Physical self-concept and gender : the role of frame of reference and social comparison among adolescent females

Louisiana State University Louisiana State University

LSU Scholarly Repository LSU Scholarly Repository

LSU Doctoral Dissertations Graduate School

2013

Physical self-concept and gender : the role of frame of reference Physical self-concept and gender : the role of frame of reference

and social comparison among adolescent females and social comparison among adolescent females

Emily Kristin Beasley

Louisiana State University and Agricultural and Mechanical College

Follow this and additional works at: https://repository.lsu.edu/gradschool_dissertations

Part of the Kinesiology Commons

Recommended Citation Recommended Citation

Beasley, Emily Kristin, "Physical self-concept and gender : the role of frame of reference and social

comparison among adolescent females" (2013).

LSU Doctoral Dissertations

. 1389.

https://repository.lsu.edu/gradschool_dissertations/1389

This Dissertation is brought to you for free and open access by the Graduate School at LSU Scholarly Repository. It

has been accepted for inclusion in LSU Doctoral Dissertations by an authorized graduate school editor of LSU

Scholarly Repository. For more information, please contact[email protected].

PHYSICAL SELF-CONCEPT AND GENDER:

THE ROLE OF FRAME OF REFERENCE

AND

SOCIAL COMPARISON

AMONG ADOLESCENT FEMALES

A Dissertation

Submitted to the Graduate Faculty of the

Louisiana State University

Agricultural and Mechanical College

in partial fulfillment of the

requirements for the degree of

Doctor of Philosophy

The School of Kinesiology

Emily K. Beasley

B.S. Mississippi State University, 2006

M.S. Mississippi University for Women, 2008

August 2013

ACKNOWLEDGEMENTS

Throughout the years, there have been several people who have had a significant impact

on my education and career. Without them, this project would not have been possible. I would

first like to thank all of the students, teachers, and volunteers who were willing to lend a helping

hand throughout the data collection process. A special thanks to the members of my dissertation

committee: Dr. Alex Garn, who has spent countless hours revising and editing my work, Dr.

Melinda Solmon, a wonderful mentor and boss whom I admire, Dr. Birgitta Baker, who was

always willing to sneak away for coffee breaks and writing inspiration, and Dr. Laura Choate

and Dr. Catherine Champagne, who provided valuable insight and feedback. I am very fortunate

to have had the opportunity to work with such an outstanding group of professors and their

advice and guidance was invaluable.

There are two friends who I appreciate deeply, Dr. Laura Helen Marks and Dr. Jasmine

Hamilton. The two of you were there when I was in need of inspiration, motivation, or a good

laugh. I am extremely lucky to have a best friend and colleague in Dr. Ashley Samson, who has

been my partner in crime for the past four years, even when she was on the opposite side of the

country. You continue to remind me that I am powerful beyond measure. Also, to the extensive

list of friends who would never let me forget the importance of simply having a good time: Dr.

Steven McCullar, Molly Fisher, K-Lynn McKey, Bryce Dudley, Zach Montgomery, and Dr.

Lisa Johnson. Whether it was for a nice tailgate, a good run, afternoon refreshments, or an

inspirational quote, I knew one of you would be ready to accept the challenge.

I would like to extend a special thanks to my dear companions at Right Lead Equestrian

Center, most of all Frank, who was always there when I needed an escape. To Alex, who has

iii

been a constant companion throughout my entire education and a furry shoulder to cry on when I

needed it. He‟s hung around just to see me through.

I would also like to acknowledge the significant contribution my family has made to my

education. To Mom, you have always pushed me to excel in whatever it was I chose to do and

encouraged me to reach my full potential. To my Nanny, who has always been a positive light

and made me believe that I was one of a kind and the best in everything (regardless if that were

actually the case). To my brother Jace, who is a daily reminder of everything that is connected

and beautiful.

To my Daddy, who has made more sacrifices for me than I can even begin to count.

Without you, I would not be the woman I am today. You have taught me the value of hard work,

determination, and ingenuity. Not to mention how to ride a horse.

To Amon Craig, you have more patience than anyone I have ever known. Your unlimited

encouragement, love, and support have kept me going when I was ready to give up. The last

three years have been an incredible journey and I look forward to our future together.

Finally, I would like to acknowledge Mrs. Jackie Harrington. I can‟t think of another

person who has had a more profound impact on my perspective on life. Because of you I know

that pain is a good thing- it reminds us that we are still alive. You have shown me what it means

to give to others (and mean it), and I will never forget the role you played in the beginning of my

college career. Those late nights and weekends spent practicing in the gym are where it all

began, although I didn‟t see it at the time. You are the reason that I am a teacher and I can only

hope that one day I will be able to follow your example and pay it forward.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ........................................................................................................... ii

ABSTRACT .....................................................................................................................................v

CHAPTER ONE: INTRODUCTION ..............................................................................................1

CHAPTER TWO: PHYSICAL SELF-CONCEPT AND FRAME OF REFERENCE FOR

GIRLS IN SAME-SEX PHYSICAL EDUCATION .......................................................................7

CHAPTER THREE: INVESTIGATING A BIG-FISH-LITTLE-POND EFFECT AND

MODERATING EFFECTS OF CLASS TYPE FOR GIRLS IN PHYSICAL

EDUCATION ....................................................................................................................37

CHAPTER FOUR: SUMMARY AND CONCLUSIONS ............................................................58

REFERENCES .............................................................................................................................65

APPENDIX

A: EXTENDED REVIEW OF LITERATURE ......................................................................75

B: STUDENT SURVEY ......................................................................................................120

C: FIELD OBSERVATION DATA COLLECTION FORM ..............................................122

D: INTERVIEW PROTOCOL .............................................................................................123

E: CHAPTER TWO EXTENDED RESULTS .....................................................................125

F: COPY OF IRB APPROVAL FORM ...............................................................................144

VITA ...........................................................................................................................................145

ABSTRACT

Quality physical education can positively influence students‟ emotional development,

specifically their perceptions of competence, self-esteem, and self-concept. Unfortunately, girls

often become less engaged and involved in physical education as they grow older and

consistently report lower physical self-concepts than males. Physical self-concept is associated

with multiple positive outcomes, yet there is only speculation addressing why females report

lower physical self-concepts than males in physical education. The overall purpose of this

dissertation was to investigate potential explanations for gender discrepancies in physical self-

concept among physical education students. One qualitative and one quantitative research study

was conducted to address this issue.

The purposes of the qualitative study were to capture female students‟ personal

interpretations of physical self-concept, frame of reference, and the physical education

environment. Results indicate that the girls enrolled in same-sex physical education classes

perceived the coed environment negatively and experienced pressure regarding physical ability

and appearance. Competition in physical education received significant attention and was

desirable only in appropriate levels and contexts. Results supported that the students used

multiple sources of information in physical self-concept development. Participants conformed to

traditional gender norms and viewed males as more aggressive and physically dominant than

females. In addition, all participants perceived same-sex physical education as a more desirable

setting.

The purposes of the quantitative study were to investigate a BFLPE and moderating

effects of class type among female students in physical education. It was hypothesized that: a)

individual ability would positively predict sport self-concept and class-level ability; b) class-

level ability would negatively predict sport self-concept (also known as a BFLPE); and c) class

structure (e.g. same sex, coeducational) would moderate the BFLPE in physical education.

Results provided evidence that participants in both class types experienced a BFLPE. Class type

did not increase or decrease the BFLPE for these students, indicating that the girls in these coed

classes were not at an increased risk for experiencing negative consequences to their sport self-

concept as a result of a BFLPE.

Overall, results provide additional information of physical self-concept development

among adolescent females in physical education. Implications for practitioners and suggestions

for future research are included.

CHAPTER ONE: INTRODUCTION

Dramatic declines in physical activity levels and participation in physical education have

contributed to an increase in overweight and obesity for children and adolescents (Cumming et

al., 2011; United States Department of Health and Human Services [USDHHS], 2011). Regular

physical activity engagement in exercise settings such as physical education can benefit students

not only physically, but cognitively, emotionally, and socially as well (Bailey, 2009; USDHHS,

2011). Despite the benefits of physical activity, there is still a pressing need to motivate students

to participate in physical education classes. Unfortunately, students‟ enjoyment and positive

attitudes toward physical education decline during the transition from childhood to adolescence

(Subramaniam & Silverman, 2007). Physical self-concept, an individual‟s perception of himself

or herself within the physical domain (Marsh, 1990), is an intrinsic factor that can help explain

student engagement in physical activity both in and out of physical education (Beasley & Garn,

in press; Carlson, 1995; Welk & Joens-Matre, 2007). Positive physical self-concept is related to

physical activity participation and both adoption and adherence (Crocker et al., 2003; Cumming

et al., 2011; Dunton, Jamner, & Cooper, 2003; Marsh, Papaioannou, & Theodorakis, 2006) and a

multitude of other important outcomes such as increased happiness, intrinsic motivation, and

physical performance and decreased depression and anxiety (Craven & Marsh, 2008; Crocker,

Sabiston, Kowalski, McDonough, & Kowlaski, 2006; Schneider, Dunton, & Cooper, 2008).

