THESIS

TEACHER TOLERANCE OF RISK IN PLAY SCALE (T-TRiPS):

EVALUATING THE PSYCHOMETRIC PROPERTIES OF A NEW MEASURE

Submitted by

Kristina M. Ihrig

Department of Occupational Therapy

In partial fulfillment of the requirements

For the Degree of Master of Science

Colorado State University

Fort Collins, Colorado

Spring 2021

Master’s Committee:

Advisor: Anita Bundy

Shelly Lane
Jennifer Aberle
Copyright by Kristina M. Ihrig 2021

All Rights Reserved

 ABSTRACT

TEACHER TOLERANCE OF RISK IN PLAY SCALE (T-TRiPS):

EVALUATING THE PSYCHOMETRIC PROPERTIES OF A NEW MEASURE

A small amount of risk is a natural component of play that provides children with

developmental benefits, increasing: competencies, autonomy and self-efficacy. The

purpose of this study is to examine validity and reliability of data produced with a

new instrument, Teacher Tolerance of Risk in Play Scale (T-TRiPS). T-TRiPS, a 25-

item instrument, was administered to 99 teachers of children with disabilities, aged 4-

12. Rasch analysis was used to analyze data. Findings support strong evidence for

unidimensionality of the construct: positive point measure correlations, goodness-of-

fit statistics in the acceptable range, and a logical item hierarchy. A strata value of

3.01 and person reliability index of 0.80 support internal reliability. T-TRiPS

produces valid and reliable data regarding teacher tolerance of risk in play. T-TRiPS

can serve as the basis for self-reflection or intervention aimed at increasing children’s

access to developmentally beneficial risky play.

ii
 ACKNOWLEDGEMENTS

I wish to express my deepest gratitude to my advisor, Dr. Anita Bundy, for your

guidance, patience, and support throughout this thesis journey. Thank you for generously sharing

your knowledge of this subject and encouraging me to follow my own passion advocating for

children’s right to play through this research. It has been an honor to learn from you. I am

especially grateful to you for helping to keep me grounded in my strengths and aware of my

progress throughout the stresses and celebrations of this process. Thank you to the other

members of my incredible thesis committee, Dr. Shelly Lane and Dr. Jen Aberle, for your

encouragement, insightful comments, and invaluable guidance along this thesis journey. I am

extremely grateful to you for the knowledge you shared and the input you provided to help me

successfully complete this research. My sincere thanks goes to my friends from RIOT for your

kindness and willingness to provide feedback at all stages of this thesis. I would also like to

acknowledge the incredible friends I’ve made through the last two years of this OT program.

Thank you for being a shoulder in tough times and a cheerleader in the excited times. Finally, I

must express profound gratitude to my dear friends and my family for providing me with

encouragement throughout the last two years of study and through the process of completing this

thesis. Thank you for believing in me when I needed to be reminded that I can do hard things and

taking every opportunity to celebrate with me. This accomplishment would not have been

possible without your unwavering love and support.

iii
 TABLE OF CONTENTS

ABSTRACT.................................................................................................................................... ii

ACKNOWLEDGEMENTS ........................................................................................................... iii

CHAPTER 1 INTRODUCTION ................................................................................................... 1

The Problem of Risky Play ............................................................................................. 1

Consequences of Limiting Risky Play ........................................................................ 3

Children with Disabilities ........................................................................................... 4

Risky Play at School ................................................................................................... 5

Risky Play at School: Fear and Responsibility ....................................................... 6

Sydney Playground Project ..................................................................................... 8

The Tolerance of Risk in Play Scale (TRiPS). ....................................................... 9

Table 1.1 ................................................................................................................... 11

Drawn to Advocate for Risky Play ............................................................................... 13

Research Questions ....................................................................................................... 15

Overview of Thesis ....................................................................................................... 15

CHAPTER 2 LITERATURE REVIEW ...................................................................................... 16

Adult-Report Measures Related to Play and Risk .................................................... 18

Measures of Adult Perceptions of Risk: Beneficial or Detrimental? .................... 19

Measures of Child Play and Behaviors as Predictors of Risk and Injury ............. 21

iv
 Measures of Adult Practices that Contribute to Risk and Injury .......................... 22

Table 2.1 ................................................................................................................... 24

Observational Measures of Play and Environment on the School Playground ........ 28

Evaluations of the School Play Environment ....................................................... 28

Measures of Children’s Play on the Playground................................................... 30

Measures of Play for Children with Autism and/or Intellectual Disability .......... 31

Table 2.2 ................................................................................................................... 33

Conclusion .................................................................................................................... 37

CHAPTER 3 JOURNAL ARTICLE ........................................................................................... 39

Imbalance of Duty of Care and Dignity of Risk: A Barrier to Development ........... 40

Risky Play for Children with Disabilities ................................................................. 40

Risk-Benefit Assessment: Teachers .......................................................................... 41

Assessing Teachers’ Tolerance of Risky Play: Instrument Development ................ 42

Method .......................................................................................................................... 43

Participants ................................................................................................................ 43

Table 3.1 ................................................................................................................... 44

Instrument ................................................................................................................. 44

Table 3.2 ................................................................................................................... 46

Procedure .................................................................................................................. 46

Data Analysis ............................................................................................................ 47

v
 Construct Validity ................................................................................................. 47

Internal Reliability ................................................................................................ 48

Results ........................................................................................................................... 48

Table 3.3 ................................................................................................................... 49

Figure 3.1. T-TRiPS Person-Item Map (Wright Map) ............................................. 51

Discussion ..................................................................................................................... 52

Recommendations to Strengthen T-TRiPS ............................................................... 54

Limitations ................................................................................................................ 56

Implications for Future Research .............................................................................. 56

Conclusions ................................................................................................................... 57

CHAPTER 4 REFLECTIONS AND CONCLUSIONS ............................................................... 58

Future Research ............................................................................................................ 59

Personal Reflections...................................................................................................... 60

REFERENCES ............................................................................................................................. 62

APPENDIX ................................................................................................................................... 72

LIST OF ABBREVIATIONS ....................................................................................................... 80

vi
 CHAPTER 1

INTRODUCTION

The Problem of Risky Play

Every day, all over the world, children are exposed to risks. But not all risks are bad. In

fact, some risk that is a natural part of children’s play provides vital opportunities for

development (Sandseter, 2012). Through active engagement in “risky play,” children learn to

interact with others, trust their own abilities, discover the boundaries of their skills and imagine

their potential (Beetham et al., 2019; Brussoni, Gibbons, et al., 2015; Garwood, 1982; Piaget,

1972; Sandseter & Kennair, 2011; Yogman et al., 2018). Children must learn to recognize and

respond appropriately to risks in order to promote their health and well-being (International

School Grounds Alliance, 2017).

Opportunities to respond to safe degrees of fear, physical, social and emotional challenge,

uncertainty, and the possibility of failure support children in learning to assess risks and manage

their responses to these situations (Sandseter, 2007; Spencer et al., 2016). Although several

variations of the definition of ‘risky play’ exist, Sandseter (2009) defined this type of play as

activities that are exciting, thrilling, and involve a risk of physical injury. Additional authors

expanded this definition by including that risky play involves uncertainty and exploration - of

one’s physical capabilities, emotions, perceptions, and environment - and often involves a degree

of relinquishing control by teachers and children (Ball, Gill, & Spiegal, 2012; Kleppe, Melhuish,

& Sandseter, 2017; Lester & Russell, 2014; Spencer et al., 2016). For the purposes of this paper,

I adopted a combination of these as my definition of risky play: activities that are exciting and

thrilling, that involve exploration (of physical, social, emotional, perceptions, and environment)

and that require children and adults to accept uncertainty and relinquish control.

1
 While risky play can result in injury or even death, it is important to understand that risk

of varying degrees is inherent in every situation and represents a spectrum of severity (Ball,

2002; Ball et al., 2012). Many advocates for risky play use terms such as “manageable risky

play” or “safe risks” to refer to a level of risky play that is severe enough to be exciting and

challenging to a child, but not so severe that likelihood of injury are obvious and expected (Ball

et al., 2012; Sandseter, 2009). Perceptions of risk are subjective, indicating the severity of risky

play will vary with individuals’ comfort with risk and perceptions of how dangerous an

experience is.

Children typically engage in risky play during physical activity and while playing

outdoors but the types of activities that are considered risky are numerous and complex. To help

understand different characteristics of risky play activities, Sandseter (2007) interviewed

children and teachers on the different risky play activities they enjoy and organized these into

eight categories (Kleppe et al., 2017; Sandseter, 2009): (A) great heights, (B) high speeds, (C)

dangerous tools, (D) dangerous elements, (E) rough-and-tumble, (F) disappear/get lost, (G) play

with impact, (H) vicarious risk. (See Table 2.1 for in-depth descriptions of the categories of risky

play.) Each of these categories houses numerous activities that comprise different severity of

risk; several activities involve a combination of different categories. Even though there is a risk

of injury inherent in risky play, the thrill and excitement of experiencing this type of play is one

reason children return to risky play experiences time and time again. In addition to being a fun

and satisfying experience, risky play has significant developmental benefits that are important to

consider.

2
Benefits of Risky Play for Children of All Abilities

Increasing recognition of risky play led researchers to seek to understand the benefits to

children’s health, development, and wellbeing. In a systematic review, Brussoni, Gibbons, et al.

(2015) found that the impact of risky play on development is positive. The most commonly

identified benefit is an increase in physical activity, but social, psychological, and cognitive

benefits have also been identified (Brussoni, Gibbons, et al., 2015; Bundy et al., 2011). Children

learn social interaction skills through opportunities to problem-solve when faced with conflict

and while engaging in rough and tumble play with peers (Pellegrini, 1988). Problem-solving

related to assessing risks and learning to manage them also promotes executive functioning,

creativity, confidence and a positive sense of self. Being able to assess a risky play opportunity

and decide how to engage while best managing the risk of injury are protective factors against

injury because children have learned the boundaries of their own skills and how to best engage

with risky environments (Brussoni, Olsen, Pike, & Sleet, 2012; Christensen & Mikkelsen, 2008).

Additionally, when provided with opportunities to engage in risky play, children face less

boredom and do not so often seek out exciting opportunities that can be beyond their ability

(Brussoni, Brunelle, et al., 2015; Sandseter, 2009). The benefits of risky play are not limited to

typically developing children; they also are impactful for children with disabilities like autism

and intellectual disability as well and should, therefore, be a priority for this population (Bundy

et al., 2015).

Consequences of Limiting Risky Play

There is growing acknowledgement of the importance of risky play; however, barriers

remain in practice (Ball, Brussoni, Gill, Harbottle, & Spiegal, 2019; Sandseter, 2014). When

risky play is prohibited or a child’s skills are more advanced than the challenges their play

3
environment can offer, children often seek out exceedingly challenging experiences that provide

them with the slight thrill of fear and uncertainty to combat boredom (Sandseter, 2007). This

thrill-seeking play can result in injury as children are ill-equipped to manage this spike in

severity of risk. Therefore, it is important to provide children with gradually increasing risky

play opportunities that support them in assessing their ability to manage the risks, and ultimately,

develop new skills.

Children with Disabilities

Despite the well documented benefits of play on development and the UN’s declaration

that all children have the right to engage in play (U.N.C.R.C., 1989), children with disabilities

experience play inequity compared to their typically developing peers (Sterman, Naughton,

Bundy , Froude, & Villeneuve, 2019). The cause of this imbalance in opportunities for play is

multidimensional, including perceptions of the children’s abilities, barriers from inaccessible

physical environments, school playground policies limiting risk to prevent injury, adult fears

related to their understanding of children’s capabilities, and lack of support from the local

governments (Bundy et al., 2015).

Autism spectrum disorder (ASD) is characterized by deficits in communication and

social interactions and is often accompanied by repetitive and restrictive interests or behaviors

(American Psychiatric Association, 2013). Intellectual disability (ID) is a broad category of

disability that is usually diagnosed in children prior to the age of 18 and is characterized by

limitations in intellectual function and challenges with developing adaptive behaviors (American

Psychiatric Association, 2013). Severity of impairments varies greatly in children with ID

presenting as varying cognitive, social interaction, and physical limitations, all of which are

important to play. Because of these challenges, children with ASD and ID are often viewed as

4
lacking the skills to engage successfully in play and are often not given the opportunity based on

concerns for injury due to these perceived deficits.

Limiting risky play is not done with mal intent, but instead is an attempt by adult care

providers to limit injury for children with disabilities, those who are considered our most

vulnerable population (Beetham et al., 2019; Newnham, 2000; Wolfensberger, Nirje, Olshansky,

Perske, & Roos, 1972). Without opportunities to take on increasingly complex challenges during

play, children with disabilities miss out on opportunities to develop skills that help them manage

risky play in addition to all of the health and well-being benefits mentioned earlier (Beetham et

al., 2019; Bundy et al., 2015). Unfortunately, adult aversion to risk, sometimes called surplus

safety, results in fewer opportunities for children to engage in risky play- this is especially true

on school playgrounds and for children with disabilities (Brussoni, Gibbons, et al., 2015;

Buchanan, 1999; Bundy et al., 2015; Spencer et al., 2016; Wyver et al., 2010).

Risky Play at School

In the school setting, recess on the playground is children’s primary opportunity for play

(Ramstetter, Murray, & Garner, 2010). During the school day, recess serves as a much-needed

break from the rigorous requirements placed on children to remain still and maintain cognitive

focus during academic instruction. Educators and experts in child development recommend

protecting a regular opportunity for unstructured outdoor play (Ball, 2002; Ball et al., 2019;

Brussoni, Brunelle, et al., 2015; Gill, 2018). Unfortunately, even when this unstructured play

time is available, recess is often considered a break for teachers as well as children, rather than

an opportunity to provide quality play experiences. Further, even though the benefits of physical

activity and play are well understood, the time available for engaging in play continues to be cut

back in favor of increasing time for academic instruction (Gill, 2018; Ramstetter et al., 2010).

