Applied Economic Research. (2005) The Great
mid? Innovations: Technology, Governance, Globalization, 3, 85–100. Indian Market: Results from the NCAER Market Information Survey of
Jones, J.M. & Zufryden, F.S. (1982) An approach for assessing demo- Households (ISBN: 8188830089). National Council of Applied
graphic and price influences on brand purchase behavior. The Journal Economic Research, New Delhi. URL http://www.thesuniljain.com/
of Marketing, 46, 36. files/thirdparty/Great%20Indian%20Bazaar.pdf
Karn, S.K., Shikura, S. & Harada, H. (2003) Living environment and Nunnally, J.C. (1978) Psychometric Theory, 2nd edn. McGraw-Hill, New
health of urban poor: a study in Mumbai. Economic and Political York.
Weekly, 38, 3575–3577, 3579–3586. Oliver, R.L. (1980) A cognitive model for the antecedents and conse-
Karnani, A. (2007) The mirage of marketing to the bottom of the pyra- quences of satisfaction. Journal of marketing Research, 17, 460–469.
mid: how the private sector can help alleviate poverty. California Oliver, R.L. (1999) Whence consumer loyalty? The Journal of Market-
Management Review, 49, 90–111. ing, 63 (special issue), 33–44
Kavkani, S.A.M., Seyedjavadain, S. & Saadeghvaziri, F. (2011) Deci- Park, J.E., Yu, J. & Zhou, J.X. (2010) Consumer innovativeness and
sion-making styles of young Iranian consumers. Business Strategy shopping styles. Journal of Consumer Marketing, 27, 437–446.
Series, 12, 235–241. Prahalad, C.K. (2005) The Fortune at the Bottom of the Pyramid, 1st edn.
Kirchgeorg, M. & Winn, M.I. (2006) Sustainability marketing for the poor- Pearson Education, Delhi.
est of the poor. Business Strategy and the Environment, 15, 171–184. Prahalad, C.K. & Hart, S.L. (2002) The fortune at the bottom of the pyra-
Kline, R. (1998) Principles and Practice of Structural Equation Model- mid. Strategy1Business, 26, 54–67.
ing. The Guildford Press, New York. Radder, L., Li, Y. & Pietersen, J.J. (2006) Decision-making styles of young
Kochuyt, T. (2004) Giving away one’s poverty. On the consumption of Chinese, Motswana and Caucasian consumers in South Africa: an
scarce resources within the family. The Sociological Review, 52, 139– exploratory study. Journal of Family Ecology and Consumer Sciences/
161. doi:10.1111/j.1467-954X.2004.00462.x Tydskrif vir Gesinsekologie en Verbruikerswetenskappe, 34, 20–31.
Lastovicka, J.L. (1982) On the validation of lifestyle traits: a review and Robinson, N., Caraher, M. & Lang, T. (2000) Access to shops: the views
illustration. Journal of Marketing Research, 19, 126–138. of low-income shoppers. Health Education Journal, 59, 121–136. doi:
Lee, C.H. (2008) The effects of price consciousness, brand consciousness 10.1177/001789690005900202
and familiarity on store brand purchase intention. Management Rzem, H. & Debabi, M. (2012) Store image as a moderator of store brand
Review, 27, 113–117. attitude. Journal of business studies quarterly, 4(1), 130–148.

International Journal of Consumer Studies 00 (2016) 00–00 15

C 2016 John Wiley & Sons Ltd
V
The poor, FMCGs, consumer behaviour A. Kumar et al.

Salkind, N.J. (ed.) (2010) Encyclopedia of Research Design [Electronic Appendix A