Physical self-concept research has consistently identified gender differences across

multiple physical activity settings, including physical education, with females reporting lower

physical self-concept than males (Caglar, 2009; Hagger, Biddle, & Wang, 2005; Hagger at al.,

2010; Schmalz & Davison, 2006). In addition, as girls develop and mature, they tend to become

less physically active and report lower physical self-concepts than younger girls (Cumming et

al., 2012). There has been considerable speculation as to why girls report lower physical self-

concepts than males in physical education, and many have suggested that discrepancies in

physical self-concept scores are a result of traditional gender stereotypes in Western society

(Caglar, 2009; Crocker et al., 2003; Klomsten, Marsh, & Skaalvik, 2005; Schmalz & Davison,

2006). This is a likely explanation considering that schools are complex social environments

where gender stereotypes are reinforced and traditionally masculine subjects such as physical

education magnify the differences between males and females (Bain, 2009; Connell, 2008; Kirk,

2003; Penney & Evans, 2002). However, simply accrediting gender differences in physical self-

concept to gender stereotypes in physical education is likely a naive, inadequate explanation. The

complete understanding of this phenomenon is far more complex and there is a need for deeper

investigation of females‟ physical self-concept construction within the gendered environment of

physical education. It is likely that female students‟ social comparisons of ability and frame of

reference within the physical education setting play a role in their physical self-concept

development (Craven & Marsh, 2008; Marsh & Craven, 2002; Marsh et al., 2008b). The

gendered environment of physical education provides a complex setting to examine physical

self-concept construction and Marsh‟s (1990) theoretical framework offers a developmental and

multidimensional perspective to explore the physical self-concept development of adolescent

females in physical education.

Overview of Self-Concept and Physical Self-Concept

Self-concept is an individual‟s self-perception that is formed through his or her

interactions with and interpretations of the surrounding environment (Marsh, 1990; Marsh &

Shalveson, 1985). It can be differentiated from other constructs, has a multifaceted and

hierarchical structure, and undergoes developmental changes (Marsh, 1990; Marsh & Shalveson,

1985). In Marsh‟s (1990) model, self-esteem (global self-concept) is divided into academic and

nonacademic self-concepts. In the nonacademic domain, self-concept is further separated into

social, emotional, and physical self-concept sub-domains (See Figure 1.1). Consequently, each

sub-domain is considered to also have a multidimensional structure (Marsh, 1990; Marsh &

Craven, 2006; Marsh & Shalveson, 1985).

Figure 1.1. The Hierarchal Structure of Self-Concept and Physical Self-Concept.

Currently, there is substantial support for the multifaceted nature of self-concept and

although global self-concept is not insignificant, it is no longer considered a useful tool for

understanding the distinct dimensions of self-concept (Craven & Marsh, 2008; Marsh & Craven,

2006; Marsh & O‟Mara, 2008; Marsh, Trautwein, Ludtke, Koller, & Baumert, 2006). Therefore,

it is strongly suggested that researchers focus on the dimensions of self-concept most pertinent to

their field (Marsh, 1990, 2006; Marsh & O‟Mara, 2008; Marsh & Craven, 2006). For researchers

Global Self-

Concept

Academic

Nonacademic

Social Emotional Physical

Physical

Ability

Strength

Activity

Levels

Endurance

Sporting

Ability

Physical

Coordination

Flexibility

Physical

Appearance

Body Fat

Health

in physical education, physical self-concept is the sub-domain most applicable to understanding

this construct within the complex physical and social environment of physical education (Craven

& Marsh, 2008; Marsh, 1990; Marsh & O‟Mara, 2008).

Physical self-concept refers to an individual‟s perception of himself or herself within the

physical domain and is separated into two specific categories: physical ability and physical

appearance (Marsh, 1990, 1996, 1999, 2002; Marsh & Craven, 2006; Marsh & Shalveson, 1985;

Peart, Marsh, & Richards, 2005). In physical education, physical self-concept maintains the

multidimensional structure that has been recognized in other physical settings (Dunton et al.,

2006; Gheris, Kress, & Swalm, 2010; Marsh, Hau, Sung, & Yu, 2007). Specifically, Marsh and

colleagues suggest that an individual‟s perception of his or her physical strength, body fat,

physical activity levels, endurance, sporting ability, physical coordination, physical health,

physical appearance, and physical flexibility all contribute to his or her overall physical self-

concept (See Figure 1.1) (Marsh, Richards, Johnson, Roche, & Treymayne,1994). There is

strong support that as children age, physical self-concept becomes more differentiated and

multidimensional (Marsh 1993; Marsh, Debus, & Bornholt, 2005; Marsh et al., 2007); indicating

that physical self-concept is also best studied from a developmental perspective (Marsh, 1993,

2002, 2006, Marsh, et al., 2007).

An individual‟s frame of reference plays an important role in the development of global

and physical self-concept (Marsh & Craven, 2002) and there is support for frame of reference

effects for physical self-concept in physical education (Gheris et al., 2010; Trautwein, Gerlach,

& Ludtke, 2008). The Internal/External Frame-of-Reference (I/E) Model provides a framework

for understanding the development of self-concept in relation to internal and external

comparisons (Craven & Marsh, 2006; Marsh, 1990; Marsh & Craven, 2002). For instance,

internal comparisons of ability within each domain can impact self-concept in other domains. A

student may have average math ability and above average physical ability and he or she will in

turn use this internal comparison of ability in two separate domains to construct self-concepts in

each area (Marsh & Craven, 2002; Marsh, Hey, Roche, & Perry; 1997). Students may also

compare current performances on a specific task with their own previous performances within

the physical domain (Gheris et al., 2010). As a result, the construction of a positive physical self-

concept is part of an internal self-evaluation process. In addition to these internal evaluations,

external comparisons, like social comparisons of ability, influence the development of physical

self-concept. As children grow older, external comparison increasingly plays a larger role in self-

concept development (Marsh, 2006; Marsh, Trautwein, Ludtke, & Koller, 2008) and the context

of physical education provides a complex environment to negotiate physical self-concept

construction. For instance, physical education classrooms often have children exhibiting skills in

a social setting and these public displays of ability provide an optimal scenario for students to

engage in external social comparisons of ability (Gheris et al., 2010; Trautwein et al., 2008).

In order to provide further explanation of social comparison effects on self-concept

development, Marsh and colleagues created the Big-Fish-Little-Pond Effect (BFLPE) framework

(Marsh et al., 2008b; Marsh & Craven, 2002; Marsh, Seaton, Trautwein, Ludtke, Hau, O‟Mara,

& Craven, 2008). The basic premise of the BFLPE is that students with high ability will have a

lower self-concept when surrounded by others with high average ability. Conversely, students

with average ability will have a higher self-concept when surrounded by others with low average

ability (Chanal, Marsh, Sarrazin, & Bois, 2005; Marsh, Seaton, et al., 2008). There is currently

limited support for the BFLPE for physical self-concept in physical education (Chanal et al.,

2005; Trautwein et al., 2008) and while preliminary evidence for the I/E Frame-of-Reference

Model and the BFLPE in physical education does exist, it is sparse and conclusions from this

research are limited.

Adolescent females in physical education tend to report lower physical self-concepts than

males, which is frequently attributed to gender stereotypes. However, a multidimensional and

developmental perspective of physical self-concept construction can provide insight into how

physical self-concept may be affected in the physical education environment. Physical education

is an ideal setting for internal comparisons of performance as well as social comparisons of

ability. Therefore, there is a substantial need for deeper investigation of how adolescent girls‟

frame of reference and the social environment of physical education influence the development

of a positive physical self-concept.

The purpose of this dissertation is to explore how adolescent females in physical

education perceive the physical education environment, their personal interpretations of physical

self-concept, and the role of social comparison in their physical self-concept development.

Chapter 2 examines how female students in physical education perceive internal and external

frames of reference and captures their personal interpretations of physical self-concept, frame of

reference, and the physical education environment. This study employed a phenomenological

framework that included extensive field observations and in-depth participant interviews with

female students enrolled in same-sex physical education classes. Chapter 3 used survey and

physical ability data to investigate a possible BFLPE for sport self-concept in a sample of

adolescent females enrolled in same-sex and coeducational physical education classes.

CHAPTER TWO: PHYSICAL SELF-CONCEPT AND FRAME OF REFERENCE

FOR GIRLS IN SAME-SEX PHYSICAL EDUCATION

Introduction

Past research has suggested that physical self-concept, an individual‟s perception of

himself or herself within the physical domain (Marsh, 1990, 2002), is a primary determinant of

adolescent girls‟ global self-concept (Beasley & Garn, in review; O‟Dea & Abraham, 1999). A

positive physical self-concept is associated with physical activity participation, skill

development, and motor learning in physical education (Guerin, Marsh, & Famose, 2004; Peart,

Marsh, & Richards, 2005) and physical education can aid in the development of positive physical

and global self-concept among young girls (Craven & Marsh, 2008; Daley & Buchanan, 1999;

Jones, Polman, & Peters, 2009). Physical self-concept researchers have consistently identified

gender differences, with females reporting lower physical self-concept than males across

multiple physical activity settings, including physical education (Caglar, 2009; Hagger, Biddle,

& Wang, 2005; O‟Dea & Abraham, 1999; Schmalz & Davison, 2006).