5
 In addition to the emphasis on academics over play, the fear of child injury on the

playground has only contributed to the problem of decreasing opportunities to engage in play,

especially risky play. Policies for playground safety also serve as a significant barrier to risky

play. In many countries, school playgrounds have become very regulated by focusing on the

characteristics of the playground structures and teacher supervision to limit injury. Policies, with

good intentions, focus on removing risk and do not consider the impact to a child’s opportunities

for play or the stress they put on teachers (Ball et al., 2019). Teachers often make decisions about

which play activities to condone and which to interrupt based on their discomfort and fear of

potential consequences for allowing risky play that could be viewed as violating play policy

(Spencer et al., 2016). We must understand the reasoning behind teachers’ fears of situations

they consider risky and support them in shifting their perspectives to promote important learning.

Risky Play at School: Fear and Responsibility

Teachers have an invaluable role in the care of children. They are called to recognize the

unique skills and experiences that shape each child while at the same time recognizing children’s

future potential. The often-opposing responsibilities of teachers to protect children from any

potential of harm while simultaneously providing rich opportunities for learning so children may

develop into independent adults can be summarized by the concepts ‘duty of care’ and ‘dignity

of risk’. The juxtaposition of duty of care and dignity of risk is especially relevant when

considering the responsibilities of teachers working with children with disabilities. Duty of care

is defined by the law in relation to school liability; teachers are responsible to supervise children

and prevent their exposure to situations that could result in injury (Newnham, 2000). Dignity of

risk is the right of a person, especially those with a disability, to experience manageable

exposure to risky situations in order to learn skills for risk management in anticipation of future

6
autonomy (Wolfensberger et al., 1972).

At school, the threat of liability following injury on the school playground is a significant

concern for most teachers. Injury may give rise to charges of negligence or failure to fulfill their

duty of care. However teachers are rarely negligent (Newnham, 2000). Instead, injuries may

result from children seeking exciting play situations when opportunities for play are not

challenging enough and therefore boring (Brussoni, Brunelle, et al., 2015). Fear of liability has

resulted in an imbalance between duty of care and dignity of risk with an emphasis on duty of

care (Little, 2010). Additionally, many teachers simply have their own personal reservations

regarding risk and believe that all risky situations are dangerous, have high likelihood of

resulting in injury, and should be avoided (van Rooijen & Newstead, 2016).

Even teachers who would allow their own children to engage in risky play feel uncertain

when supervising students engaging in risky play on the playground and interrupt the play in

response. The cultural shifts towards risk aversion and surplus safety (Buchanan, 1999; Gill,

2018; van Rooijen & Newstead, 2016; Wyver et al., 2010) make it difficult for teachers to feel

they can allow risky play, without fear of consequence, even if they perceive associated benefits.

In one study, Little (2010) interviewed teachers on their perspectives of risky play and

safety during outdoor play. Even though many teachers acknowledged the value of risky play,

they feared litigation and lack of support from school administrators and child health and safety

regulations Regulatory factors that limit play include playground structure guidelines (e.g.,

height, fall material, surface texture), teacher to child ratios on the playground, and school-

specific play policy like restricting children to only going down a slide, feet first, on their bottom

(Ball, 2002; Little & Eager, 2010; van Rooijen & Newstead, 2016). Little (2010) acknowledged

the contradictory perceptions held by teachers (i.e., on one hand children are competent and

7
capable, but on the other hand, they are vulnerable and teachers must protect them) that cause

conflict between teachers’ pedagogical beliefs and their actions in the face of risk, furthering the

imbalance between duty of care and dignity of risk.

In recent years, with increasing recognition of the benefits of risky, many education

programs have attempted to prioritize opportunities for risky play at school during play outdoors

(Brussoni, Gibbons, et al., 2015; Brussoni et al., 2012; Wyver et al., 2010). Thus, educators need

to adopt a more nuanced view of risk, beyond good or bad, and understand that risky play holds

opportunities for development including confidence, healthy lifestyle, resilience, social skills,

and problem solving (Brussoni, Brunelle, et al., 2015; Cooke, Wong, & Press, 2019).

Nonetheless, several researchers (Cooke et al., 2019; Spencer et al., 2016) have found that, while

societal understandings of risky play are shifting towards valuing these benefits, the initial

reaction of teachers when supervising a child in a risky play situation is still to interrupt the play

to protect themselves from negative consequences. In order for a child to truly reap the benefits

of learning through the dignity of risk, their care providers must cease overprotection and instead

seek a realistic balance with their duty of care.

Sydney Playground Project

My research takes place within the context of a larger cluster trial (CT) called The

Sydney Playground Project (SPP). SPP featured a 2-part intervention: (1) introduction of loose

parts materials to school playgrounds and (2) risk-reframing workshops to help adults

acknowledge and change their beliefs of risk with the end goal of promoting risky play (Bundy et

al., 2015). Risk-reframing workshops provided opportunities for adults to discuss their beliefs

about risky play as well as educate them about: its benefits, what to do when faced with

discomfort while supervising play, ways of scaffolding learning for children during risky play,

8
and how to evaluate the severity of a risky play situation (Bundy et al., 2015; Niehues et al.,

2013; Spencer et al., 2016). These trainings helped teachers feel more comfortable with

balancing duty of care and dignity of risk and increasing children’s opportunities for risky play

(Bundy et al., 2015; Niehues et al., 2013; Spencer et al., 2016). Three iterations of the SPP

occurred: a pilot, a cluster randomized trial in mainstream primary schools and a cluster trial in

substantially separate programs for children with autism and intellectual disability in greater

Sydney, Australia.

The SPP team developed an instrument to measure adult risk tolerance and help

determine the impact of the SPP. During the risk reframing sessions of the first CT, the

researchers administered the Tolerance of Risk in Play Scale (TRiPS) to teachers and parents (7

out of 100 were teachers) to examine evidence for construct validity and reliability and provide a

baseline measure of their tolerance of risky play (Hill & Bundy, 2014).

The Tolerance of Risk in Play Scale (TRiPS). The Tolerance of Risk in Play Scale

measured caregivers’ tolerance of risk when considering various risky play activities. The

original measure includes several yes/no items based on categories of risky play, risk propensity,

parental overprotection and risk tolerance in supervision and childhood injury risk (Morrongiello

& Corbett, 2006; Sandseter, 2007, 2009). The items included range from “very risky,” and

therefore difficult to endorse, to those that are “not risky at all,” and therefore easy to endorse.

TRiPS produced valid and reliable data (Hill & Bundy, 2014). See Table 1.1 for psychometric

data and Figure 1.1 for a complete list of TRiPS items and their corresponding risky play

categories. During its use in the SPP however, it became clear that this measure was not a good

reflection of the roles of teachers. The responsibilities of teachers and play situations they

supervise while at school differ greatly from those of parents. Additionally, the TRiPS was

9
written in consideration of an individual (or few) child’s play behaviors and is less applicable for

teachers who are responsible for supervising a group of children. There are few instruments

available that evaluate how adult caregivers (parents and teachers) feel about risky play, and

none that evaluate teachers’ tolerance of risky play on the playground. A new measure

specifically for teachers of children of all abilities and for the school playground context is

needed. In order to shift teachers’ perspectives of risk and support them in providing manageable

risky play opportunities for children, researchers require a valid and reliable tool to identify a

teacher’s current tolerance (or intolerance) of risky play. The ability to measure teacher tolerance

of risky play may serve as a mechanism for change by offering opportunities for intervention as

well as by identifying tangible increases in a teacher’s risk tolerance following intervention. A

new instrument may support the effort to promote autonomy, health, and well-being for all

children beginning with the ability to accurately measure teacher tolerance of risk in play.

Modifications to the TRiPS (Hill & Bundy, 2014) to better suit the roles and situations of

teachers resulted in a new instrument called the Teacher Tolerance of Risk in Play Scale (T-

TRiPS). T-TRiPS will help teachers acknowledge their perceptions of risk and the impact of

their intolerance of risky play on children’s development. The purpose of this research is to

evaluate data produced with the T-TRiPS for evidence of validity and reliability.

10
Table 1.1
Psychometric Properties of the Tolerance of Risk in Play Scale (TRiPS) (Hill & Bundy, 2014)
Purpose Description Psychometric data
Measure adult 32 items with Internal validity:
caregiver dichotomous response goodness of fit statistics for all items within
tolerances of (yes / no) ranging from acceptable values:
risk during “very risky” to “not MnSq: <1.5
children’s play risky at all” ZSTD: <2
2
Primary school External validity:
aged children Risk Averse Children 3 to 4 years:
M = -1.62 M = -0.13
SD = 1.68 SD = 1.25
Somewhat Risk Children 5 to 7 years:
Averse M = 0.20
M = -0.37 SD = 1.78
SD = 1.30 Children 8 to 10 years:
Somewhat Risk M = 1.5
Tolerant SD = 1.78
M = 1.20 Children 11 to 13 years:
SD = 1.81 M = 1.95
Risk Tolerant SD = 1.82
M = 2.64
SD = 2.1
Person Separation Index 2.56
Person Reliability Index.87

11
Figure 1.1. Tolerance of Risk in Play Scale Item Map and Corresponding Category of
Risky Play (Hill & Bundy, 2014; Sandseter, 2009)

12
 Drawn to Advocate for Risky Play

Before I was an occupational therapy student, I was an early childhood educator for 5

years at a local play-based lab school called the CSU Early Childhood Center (ECC). The ECC

is inspired by the Reggio Emilia approach to teaching which informed my professional practice

and interactions with children. I learned alongside the children, viewing them as capable and

competent, and the environment as an important source of learning that supports or limits

development (Malaguzzi, 1994).

During my time at ECC, I worked with many age groups--the school provides care for

children ages 6 weeks to 6 years--but spent the most time teaching older preschool children: the

4-6-year-olds. I witnessed firsthand the incredible impact play has on learning and development.

For 5 years, I tracked immense change in my students’ development in all educational domains,

from traditional academic skills to motor development to cognitive and social-emotional skills.

Even though I attributed these changes to their engagement in play and the belief of our school

that all children are competent and capable, the limitations of licensing and societal fears

surrounding injury were ever-present and often restricted the play opportunities I was able to

offer my students. Many times, my fear of children becoming injured and the consequences that

could follow, not to mention licensing restrictions, led me to limit play experiences that children

considered valuable and exciting. A key example of this is the opportunity to climb. Even though

we had a wonderful tree in the yard with branches perfectly spaced for young children to climb, I

had to redirect the children to climb on the playground structure because it was approved by

licensing and was considered safe. I understood the value of the challenge that comes with

climbing trees, but I feared children getting hurt and my own liability as a result of such play.

Now that I am an occupational therapy student, I recognize the importance of play on

13
development from a new perspective and value it as one of the key occupations of childhood

(Parham & Fazio, 2008). The role of occupational therapists is to empower and support our

clients to participate in the occupations (all the ways people spend their time) that they find

meaningful in order to promote their health, well-being, self-efficacy, and quality of life. Upon

reflection of my time as a teacher from this occupational therapy perspective, I recognize the

barrier to many types of play activities is a serious risk to the health and wellbeing of children.

In a written address to educators in Reggio Emilia, Italy titled “The Image of the Child--

Where Teaching Begins”, Loris Malaguzzi (1994) wrote about the powerful transaction between

children and available environments and opportunities. He called on teachers to recognize the

power of play experiences when children are allowed to be children and engage in social

relationships, experiment with materials, and follow their interests without adult interruption.

“What we want to do is activate within children the desire and will and great pleasure that comes

from being the authors of their own learning” (Malaguzzi, 1994, p. 3). Malaguzzi (1994, p. 5)

concluded his address with a powerful challenge for teachers: “Instead of always giving children

protection, we need to give them the recognition of their rights and of their strengths.” I believe

one way of supporting children’s learning through risky play is by helping teachers gain

knowledge and strategies to face feelings of uncertainty and refrain from interrupting play and,

instead, empower children to acknowledge their own strengths and boundaries related to risk.

All children, regardless of ability and background, have the right to play (U.N.C.R.C.,

1989). During the United Nation’s Convention on the Rights of the Child in 1989, people all over the

world received a call to action to protect this right and provide children with safe and nurturing

opportunities for learning and development through play (U.N.C.R.C., 1989). Through this research,

I respond to this call and uphold my responsibility to advocate for children’s rights to play by arguing

that the best possible opportunity for development lies in benefits gained through opportunities for

14
risky play. I chose to participate in this research to help address this problem and remove barriers

to risky play by contributing a tool that will help teachers think differently about their own

tolerance of risky play and help identify ways to promote this important right for all children.

Research Questions

The aim of this study is to assess the psychometric properties of data gathered with the

Tolerance of Risk in Play Scale for Teachers (T-TRIPS). This aim is guided by the following

research questions:

1) Do data collected by this instrument demonstrate evidence of construct validity?

2) Do data collected by this instrument demonstrate evidence of internal reliability?

Overview of Thesis

In this chapter, I presented the background to the research, explained the rationale for and

significance of the study, introduced the research that gave rise to this study and my own

background, and identified the research aims and questions.

In Chapter 2, I presented a review of available instruments to assess adult perspectives of risk

and play as well as instruments to assess the play environment on school playgrounds.

Chapter 3 comprises includes a manuscript formatted for the journal to which I intend to

submit it.

In Chapter 4, I discuss the findings and implications for future research as well as personal

reflections of the thesis experience.

15
 CHAPTER 2

LITERATURE REVIEW

This critical synthesis of the literature includes two sections and several sub-sections. The

first evaluates existing adult-report measures related to children’s play and risk. The second

focuses on available observation-based tools that evaluate the school playground and children’s

play on school playgrounds.