Version]. Sage, Thousand Oaks. URL http://books.google.co.in/book-
s?id5b_1yAwAAQBAJ&printsec5frontcover&source5gbs_ge_sum-
Research instrument
mary_r&cad50#v5onepage&q&f5false
Sarin, A. & Jain, R. (2009) Effect of mobiles on socio-economic life of
1. Consumer Decision-Making Styles
urban poor. Indian Institute of Management Ahmedabad Working
(i) Price consciousness(PC)
Paper. URL http://www.iimahd.ernet.in/assets/snippets/workingpa-
perpdf/2009-02-05Sarin.pdf PC1. It is important for me that I purchase TEA* at low price.
Schiffman, L.G. & Kanuk, L.L. (2004) Consumer Behaviour, 8th edn. PC2. In my opinion, there is not much difference between a higher
Pearson Education, Delhi. priced TEA* and a lower priced TEA*.
Shukla, R. (2010) How India Earns, Spends and Saves: Unmasking the PC3. Before purchasing TEA*, I enquire TEA* prices at a number of
Real India. Sage, New Delhi. nearby stores.
Silverthorne, S. (2007) The business of global poverty. Harvard Business PC4. During a TEA* purchase, I do a price-quantity comparison of vari-
School Working Knowledge. Research & Ideas. URL http://hbswk. ous TEA* brands.
hbs.edu/pdf/item/5529.pdf PC5. I bargain whenever I purchase TEA*.
Sinha, I. & Batra, R. (1999) The effect of consumer price consciousness (ii) Quality consciousness(QC)
on private label purchase. International journal of research in market- QC1. Before making a TEA* purchase, I try to confirm its quality.
ing, 16(3), 237–251. QC2. I do take its quality assurance from the seller.
Smith, S.C. (2005) Ending Global Poverty: A Guide to What Works. Pal- QC3. I can adjust with quantity of TEA* but never compromise on the
grave Macmillan, New York. quality of TEA*.
Sproles, E.K. & Sproles, G.B. (1990) Consumer decision-making styles QC4. I do not hesitate visiting a number of stores in search of the best
as a function of individual learning styles. Journal of Consumer quality of TEA*.
Affairs, 24, 134–147.
(iii) Brand consciousness(BC)
Sproles, G.B. (1985) From perfectionism to faddism: measuring consum-
BC1. In quality term, each TEA* brand is different from other TEA*
ers’ decision-making styles. In Proceedings, American Council on
brands.
Consumer Interests, pp. 79–85.
BC2. I purchase the TEA* brands that are regularly advertised.
Sproles, G.B. & Kendall, E.L. (1986) A methodology for profiling con-
BC3. I prefer to purchase lesser amount of a popular TEA* brand even
sumers’ decision-making styles. Journal of Consumer Affairs, 20,
267–279. when I could purchase larger quantityof an unknown TEA* brand with
Subrahmanyan, S. & Gomez-Arias, J.T. (2008) Integrated approach to the same amount of money.
understanding consumer behavior at bottom of pyramid. Journal of BC4. I do adjust my household grocery budget so that I can buy well
Consumer Marketing, 25, 402–412. known TEA* brands.
Tai, S.H. (2005) Shopping styles of working Chinese females. Journal of (iv) Brand loyalty
Retailing and Consumer Services, 12, 191–203. BL1. Once I like a TEA* brand, I continue buying it.
Tate, R.S. (1961) The supermarket battle for store loyalty. The Journal of BL2. Some TEA* brands are my favourite ones which I regularly
Marketing, 25, 8–13. purchase.
Tooley, J. & Dixon, P. (2006) De facto’ privatisation of education and BL3. Whenever I find my favourite TEA* brands unavailable at a particu-
the poor: implications of a study from sub-Saharan Africa and India. lar store, I do not hesitate to look forthem at other neighbouring stores.
Compare, 36, 443–462. ‘ BL4. I frequently change TEA* brands.**
Tsai, H.T. & Huang, H.C. (2007) Determinants of e-repurchase inten- 2. Post purchase behaviour
tions: an integrative model of quadruple retention drivers. Information (i) Repurchase Intentions
& Management, 44, 231–239. PPB1. If I am satisfied with a particular TEA* brand, I continue buying it.
Viswanathan, M. (2007) Understanding product and market interactions PPB2. If I am satisfied with a particular TEA* brand, I continue purchas-
in subsistence marketplaces: a study in South India. In Product and
ing it from the same store.
Market Development for Subsistence Marketplaces (ed. by J.A. Rosa
(ii) Recommendations to the others
& M. Viswanathan), Advances in international management, Vol. 20,
PPB3. If I am satisfied with a TEA* brand, I recommend it to others.
pp. 21–57. Emerald Group Publishing Ltd.
PPB4. If I am dissatisfied with a TEA* brand, I mention it to others.
Wells, W.D. (1975) Psychographics: a critical review. Journal of Market-
ing Research, 196–213.
Wesley, S., LeHew, M. & Woodside, A.G. (2006) Consumer decision-
making styles and mall shopping behavior: building theory using 1
Note: All the variables/items were on a five point Likert Scale
exploratory data analysis and the comparative method. Journal of
(1 5 strongly disagree and 5 5 strongly agree).
Business Research, 59, 535–548. 2
*Data was collected for five FMCGs (bathing soap, cooking oil, tea,
Ye, L., Bose, M. & Pelton, L. (2012) Dispelling the collective myth of
toothpaste and washing soap) separately,
Chinese consumers: a new generation of brand-conscious individual- 3
**Negative statement(s)
ists. Journal of Consumer Marketing, 29(3), 190–201.

16 International Journal of Consumer Studies 00 (2016) 00–00

C 2016 John Wiley & Sons Ltd
V
A. Kumar et al. The poor, FMCGs, consumer behaviour

Appendix B
Structural model

Source: Primary data.

International Journal of Consumer Studies 00 (2016) 00–00 17

C 2016 John Wiley & Sons Ltd
V