Although gender differences in physical self-concept are well documented, there is

currently only speculation as to why these differences are so distinct within physical education.

Many researchers attribute these gender discrepancies to Western society‟s gender stereotypes

(Asci, 2002; Klomsten, Marsh, & Skaalvik, 2005; Klomsten, Skaalvik, & Espenes, 2004;

Schmalz & Davison, 2006). While this seems to be a reasonable conclusion, simply attributing

gender differences in physical self-concept among physical education students to gender

stereotypes may not account for the complexity of this phenomenon. It is likely that gender

stereotypes in physical education and female students‟ frame of reference regarding their

physical capabilities and appearance are closely related to physical self-concept development.

Physical Self-Concept

According to Marsh‟s (1990) theoretical framework, self-concept has a multidimensional

and hierarchal structure. In this framework, self-esteem, identified as global self-concept, is

divided into academic and nonacademic self-concepts. In the nonacademic domain, self-concept

is further separated into social, emotional, and physical self-concept sub-domains. Physical self-

concept is a multidimensional facet of nonacademic self-concept (Craven & Marsh, 2008;

Marsh, 1990) and the two primary sub-domains of physical self-concept are physical ability and

physical appearance, which are further divided into nine separate sub-categories. More

specifically, Marsh and colleagues (Marsh, Richards, Johnson, Roche, & Treymayne,1994)

suggest that an individual‟s perception of his or her physical strength, body fat, physical activity

levels, endurance, sporting ability, physical coordination, physical health, physical appearance,

and physical flexibility all contribute to his or her overall physical self-concept (See Figure 2.1).

Self-concept is constructed in relation to the surrounding environment (Marsh, 1990; Marsh &

Shalveson, 1985), and the Internal/External Frame-of-Reference (I/E) Model (Marsh, 1986)

provides a valuable theoretical perspective to examine the development of these multiple facets

of physical self-concept in physical education.

Internal /External frame-of-reference model. Previous inquiry of self-concept

development within the academic domain has established that the construction of self-concept is

related to both internal and external factors (Marsh & Craven, 2002). Marsh‟s (1986)

Internal/External (I/E) Frame-of-Reference Model provides a theoretical framework for

understanding the development of self-concept in relation to internal and external comparisons

(Craven & Marsh, 2006; Marsh, 1990; Marsh & Craven, 2002). The I/E Frame-of-Reference

Model is based on the distinctiveness of and potential interactions between internal and external

frames of reference (Marsh & Yeung, 1998). Theorists‟ posit that individuals use: (a) internal

comparisons of ability (e.g. comparing current performances to past performances) in different

domains to construct self-concept within specific domains; and (b) external comparisons of

ability (e.g. comparing personal performance on a task to a peer‟s performance on the same task)

in order to construct domain-specific self-concepts (Craven & Marsh, 2006; Marsh & Craven,

2002).

Figure 2.1. The Hierarchal Structure of Self-Concept and Physical Self-Concept.

Since the original development of the model, Skaalvik and Skaalvik (2002) have

identified more specific internal and external frames of reference and sources of information

within school settings. The five sources of information identified for both frames of reference

are: (a) direct observations of achievement, (b) the teachers‟ responses and comments, (c)

responses and comments from classmates, (d) responses and comments from individuals outside

Global Self-

Concept

Academic

Nonacademic

Social Emotional Physical

Physical

Ability

Strength

Activity

Levels

Endurance

Sporting

Ability

Physical

Coordination

Flexibility

Physical

Appearance

Body Fat

Health

the class, and (e) grades. For external comparisons, students may use school-average ability,

class-average ability, students within their class, or students outside of their class as frames of

reference. For instance, students may overhear the teacher complimenting another student‟s

achievement in a particular subject and use this external source of information to form his or her

respective domain-specific self-concept. Likewise, the frames of reference for internal

comparisons are focused on achievements in different school subjects at the same point in time,

in the same subject over an extended period of time, in different subjects with respective goals

and aspirations, and in different subjects with subsequent effort in each (Skaalvik & Skaalvik,

2002). Thus, a student may compare his or her grade in a particular subject to grades received in

the same subject in a previous year and use this internal comparison to construct his or her self-

concept in that area.

The I/E Frame-of-Reference Model is a robust framework that has demonstrated cross-

cultural generalizability within academic settings (Marsh & Yeung, 1998; Marsh, Kong, & Hau,

2001). According to a meta-analysis conducted by Moller and colleagues, this framework is

applicable across gender, age groups, cultures, and various achievement settings (Moller,

Pohlmann, Koller, & Marsh, 2009). Therefore, this model is well-suited for examination of the

role frame of reference may play in physical self-concept development in physical education.

Unfortunately, investigation of the I/E Frame-of-Reference Model in the physical domain is

extremely limited. Although incomplete, previous studies have begun to provide preliminary

support for frame of reference effects in the development of physical self-concept in physical

education (Gheris, Kress, & Swalm, 2010; Trautwein, Gerlach, & Ludtke, 2008). For example, it

has been documented that students often compare their current performance on a specific task

with their own previous performances (Gheris et al., 2010). In addition, Marsh (1993) reported

that social comparisons of running ability among students impacted their own self-perceptions in

the physical education class.

The context of physical education provides a complex setting to negotiate the

development of physical self-concept. For instance, PE classrooms often have children

exhibiting skills in social settings and such public displays of ability provide an optimal scenario

for external social comparisons (Gheris et al., 2010; Trautwein et al., 2008). It is likely that in

physical education, both internal and external frames of reference work in combination with one

another, and in relation to the different sources of information outlined by Skaalvik and Skaalvik

(2002). For instance, a physical educator may provide corrective feedback regarding a student‟s

performance of a particular skill, such as throwing (e.g. “Remember to turn your side to the

target.”). The student may perceive that he or she is exerting the same amount of effort to

execute the skill as if completing an exam in math. However, if feedback from the math teacher

is positive regarding his or her performance (e.g. “You did a great job on this exam!”), the

student may form negative internal perceptions of his or her ability in physical education and

physical self-concept may be negatively influenced. In a similar manner, a different student may

overhear the physical education teacher‟s corrective feedback regarding throwing and observe

his or her classmate working to correct a skill that he or she has already mastered. This student

may use this external comparison and evaluate himself or herself more positively, which would

have a positive impact on physical self-concept.

Given the above examples, it is reasonable to conclude that the I/E Frame-of-Reference

Model can be readily applied to a physical education setting. It is inferred that perceptions of

ability play a major role in physical self-concept development (Marsh et al., 2010), yet research

of frame of reference in physical education has centered on students‟ actual ability or utilized

only quantitative data without taking students‟ actual perceptions and interpretations of ability

into consideration (Chanal, Marsh, Sarrazin, & Bois, 2005; Trautwein et al., 2008). Considering

the lack of extant research in this area, there is a substantial need to examine students‟ physical

self-concepts and perceptions of the physical education environment from a qualitative

perspective. A deeper understanding of students‟ personal interpretations of physical self-

concept development in physical education can provide physical educators with valuable

information that could empower them to become more effective educators. A phenomenological

methodology provides a unique perspective that allows the researcher to become immersed in the

physical education experiences of participants and observe the environment firsthand. Therefore,

the purpose of this study was to provide in-depth descriptions of female students‟ personal

interpretations of physical self-concept, frame of reference, and the physical education

environment using a phenomenological approach

Methods

Qualitative Design

Considering the extensive role self-perceptions play in physical self-concept

development, this study employed a phenomenological methodology in order to closely examine

students‟ personal perceptions and views (van Manen, 2011). One advantage of using a

phenomenological approach is its ability to describe a phenomenon and how individuals interpret

this phenomenon (Patton, 2002). In addition, such methodology allows the researcher to become

immersed in the experiences of the study participants (van Manen, 1990, 2011). Consequently, in

this study, asking students to describe their experiences, perceptions, and feelings in-depth

provided deeper insights into their personal perceptions of frame of reference, the physical

education environment, and physical self-concept.

According to van Manen (1990, 2011), a phenomenological methodology involves three

primary data collection strategies: (a) observation, (b) field notes, and (c) interviewing. In a

phenomenological approach, the researcher plays an essential role and must integrate him or

herself into the participants‟ environment to gain an in-depth understanding of the phenomenon.

Through conducting detailed observations the researcher attempts to describe, explain, and

interpret participant experiences. In order to understand how individuals interpret events or

environments, the researcher must also observe these events and environments firsthand (Patton,

2002). This experience allows the researcher an additional viewpoint other than simply that of

the participant (van Manen, 1990, 2011). In a similar manner, researchers then conduct in-depth

interviews to gain insight into the participants‟ interpretations of the phenomenon (Patton, 2002).

Finally, throughout the entire process, the researcher attempts to describe and understand the

phenomena at hand through both writing and rewriting field notes and his or her personal

interpretations of the phenomenon (van Manen, 1990, 2011).