- Adult-report Measures Related to Play and Risk

o Adult Perceptions of Risk – Beneficial or Detrimental?

o Child Play and Behaviors as Predictors of Risk and Injury

o Adult Practices that Contribute to Risk and Injury

- Observational Measures of Play and Environment on the School Playground

o Evaluations of the School Play Environment

o Children’s Play on the Playground

o Play for Children with ASD/ ID

To find relevant literature for this study, I completed a search of electronic databases,

university textbooks, and websites. In addition, I engaged in handsearching the reference

sections of related literature and theses. Electronic databases that were searched include: ERIC,

CINAHL, MEDLINE, Academic Search Premier, and PubMed. Search terms included: risk,

tolerance, attitudes, perceptions, measure, instrument, play, children, teacher, playground, fear,

risk assessment, risky play, tool, questionnaire, measurement, survey, questionnaire, attitude to

risk, risk reframing, playthings, child, children with disabilities, children with autism. I searched

websites including Google Scholar (http://scholar.google.com), Shirley Ryan AbilityLab

(https://www.sralab.org/

16
rehabilitation-measures), and Sydney Playground Project (https://www.sydneyplayground

project.com/publications) for literature related to teachers’ risky play tolerance.

Inclusion criteria for instruments and observation tools in this literature review are: (1)

those intended for use in reference to children 4- to 14-years-of-age, (2) those to be completed by

adult caregivers (teachers and parents) , and (3) those that assess children’s play and play

behaviors, risky play, adult opinions of risky play, adult behaviors that contribute to risk, and

perspectives of the play environment. Few measures and observation tools exist that are

specifically intended for children with autism spectrum disorder (ASD) and intellectual disability

(ID); however, several are appropriate for use with children of all abilities. I included those.

Additionally, I included adult-report instruments intended for use in various environments

(playground, home, and community).

Topics of play and risk that are beyond the scope of this review are those that are:

indirectly relate to play; observational tools focusing only on child physical activity levels; child-

report measures of play and risk tolerance; adult-report measures of a child’s capacity,

playfulness, or skill related to play; and observational tools intended for use anywhere except the

school playground. I was unable to access some instruments mentioned in the literature. I

attempted to access these instruments through interlibrary loan at Colorado State University and

conducted an extensive search of the sites included above; many instruments were published

prior to digital publication and were unavailable.

Before discussing available instruments, I wish to acknowledge the qualitative studies

found during this literature review that examine teacher’s perspectives or beliefs of risky play

through various data collection methods-- interviews, focus groups, email correspondence, and

reflection by researchers (Cevher-Kalburan, 2014; Lindqvist, Nordänger, & Landahl, 2009;

17
Little, 2010; Little, Sandseter, Hansen, & Wyver, 2012; McFarland & Laird, 2017; Niehues et

al., 2013; Sandseter, 2011; Sandseter, 2012, 2014; Spencer et al., 2016). I mention these studies

only briefly as they, too, are beyond the scope of this review. Researchers studied a variety of

topics related to risky play using qualitative methods to understand teacher perspectives

including: tolerance of risky play by the children in their care, beliefs of the value of risky play,

perspectives of barriers to risky play, and perspectives of consequences associated with risky

play. While these studies align with the purpose of this thesis and offer valuable information on

teacher perspectives of risky play, they did not use a standardized instrument to measure adult

risk-tolerance. Additionally, many of these studies had small sample sizes and included time-

consuming procedures, which limit the generalizability or replicability of these methods in future

studies. Some studies used tools specifically developed for their research including “self-report

scale”, “survey”, or “tool” (Sandseter, 2014), but the majority of these were difficult to find and

if found, they have not been evaluated for their psychometric properties.

Adult-Report Measures Related to Play and Risk

I found several adult-report measures that evaluate the behaviors of children and adults

related to play and risk. Few considered risk to be a benefit; most considered risk to be a definite

precursor to injury. Measures that describe risk as detrimental are intended to aid in mitigating

any possibility of risky experience, therefore removing any possibility of injury. Other

instruments assess a child's behaviors as a precursor to injury; some assess adult behaviors and

consider them a precursor to injury. Several measures were intended for a specific age group and

are not appropriate for use with children in a primary school setting; none are intended

specifically for children with disabilities.

18
 Two measures I reviewed in this section evaluate two different constructs within the

same instrument: perceptions of risk and adult practices that contribute to risk. Table 2.1 contains

a brief description of these measures along with their psychometric data. While several adult-

report measures of children’s play and risk exist, none of the available instruments assesses

teacher tolerance of risky play on the school playground.

Measures of Adult Perceptions of Risk: Beneficial or Detrimental?

Adults’ fear of the consequences of child injury and a societal tendency towards risk

avoidance has resulted in surplus safety (Buchanan, 1999; Wyver et al., 2010); many measures

were developed that view risk as dangerous and some measure adult perceptions that risks are

detrimental. For example, the Worry Assessment and Risk Estimation Scale (WARE) (Will,

Lorek, Sabo, & Kidd, 2009) considers risk as a direct cause of injury.

Nonetheless, some measures that capture adults’ perspectives that risk could be beneficial

do exist. The Risk Engagement and Protection Survey (REPS) (Olsen, Ishikawa, Masse, Chan, &

Brussoni, 2018), the Parental Perception of Positive Potentiality of Outdoor Autonomy for

Children (PPOAC) (Prezza, Alparone, Cristallo, & Luigi, 2005), and an untitled scale used by

(Sandseter, 2014) all perceive mild risk or risky play as potentially beneficial for child

development and assess adult perceptions of risk from that perspective. The REPS and the

WARE each measure two different constructs, therefore, I discuss them in two sections of this

literature review.

The REPS was administered to fathers visiting the hospital with their child and assesses

two seemingly contrasting concepts (Olsen et al., 2018): a father’s approach to protect his child

from injury and his approach to engaging his child in healthy risk-taking. Olsen et al. (2018)

discovered that while the subscale for each concept is psychometrically sound, the two concepts

19
have a weak correlation and thus belong to two independent constructs. Additionally, REPS

assesses the fathers’ perceptions that risky physical play has developmental value but does not

consider other types of risky play (Sandseter, 2009) and therefore provides an incomplete picture

of risky play perceptions.

The PPOAC is intended to assess the degree to which parents perceive their children

playing outdoors alone in their neighborhood is a worthwhile risk that may promote autonomy

(Prezza et al., 2005). While this instrument’s assumption that risky situations could positively

impact child development is promising for shifting perspectives of risky play, it provides only a

limited glimpse into adult perceptions of risk and does not touch on different types of risky play.

Additionally, the PPOAC does not evaluate the adult caregiver’s tolerance of risky play (Prezza

et al., 2005).

Following her work to understand risky play, Sandseter (2014) conducted a new study

using a custom scale to collect data on teacher’s perceptions of risky play. This new scale,

however, has not been proven to produce data that are psychometrically sound and is therefore

unable to be used in future research as a standardized measure.

Unlike the previous three measures, the WARE assesses levels of parent concern for risk

and the relationship between their concerns and potential injurious situations (Will et al., 2009).

This measure is intended for use with 6-year-old children. Notably, the authors of the WARE

intentionally chose the items to cover a risk continuum, indicating their understanding that risk

has different degrees of severity. However, unlike other positive risk scholars, the authors of the

WARE view risk as detrimental (Ball, 2002; Ball et al., 2012; Brussoni, Brunelle, et al., 2015).

Measures of adult perspectives of risk as being either beneficial or detrimental do

demonstrate some understanding of the benefits of risky play; however, with the exception of the

20
tool used by Sandseter (2013), they only scratch the surface of this important topic. These

measures consider only one type of risky play and or only take a narrow, negative assumption of

risk. This narrow consideration is in stark contrast to risky play scholars’ arguments that (1) there

are various severities of risk and (2) several unique categories of risky play exist, each with

specific benefits to child development (Ball, 2002; Sandseter, 2009). In addition, these measures

are intended to consider children of specific limited age ranges; none are appropriate for all

children in a primary school setting (4-12 years). The scale developed by Sandseter (2013) is the

only tool in this review that specifically measures teacher perspectives of risky play activities in

a school setting; all others are intended for parents in a variety of environments that differ greatly

from the school playground. Unfortunately, Sandseter’s (2013) scale is not a standardized

measure. Instruments like the ones mentioned above are impractical for teachers to use on a

playground because they refer to a much smaller number of children than the entire class for

which the teacher would be responsible. There are significant limitations with all of these

measures, but it is important to acknowledge the efforts of the authors towards valuing the

benefits of risk.

Measures of Child Play and Behaviors as Predictors of Risk and Injury

Researchers have attempted to understand how children spend their time and measure the

behaviors children engage in that can potentially result in dangerous situations and injury.

Unfortunately, measures like the Play Activity Questionnaire (PQ) (Finegan, Niccols, Zacher, &

Hood, 1991), the Sensation Seeking Scale (SSSC) (Morrongiello & Lasenby, 2006) and the

Injury Behaviors Checklist (IBC) (Speltz, Gonzales, Sulzbacher, & Quan, 1990) place the cause

of injury within the child rather than an interaction between the child and the physical and social

environment. The PQ (Finegan et al., 1991) is intended to assess children’s preferences for

21
different types of play activities and identify gender preferences for these activities. While this

measure provides adults with a glimpse of how children spend their time and the item ‘rough and

tumble’ play equates to one category of risky play (Sandseter, 2009), the PQ does not consider

risk or adult perception of risky play.

The SSSC (Morrongiello & Lasenby, 2006) is grounded in the assumption that sensation

seeking behaviors will result in a child engaging in physically risky play and potentially

becoming injured. The Injury Behaviors Checklist (IBC) (Speltz et al., 1990) also evaluates the

behaviors of a child that could result in injury. Both measures assume the cause of injury resides

within the behaviors and actions of the child and both measures view risky situations as being

inherently negative. Additionally, the PQ and the SSSC are parent-report and are inappropriate

for use by teachers because of the stark differences in activities and environments experienced

between parents and teachers. These measures fall short of addressing adult’s perceptions of

risky play, especially those of teachers.

Measures of Adult Practices that Contribute to Risk and Injury

In an attempt to mitigate injury, researchers have tried to understand adults’ roles in risky

situations by measuring parents’ behaviors related to risk and injury. Instruments including the

Parent Supervision Attributes Profile Questionnaire (PASPQ) (Morrongiello & Corbett, 2006),

the Supervision Attributes and Risk-taking Questionnaire (SARTQ) (Morrongiello, Corbett, &

Kane, 2011), the Worry Assessment and Risk Estimation Scale (WARE) (Will et al., 2009), and

the ‘protection’ component of the Risk Engagement and Protection Scale (REPS) (Olsen et al.,

2018) place blame on parents for child injury instead of considering the interaction between the

parent, the child, and the environment.

22
 The PASPQ is intended for parents of preschool-aged children and the SARTQ is

intended for primary school-aged children. In an attempt to identify potential cause of injury,

both measures evaluate supervision-related behaviors and beliefs of parents as predictors of their

children’s risk of injury (Morrongiello & Corbett, 2006; Morrongiello, Corbett, & Kane, 2011).

Notably, both the PASPQ and the SARTQ evaluate parents’ tolerance of their children’s risk-

taking, albeit with a detectable negative perception of risk. Both the PASQ and the SARTQ

consider all risk to be dangerous and detrimental, therefore assuming the need to minimize risk

to protect the child from injury. These measures view the potential cause of injury as residing

within the parent because of their behaviors, even when asked to consider the child’s risky

behaviors.

The other half of the REPS assesses fathers’ approach to protecting their children from

injury (Olsen et al., 2018) by measuring the degree to which fathers engage in various behaviors,

for example, knowing what their child is doing at all times and making sure the home is free of

all hazards (Olsen et al., 2018). While some items in this measure do address father comfort with

risky situations, child injury is the assumed result if a father fails to fully engage in the item

behaviors. There is no middle ground to allow for engaging in manageable risky play.

Similar to previous subsections of this review, these measures have significant limitations

for providing safe opportunities for risky play. A chief issue of these measures is the assumption

that a parent’s behaviors are directly responsible for risk resulting in negative and dangerous

consequences. Additionally, none of the above measures are intended to be used by a teacher in

consideration of a group of children; none of the measures are appropriate for a school

playground, and none evaluate an adult’s tolerance of risky play.

23
Table 2.1

Adult-report Measures of Play and Risk

Instrument Age Reporter & Purpose Description Psychometrics

(Author) Environment
Adult Perceptions of Risk – Beneficial or Detrimental?
*Risk Engagement 6-12 Assess fathers’ 8 items: protection from Injury protection
and Protection years approach to protecting injury α = 0.75
Survey (REPS) their children from r = 0.88
(Olsen et al., 2018) injury and engaging 6 items: risk engagement
children in risks. (attitudes towards child risk Risk engagement
engagement AND benefits α = 0.77
While at hospital for an of risk engagement) r = 0.48
injury / non-injury
reason Low correlation
between injury
protection and risk
engagement factors
indicate they are
separate constructs
Parental Perception 9-10 Parents Assess parent 5-item scale α = 0.74
of Positive years perception that Higher scores indicate
Potentiality of Outdoor play children playing alone positive perceptions of
Outdoor Autonomy environments outdoors is a potentiality to develop
for Children within worthwhile risk that autonomy
(PPOAC) neighborhood will promote their
(Prezza et al., 2005) child’s autonomy

24
Questionnaire of Teachers Assess practitioner’s 11 item scale, each item Not tested
practitioner’s attitudes of risky play with two statements (one
personal risk-taking more risk tolerant, one more
and perceptions of risk averse)
children’s risk-
taking (scale)
(Sandseter, 2014)
Child Play and Behaviors as Predictors of Risk and Injury
Play Activity 4-12 Parents Assess children’s 15 items (active and Moderate level of
Questionnaire (PQ) years interest in different adventurous, athletic, rough- parent agreement
(Finegan, Niccols, types of play and and-tumble, and quiet) r(79) = 0.70, p <
Zacher, & Hood, distinguish between 0.001
1991) boys’ and girls’ 7-point scale α = 0.81
preferred play
activities

25
Injury Behaviors 2-5 Parent or Measure of behavioral 24 items describing various α =0.87
Checklist years teacher characteristics that specific risky behaviors r =0.81 (p < .01)
(Speltz et al., 1990) predict injury in
children. Possible total score range 0- Validity of high injury
96 liability:
M = 33.7
SD = 13.4
Validity of moderate
injury liability:
M = 24.4
SD = 9.7
Validity of low injury
liability:
M = 22
SD = 10.5