Participants

Participants in this study were ten female students (N= 10) enrolled in one ninth grade

(n=6) and one tenth grade (n=4) physical education class at a local all-girls private high school.

The girls reported their ethnicity as Caucasian (100%), with a mean age of 14.7 years (SD=.48).

All of the students were self-described as physically active and were involved in various

structured and informal physical activities outside of school (e.g. running, soccer, basketball,

bowling, etc.). In addition, all of the girls were consistently observed participating during the

physical education classes.

Setting

Classes took place across two consecutive semesters (spring and fall) and two

participants were enrolled in both classes. Classes were taught by one female certified physical

education teacher and met daily for 50 minute time periods. The physical education program at

the school was based on a multi-activity model and each class participated in at least three

physical education curricular units per semester. Field notes provided rich descriptions of several

of the units and the atmosphere of the class during each unit. Units included individual lifetime

activities such as badminton, archery, Pickleball, and fitness exercises. Team sport activities like

basketball and softball were also offered. In addition, nine weeks of the semester were spent in

health units where topics such as CPR and eating disorders were covered. All students were

required to dress out in school-issued uniforms (identical t-shirts and athletic shorts of their

choice) during physical education classes. Students were required to take two semester credits of

physical education before graduation and received a letter grade for the course. Grades were

based on participation, dressing out, and scores on written exams.

It is important to note that throughout the school there was an overwhelming sense that

all-girls schools are preferable to coed settings. This culture and belief that same-sex classes are

“better” was reinforced by both students and teachers, especially with regard to physical

education. Therefore, the overall perception that all-girls classes are significantly more beneficial

to female students than coed classes was taken into consideration when interpreting the results of

this study.

Role of the Researcher

All observations, field notes, and interviews were completed by one female researcher

who was familiar with the school, administration, and faculty. She had facilitated field teaching

experiences and observations at this school for physical education teacher education students

from a local university throughout the two years prior to data collection. She also accompanied

university students during observations and field teaching experiences. In addition, the

researcher was a former physical education teacher and viewed the physical education

environment from an educator‟s perspective. She became a fixture in the classes and throughout

the course of field observations and interviews, students would often ask their physical education

teacher about the researcher when she was not present. Therefore, participants in this study were

very comfortable with her and accustomed to her presence in their physical education classes. As

a result, participants were open and forthcoming throughout the data collection process. This

open relationship with both the participants and the teacher facilitated follow-up interviews and

discussion where the researcher was able to spend a significant amount of time discussing

emerging themes as a method to reduce researcher bias.

Procedures

Permission to conduct this study was granted by the University Institutional Review

Board. Students were provided with a full description of the study procedures prior to

participation and informed consent was collected from parents. Likewise, all participants were

required to submit a signed child assent form prior to participation. Participants first completed

questionnaires containing demographic questions and items evaluating physical self-concept.

Survey responses were used as selection criteria for interviews. In addition, participants were

observed during physical education classes.

Measures

Participants were selected for interviews based upon their responses to the global

physical self-concept scale (M=4.01, SD=1.54) of the PSDQ-S (Marsh et al., 2010) (See

Appendix B). Purposeful sampling took place in order to recruit students with a wide range of

scores on the PSDQ-S (1.00-5.67). Students (N=12) from both semesters (spring and fall) were

invited to participate in interviews, however, two of the invited students declined citing the time

commitment as a barrier to participation

PSDQ-S. The PSDQ-S (Marsh et al., 2010) is a psychometrically sound 40 item

instrument containing items from the PSDQ (Marsh, 1996) and it has good reliability, test-retest

stability over both short and long term, a well-defined factor structure that is invariant over

gender, and both convergent and discriminant validity (Marsh, 1997, 2002; Nigg, Norman,

Rossi, & Benisovich, 2001). The PSDQ-S is a valid and reliable measure with reliability

coefficients ranging from .84-.91 (Marsh et al., 2010). Items from the global physical subscale,

which require individuals to respond to the declarative statements 1) “Physically, I feel good

about myself”, 2) “Physically, I am happy with myself”, and 3) “I feel good about who I am

physically” on a 6-point Likert scale ranging from 1 (not at all true) to 6 (very true), were used to

select participants for interviews.

Data Collection

Field Observations. Field observations (N=30) were conducted and each observation

lasted for the duration of one physical education class period (50 minutes). Observations spanned

across the spring (n=14) and fall (n=16) semesters. During observations, extensive field notes

were taken and the researcher provided detailed descriptions of the physical education

environment, social context, activities, students‟ interactions with one another, and interactions

with the teacher. Immediately following each observation, field notes were typed, analyzed, and

separated into the following categories: Class Description, Additional Notes, Perceptions and

Thoughts, and Further Questions (See Appendix C).

Interviews. A total of nineteen participant interviews (N=19) were completed. Each

interview lasted approximately 30 minutes (M= 30.16; SD=9.36) and was audio recorded.

Interviews took place before school, during lunch periods, or after school and were conducted in

an empty health classroom located in the school gymnasium. The majority of participants

completed two interviews; however, one student (Morgan) completed the interview protocol

during one session and declined to participate in a second interview. Participants responded to

questions such as: “What do you like/dislike about your physical education class?”, “How would

you describe your abilities in PE?”, and “How would you describe the atmosphere of your PE

class?”. In addition to the interview protocol (See Appendix D), discussion was prompted based

on events taking place in the physical education class (e.g. “Can you tell me more about what

you did in class today?”).

Data Analysis

Phenomenological themes are understood as “the structures of experience” (van Manen,

1990, p.79) and an integral component of a phenomenological research design is thematic

analysis. It is important that those who were engaged in the experience play a significant role and

provide personal insight during the coding process. Thus, field notes and interview transcriptions

were analyzed line-by-line by the primary researcher. Emerging themes identified during the

reflective process (van Manen, 2011) were then discussed with participants during follow-up

interviews. An inductive analysis approach was employed to identify categories, sub-themes, and

higher-order themes (Patton, 2002). Interviews, field notes, and member checks were used in

order to triangulate the data and reduce researcher bias. In addition, informal collaborative

reflection with a second researcher took place during revision of the initial themes (van Menen,

2011).

Results

Three major themes developed from data analysis: (a) Social Risk, (b) A Favorable

Competitive Environment, and (c) Points of Perception. Multiple aspects of Social Risk in

coeducational (coed) physical education were identified, specifically anxiety and perceived

pressures regarding physical ability and appearance. However, same-sex physical education was

perceived to negate these concerns and was viewed as a positive alternative. Competition was

frequently discussed in terms of A Favorable Competitive Environment. Overall, participants

enjoyed and valued competition but it was considered to be activity dependent, unpleasant in

certain circumstances, and desirable only in appropriate amounts. Finally, participants outlined

two Points of Perception, “firsthand accounts” and “external feedback”, which provided a basis

for physical self-concept development.

Social Risk

“Social Risk” was a major theme that emerged from data analysis. All participants

(100%) indicated that physical education created a social environment that was distinct from

other classes and some (60 %, 6/10) saw the social atmosphere as the most positive aspect of

physical education. However, the social environment was not always viewed positively. The

class structure, specifically a coed setting, was perceived to have potential negative social

consequences.

Appearance in coed classes. Participants addressed two primary “Social Risks” in coed

physical education; anxiety regarding appearance (50 %, 5/10) and physical ability (80%, 8/10).

Participants drew on past experiences and compared their current class to coed classes they

experienced during middle school. For example, Vanessa (all names are pseudonyms) (mid-

range physical self-concept) discussed how important appearance could be in coed physical

education when she said:

The girls would, before they go out to PE they would go and fix their hair and make sure

that nothing was wrong or whatever because of the guys. But here it‟s just like, you get

changed and then you go out and you don‟t really have to worry about if your hair is out

of place or if you haven‟t shaved or something like that. That really doesn‟t go against

you I guess. Just cause all-girls, like, some girls haven‟t shaved or whatever so you

don‟t have to worry about being criticized.

Dee (high physical self-concept) expressed similar feelings and mentioned that in an all-female

environment the girls didn‟t “worry as much about sweating”. Laura (mid-level physical self-

concept) said she tries harder in an all-girls class because “no one really cares about your

appearance” and she believed that in coed classes the girls “don‟t try as hard” and “they only

cared about their appearance”. The perceived pressure to appear a particular way in coed classes

was discussed by several participants. Beth (low physical self-concept) considered the coed

atmosphere to set different expectations and demands on girls with regard to appearance and

said, “I think the pressure would be higher in a coed school because there‟s guys everywhere and

you have to worry about what you look like and how skinny you are, and make-up and clothes,

stuff like that”. It is interesting to note that throughout the course of field observations, there

was never an instance when a student did not dress in the required physical education uniform

(school issued top and athletic shorts) and between two classes (n=46), only one student was ever

noted to wear visible make-up.

Ability in coed classes. Participants also discussed in detail perceived “Social Risk”

associated with ability in coed physical education. There is a substantial threat for students who

are considered to be less skilled in a coed class and according to Morgan (low physical self-

concept), the less skilled students are the girls. She stated, “I feel like guys have their own level

and girls have their own level of skills”. Alicia (high physical self-concept) articulated similar

feelings and said, “I think for coed you have more of the varied skill set… and I think there‟s big

differences in coed PE.” According to Amanda (mid-level physical self-concept), “some guys

have different abilities than girls” and boys are “more athletic”.