The effect of injury

liability on IBC scores
was significant
( 22.33, p < .001)
Sensation Seeking 7-12 Parent- and Measure sensation 32 items of sensation Internal reliability:
Scale for Children years child-report seeking in relation to seeking and tendency in Boredom
(SSSC) physical risk taking. five categories of behavior Susceptibility
(Morrongiello & (boredom susceptibility, α =0.70
Lasenby, 2006) thrill seeking, behavioral Behavioral Intensity
intensity, novelty seeking, α =0.83
and inhibitory control) Behavioral Inhibition
α =0.83
Novelty Seeking
α =0.45
Thrill Seeking
α =0.85

26
 Adult Practices that Contribute to Risk and Injury
Worry Assessment 6 Assess levels of 21 injury and health hazard Internal Consistency:
and Risk Estimation years parental concern and scenarios, intended to cover α = 0.89
Scale (WARE) perceptions of risk a risk continuum
(Will et al., 2009) associated with injury
hazards. 11-point scale
Parent Supervision 2-5 Parents Assess parent 29 items addressing (1) α = > 0.70 for all
Attributes Profile years supervision in relation protectiveness, (2) subscales
Questionnaire to child risk of injury supervision beliefs, (3) r = > 0.70 for all
(PASPQ) tolerance for children’s risk subscales
(Morrongiello & taking, (4) belief in fate Goodness of Fit Index
Corbett, 2006) determining safety = 0.93
Comparative Fit Index
5-point scale = 0.96
Standardized Root
Mean Square Error of
Approximation = 0.06
x2 = 1.79
Supervision 7-10 Parents Assess perceptions of 3 subscales: Overall α = 0.73
Attributes and Risk- years supervision and risk- - parent need for
taking Questionnaire taking in relation to psychological control C α = 0.72
(SARTQ) risk of injury (C)=15 items
(Morrongiello et al., - beliefs in the value of VS α = 0.78
2011) Assumes the need to supervising children at
minimize risk these ages (VS)= 17 items CRT α = 0.75
- children’s risk-taking
(CRT)= 18 items

5-point scale
*These instruments fit within two categories: adult perceptions of risk and adult practices that contribute to risk

27
Observational Measures of Play and Environment on the School Playground

To understand the role of teachers and risky play experiences on the school playground,

researchers have developed several measures to evaluate affordances of the environment, how

children spend their time, and the dynamic between the two during play at school. I identified

playground observational measures that evaluate characteristics of the play context, measures of

children’s play and behaviors, and two measures specific to children with disabilities. I only

found one measure that acknowledges the existence of risky situations present in the playground

environment and evaluates the impact of those situations on play. I did find one measure that

evaluates play policy and opportunity related to play participation. I identified some measures

that evaluate the play environment and several observational measures that assess children’s

levels of physical activity, types of play, and social interactions on the playground. Table 2.2

comprises a brief description of these measures along with information about psychometrics.

While several playground-specific measures exist, none evaluate teacher perspectives of

risky play. Nonetheless, these measures provide insights into the play behaviors of children,

which can then help inform opportunities for safely managing risky play (Cooke et al., 2019;

Niehues et al., 2013; Spencer et al., 2016).

Evaluations of the School Play Environment

Researchers have attempted to measure the quality of the play environment and the

resulting activities and behaviors of children through use of observation measures like the

School Physical Activity Policy Assessment (S-PAPA) (Lounsbery, McKenzie, Morrow, Holt, &

Budnar, 2013) and the Great Recess Framework Observation Tool (GRF-OT) (Massey, Stellino,

Mullen, Claassen, & Wilkison, 2018). Both measures focus on the availability and quality of

physical and social environment variables and the play activities and behaviors that take place on

28
primary school playgrounds. S-PAPA is unique in that it considers school policies of physical

activity alongside characteristics of physical education and school recess with the intention of

identifying opportunities to increase physical activity (Lounsbery et al., 2013). The GRF-OT is

intended to assess the quality of the playground environment and resulting child play behaviors,

particularly to detect change in the playground environment following intervention (Massey et

al., 2018). Notably, both measures emphasize the physical components of a playground but only

briefly consider social components such as adult supervision or interaction with peers and

teachers. These measures address important concepts related to children’s opportunities for play

on the playground but fail to provide a complete picture of the characteristics contributing to

risky play situations and do not assess tolerance of risky play.

Physical play is not the only type of play, nor the most important, that children engage in

on the school playground; however, S-PAPA’s primary outcome is physical activity (Ball et al.,

2012; Gill, 2018; Lounsbery et al., 2013). Additionally, the S-PAPA is concerned about policies

surrounding play and the impact of policies on promoting opportunities for physical activity. The

GRF-OT does expand its focus from physical activity to other play behaviors with peers and

appropriate use of materials; only one item on the GRF-OT evaluates teacher interaction with

children during play, and two items assess behaviors that are included in risky play categories

(Sandseter, 2009)- solving social conflicts with peers and appropriate, and ‘safe’ use of materials

(Massey et al., 2018). An emphasis on safety is evident in both measures and any mention of risk

is considered negative and dangerous; neither include perspectives on the developmental benefits

of risky play.

One other tool is worthy of consideration, however, it is a protocol, not a measure. The

Risk-Benefit Assessment (RBA) (Ball et al., 2012) was developed as a strategy for evaluating the

29
beneficial or detrimental consequences of environmental conditions that are considered risky.

The protocol acknowledges that there are varying degrees of risk and many risky situations

provide opportunities for child development of physical, social, and cognitive skills. While the

Risk-Benefit Assessment documents the assessor's understandings of the pros and cons of risky

play, it is a protocol for assessing the environment, designed to take place before play begins and

does not capture teachers’ subjective tolerance of risky play.

All three of these measures focus on the objective details of the playground environment;

however, only the RBA assumes that children want to engage in risky play, and none evaluates

teachers’ perspectives of risky play. An understanding of risky play opportunities afforded by the

environment alone is not enough; instead, adults must gaining an understanding of how children

want to spend their time on the playground which subsequently may assist teachers in identifying

ways to support manageable risky play experiences.

Measures of Children’s Play on the Playground

Researchers have developed systematic observation measures to evaluate what children

do, both play and non-play, on the school playground. The System for Observing Play and

Leisure Activity in Youth (SOPLAY) (McKenzie, 2002) assesses physical activity level and

group size during play on the playground. The System for Observing Children’s Activities and

Relationships during Play (SOCARP) (Ridgers, Stratton, & McKenzie, 2010) evaluates

children’s physical activity, group size, types of play activities, and social behaviors during play

on the playground. The System for Observing Outdoor Play (SOOP) (Engelen et al., 2017)

measures any change in play following interventions or changes to the playground environment

and also considers peer interaction, types of play activities, and group size. All three of these

measures focus on children’s play as the primary outcome and acknowledge social and physical

30
environmental variables, however, they emphasize how much a child is moving around on the

playground over other important types of play or the quality of the play experience. Only the

SOOP documents the location of children and the presence of teachers on a map, but it does not

evaluate impact of teacher-child interaction on play behaviors. One key limitation of these

measures is that they are time sampling techniques--a researcher documents one area of a

playground in a snapshot of time. While this method is effective for understanding transactional

relationships among children, the environment, and teachers by tracking patterns of play, group

sizes, and materials, it only considers these briefly. None includes types of risky play and none

evaluates teachers’ (or children’s) level of tolerance of risky play on the playground.

Measures of Play for Children with Autism and/or Intellectual Disability

Observation measures have been developed that are specifically intended to assess the

play behaviors of children with disabilities, autism spectrum disorder (ASD) and intellectual

disability (ID), on school playgrounds. In a recent study, Grady-Dominguez et al. (2019)

described a new purpose-built iPad application intended to measure characteristics of play and

play behaviors of children with ASD and ID on school playgrounds; the assessment produced

psychometrically sound data. Like the three observational measures mentioned above, this

measure uses a digitally-prompted time sampling technique but this measure does prompt

examiners to document the play activities, social interactions between peers, and characteristics

of the environment. This measure evaluates the difficulty of the physical and social

characteristics of the playground environment and the ability of the environment to facilitate

various complexities of play behaviors of children with disabilities, but it does not include risky

play specifically. It also does not evaluate teachers’ tolerance of the various play behaviors,

which might reflect their risky play tolerance.

31
 One playground observation measure specifically related to children with disabilities is

the Playground Observation Checklist (POC) (Ingram, Mayes, Troxell, & Calhoun, 2007).

Unlike the other measures in this section, this instrument is intended to aid members of a school

disability assessment team in diagnosing children as having ASD or ID based on the play

behaviors they demonstrate on the playground (Ingram et al., 2007). Children with autism have

expected play behaviors according to past research and this measure has strong predictive power

to diagnose autism based on those expectations. That being said, this instrument is not

appropriate for evaluating a child with autism’s preferred play activities on the playground

because it does not document preferences. Furthermore, this instrument, like others, does not

include risky play activities and does not evaluate teachers’ levels of risky play tolerance. Both

of these features could help promote opportunities for children with ASD and ID to engage in

developmentally beneficial risky play experiences.

32
Table 2.2

Observational Measures of the School Playground

Instrument Age Purpose Description Psychometrics

(Author)
School Play Environment
School Physical Primary Measure school policy and Items include: 89/96 items w/ moderate
Activity Policy school environmental variables - background questions (7 to almost perfect
Assessment related to the quantity and items) agreement
(S-PAPA) quality of physical activity in - physical education (47 (Kappa = 0.42–0.87)
(Lounsbery et al., children at school. items)
2013) - recess (27 items) 7/96 items w/ fair
To be completed by school - other before, during, after agreement
PE teachers school programs (15 (Kappa = 0.13–0.39)
items)

Open-ended, dichotomous,
multichotomies, and checklist
Great Recess Primary Observational tool to assess 17 items of recess Significant overall
Framework school the quality of the playground environment: safety and (p < 0.001)
Observational Tool context (environmental, structure, student behaviors,
(GRF-OT) social, behavioral) and the adult engagement/ Convergent validity:
(Massey et al., 2018) behaviors that take place supervision, transitions. structure and safety
during recess (p < 0.001, β = 0.272)
4-point scale adult engagement/
supervision
(p < 0.001, β = 0.246)
student behaviors
(p = 0.024, β = 0.102)
interrater reliability:

33
 strong agreement with an
ICC (2,1) of 0.951 (95%
CI, 0.932, 0.964).

Test-retest reliability:
(ICC = 0.949, 95% CI,
0.882, 0.979) w/ 3-day
average across 2 time point
Risk-benefit n/a Protocol to evaluate the Four levels: (1) policy Not reported
Assessment potential risks and associated framework, (2) risk-benefit
(Ball et al., 2012) benefits of environmental assessment, (3) technical
conditions and paly activities inspection, (4) dynamic risk-
on playgrounds . Used to benefit assessment.
determine if play policy
needs to change to mitigate Keep detailed documentation
risk. of evaluation of risks and
associated benefits to make
Teachers, school admin, judgements about the play.
policy makers
Children’s Play on the Playground
System for Observing School- Systematic observation tool Momentary time sampling Interrater reliability:
Play and Leisure aged (K-12) to measure the number of technique: observe activity of contextual variables
Activity in Youth students and their levels of participating children = 88-97%
(SOPLAY) physical activity during play (‘sedentary’, ‘walking’, or intraclass correlations
(McKenzie, 2002) on the playground ‘very active’ and type of = 0.76 - 0.98
activity), alternating
boys/girls Interobserver agreements:
(IOA=80%)
Contextual conditions: time,
temp, area/ equipment Intraclass correlations:
accessibility, presence of (R=0.75)
supervision, presence/ type of
organized activity

34
System for Observing School- Systematic observation tool Time sampling technique: 10 Interrater reliability = 88-
Children’s Activities aged (K-12) to assess children’s physical sec observation followed by 90%
and Relationships activity, social group size, 10 sec record period for each
during Play types of activities, and social child Concurrent validity with
(SOCARP) behaviors during play accelerometry = r =0.67, p
(Ridgers et al., 2010) <0.01
System for Observing Primary Systematic observation Counter-clockwise scanning Interrater reliability = 95%
Outdoor Play school-aged instrument to assess changes areas of playground 1 x/ min
(SOOP) in social and creative play for 20 min
(Engelen et al., 2017) and use of loose parts recorded on map:
materials. Documented: - number of kids (boys/girls)
number of children playing - presence of teachers
together, number of children
in an area, types of 10 categories (active play;
activities/play construction; creative/
imaginative play; eating,
Assess play and non-play moving-not playing; inactive
activities, groups, number of (sitting or standing) play;
children playing together, and inactive – not social (alone or
presence of teachers. no interaction); and inactive
– social, sports)
Play for Children with ASD / ID
*Purpose-built iPad Primary Observation tool to measure 19 item tool Infit MnSq = 0.83-1.19
application to school-aged characteristics of play and Researcher scored one Outfit MnSq = 0.46-1.45
measure what play behaviors of children quadrant of the playground at
children do on the with autism and intellectual a time, moving clockwise Point Measure Correlation
playground disabilities on the school = 0.11-0.61
(Grady-Dominguez et playground
al., 2019) Test reliability = 0.60-0.67

Session separation index =

1.23

35
 Playground 5 – 11 years Assess children’s play and Completed by members of Interrater reliability: 100%
Observation social behaviors in school assessment team.
Checklist (POC) playground environments (F = 102.7, p < 0.0001)
(Ingram et al., 2007) 10 item checklist
Tool to aid identification of (engages in social play; not Positive predictive power
children with disabilities socially isolated; respects for autism: 69%
(ASD and ID) boundaries/personal space;
not exhibit socially ‘Uses playground
inappropriate behavior; equipment functionally’
follows game rules; responds did not differentiate
to winning/losing; initiates between groups (x2= 0.3–
conversation w/ peer; 4.2, p > 0.04)
sustains conversation w/
peer; does not exhibit gross
motor incoordination; uses
playground equipment
functionally)

Dichotomous scale (‘present’

or ‘absent’)
*This instrument fits within two categories: child’s play/ play behaviors and play of children with ASD / ID

36
 Conclusion

My review identified a glaring gap in appropriate available tools to assess teacher

tolerance of risky play in a school setting. Few existing measures have a positive perspective of

risky play and consider the potential benefits risky play can have on child development (Olsen et

al., 2018; Prezza et al., 2005). Many measures identify the cause of risk and injury as originating

within the behaviors of the child or parent and categorize risk as a negative, dangerous, and an

inevitable precursor to child injury (Morrongiello & Corbett, 2006; Morrongiello, Corbett, &

Kane, 2011; Will et al., 2009). This latter category of measures assumes all possibilities of

engaging with risk must be limited, or better, eliminated.

In addition to measures of attitudes towards risky play, I evaluated several playground

observation measures in an attempt to identify any that could provide insight into the transaction

between children’s play and characteristics of the environment, available materials, opportunities

for peer interaction, supervision, and activities. Only one protocol for evaluating the play

environment included an assessment of the possible benefits and detriments of risk and risky

play (Ball et al., 2012). Other playground-specific measures are intended to assess activity levels,

types of play, and social interactions of children during recess, but do not identify opportunities

for risky play (Engelen et al., 2017; Grady-Dominguez et al., 2019; Massey, Stellino, Hayden, &

Thalken, 2019; McKenzie, 2002; Ridgers et al., 2010). I found only two observation measures

specific to children with disabilities for use on school playgrounds (Grady-Dominguez et al.,

2019; Ingram et al., 2007). One measure is intended to aid in diagnosis of ASD or ID though

observation of play on the playground and is therefore not relevant to my research questions.

One playground measure specifically evaluates the school’s play policies related to child play

behaviors and availability of play resources, but it does not assess teacher tolerance of risk or the

37
impact of teacher tolerance on children’s opportunities for risky play (Lounsbery et al., 2013).

While many available playground observation measures consider the characteristics of the

environment in relation to play, they fail to consider the role of the environment in providing

opportunities for manageable risky play.

Available research on risky play advocates for offering manageable risky play

experiences to children of all abilities on the school playground, but also identifies teacher risk

aversion, fear, and lack of understanding of how to offer manageable risks as a barrier to this

important developmental opportunity (Ball et al., 2012; Brussoni, Brunelle, et al., 2015; Brussoni

et al., 2012; Little et al., 2012; Sandseter, 2011; Sandseter, 2012, 2014; Wyver et al., 2010).

Unfortunately, there are no measures currently available that appropriately evaluate teacher

tolerance of risky play on the school playground in order to help teachers reframe their

understanding of risk and develop strategies for implementing safe risky play. The shortcomings

of available measures indicate there is a gap in the literature- there is a need for a new measure

for primary school teachers, in reference to children of all abilities, and one that is appropriate

for use in the unique physical, social, and behavioral context of a school playground. The

Teacher Tolerance of Risk in Play Scale (T-TRiPS) was developed from the original TRiPS (Hill

& Bundy, 2014) to fill this gap and establish a means for evaluating teacher tolerance of risk in

play. This instrument must be evaluated to determine if the data produced suggests evidence for

construct validity and reliability. A psychometrically sound instrument to measure teacher

tolerance of risk in play will help increase opportunities for risky play experiences for children

and help teachers feel confident in their facilitation of such play.

38
 CHAPTER 3

JOURNAL ARTICLE

Teacher Tolerance of Risk in Play Scale (T-Trips):

Evaluating the Psychometric Properties of a New Measure

Much of children’s active play inherently holds the possibility of minor injuries, like

bumps and bruises, and thus is considered to be “risky.” Risky play includes exciting and

thrilling activities, involves exploration of physical and social environments, and includes

exploration of emotions and perceptions of the world (Ball, 2002; Kleppe et al., 2017; Lester &

Russell, 2014; Sandseter, 2009). Because risky play includes potential injury, it requires children

and adults to accept uncertainty and relinquish control (Ball, 2002; Kleppe et al., 2017; Lester &

Russell, 2014; Sandseter, 2009).

Previous researchers have identified the following risky play categories: (a) great heights,

(b) high speeds, (c) dangerous tools, (d) dangerous elements, (e) rough-and-tumble, (f)

disappear/get lost, (g) play with impact, and (h) vicarious risk (Kleppe et al., 2017; Sandseter,

2009). The above categories focus on physical play; however, other risky play activities exist

that do not fit within these categories such as challenges during social problem solving or

experimenting with a child’s own perceptions of their abilities.

While many adults attempt to avoid any possibility of injury for children, engaging in

risky play promotes development including physical activity, executive functioning, skills for

social interaction, problem-solving, creativity, self-efficacy, and self-esteem (Brussoni, Gibbons,

et al., 2015; Brussoni et al., 2012; International School Grounds Alliance, 2017; Wyver et al.,

2010). Children of all abilities develop an understanding of their own skills and boundaries

through opportunities to independently manage play (Ball, 2002). Challenging play experiences

39
help children become aware of, and increase their competencies for, managing risky situations

without relying on adults (Brussoni, Brunelle, et al., 2015; Bundy et al., 2015).

Imbalance of Duty of Care and Dignity of Risk: A Barrier to Development

Duty of care refers to a teacher’s responsibility to protect children from harm.

Alternatively, dignity of risk implies the right to gain skills to manage risk and develop

autonomy (Wolfensberger et al., 1972). At school, recess provides an essential opportunity for

play that supports development of autonomy (Bundy et al., 2015; Hyndman, 2015; Ramstetter et

al., 2010). However, teachers’ fears of liability (Newnham, 2000; Niehues et al., 2013) often lead

to play supervision practices that reflect an imbalance between duty of care and dignity of risk.

Over-emphasis on duty of care results in teachers attempting to eliminate risky play on

playgrounds (Spencer et al., 2016). Even though many teachers recognize benefits associated

with risky play, fear of the consequences of child injury often results in attempts to eliminate any

form of risk, no matter how minor, without regard for possible benefits (Little , 2016; Wyver et

al., 2010). Consequently, children miss opportunities to experience its benefits (Ball et al., 2019;

Newnham, 2000; U.N.C.R.C., 1989). While most people who work with children have children’s

best interests at heart and enact protection strategies with good intentions, teachers’ tendency

towards surplus safety, fear of social consequences and lack of understanding of the benefits

often threaten children’s right to engage in risky play.

Risky Play for Children with Disabilities

Risky play situations on school playgrounds are especially limited for children with

disabilities. Children with disabilities, particularly those with cognitive limitations such as

judgement, inhibition management, and problem solving (e.g., children with autism spectrum

disorder [ASD] or intellectual disability [ID]), often experience greater interruptions to

40
developmentally important risky play than do their typically-developing peers (Bundy et al.,

2015).

In order to develop the important skills associated with risky play, children with

disabilities require repeated exposure to a variety of situations with ever-increasing levels of

difficulty (Bundy et al., 2015). Thus, risky play is especially important for children with

disabilities (Bundy et al., 2015; Gilman, 2007). Nonetheless, despite its benefits, many adults

perceive risky play as dangerous, unnecessary, and offering only negative consequences,

particularly for children with disabilities. Low tolerance of risky play leads adults to limit

opportunities for these developmentally meaningful play experiences (Beetham et al., 2019;

Bundy et al., 2015; Little et al., 2012; Scott, Jackson, & Backett-Milburn, 1998)

In a study of teachers’ responses to risky play while supervising children with disabilities

on the playground, Spencer et al. (2016) found that adults often justified their interruptions of

play by citing duty of care coupled with the perception that children with disabilities are

incapable of managing risky situations or understanding the consequences of their actions.

Findings from this study suggested not only that teachers view a child’s disability as a potential

risk for injury but also that teachers are strongly influenced by fears of social consequences such

as having to report injuries to parents or being perceived as negligent (Spencer et al., 2016).

Consequently, teachers often determine safety rules based on the least capable child in the group

and apply these to the entire group, eliminating any opportunity for children to develop new risk

management skills regardless of individual children’s varying capabilities (Ball et al., 2012;

Bundy et al., 2015).

Risk-Benefit Assessment: Teachers

Risk-benefit analysis allows teachers to consider the benefits and dangers of various play

41
situations. Ball and colleagues (2012) defined a difference between “good risks” (i.e.,

experiences that hold developmental value and have few unforeseen negative outcomes) and

“bad risks” (i.e., experiences that are too complex for children to assess independently and have

no developmental benefit like broken, sharp objects). Opportunities for good risk not only

promote development but help to mitigate injury. Attempting to mitigate or eliminate risk not

only has negative consequences for development, but often results in boredom, leading children

to seek out exceedingly risky play (Brussoni, Brunelle, et al., 2015; Brussoni et al., 2012;

Christensen & Mikkelsen, 2008; Wyver et al., 2010). Thus, risk avoidance strategies can result in

more serious injury than supporting children to manage risks through appropriate risky play

opportunities.

Assessing Teachers’ Tolerance of Risky Play: Instrument Development

Teachers are the gatekeepers of children’s opportunities for, and access to, risky play at

school. Healthy teacher tolerance of risky play is important for promoting developmental

opportunities on the playground. A valid and reliable instrument to accurately measure teacher

tolerance of risky play would help teachers understand their risk tolerance and measure change

as needed. A review of available literature revealed few instruments to evaluate adult caregivers’

perceptions of risky play, and none to measure teachers’ tolerance. One instrument, the

Tolerance of Risk in Play Scale (TRiPS) (Hill & Bundy, 2014), measures parents’ tolerance of

risk in play. While the TRiPS produced valid and reliable data for parents, it is not suitable for

use with teachers as many items do not reflect the kind of play that occurs at school (Hill &

Bundy, 2014). Therefore, the purpose of this study was to examine evidence for construct

validity and reliability of the Teacher Tolerance of Risk in Play Scale (T-TRiPS).

42
 Method

This study was part of a larger investigation known as the Sydney Playground Project

(SPP) (Bundy et al., 2015). SPP comprised two interventions: risk reframing for adult

participants and addition of loose materials to the school playground. Approval for this research

was granted by the Human Research Ethics Committee at The University of Sydney and the trial

was registered with the Australian and New Zealand Clinical Trial Registry

(#ACTRN12614000549628) (Bundy et al., 2015).

Participants

A total of 99 preschool and primary school teachers of children, ages 4-12 years and

diagnosed with ASD and/or ID, completed the instrument. Teachers ranged in age from younger

than 20 years to older than 56 years and included lead and assistant teachers. Duration of

teachers’ experience teaching children with disabilities ranged from 4 months to 29 years (M =

10.14; SD = 4.99). See Table 3.1. Of the 99 participants, 48.4% reported growing up in Australia.

43
Table 3.1

Teachers’ years of experience and age of students

Teacher Experience N Student Ages n

< 1- 4 yrs. 30
4 - 6 yrs. 3
7 - 9 yrs. 10
10 - 12 yrs. 4
multiple 13
5 - 9 yrs. 29
4 - 6 yrs. 4
7 - 9 yrs. 9
10 - 12 yrs. 2
multiple 14
10 -19 yrs. 28
4 - 6 yrs. 7
7 - 9 yrs. 6
10 - 12 yrs. 1
multiple 14
≥ 20 yrs. 11
4 - 6 yrs. 1
7 - 9 yrs. 1
10 - 12 yrs. 2
multiple 7
unknown 1
7 - 9 yrs. 1
Note: Teacher experience M=10.14; SD=4.99

Instrument

The T-TRiPS includes 25 items with dichotomous answers (find abbreviated items Table

3.2 and full instrument in the Appendix) and 5 short answer questions probing teachers’ personal

experiences of risky play. Items reflect the six categories of risky play identified by Sandseter

(2009) (great heights, high speed, dangerous tools, dangerous elements, rough-and-tumble,

disappear/get lost). The instrument can be completed in approximately 15 minutes and is

typically is completed while teachers are not on the playground but considering the play of the

44
children in their classroom.

The Teacher Tolerance of Risk in Play Scale (T-TRiPS) was adapted from the original

Tolerance of Risk in Play Scale (TRiPS) (Hill & Bundy, 2014) by modifying or removing items

as they apply to the context of the teacher. The T-TRiPS items reflect activities teachers

commonly supervise their students doing during a school day. The TRiPS produced valid and

reliable data. All items fit the expectations of the Rasch model and unexplained variance in a

principal components analysis of residuals was only the strength of about 2 ½ items,

demonstrating good evidence for unidimensionality (construct validity). Reliability of TRiPS

data was demonstrated by a person reliability index of .87 and a person separation index value of

2.63, indicating the instrument is capable of separating participants into >2 levels of risk

tolerance.

45
 Table 3.2

T-TRiPS items and corresponding Categories of Risky Play (Sandseter, 2009)

T-TRiPS items Categories of Risky Play
1) Play in classroom unsupervised? Disappear/ get lost
2) Play chase with one another? Rough & tumble play
3) Go head-first down a dip/slide? High speeds / rough & tumble
4) Continue playing after a scrape? Rough and tumble
5) Challenging activities if may fail?
6) Climb a tree within your reach? Great heights
7) Run in a place with open fire/ heater? Dangerous elements
8) Play-fight with sticks? Dangerous tools
9) Walk on slippery surface if may fall? Dangerous elements
10) Use 'adult tools' unsupervised? Dangerous tools
11) Engage in rough and tumble play? Rough & tumble
12) Jump down a height of 3-4 meters? Great heights
13) Care for pets in the classroom?
14) Swim close to the pool edge? Dangerous elements
15) Resolve disagreements push/poking? Rough & tumble play
16) Play on equipment may break bone?
17) Play-fight, testing who is strongest? Rough & tumble play
18) Wait to see how manage challenges?
19) Ride a toy/bicycle down a steep hill? High speeds
20) Climb as high as they want in a tree? Great heights
21) Play in playground unsupervised? Disappear / get lost
22) Resolve disagreements shouting?
23) Go new environment alone if watch? Disappear/ get lost
24) Balance on surface 2 meters high? Great heights
25) Encourage to take risks if fun?

Procedure

Participants completed the T-TRIPS during the pre-intervention phase of SPP risk

reframing workshops. Teachers chose between an electronic (iPad with the instrument

administered via Survey Monkey) and pencil-and-paper versions of T-TRiPS. A research

assistant entered data from paper assessments into the Survey Monkey platform.