Even girls like Michelle (high physical self-concept), who consider themselves skilled,

may not have sufficient ability to succeed in a coed setting. She said, “I could always like, hang

ya know? Like up there, but maybe not in like, basketball. Because boys tend to overpower in

those kinds of sports.” The idea that boys overpower girls in coed physical education and take

control of the class was consistent among other participants. Dee said the boys would “be

playing hard and running fast so the girls didn‟t have time to reach it [the ball]”. She also

commented that the girls were “letting the guys do whatever they want” and “weren‟t

participating as much [as the boys]”. Amanda also discussed how girls‟ participation can be

negatively impacted in coed classes and said, “I feel like guys are more prone to exclude girls

cause they‟re like „Oh she can‟t do it, she can‟t do this, she can‟t do that‟.” Feelings of

inadequacy, awkwardness, and embarrassment regarding ability in a coed environment were

common threads for the girls in this study. Beth commented, “Guys would get mad if you didn‟t

win or do something right.” Heather (high physical self-concept) discussed how insufficient

knowledge and ability in physical education could be detrimental for girls and said:

With coed, the guys know how to play sports, so when you would ask a question, you

didn‟t want to seem stupid. In all-girls, most of the girls in my class don‟t really know all

the knick-knack rules of the sport, so you don‟t really feel like an outcast asking a

question.

Surprisingly, interview data suggested that in coed classes the some of the girls were not

only worried about a lack of ability, but also appearing to be overly skilled. According to

participants (30%, 3/10), there are negative consequences for girls who are “athletic” in coed

physical education. They appear to be “trying too hard” and are at risk for name calling. For

example, Dee expanded on this idea when she stated that there was “a really athletic, competitive

girl at my old school and they called her a man”. Heather had a similar experience and said,

“There was this one girl, everyone called her „the beast‟ because she was just amazing at

everything we did and she was better than some of the guys.” Morgan explained perceived

pressures regarding appearance and ability best and said, “Guys are supposed to be better and

manlier looking than girls are.” She also stated, “When guys are there you don't wanna look like,

not manly but, like you‟re better than they are and so you don't wanna try as hard.”

Field notes provided evidence of a wide range of skill levels in each class. However, the

students observed to be the most skilled were often cheered on by classmates and provided with

positive feedback. There were no instances reported or observed where high skilled students

experienced negative consequences regarding ability. Thus, all participants (100%) expressed

that same-sex physical education classes reduced these concerns. According to Beth, there are

advantages to an all-female environment because, “You‟re not as embarrassed to do different

things cause there are not guys watching you.” Heather had a similar attitude and commented,

“You don‟t have to feel embarrassed. Like, if you can‟t kick the ball across the gym, you don‟t

feel embarrassed, because girls don‟t always care about that kind of stuff.” Rebecca (low

physical self-concept) preferred an all-girls class because she felt less pressure to perform. She

said, “I don‟t have to be impressive. Stuff like that. And people don‟t have to impress me or

anybody else.” Beth liked her same-sex class better than coed classes because “they [the girls]

don‟t really care if you win or not”. These results suggest that unlike a coed setting where there

was perceived pressure to appear and perform a particular way, the all-girls environment is one

with no fear of negative consequences associated with not meeting unspoken expectations.

A Favorable Competitive Environment

The second theme that emerged from participant interviews was “A Favorable

Competitive Environment”. All participants (100%) discussed both positive and negative

feelings toward competition and the majority (80 %, 8/10) stated that they valued and enjoyed

competition in physical education. The majority of girls (80%, 8/10) also said competition was

appropriate for physical education and there was a consistent belief that it is an enjoyable tool to

become more active, involved, and skilled in certain circumstances. However, there was a

consistent theme that competition is activity dependent, unpleasant in a coed setting, and only

appropriate in certain amounts.

Activity dependent. Over half (75%, 6/8) of the participants who believed competition is

acceptable in a physical education environment said it was activity dependent. The three most

popular activities mentioned as favorites during interviews were archery (50%, 5/10), softball

(40%, 4/10), and Pickleball (40%, 4/10), and for each some degree of competition was involved.

The archery unit culminated with an individual tournament, during the softball unit the teacher

recorded team scores, and the Pickleball unit consisted of a partner tournament. Alicia described

softball as “kind of competitive” because the teacher kept score, but “people were more

competitive” when it came to Pickleball. The Pickleball tournament was frequently discussed

and was considered a popular activity throughout the school. Each year the school would hold

one tournament and overall winners received a t-shirt. Michelle agreed that competition was

heightened during the Pickleball tournament and stated, “It got really competitive.” Student

perceptions of competition during activities varied and competition levels often fluctuated

between units. Morgan addressed this variation in her class when she said, “They [the teachers]

teach everyone how to play sports and some are more competitive than others, but there‟s a

variety of different things.”

The physical education students at this school had the opportunity to participate in a

variety of activities over the course of the semester and throughout field observations it was

often noted that the atmosphere of the class would change depending on the activity. The most

evident shift was observed during the archery unit. The first half of the unit had been spent

practicing and the second half was devoted to individual competition. In the first lesson where

scores were recorded, an obvious transition in the atmosphere occurred. During practice rounds

the students were primarily concerned with socializing and conversation was often focused on

irrelevant topics (e.g. boys, music videos, movies). However, when the teacher began keeping

score, the focus of conversation shifted and was concentrated on topics related to the lesson such

as shooting, performance, and scorekeeping. Overall, the class appeared to become more

engaged in the activity. In contrast, there were several activities, such as the fitness unit, where

little to no competition was observed. Therefore, competition in this physical education class was

often activity dependent.

Unpleasant in coed settings. Participants also discussed how competition was

unpleasant in a coed setting. When asked to describe the negative aspects of competition

Amanda said, “Well I used to go to a coed school so it was a little bit more uncomfortable with

guys, but it‟s really not here cause you can kindof be yourself.” Another participant, Michelle,

described the differences between coed and same-sex classes and said, “It‟s a little different

because in coed PE it was more competitive, but in this one it‟s not as competitive cause

everyone‟s kindof laid back”.

Often participants (50%, 5/10) described males in physical education as “aggressive” or

“rough”. For example, when explaining why she believes boys and girls should participate in

activities separately Rebecca said, “Cause some boys and girls aren‟t the same. It‟s like you‟re

playing, such as a mouse playing a bear.” The fear of being overpowered or even injured when

participating in competitive physical activities with males was mentioned by Vanessa, who said

she would not want to play a sport with boys because “they get pretty aggressive if you‟re

playing on the same team”. She said in that situation she would opt out of the game and “let them

do all the work”. Throughout participant interviews, the idea that an all-girls physical education

class is a more positive setting for competition was frequently discussed. Morgan mentioned that

the all-girls environment was well-suited for competition because “you can be yourself really

and do what you want to do and be competitive”. Amanda, who described herself as “pretty

competitive”, also suggested that the environment of the class was a positive outlet for

competition and referred to it as a “safe environment to be able to play” and a “safe place to

interact.” Although the belief that competition in a coed physical education setting was

unpleasant and could be physically or emotionally damaging to girls was prevalent, for the girls

in this study same-sex physical education was a solution to their concerns.

Acceptable levels. Although the majority of girls (80%, 8/10) preferred competition and

felt their own physical education environment to be a safe and comfortable outlet for it, a large

percentage (60%, 6/10) discussed an acceptable level of competition. For example, although

Pickleball was often viewed in a positive light, Morgan offered another viewpoint when she said,

“Some people do get really, really, into it. I got like really into it too, and it‟s to the point to

where it‟s not fun cause you‟re just trying too hard.” Michelle also suggested that there is an

acceptable level of competition and commented, “If you don‟t have competition it‟s not really

that fun, but if you have too much competition then people get mad and it‟s just a lot a stupid

stuff going on.”

When describing the acceptable level of competition for physical education, participants

discussed negative aspects to either not having enough competition or being immersed in an

overly competitive environment. The physical education classes the participants were enrolled in

were described as “not really competitive”. For half of the participants (50%, 5/10), a lack of

competition was even perceived negatively and they expressed a desire for more competition. If

there appeared to be a lack of competition in particular sports (e.g. basketball or badminton) that

were intended to be competitive, the girls who were self-described to be “taking the game

seriously” expressed a sense of frustration. Participants mentioned that in basketball there were

other students who “really didn‟t care” or who were “lazy” and without competition they

couldn‟t “get a good game” and the games would be “so boring”. Overall, there was a general

theme that an appropriate level of competition was enjoyable when experienced in a safe and

positive environment.

Points of Perception

Finally, participants in this study identified firsthand accounts of achievement and

external feedback as “Points of Perception” related to physical self-concept development.

Firsthand accounts of others‟ achievements and their own individual accomplishments were

often identified by the girls as sources of comparison for multiple aspects of physical self-

concept. In addition, participants used external feedback from others in order to construct their

physical self-concepts.