46
Data Analysis

Rasch analysis was employed using Winsteps version 4.3.1 software (Linacre, 2018) to

transform raw data to interval-level data and examine the evidence for construct validity and

internal reliability. The analysis reveals the degree to which data from items and people fit the

assumptions of the Rasch model. As applied to this study, (1) easier items (reflecting less risk)

are easier for all people to approve; and (2) participants with greater tolerance will be more likely

to endorse more risky items than people with less tolerance (Bond & Fox, 2015).

Construct Validity

Rasch allows for consideration of one human attribute, or latent trait, (i.e., the

“construct,” in this case, tolerance of risky play). The model organizes items and people along a

single continuum based on strength of the latent trait within each item and person (Bond & Fox,

2015). Several findings from Rasch analysis were examined. To determine if data from

individual items corresponded positively with total measure scores, point measure correlations

(PMC) were examined and only items with positive PMCs, indicating the items contribute

positively to the construct, were retained. Goodness of fit statistics, expressed in mean square

(MnSq) and z (ZSTD) values, provide evidence of the degree to which the items conform to the

expectations of the Rasch model. MnSq values between 0.5 and 1.5 and ZSTD values ± 2 were

accepted; the researchers expected that 95% of items would meet those criteria.

To determine the degree to which the range of T-TRiPS item difficulty matched the range

of teacher tolerance, the Wright Map - a visual representation of the item and person hierarchies-

was examined. The researchers expected to find similar item and participant means and ranges

and small gaps between item difficulty. Large gaps in the item hierarchy suggest that the items

do not adequately cover the range of risk tolerance observed in the sample population and

47
cannot, therefore, yield precise measurement of teacher tolerance at the level of a gap.

Correspondingly, items that are very close together along the construct represent a similar level

of risk tolerance and may therefore provide redundant information. See Figure 3.1. The

researchers also examined the hierarchy of the items to determine if they represent a logical

progression, as suggested in related literature.

In order to test the strength of any additional dimensions in the data, Winsteps provides a

principal components analysis of standardized model residuals (PCA). Eigenvalues associated

with each contrast reveal the strength of a second underlying construct. Eigenvalues less than 3

(i.e., the strength of three items) were accepted as evidence that any underlying construct is not

large enough to distort the main construct, indicating unidimensionality.

Internal Reliability

Two sources of evidence for internal reliability were considered: strata, calculated from

the person separation index, and the person reliability index. Strata indicate the levels of teacher

tolerance of risky play revealed in the data; at least two levels are needed. Strata were calculated

using the equation (H= (4G+1)/3), where H = strata and G = person separation index. The person

reliability index, a Rasch equivalent to Cronbach’s alpha, suggests the ease with which measures

would be reproduced with a similar sample. Person reliability index scores ≥0.8 were accepted.

Results

The evidence for construct validity of the data was excellent. Specifically, each item

correlated positively (PMC) with overall measure scores. (See Table 3.3). Further, goodness of

fit statistics for all 25 items were within the acceptable range, indicating that the items conform

to the expectations of the Rasch model.

48
 Table 3.3

Rasch Output for T-TRiPS Items

Infit Outfit

Item description Meas. S.E. PMC MnSq ZSTD MnSq ZSTD

1) Play in classroom unsupervised? 1.09 0.24 0.27 1.26 2.24 1.47 1.76
2) Play chase with one another? -3.76 0.55 0.31 0.81 -0.32 1.49 0.78
3) Go head-first down a dip/slide? -0.77 0.24 0.50 0.94 -0.52 0.85 -0.53
4) Continue playing after a scrape? -3.49 0.50 0.32 0.90 -0.14 0.79 -0.03
5) Challenging activities if may fail? -3.26 0.46 0.15 1.15 0.52 1.60 0.96
6) Climb a tree within your reach? -0.26 0.23 0.46 1.00 0.03 1.04 0.27
7) Run in a place with open fire/ heater? 3.11 0.39 0.34 1.04 0.23 0.83 -0.11
8) Play-fight with sticks? 2.84 0.36 0.25 1.16 0.71 1.37 0.80
9) Walk on slippery surface if may fall? 0.11 0.23 0.54 0.88 -1.37 0.85 -0.73
10) Use 'adult tools' unsupervised? 2.60 0.33 0.45 0.94 -0.22 0.62 -0.79
11) Engage in rough and tumble play? -0.05 0.23 0.55 0.88 -1.31 0.79 -1.10
12) Jump down a height of 3-4 meters? 2.72 0.34 0.36 1.00 0.10 1.21 0.56
13) Care for pets in the classroom? -1.15 0.26 0.30 1.23 1.65 1.22 0.75
14) Swim close to the pool edge? -0.96 0.25 0.45 1.02 0.21 0.85 -0.45
15) Resolve disagreements push/poking? 0.69 0.23 0.43 1.07 0.71 0.97 -0.07
16) Play on equipment may break bone? 0.32 0.23 0.34 1.20 2.11 1.26 1.29
17) Play-fight, testing who is strongest? 1.88 0.28 0.38 1.09 0.61 0.98 0.06
18) Wait to see how manage challenges? -3.24 0.46 0.30 1.01 0.16 0.78 -0.10
19) Ride a toy/bicycle down a steep hill? 0.79 0.24 0.46 1.01 0.17 0.91 -0.38
20) Climb as high as they want in a tree? 2.39 0.31 0.53 0.80 -1.00 0.55 -1.12
21) Play in playground unsupervised? 1.87 0.28 0.52 0.86 -0.98 0.70 -0.86
22) Resolve disagreements shouting? -0.68 0.24 0.43 1.02 0.21 1.19 0.80
23) Go new environment alone if watch? -0.34 0.24 0.55 0.86 -1.42 0.80 -0.94
24) Balance on surface 2 meters high? 1.02 0.24 0.56 0.83 -1.64 0.76 -1.02
25) Encourage to take risks if fun? -3.47 0.50 0.44 0.78 -0.50 0.27 -1.11

As shown in Figure 3.1, the mean of item difficulty and teacher tolerance were similar

(item mean = 0; person mean = 0.19); while the person range (-4.09 – 4.56) was greater than the

item range (-3.76 – 3.11), most teachers’ scores fell within the item range. Items expected to be

more difficult to endorse did, in fact, earn higher measure scores, indicating logic of the

construct. See Table 3.3. However, some gaps existed along the item hierarchy and 8 items

49
achieved the same level of difficulty (i.e., located on the same line), suggesting item redundancy.

The PCA revealed that person and item measures explained 48.5% of variance in TRiPS scores.

After removing the Rasch construct, the largest remaining contrast had an eigenvalue of 2.29,

providing sufficient evidence for unidimensionality of the construct.

The evidence for internal reliability was also strong. T-TRiPS data yielded 3.01 strata,

suggesting that the instrument reliably discriminated more than 3 levels of tolerance of risky

play. Winsteps generated a person reliability index of 0.80.

50
Figure 3.1. T-TRiPS Person-Item Map (Wright Map)

51
 Discussion

The purpose of this study was to examine construct validity and internal reliability of data

collected with T-TRiPS, an instrument to measure teacher tolerance of risky play. Results

suggested the strength of the instrument; all items contribute to a logical unidimensional

construct. Specifically, Rasch analysis revealed that all items form a logical hierarchy and

yielded acceptable goodness of fit statistics.

The T-TRiPS item hierarchy reflected risky play literature. Items located toward the top

of the construct (Figure 3.1) are, understandably, much more difficult for teachers to endorse

(i.e., “run in a place with an open fire”, “use adult tools unsupervised”, and “climb as high as

they wanted in a tree”). Our findings supported Cevher-Kalburan’s (2014) research in which

many teachers identified children’s play with “dangerous tools” and “dangerous elements” as

more difficult to endorse than other risky play activities. Endorsing items located in the middle

of the T-TRiPS construct (i.e., “ride a bicycle down a steep hill”, “walk on a slippery surface”,

“rough and tumble”, “climb a tree within your reach”, and “resolve disagreements independently

if not hitting”) requires moderate risk tolerance. Endorsing items at the bottom of the hierarchy,

represented in Figure 3.1, (i.e., “continue playing challenging activities if they may not

succeed?”, “continue playing after a scrape”, and “play chase with one another”) requires very

little risk tolerance. A teacher whose measure score is at the bottom of the range is quite risk

averse and may benefit from intervention to increase their tolerance.

Findings from a study of teacher responses to uncertainty while supervising children with

disabilities on the playground suggested teachers have greater difficulty tolerating risky play in

groups of children compared to risky play by a single child (Spencer et al., 2016). Interestingly,

according to the item hierarchy represented in Figure 3.1, group play items were fairly evenly

52
distributed along the continuum with some like “play chase with others” fairly easy to endorse,

others involving social problem-solving like “resolving disagreements” both with and without

physical contact and “rough and tumble play” posing moderate difficulty to endorse. Supporting

the findings of Spencer et al. (2016), Cevher-Kalburan (2014) found teachers are uncomfortable

with possible consequences of risks, especially “hurting themselves and peers”. The two

difficult-to-endorse play-fight items support this research as they are located at the top of the

hierarchy and require significant risk tolerance.

Several items are located on the same line along the construct indicating that endorsing

them requires the same level of risk tolerance. However, the items represent different categories

of risky play, therefore, we recommend retaining all items. For example, the items “go headfirst

down a slide” and “resolve disagreements independently if not hitting” are located on the same

difficulty level. While these two items require the same level of tolerance of risk to endorse, they

offer substantially different information.

Findings demonstrated the T-TRiPS is able to separate participants into three levels of

risk tolerance: high, moderate, and low. This strength of the T-TRiPS to determine the level of a

teacher’s tolerance of risky play will allow researchers and interventionists to differentiate

between participants more precisely than simply determining who is more or less tolerant. The

ability to identify risk averse teachers from those with moderate and even high tolerance will

allow interventionists and researchers to prioritize interventions as well as more precisely

measure change in the future. Additionally, the strong person reliability index indicated

researchers can expect future research with new participants (e.g. teachers of children of varying

abilities or from different regions of the world) will yield similar results.

53
Recommendations to Strengthen T-TRiPS

Findings from this research serve as evidence of the strength of the T-TRiPS as a valid

and reliable instrument that will provide researchers and practitioners with the important

opportunity to measure teacher tolerance of risk in play. The following recommendations for

improving this already strong instrument will support accuracy of measurement and provide

opportunities for continued research.

While items are logically distributed along the construct, there is one notable gap along

the hierarchy (Figure 3.1) suggesting addition of new items is warranted. Gaps in item

distribution represent points where precise measurement of a teacher’s tolerance is not possible.

The most concerning gap is located approximately 1 standard deviation below the item mean

(0.0) between item 13, “care for pets in classroom,” and item 18, “wait to see how students

manage challenges before getting involved.” Several participants have risky play tolerance at the

same level as this gap, represented in Figure 3.1 by ‘X’, indicating they were not measured

precisely. Participants at this level of the construct are fairly intolerant of risks and could

therefore be an appropriate target group for intervention to improve risk tolerance. Introduction

of new items or modifying existing items to fill these gaps will improve measurement precision

and support intervention (Bond & Fox, 2015).

Another example of a gap, but one that is not very concerning, is located nearly one

standard deviation above the item mean (S), between item 21, “play on the playground

unsupervised,” and item 1, “play in classroom unsupervised.” Because participants adjacent to

this gap (and the two smaller gaps above) are already fairly risk tolerant, there is less need to

create additional items to fill these gaps.

54
 Many T-TRiPS items reflect the categories of risky play identified by Sandseter (2009),

see Table 3.2. However, the item “use adult tools unsupervised” is the only item representing the

category “dangerous tools” and is located more than one standard deviation above the mean

indicating it is fairly difficult to endorse. Another easier item representing this category would be

beneficial.

Kleppe et al. (2017) defined two new risky play categories, “play with impact” and

“vicarious risk” after this instrument was developed; therefore, these categories are not

represented. Assuming such items fit within the same unidimensional construct, new items

representing these categories as well as items reflecting social, emotional, and intellectual risk,

might provide more thorough information on teacher tolerance of risk in play. Including new,

non-physical items in the T-TRiPS (e.g. social conflict or trying novel activities) could support

the growing body of knowledge of risky play. This research is especially important as children

with disabilities may struggle with these types of play and would benefit from opportunities to

engage in these types of risky play activities (Bundy et al., 2015).

An additional opportunity to improve the T-TRiPS lies in modifying the language of the

instructions to consider the majority of students. Currently the instructions state “Think about

what you would do for the majority of your students when responding to each question” (see

Appendix). While the concept of considering the majority is clear enough, we do not know the

extent to which teacher responses were influenced by the least capable students. Safety rules are

often determined by the capabilities of the least skilled child in the group with blanket

expectations given to the whole. “If this child cannot do a particular activity, then no one should

be allowed to do it.” With more restrictions on play, more skilled children may seek excessively

risky situations to combat boredom, resulting in injury. I propose modifying the instructions of

55
the T-TRiPS (Appendix) from “think about the majority of your students” to something more

explicit like “think about the majority of your students excluding the most and least capable” and

include “majority of students” within each T-TRiPS item. This simple modification will help

guide teachers to consider the whole group when completing the T-TRiPS and consider the

individual skills in risk-benefit assessment to improve risk tolerance.

Limitations

While findings from this research provide strong evidence that the T-TRiPS produces

data that are valid and reliable, like all research, this study had limitations. Firstly, the sample

used in this study is fairly small and only included teachers of children with ASD and/or ID.

Additional research is needed with a broader range of teachers of children both with disabilities

and those who are typically developing. Secondly, the T-TRiPS used in this study did not ask for

teacher or student gender. Not knowing if participants and children were male or female

prevented us from determining the extent to which gender was a factor in teacher tolerance of

risk. Sandseter (2014) indicated male teachers experience greater excitement when engaging in

risks and are therefore more risk tolerant than female teachers. Additionally, adults are often

more tolerant of risky play by male children than female children (Sandseter, 2014).