Firsthand accounts of others. The majority of participants (80%, 8/10) discussed

multiple aspects of physical self-concept in relation to firsthand accounts of others‟ ability,

others‟ appearance, or their own performances. Field observations supported this finding. For

instance, the structure of the archery unit facilitated direct observations of classmates. Students

watched one another shoot and were often overheard cheering and making encouraging remarks

such as “good job” and “that was a good shot”. In one instance, a student told her group member

“good job” when she hit the target, yet later referred to herself as “terrible” when she missed her

own shot. She observed her peer hit the target and used this firsthand account of another‟s

performance to provide information regarding her own competence. Throughout the unit, it was

not uncommon to observe the students encouraging their classmates, yet often making degrading

comments regarding their own personal performances.

For some participants (40%, 4/10), the most common source for firsthand accounts of

physical ability were family members. The physical self-concept domains of flexibility, strength,

and sports ability were all described via firsthand accounts of family members‟ ability in these

areas. For example, both Beth and Amanda described themselves as “not as flexible” as their

sisters and Alicia evaluated her sports ability positively because her family is “really athletic”.

Although participants described themselves in relation to others in several of the physical self-

concept sub-domains, they frequently discussed their appearance in relation to that of friends

(20%, 2/10) and other girls in general (70%, 7/10). Morgan described her friends as the opposite

of herself when she said “My friends are just really tiny and they‟re not as tall as me.”

Field notes validated friends as a source of firsthand comparisons for appearance. During

a “Power Walk” lesson, two girls were observed comparing their stomachs and thighs and

pinching body fat on their midsections. Throughout the conversation, they were overheard

commenting “I‟m insecure in a swimsuit now”, “I need to lose a few pounds”, and “I need to get

back in shape”. When firsthand accounts of other girls in general were the primary sources of

information, comparisons were based on appearance, body shape, and body size. When asked to

describe her appearance Heather stated, “I would say that there are prettier girls and then there

are like…I‟m better looking than some girls.” Vanessa communicated similar feelings when

discussing various aspects of her own body, such as muscle definition, and commented that her

legs were “not as skinny as somebody else‟s”. She also expressed conflicting emotions related to

comparing herself to others who had different body shapes and sizes:

It can be bad for your…. I guess if you‟re looking at someone who is really skinny it

can be bad. Like, I really wanna look like her but I can‟t, just because how my body is

and how I‟m built, because I play soccer and all these other sports so much.

Morgan also talked about comparing herself to others and the idea of being “skinny”.

She said, “It‟s just, everyone is skinny I guess and you fit in more [if you are also

skinny]”. Another participant, Beth, was very honest regarding her firsthand accounts of others‟

appearance when she admitted, “I do look at the super-skinny and wish that I had that body.” It

is interesting to note that even when discussing endurance, Rebecca referred back to her body

shape and size and compared herself to other girls when she said, “Well, I‟m not one of those

really, really, skinny girls that can run, you know.” For the girls in this study, firsthand accounts

of others‟ physical appearance were the most frequently discussed sources of information.

Firsthand accounts of ability. All participants (100%) discussed firsthand accounts of

personal performance as “Points of Perception” for physical ability. For example, Michelle

expressed her feelings of coordination when she commented, “Coordination, I have pretty good

hand and feet coordination, like soccer, but there are a couple of things, like Pickleball, that I‟m

not good at.” All participants (100%) discussed flexibility in relation to firsthand accounts of

their own personal achievements. They made comments such as “I can only do a left split” or “I

can‟t touch my toes”. Participants also used firsthand accounts of personal performance when

comparing their academic and physical abilities. For example, Dee compared her achievements

in the physical domain to her academic abilities and said, “Well, I don‟t really exercise that much

so I wouldn‟t say I‟m really an athlete, but I mean, I‟m really good at math.”

External feedback. The second “Point of Perception” for participants in this study (50%,

5/10) was external feedback, specifically feedback from family members (20%, 2/10), coaches

(20%, 2/10), and friends (10%, 1/10). For instance, Vanessa said she knows how to describe her

abilities in soccer “by what my parents tell me”. Beth discussed comments from her friends when

asked about her appearance and said, “Friends tell me I‟m pretty and stuff. I mean, I don‟t really

agree with them.” Although it was not discussed during participant interviews, field observations

provided additional evidence that comments from others may be a source of information in the

physical education environment. Throughout the archery unit the girls consistently provided

positive feedback to one another regarding their performances. For example, one was overheard

saying “Becky is a pro at this” and later in the same lesson Becky told her group “I‟m finally

good at something”. Comments and responses regarding performance were not limited to verbal

feedback. During the badminton tournament, one partner group was consistently observed giving

each other high-fives when they scored a point. In addition, the teacher was often noted to

provide positive performance, corrective, and nonverbal feedback to students. Therefore,

although participants may not have been aware of it, they were often receiving feedback from

their peers and the teacher.

Discussion

This study examined adolescent females‟ physical self-concept from a phenomenological

perspective. Results indicate that female students in same-sex physical education classes may

perceive coed classes as places where they are at risk for emotional and physical harm.

Participants in this study discussed perceived pressures regarding both physical ability and

appearance in coed physical education. Results also highlight the role of competition in physical

education and how it impacts the class atmosphere. Participants complied with traditional gender

expectations and perceived males as aggressive, rough, and dangerous during competitive

activities. All participants viewed same-sex classes as a solution to their concerns regarding

physical ability, physical appearance, and competition in physical education. In addition, results

indicate that female physical education students use multiple sources of information to construct

their own physical self-perceptions of ability and appearance.

Participants in this study indicated a significant perceived social risk associated with coed

physical education, specifically concerns with regard to their appearance. These results are not

surprising considering the role of the body in physical education and the girls‟ perceptions that

they were often “being watched” by the boys during their coed classes. For example, physical

education often requires students to put their bodies on display and use them as a means of

demonstrating knowledge and skill in the physical domain (Azzarito & Solmon, 2006, 2009;

Larsson, Fagrell, & Redelius, 2009). In addition, students are typically expected to wear a

school-issued uniform, which was identified as a negative aspect of coed physical education,

perhaps because it can be uncomfortable or show certain parts of the body. In physical education

students may be required to wear shorts that expose their legs and if the uniform is ill-fitting, it

can emphasize feminine aspects of the body such as the breasts, hips, and waist. As girls

progress through adolescence, they often report increased levels of body dissatisfaction,

specifically with their hips, thighs, and waist (Rosenblum & Lewis, 1999) and if students are

uncomfortable in their uniforms these feeling may be amplified. Participants indicated that the

coed setting was a site where “guys are watching you” and expressed significant concern

regarding the presentation of their bodies in coed classes. Their concern was perhaps a result of

perceived male observation and increased feelings of discontent with feminine-typed body parts

(Azzarito & Solmon, 2006; Lawler & Nixon, 2011; Rosenblum & Lewis, 1999). Anxiety was

directed toward how the body was perceived in coed physical education and centered on

traditionally stereotypical attributes of femininity such as wearing make-up, having a nice-

looking hairstyle, or having shaved legs. Participants also expressed concern with appearing less

feminine, which included sweating during physical activity. Therefore, for girls like the ones in

this study who may feel watched by boys in their classes (Flintoff & Scranton, 2001; Webb,

McCaughtry, & McDonald, 2004), public displays of the body in combination with perceptions

of male observation in physical education can create negative feelings of anxiety and pressure.

These results hold many implications for practitioners. The internalization of appearance

ideals is associated with increased body dissatisfaction among adolescent girls (Lawler & Nixon,

2011) and appearance is an important sub-domain of physical self-concept (Marsh et al., 1994).

Therefore, increased anxiety regarding how the body is presented, how it meets appearance

ideals, and how it is perceived by others (i.e. male classmates) in coed physical education could

potentially negatively impact female students‟ physical self-concepts. The environment of the

coed class can sometimes create an increased awareness of the body and encourage conformity

to gendered body ideals for young girls. Teachers should be aware that female students may feel

as if they are being watched by male classmates and keep this in mind when asking students to

perform skills in front of the class or participate in activities that put the body on display in front

of others. Teachers should also consider how physical education uniforms can create unease and

anxiety for some students. For example, some girls may be uncomfortable wearing shorts if they

feel like they need to shave their legs. Physical educators can focus on creating an environment

where students feel comfortable, safe, and accepted to counteract female students‟ perceptions of

surveillance and feelings of unease resulting from gendered body expectations in coed classes.

Participants also identified negative consequences associated with possessing either too

much or not enough skill in coed physical education. Consistent with previous research, there

were negative social consequences such as name calling, harassment, and exclusion from

activities, associated with the girls‟ physical ability (Cockburn & Clarke, 2002; Hills & Croston,

2012). For example, participants indicated that they may be at risk of being called a “man” or

“beast‟ if they were considered to be “athletic” or would even be excluded from activities

entirely. These results are consistent with previous research stating that girls who are viewed as

“athletic” or are high-skilled in physical activity and sport are at risk for having their sexuality

questioned and may be called names like “tomboy”, “butch”, or “lesbian” (Clarke, 2002; Gorely,

Holroyd, & Kirk, 2003). In their coed classes, the girls in this study were attempting to balance

displays of skillfulness with feminine and masculine expectations. Girls tend to value skillfulness

and have a desire to appear strong and competent in physical activities (Azzarito, 2010; Lee,

2009), yet a desire to avoid displays of masculinity in physical education is not uncommon for

girls (Azzarito & Solmon, 2009), perhaps due to the perceived social cost of demonstrating

athleticism.

In contrast, when girls in coed physical education don‟t possess or demonstrate adequate

skill, they are often excluded from activities. Coed classes frequently participate in male-

stereotyped sports (Brown & Rich, 2002; Gorely et al., 2003) that require students to perform

masculine or feminine traits and girls may either choose not to participate or allow the high-

skilled males to dominate during these activities. For example, according to one participant, in a

masculine activity like basketball, students who demonstrate conventionally masculine

characteristics such as aggressiveness, speed, and power will be more successful, and these

students are typically the males in the class As Michelle indicated, girls who do not possess these

traits may consider themselves to be unskilled in physical education, which can negatively

impact their physical self-concepts. In addition, if these girls are using an external frame of

reference to evaluate their physical abilities, they will likely perceive themselves as inadequately

skilled.

Competition in physical education was also a key theme identified in this study. In a coed

setting, competition was perceived negatively and influenced girls‟ participation in activities

during class. These results are consistent with previous research indicating that girls in coed

classes are less active than boys (McKenzie, Prochaska, Sallis, & Lamaster, 2004). Participants

mentioned “not being on the same level as boys” in physical education and expressed a fear of

physical harm when competition was involved and girls played alongside or against boys. The

girls reflected that males tended to dominate during competitive activities like basketball and

displayed “rough” or “aggressive” behaviors. A male-model of physical education that includes

traditionally masculine and competitive activities and reinforces a dominant gender order is often

common in coed settings (Kirk, 2002, 2003). For males, such an opportunity to demonstrate their

skill, assertiveness, and masculinity during these competitive activities translates into physical

and social capital in the physical education environment. In the traditionally masculine culture of

coed physical education, teachers often provide male-stereotyped activities which value

aggressiveness, pain tolerance, and strength (Brown & Rich, 2002; Kirk, 2002, 2003). For girls

like the ones in this study, activities that are often dominated by the males in their classes and

encourage these traditionally masculine characteristics can create fear and anxiety. Although

teachers may not necessarily provide activities that involve physical contact, they should

consider physical closeness and competition when making curricular choices. For example,

participants often discussed basketball as a male-dominated contact sport that boys and girls

should participate in separately. Participation in such activities creates unease for girls regarding

their skill level and their physical or emotional safety. If teachers desire to implement

competitive activities in their classes, they should consider less traditional alternatives, like

ultimate Frisbee, that are not gender stereotyped. Modifying equipment (e.g. a Frisbee instead of

a basketball) and rules (e.g. everyone must touch the Frisbee before the team scores) allows the

teacher to create a more inclusive environment that will promote gender equity. In addition, these

modifications can create balance for a wide range of skillfulness in the class and produce a more

welcoming atmosphere for all students, regardless of skill level.

While all of the participants in this study preferred same-sex physical education and

identified it as a positive alternative to coed classes, it is important to note that the culture of the

school strongly advocated for an all-girls environment. However, the environment and

atmosphere of the girls‟ previous coed physical education classes may have also contributed to

their preferences for all-girls physical education. For instance, unfortunately teachers often

acknowledge male dominance in their classes as something that is expected and may

consequently avoid challenging traditional stereotypes or favor boys over girls (Chalabaev,

Trouilloud, & Jussim, 2009; Chepyator-Thomas & Ennis, 1997; Larsson et al., 2009). Students

and teachers are both aware of the gender order in physical education and generally enforce

dominant gender stereotypes associated with hegemonic masculinity (Brown & Rich, 2002;

Kirk, 2003; Larsson et al., 2009).Therefore, it is reasonable to assume that conformity to these

stereotypes simply creates an environment that influences the experiences students have in the

class. For instance, a dominant masculine culture in physical education may regulate

opportunities and access to activities, actual and perceived skillfulness, use and self-evaluation of

the body, and teacher-student interactions and relationships (Azzarito & Solmon, 2009;

Chepyator-Thompson & Ennis, 1997; Klomsten et al., 2005; Lee, 2009), all of which can

influence physical self-concept. Thus, the traditional culture of coed physical education has the

potential to create an environment that can negatively impact female students‟ physical self-

concepts. Participant interviews indicated that their previous coed classes did conform to a

traditional, masculine model of physical education, therefore, this culture likely contributed to

their perceptions that same-sex physical education was “better”. Past research has indicated that

gender identity and expression is fluid and that teachers can influence the gendered culture of

physical education by attempting to provide gender equity in their classes (Azzarito & Katzew,

2010; Azzarito & Solmon, 2006). Therefore, teachers in coed classes should have an awareness

of the hidden curriculum, avoid traditional gender stereotyped activities, and structure

competitive activities in such a way that all students have an equal opportunity for success. By

doing so, they may be able to create an atmosphere that fosters positive physical self-concept

development and negates some girls‟ negative feelings toward coed physical education.

Finally, results of this study provide support for the I/E Frame-of-Reference Model for

understanding the development of physical self-concept in relation to internal and external

comparison (Craven & Marsh, 2006; Marsh, 1990; Marsh & Craven, 2002) and the five sources

of information for self-concept development (Skaalvik & Skaalvik, 2002) in the physical

domain. Participants identified internal (i.e. prior experiences and comparisons of ability

between domains) and external (i.e. observations of others and external feedback) frames of

reference when describing their physical self-concepts. Firsthand accounts of others‟

accomplishments (e.g. family members, friends, other girls in general) and personal

performances are similar to the first source of information, “direct observations of achievement”.

In addition, results support that participants used external feedback from family members,

coaches, friends, classmates, and the teacher in the self-evaluation process. These findings

provide evidence for three other sources of information, “teachers‟ responses and comments”,

“responses and comments from classmates”, and “responses and comments from others outside

the class”. There was no evidence that participants used grades as a source of information to

construct physical self-concept. This is not surprising considering that grades in this physical

education class were based on participation, dressing out, and scores on written exams.

Therefore, students‟ grades did not provide them with any information regarding their physical

ability in the class.

The results of this study have many implications for teachers in physical education. A

deeper understanding of how students construct physical self-concept in physical education

makes it possible to design curriculum and structure classes in ways to promote positive physical

self-concept development. Teachers should keep in mind that students are comparing their

personal achievements to those of others (Craven & Marsh, 2008; Skaalvik & Skaalvik, 2002)

and avoid placing students on display, emphasizing personal ability, and providing negative or

degrading feedback in front of the class. Instead, teachers can encourage students to focus on

personal improvement and task mastery in order to create a positive learning environment that

does not highlight social comparison of ability among classmates (Solmon, 2003). In addition,

teachers must be aware of the types of feedback they are providing and strive to give students

constructive and corrective feedback regarding skill performance. Finally, teachers should

consider students‟ prior experiences in the physical domain, experiences outside the class, and

achievements in other areas. If teachers are aware of any preconceived notions regarding

physical ability that students bring with them into the class, they will have a better chance of

positively impacting student learning and skill development.

In conclusion, this study makes a significant contribution to existing literature by

highlighting current issues experienced by girls in coed and same-sex physical education. This

study also provides support for the I/E Frame-of-Reference Model and evidence for four sources

of information (direct observations of achievement, teachers‟ responses and comments,

responses and comments from classmates, and responses and comments from others outside the

class) for physical self-concept development in physical education. Limitations of this study

include the lack of diversity and the narrow focus of a same-sex physical education setting. In

addition, the fact that the overall culture of the school promoted the idea that all-girls classes are

preferable to coed classes should be considered as a limitation. Future research should include

additional demographics and cultures in addition to investigating both coed and same-sex

physical education classes.

CHAPTER THREE: INVESTIGATING A BIG-FISH-LITTLE-POND EFFECT AND

MODERATING EFFECTS OF CLASS TYPE FOR GIRLS IN PHYSICAL EDUCATION

Introduction

A primary goal of physical education is to promote the development of positive physical

and global self-concepts (National Association for Sport and Physical Education [NASPE],

2011) and research indicates that a positive physical self-concept results in favorable outcomes

among students in physical education (Chanal, Sarrazin, & Bois, 2005; Marsh, Trautwein,

Ludtke, Koller, & Baumert, 2006; Trautwein, Gerlach, & Ludtke, 2008). However, females often

report a lower physical self-concept than males in physical education (Asci, 2002; Hagger,

Biddle, & Wang, 2005; O‟Dea & Abraham, 1999) and it is likely that the significant role of

social comparison in physical self-concept development (Marsh et al., 2008a) contributes to

these gender discrepancies. Marsh‟s and colleagues‟ theoretical framework for the Big-Fish-

Little-Pond-Effect (BFLPE) provides an explanation for the effects of social comparison on self-

concept and offers a unique perspective that could provide insight as to why females typically

report a lower physical self-concept in physical education (Marsh & Craven, 2002; Marsh et al.,

2008a).