Implications for Future Research

Future research is needed to create additional items and subsequently ensure the revised

T-TRiPS produces data that represent a unidimensional construct. Another possible opportunity

for future research lies in creating an abbreviated version of the T-TRiPS to use as a screening

tool. Participant scores on a brief screening tool could help researchers, interventionists, and

teachers identify those at greatest risk of hindering children’s right to risky play and therefore

those most appropriate for in-depth assessment and intervention.

56
 Conclusions

The strength of these findings suggest the T-TRiPS can be used not only to help teachers

understand their own risk tolerance, but also as a basis for intervention to help them shift

perspectives and increase risk tolerance. Items located just above a teacher’s tolerance level can

provide opportunities for goals and ideas for intervention to improve risk tolerance. Additionally,

these findings suggest T-TRiPS could be used in future research to provide insights into short-

and long-term impacts of varying degrees of risk tolerance on child development.

57
 CHAPTER 4 REFLECTIONS AND CONCLUSIONS

Completing this thesis not only reinforced many concepts from my coursework but also

provided me with knowledge and skills that will strengthen my future career as an occupational

therapist. Through this research I have learned that several factors influence teacher tolerance of

risk in play as well as the importance of advocating for opportunities to engage in risk for all

children through compassionate education of teachers. This research is an important step towards

increasing children’s access to their right to experience risky play.

Several notable findings support Teacher Tolerance of Risk in Play Scale (T-TRiPS) as

an instrument that produces psychometrically sound data. All items measure a unidimensional

construct of tolerance of risk in play. A variety of play activity items represent a hierarchy of

difficulty ranging from ‘easy to endorse’ to ‘difficult to endorse.’ The placing of these items

along the hierarchy is consistent with previous risky play research and is representative of

activities teachers often supervise on the playground. T-TRiPS reliably discriminated more than

three levels of tolerance of risky play (low, moderate, and high) which may make it sensitive to

change as a result of interventions aimed at increasing teacher tolerance of risk.

The ability to measure teacher tolerance of risk in play has important implications for

child development and opens several opportunities for children of all abilities. Children with

disabilities often are considered more vulnerable than their typically developing peers, increasing

the already over-protective practices of teachers designed to prevent injury on the school

playground (Spencer et al., 2016). Being able to assess a play opportunity and decide how to

engage while best managing risk is a protective factor; children with that ability have learned the

boundaries of their own skills (Brussoni et al., 2012; Christensen & Mikkelsen, 2008).

58
 Adults are responsible for providing opportunities for development through play,

however adults’ misunderstanding of risky play, perceptions of the limited abilities of children,

and fear of consequences to themselves as well as children interfere with their providing risky

play opportunities (Ball et al., 2019; Brussoni, Brunelle, et al., 2015; Lester & Russell, 2014;

Spencer et al., 2016). T-TRiPS can help adults reflect on their own tolerance of risk and may

ultimately help them shift these perspectives and increase opportunities for risky play

experiences.

Future Research

While findings from this research demonstrated the strength of this instrument, there are a

few opportunities for modification to improve the instrument. One notable gap is present in the

item hierarchy at a level that would require fairly low tolerance of risk in play to endorse items.

The lack of items in this region of the hierarchy makes it impossible to precisely measure

teachers whose tolerance falls in this general range. Therefore, I propose new items be added to

fill this gap. Through my review of risky play literature and reviewing the results, I also propose

adding new items that represent risky play categories ‘play with impact’ and ‘vicarious risk’ that

Kleppe and colleagues identified after the T-TRiPS was developed. Additionally, I would modify

some existing items to be either easier or more difficult to endorse by adjusting characteristics

like level of adult supervision (e.g. “use adult tools while you are close by” or “use adult tools

while you provide hand-over-hand support”) and involvement of peers (e.g. “go head-first down

a slide without peers close by” or “go head-first down a slide with peers playing at the bottom”).

I also would modify the instructions of the instrument to help participants consider the whole

group of children rather than the least or most capable child in their care. This effort may help

shift away from the tendency to make safety rules for the group based on the capabilities of one

59
child. I would include an opportunity for participants to self-identify their gender as well as

report the gender of the children in their care in order to determine the extent to which gender

may play a role in tolerance of risky play. Finally, while our small sample size is acceptable for

Rasch analysis, a larger sample size of teachers from a variety of backgrounds teaching a variety

of children will help continue this research and support the goal of ultimately using the T-TRiPS

in practice. Overall, our findings suggest this instrument produces psychometrically sound data

and will serve as an important tool to help teachers understand their tolerance of risk in play and

serve as a mechanism for change to improve opportunities for all children to engage in risky play

experiences.

Personal Reflections

After teaching young children and experiencing my own fears of social consequences and

litigation, this research helped me understand the profound impact gaining knowledge can have

to shift perspectives. My past experiences as a teacher in a Reggio Emilia-inspired early

childhood education center helped me view children as competent and capable, but I found I did

not always practice this belief. My hope is that this instrument will help others acknowledge their

own barriers to risky play tolerance and work to shift their perspectives and provide the children

in their care with these important risky play experiences.

Through this thesis process and learning about the many factors that influence teacher

tolerance of risk in play, I am inspired to share my findings from the T-TRiPS and support

teachers in acknowledging and changing their perspectives of risky play. Through my role as an

occupational therapist, I plan to partner with adults who supervise play to educate them on the

importance of risk in play to develop children’s skills. In the future I would like to present at

conferences and schools to teachers and other professionals working with children on the

60
benefits of risky play. In these situations, I hope to use the T-TRiPS to measure their tolerance of

risk in play and provide opportunities to shift their perspectives of risky play through workshops

and discussions of risk. I will use the knowledge I’ve gained through this research to continue re-

centering my perspectives of children as competent and capable and use my role as an OT to

advocate for children’s right to engage in risky play.

61
 REFERENCES

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental

disorders: DSM-5 (Fifth ed.). Arlington, VA: American Psychiatric Association.

Ball, D. (2002). Playgrounds- risks, benefits, and choices Ball, D..pdf>. Retrieved from

London, Health and Safety Executive, Middlesex University:

Ball, D., Brussoni, M., Gill, T., Harbottle, H., & Spiegal, B. (2019). Avoiding a dystopian future

for children's play. International Journal of Play, 8(1), 3-10dys.

doi:10.1080/21594937.2019.1582844

Ball, D. J., Gill, T., & Spiegal, B. (2012). Managing risk in play provision: Implementation

guide. Retrieved from http://www.playengland.org.uk/resources/managingrisk-in-play-

provision-implementation-guide.aspx

Beetham, K. S., Sterman, J., Bundy , A. C., Wyver , S., Ragen, J., Engelen, L., . . . Naughton, G.

(2019). Lower parent tolerance of risk in play for children with disability than typically

developing children. International Journal of Play, 1-12.

doi:10.1080/21594937.2019.1643980

Bond, T. G., & Fox, C. M. (2015). Applying the Rasch Model: Fundamental measurement in the

human sciences (third ed.). New York: Routledge.

Brussoni, M., Brunelle, S., Pike, I., Sandseter, E. B., Herrington, S., Turner, H., . . . Ball, D. J.

(2015). Can child injury prevention include healthy risk promotion? Inj Prev, 21(5), 344-

347. doi:10.1136/injuryprev-2014-041241

Brussoni, M., Gibbons, R., Gray, C., Ishikawa, T., Sandseter, E. B., Bienenstock, A., . . .

Tremblay, M. S. (2015). What is the relationship between risky outdoor play and health

62
 in children? A Systematic Review. International Journal of Environmental Research and

Public Health, 12(6), 6423-6454. doi:10.3390/ijerph120606423

Brussoni, M., Olsen, L. L., Pike, I., & Sleet, D. A. (2012). Risky play and children's safety:

balancing priorities for optimal child development. International Journal of

Environmental Research and Public Health, 9(9), 3134-3148. doi:10.3390/ijerph9093134

Buchanan, C. (1999). Building Better Playgrounds: A project for parents? UAB Magazine, 19(3).

Retrieved from http://main.uab.edu/show.asp?durki=25353

Bundy , A., Wyver , S., Beetham, K., Ragen, J., Naughton, G., Tranter, P., . . . Sterman, J.

(2015). The Sydney playground project- levelling the playing field: a cluster trial of a

primary school-based intervention aiming to promote manageable risk-taking in children

with disability. BMC Public Health, 15(1), 1125. doi:10.1186/s12889-015-2452-4

Bundy , A. C., Naughton, G., Tranter, P., Wyver , S., Baur, L., Schiller, W., . . . Brentnall, J.

(2011). The sydney playground project: popping the bubblewrap - unleashing the power

of play: a cluster randomized controlled trial of a primary school playground-based

intervention aiming to increase children’s physical activity and social skills. BMC Public

Health, 11, 680.

Cevher-Kalburan, N. (2014). Developing pre-service teachers’ understanding of children’s risky

play. Journal of Adventure Education and Outdoor Learning, 15(3), 239-260.

doi:10.1080/14729679.2014.950976

Christensen, P., & Mikkelsen, M. R. (2008). Jumping off and being careful: children's strategies

of risk management in everyday life. Sociol Health Illn, 30(1), 112-130.

doi:10.1111/j.1467-9566.2007.01046.x

63
Cooke, M., Wong, S., & Press, F. (2019). Towards a re-conceptualisation of risk in early

childhood education. Contemporary Issues in Early Childhood.

doi:10.1177/1463949119840740

Engelen, L., Wyver , S., Perry, G., Bundy , A., Chan, T. K. Y., Ragen, J., . . . Naughton, G.

(2017). Spying on children during a school playground intervention using a novel method

for direct observation of activities during outdoor play. Journal of Adventure Education

and Outdoor Learning, 18(1), 86-95. doi:10.1080/14729679.2017.1347048

Finegan, J., Niccols, G. A., Zacher, J. E., & Hood, J. E. (1991). The play activity questionnaire:

A parent report measure of children's play preferences. Archives of Sexual Behavior,

20(4), 393-408. doi:10.1007/BF01542619

Finegan, J. K., Niccols, G. A., Zacher, J. E., & Hood, J. E. (1991). The Play Activity

Questionnaire: A Parent Report Measure of Children's Play Preferences. Archives of

Sexual Behavior, 20(4), 393-408.

Garwood, S. G. (1982). Piaget and play: Translating theory into practice. Topics in Early

Childhood Special Education.

Gill, T. (2018). Playing it Safe? A global white paper on risk, liability and children's play in

public space.

Gilman, S. (2007). Including the Child With Special Needs: Learning From Reggio Emilia.

Theory Into Practice, 46(1), 23-31. doi:10.1207/s15430421tip4601_4

Grady-Dominguez, P., Bundy , A., Ragen, J., Wyver , S., Villeneuve, M., Naughton, G., . . .

Beetham, K. (2019). An observation-based instrument to measure what children with

disabilities do on the playground: a Rasch analysis. International Journal of Play, 8(1),

79-93. doi:10.1080/21594937.2019.1580340

64
Hill, A., & Bundy , A. C. (2014). Reliability and validity of a new instrument to measure

tolerance of everyday risk for children. Child: Care, Health, and Development, 40(1), 68-

76. doi:10.1111/j.1365-2214.2012.01414.x

Hyndman, B. (2015). Where to next for school playground interventions to encourage active

play? An exploration of structured and unstructured school playground strategies.

Journal of Occupational Therapy, Schools, & Early Intervention, 8(1), 56-67.

doi:10.1080/19411243.2015.1014956

Ingram, D. H., Mayes, S. D., Troxell, L. B., & Calhoun, S. L. (2007). Assessing children with

autism, mental retardation, and typical development using the Playground Observation

Checklist. Autism, 11(4), 311-319. doi:10.1177/1362361307078129

International School Grounds Alliance. (2017). Declaration on Risk in Play and Learning.

Kleppe, R., Melhuish, E., & Sandseter, E. B. H. (2017). Identifying and characterizing risky play

in the age one-to-three years. European Early Childhood Education Research Journal,

25(3), 370-385. doi:10.1080/1350293x.2017.1308163

Lester, S., & Russell, W. (2014). Turning the world upside down: Playing as the deliberate

creation of uncertainty. Children (Basel), 1(2), 241-260. doi:10.3390/children1020241

Linacre, J. M. (2018). Winsteps Rasch Measurement Computer Program: User's Guide (Vol.

4.3.1). Beaverton, OR: Winsteps.com.

Lindqvist, P., Nordänger, U. K., & Landahl, J. (2009). Insurance and Assurance: Teachers'

Strategies in the Regimes of Risk and Audit. European Educational Research Journal,

8(4), 508-519. doi:10.2304/eerj.2009.8.4.508

Little, H. (2010). Finding the balance: Early Childhood practitioners’ views on risk, challenge

and safety in outdoor play settings.

65
Little, H. (2016). Promoting risk-taking and physically challenging play in Australian early

childhood settings in a changing regulatory environment. Journal of Early Childhood

Research, 15(1), 83-98. doi:10.1177/1476718x15579743

Little, H., & Eager, D. (2010). Risk, challenge and safety: implications for play quality and

playground design. European Early Childhood Education Research Journal, 18(4), 497-

513. doi:10.1080/1350293x.2010.525949

Little, H., Sandseter, E., Hansen, B., & Wyver , S. (2012). Early Childhood Teachers' Beliefs

about Children's Risky Play in Australia and Norway. Contemporary Issues in Early

Childhood, 13(4), 300-316. doi:10.2304/ciec.2012.13.4.300

Lounsbery, M. A. F., McKenzie, T. L., Morrow, J. R., Holt, K. A., & Budnar, R. G. (2013).

School Physical Activity Policy Assessment. Journal of Physical Activity and Health, 10,

496-503.

Malaguzzi, L. (1994). Your image of the child: Where teaching begins. Child Care Information

Exchange.

Massey, W. V., Stellino, M. B., Hayden, L., & Thalken, J. (2019). Examination of the

Responsiveness of the Great Recess Framework-Observational Tool. International

Journal of Environmental Research and Public Health, 17(1).

doi:10.3390/ijerph17010225

Massey, W. V., Stellino, M. B., Mullen, S. P., Claassen, J., & Wilkison, M. (2018). Development

of the great recess framework - observational tool to measure contextual and behavioral

components of elementary school recess. BMC Public Health, 18(1), 394.

doi:10.1186/s12889-018-5295-y

66
McFarland, L., & Laird, S. G. (2017). Parents’ and early childhood educators’ attitudes and

practices in relation to children’s outdoor risky play. Early Childhood Education Journal,

46(2), 159-168. doi:10.1007/s10643-017-0856-8

McKenzie, T. L. (2002). System for Observing Play and Leisure Activity in Youth (SOPLAY).

San Diego, CA: San Diego State University

Morrongiello, B. A., & Corbett, M. (2006). The Parent Supervision Attributes Profile

Questionnaire: a measure of supervision relevant to children's risk of unintentional injury.

Inj Prev, 12(1), 19-23. doi:10.1136/ip.2005.008862

Morrongiello, B. A., Corbett, M. R., & Kane, A. (2011). A measure that relates to elementary

school children's risk of injury: the supervision attributes and risk-taking questionnaire

(SARTQ). Inj Prev, 17(3), 189-194. doi:10.1136/ip.2010.028548

Morrongiello, B. A., Corbett, M. R., & Kane, A. . (2011). A measure that relates to elementary

school children’s risk of injury: The Supervision attributes and risk-taking questionnaire

(SARTQ). Injury Prevention 17, 189–194.

Morrongiello, B. A., & Lasenby, J. (2006). Finding the daredevils: development of a Sensation

Seeking Scale for children that is relevant to physical risk taking. Accid Anal Prev, 38(6),

1101-1106. doi:10.1016/j.aap.2006.04.018

Newnham, H. (2000). When is a teacher or school liable in negligence? Australian Journal of

Teacher Education, 25(1). doi:10.14221/ajte.2000v25n1.5

Niehues, A. N., Bundy , A., Broom, A., Tranter, P., Ragen, J., & Engelen, L. (2013). Everyday

uncertainties: reframing perceptions of risk in outdoor free play. Journal of Adventure

Education & Outdoor Learning, 13(3), 223-237. doi:10.1080/14729679.2013.798588

67
Olsen, L. L., Ishikawa, T., Masse, L. C., Chan, G., & Brussoni, M. (2018). Risk Engagement and

Protection Survey (REPS): developing and validating a survey tool on fathers' attitudes

towards child injury protection and risk engagement. Inj Prev, 24(2), 106-112.

doi:10.1136/injuryprev-2017-042413

Parham, L. D., & Fazio, L. S. (2008). Play in Occupational Therapy for Children (2nd ed.). St.

Louis, Missouri: Mosby Inc.

Pellegrini, A. D. (1988). Elementary-school children’s rough-and-tumble play and social

competence. Developmental Psychology, 24(6), 802–806. https://doi-

org.ezproxy2.library.colostate.edu/10.1037/0012-1649.24.6.802

Piaget, J. (1972). Some Aspects of Operations. In M. Piers (Ed.), Play and Development (1st

ed.). New York: W.W. Norton & Company Inc.

Prezza, M., Alparone, F. R., Cristallo, C., & Luigi, S. (2005). Parental perception of social risk

and of positive potentiality of outdoor autonomy for children: The development of two

instruments. Journal of Environmental Psychology, 25(4), 437-453.

doi:10.1016/j.jenvp.2005.12.002

Ramstetter, C. L., Murray, R., & Garner, A. S. (2010). The crucial role of recess in schools

Journal of School Health 80(11), 517-526.

Ridgers, N. D., Stratton, G., & McKenzie, T. L. (2010). Reliability and Validity of the System

for Observing Children’s Activity and Relationships During Play (SOCARP). Journal of

Physical Activity and Health, 7, 17-25.

Sandseter, E. B. (2007). Categorizing risky play - How can we identify risk-taking in children’s

play? European Early Childhood Education Research Journal, 15(2), 237-252.

doi:10.1080/13502930701321733#.VTDjW2ccSUk

68
Sandseter, E. B. (2009). Characteristics of risky play. Journal of Adventure Education &

Outdoor Learning, 9(1), 3-21. doi:10.1080/14729670802702762

Sandseter, E. B. (2011). Children's risky play in early childhood education and care.

Sandseter, E. B., & Kennair, L. E. O. (2011). Children's risky play from an evolutionary

perspective: The anti-phobic effects of thrilling experiences. Evolutionary Psychology,

9(2), 257-284.

Sandseter, E. B. H. (2012). Restrictive safety or unsafe freedom? Norwegian ECEC practitioners'

perceptions and practices concerning children's risky play. Childcare in Practice, 18(1),

83-101. doi:10.1080/13575279.2011.621889#.VS9wTmccSUk

Sandseter, E. B. H. (2014). Early childhood education and care practitioners' perceptions of

children's risky play; examining the influence of personality and gender. Early Child

Development and Care, 184(3), 434-449. doi:10.1080/03004430.2013.794797

Scott, S., Jackson, S., & Backett-Milburn, K. (1998). Swings and roundabouts: Risk anxiety and

the everyday worlds of children. Sociology, 32(4), 689-705.

Speltz, M. L., Gonzales, N., Sulzbacher, S., & Quan, L. (1990). Assessment of Injury Risk in

Young Children: A Preliminary Study of the Injury Behavior Checklist. Journal of

Pediatric Psychology, 15(3), 373-383.

Spencer, G., Bundy , A., Wyver , S., Villeneuve, M., Tranter, P., Beetham, K., . . . Naughton, G.

(2016). Uncertainty in the school playground: shifting rationalities and teachers’ sense-

making in the management of risks for children with disabilities. Health, Risk & Society,

18(5-6), 301-317. doi:10.1080/13698575.2016.1238447

69
Sterman, J. J., Naughton, G. A., Bundy , A. C., Froude, E., & Villeneuve, M. A. (2019). Planning

for outdoor play: Government and family decision-making. Scand J Occup Ther, 26(7),

484-495. doi:10.1080/11038128.2018.1447010

U.N.C.R.C. (1989). Convention on Rights of the Child.

van Rooijen, M., & Newstead, S. (2016). Influencing factors on professional attitudes towards

risk-taking in children’s play: a narrative review. Early Child Development and Care,

187(5-6), 946-957. doi:10.1080/03004430.2016.1204607

Will, K. E., Lorek, E. J., Sabo, C. S., & Kidd, D. G. (2009). Measuring Injury Risk Perceptions:

Feasibility of a Risk Estimation Scale. American Journal of Health Behavior, 33(6), 639-

649.

Wolfensberger, W. P., Nirje, B., Olshansky, S., Perske, R., & Roos, P. (1972). The Dignity of

Risk. In The Principle of Normalization In Human Services. Toronto, Canada: National

Institute on Mental Retardation.

Wyver , S., Bundy , A., Naughton, G., Tranter, P., Sandseter, E. B., & Ragan, J. (2010). Safe

outdoor play for young children: Paradoxes and consequences. In S. Howard (Ed.),

Proceedings of the AARE International Education.

Wyver , S., Tranter, P., Naughton, G., Little, H., Sandseter, E., Hansen, B., & Bundy , A. (2010).

Ten Ways to Restrict Children's Freedom to Play: The Problem of Surplus Safety.

Contemporary Issues in Early Childhood, 11(3), 263-277.

doi:10.2304/ciec.2010.11.3.263

Yogman, M., Garner, A., Hutchinson, J., Hirsh-Pasek, K., Golinkoff, R. M., AAP COMMITTEE

ON PSYCHOSOCIAL ASPECTS OF CHILD AND FAMILY HEALTH, & MEDIA, A.

70
C. O. C. A. (2018). The power of play: A pediatric role in enhancing development in

young children. Pediatrics, 142(3), 1-17.

71
 APPENDIX

Teacher Tolerance of Risk in Play Scale (T-TRiPS)

1. Purpose of the survey

This survey is part of the Sydney Playground Project, a scientific research project conducted with Ethics approval
from the University of Sydney. All information gathered from this survey is anonymous and will be used only for
research purposes.

2. Information about the child

Think about what you would do for the majority of your students when responding to each question.

3. Information about the child (cont.)

How long have you been teaching/in this role for? (including other schools)
฀
฀

*What is the age group of the children you currently teach/support?

4-6

7-9

10-12

It varies

*How often do you encourage your students to take everyday risks?

Never Seldom Sometimes Often
Please tick 



 



 



 ฀





4. Questionnaire

Think about all students when responding to the following questions. If the answer is ever “yes”, select “yes".

*Do you trust your students to play in the classroom without constant supervision?
 Yes




 No




Comment (optional)

฀
฀

72
The Tolerance of Risk in Play Scale (staff)
*Do you let your students play chase with one another?
 Yes




 No




Comments (optional)

฀
฀

*Would you let your students go head first down a slippery dip/slide?
 Yes




 No




Comment (optional)

฀
฀

*Do you allow your students to continue playing after s/he gets a scrape?
 Yes




 No




Comment (optional)

฀
฀

*Do you let your students continue to play very challenging activities even when you
know they may not succeed?
 Yes




 No




Comment (optional)

฀
฀

5. Questionnaire (cont.)

*Would you let your students climb a tree or other surface within your reach?
 Yes




 No




Comment (optional)

฀
฀

73
The Tolerance of Risk in Play Scale (staff)
*Would you let your students run in a place where there was an open fire or portable
heater?
 Yes




 No




Comment (optional)

฀
฀

*Do you allow your students to play-fight with sticks?

 Yes




 No




Comment (optional)

฀
฀

*Would you let your students walk on a slippery surface if there was a chance they
may fall?
 Yes




 No




Comment (optional)

฀
฀

*Do you let your students use 'adult tools' (e.g. hammer and nail, knife, scissors)
unsupervised?
 Yes




 No




Comment (optional)

฀
฀

6. Questionnaire (cont.)

*Do you allow your students to engage in rough and tumble play?
 Yes




 No




Comment (optional)

฀
฀

74
The Tolerance of Risk in Play Scale (staff)
*Would you let the majority of your students jump down from a height of 3-4 metres?
 Yes




 No




Comment (optional)

฀
฀

*Would you allow your students to care for pets in the classroom?
 Yes




 No




Comment (optional)

฀
฀

*Would you let your students swim close to the pool edge while you were watching
from the side?
 Yes




 No




Comment (optional)

฀
฀

7. Questionnaire (cont.)

*Do you let your students resolve disagreements (without stepping in) if they are
pushing or poking one another?
 Yes




 No




Comment (optional)

฀
฀

*Would you allow your students to play on equipment if you thought there was the
potential s/he may break a bone?
 Yes




 No




Comment (optional)

฀
฀

75
The Tolerance of Risk in Play Scale (staff)
*Do you allow your students to play-fight, testing who is strongest?
 Yes




 No




Comment (optional)

฀
฀

8. Questionnaire (cont.)

*Do you wait to see how well your students manage challenges before getting
involved?
 Yes




 No




Comment (optional)

฀
฀

*Would you let your students climb as high as they wanted in a tree or on another
surface?
 Yes




 No




Comment (optional)

฀
฀

*Would you allow your student to ride a toy/bicycle down a steep hill?
 Yes




 No




Comment (optional)

฀
฀

9. Questionnaire (cont.)

76
The Tolerance of Risk in Play Scale (staff)
*Would you let your students climb as high as they wanted in a tree?
 Yes




 No




Comment (optional)

฀
฀

*Do you let your students play in the playground unsupervised?

 Yes




 No




Comment (optional)

฀
฀

*Do you let your students resolve disagreements (without stepping in) if the children
are shouting but not hitting one another?
 Yes




 No




Comment (optional)

฀
฀

*Would you let your student(s) go off on their own in a new environment if you were
able to watch them from afar?
 Yes




 No




Comment (optional)

฀
฀

*Would you let your students balance on a fallen tree or other narrow surface 2 metres
above the ground?
 Yes




 No




Comment (optional)

฀
฀

77
The Tolerance of Risk in Play Scale (staff)
*Do you encourage your students to take some risks if it means having fun during
play?
 Yes




 No




Comment (optional)

฀
฀

10. Your experience of risk (Cont.)

Share an example of a time when you allowed your student to do something that made
you feel uncomfortable? Please describe.
฀

What were the benefits associated with allowing this child to do that?
฀

Can you think of something you used to do regularly as a child that may be considered
'risky'? Please describe.
฀

What benefits did you get from doing that?

฀

Would you allow your students to do that?

 Yes




 No




Comment (optional)

฀
฀

11. Your experience of risk (cont.)

Describe what you fear most for your students and why?
฀

78
The Tolerance of Risk in Play Scale (staff)
12. Information about you

*What is your age?

 20 years or below




 21-25 years




 26-30 years




 31-35 years




 36-40 years




 41-45 years




 46-50 years




 50-55 years




 56 years and above





In what country did you grow up?

฀

13. Survey complete

Thankyou for completing this survey.

For more information about The Sydney Playground Project, please visit our website: http://sydney.edu.au/health-
sciences/sydney-playground-project/

79
 LIST OF ABBREVIATIONS

ASD Autism Spectrum Disorder

GRF-OT Great Recess Framework Observation Tool

IBC Injury Behaviors Checklist

ID Intellectual Disability

MnSq Mean square

PASPQ Parent Supervision Attributes Profile Questionnaire

POC Playground Observation Checklist

PPOAC Parental Perception of Positive Potentiality of Outdoor

Autonomy for Children

PQ Play activity Questionnaire

RBA Risk-Benefit Assessment

REPS Risk Engagement and Protection Survey

SARTQ Supervision Attributes and Risk-taking Questionnaire

SOCARP System for Observing Children’s Activities and Relationships

during Play

SOOP System for Observing Outdoor Play

SOPLAY System for Observing Play and Leisure Activity in Youth

S-PAPA School Physical Activity Policy Assessment

SPP Sydney Playground Project

SSSC Sensation Seeking Scale

TRiPS Tolerance of Risk in Play Scale

T-TRiPS Tolerance of Risk in Play Scale for Teachers

80
WARE Worry Assessment and Risk Estimation Scale

ZSTD Z standardized scores

81