Physical Self-Concept

According to Marsh‟s (1990) theoretical framework, physical self-concept is an

individual‟s perception of himself or herself within the physical domain. Physical self-concept is

a sub-domain of nonacademic self-concept, which is one component of global self-concept. The

hierarchal and multidimensional structure of global self-concept (self-esteem) lends itself to

examination from a domain-specific perspective (Marsh, Richards, Johnson, Roche, &

Treymayne, 1994). Therefore, physical self-concept, specifically, an individual‟s perception of

his or her physical strength, body fat, physical activity levels, endurance, sporting ability,

physical coordination, physical health, physical appearance, and physical flexibility, is the

component of self-concept that is most applicable to the physical domain (Marsh, 1990; Marsh &

Craven, 2006) (See Figure 3.1).

Figure 3.1. The Hierarchal Structure of Self-Concept and Physical Self-Concept.

The Big-Fish-Little-Pond-Effect (BFLPE)

The construction of self-concept is related to both internal and external factors and social

comparison is one external factor that influences self-concept development (Marsh & Craven,

2002). In order to provide further explanation of the role of social comparison in self-concept

construction, Marsh and colleagues developed the Big-Fish-Little-Pond Effect (BFLPE)

framework (Marsh & Craven, 2002; Marsh et al., 2008). The basic premise of the BFLPE is that

self-concept is negatively impacted by school and class average ability. In other words, students

Global Self-

Concept

Academic

Nonacademic

Social Emotional Physical

Physical

Ability

Strength

Activity

Levels

Endurance

Sporting

Ability

Physical

Coordination

Flexibility

Physical

Appearance

Body Fat

Health

with high ability will have a lower self-concept when surrounded by others who also have a high

ability. Conversely, students with low averages of ability will have a higher self-concept when

surrounded by others with low averages of ability (Chanal et al., 2005; Marsh et al., 2008a;

Seaton, Marsh, Yeung, & Craven, 2011). Unfortunately, all students are susceptible to a BFLPE

regardless of ability level (Seaton et al., 2011), and are subject to having social comparisons of

ability impact their self-concept (Thijs, Verkuyten, & Helmond, 2010).

There is currently substantial support for a BFLPE within the academic domain and the

framework has demonstrated internal and external validity as well as cross-cultural

generalizability for various sub-domains of academic self-concept (Marsh et al., 2008a;

Nagengast & Marsh, 2012; Seaton, Marsh, & Craven, 2010). The BFLPE model is considered to

be domain-specific (Marsh et al., 2008a) and previous researchers have identified a stronger

BFLPE for some domains (e.g. verbal self-concept, mathematics self-concept) than others (e.g.

science self-concept) (Seaton et al., 2011). The domain specificity of the model is attributed to

the multidimensional structure of self-concept. Therefore, based on the multidimensionality of

self-concept and the BFLPE framework, it is likely that a BFLPE could be observed in the

physical domain. However, there is currently very limited support for a BFLPE for physical self-

concept in physical education. To date, research has been limited to examining gymnastics and

sport self-concepts, but these initial findings have provided initial evidence for a BFLPE in the

physical domain (Chanal et al., 2005; Trautwein et al., 2008). For example, students surrounded

by classmates with superior gymnastics skills have reported lower gymnastics self-concept

(Chanal et al., 2005). In addition, Marsh (1993) reported that students compared their personal

fitness in relation to classmates of the same gender and age when constructing their fitness self-

concepts. Specifically, social comparisons of running ability among students impacted their own

fitness self-perceptions in the physical education class (Marsh, 1993). Finally, Trautwein and

colleagues (2008) have provided support for frame of reference and a BFLPE in physical

education and noted that class average ability could impact both sport self-concept and physical

activity levels of students. These studies, although few, add additional support to the underlying

theoretical basis for a BFLPE in the physical domain (Marsh et al., 2008a).

Despite their substantial contributions, these studies do have limitations; primarily with

regard to measures of student ability in physical education. Individual student and class-average

ability play a critical role in the theoretical framework of the BFLPE (Marsh et al., 2008a), yet

research investigating frame of reference and a BFLPE in the physical domain has used fitness

test scores (Marsh, 1993), skill-related fitness assessments measuring balance, speed, and

coordination (Trautwein et al., 2008), and measures of student perceptions of ability (Margas,

Fontayne, & Brunel, 2006). To our knowledge, only one study (Chanal et al., 2005), has used

any measure of students‟ physical ability to execute a specific sport-related skill. In addition,

none of the studies outlined above have been entirely framed in Marsh‟s multidimensional

framework of physical self-concept or used domain-specific items from the Physical Self-

Description Questionnaire (PSDQ, Marsh, 1996) to evaluate physical self-concept. According to

Marsh and colleagues (2008a), in order to examine a BFLPE, it is essential to use ability

measures that are comparable between individuals, classes, and schools in addition to reliable

and valid measures of physical self-concept. Thus, there is a substantial need for more research

that utilizes physical ability assessments and multidimensional measures of physical self-concept

to investigate the BFLPE framework in physical education.

A BFLPE in Physical Education

Social comparison is the foundation of a BFLPE (Marsh & Craven, 2002; Marsh et al.,

2008a; Thijs et al., 2010) and the environment of physical education provides an optimal setting

for students to make external social comparisons of ability to form their own self-perceptions.

These social comparisons may provide an explanation for gender discrepancies in physical self-

concept scores among physical education students. For instance, physical education classes

frequently require students to perform sport-specific skills in front of one another and teachers

often use ability grouping in their classes. In addition, physical education is considered a

traditionally masculine subject where gender norms are reinforced (Bain, 2009; Connell, 2008;

Kirk, 2003; Penney & Evans, 2002). Students often participate in sports that are considered to be

traditionally male sports and if students conform to these traditional gender norms and males are

viewed as more skilled in sport and physical education than girls (regardless of actual ability),

girls will use this external comparison of ability to construct their physical self-concepts,

specifically their sport self-concept, which will likely be lower than that of the boys who are

viewed as more skilled. Thus, due to the sport-based nature of many physical education

programs and conformity to traditional stereotypes, differences in physical self-concept may be a

result of differences in the sport self-concepts of students. Conversely, girls enrolled in same-sex

physical education classes are not exposed to males as objects for external comparison. It is

possible that gender differences in sport self-concept among students enrolled in same-sex

classes might be diminished when compared to those of their coeducational counterparts.

Therefore, it is possible that the type of physical education class students are enrolled in (i.e.,

same-sex; coeducational) may moderate the BFLPE in physical education. Students‟ frame of

reference for sport self-concept and external social comparisons may be closely tied to the

physical education environment and may be influenced by the gender make-up of the class.

The purpose of this study was to investigate a BFLPE and the moderating effects of class

type among female students enrolled in same-sex and coed physical education classes. This study

was based on extensive research examining a BFLPE for academic self-concept (Marsh et al.,

2008a; Seaton, Marsh, & Craven, 2009; Seaton et al., 2011; Thijs et al., 2010) and previous

studies investigating a BFLPE for activity-specific physical self-concept (Chanal et al., 2005)

and sport self-concept (Trautwein et al., 2008). It was hypothesized that: a) individual ability

would positively predict sport self-concept; b) class-level ability would negatively predict sport

self-concept (also known as a BFLPE); and c) class structure (e.g. same sex, coeducational)

would moderate the BFLPE in physical education. Specifically, same-sex classes would reduce

the negative relationship between class ability and sport self-concept (i.e., BFLPE) when

compared to co-educational classes.

Methods

Quantitative Design

Through their inquiry in the academic domain, Marsh and colleagues (Marsh et al., 2008)

have concluded that a thorough investigation of a BFLPE requires: 1) a multilevel research

design, 2) an objective measure of individual ability, 3) ability measures that are comparable

across schools, 3) a valid measure of self-concept, and 4) tests of effects of school-average

ability on self-concept after controlling for effects of individual ability. Unfortunately, this

methodology poses several problems for investigations in a physical education setting.

As a general rule, multilevel analyses should have a minimum of 20 groups at the highest

level (Kreft & De Leew, 1998). Often, this is not feasible and previous research has been unable

to meet these requirements (Chanal et al., 2005). In addition, only 19 states in the country

currently require assessments in physical education and of those, only 14 require physical fitness

assessments. Likewise, there are currently no states that require skill testing (NASPE &

American Heart Association [AHA], 2010). Therefore, a standardized measure of student ability

for physical education students that would be comparable across schools is not readily available.

Due to the limited amount of existing research and potential methodological

considerations, this study was modeled after research conducted in the physical domain

examining the BFLPE (Chanal et al., 2005) and frame of reference effects (Trautwein et al.,

2008) as well as previous inquiry within the academic domain (Marsh et al., 2008a). Therefore,

a quantitative, cross-sectional research design was used to examine the relationships among

individual ability, class ability, class structure, and physical self-concept.

Participants and Setting

Participants for this study were female students (N=97) enrolled in required physical

education classes at two middle schools in the Southeastern United States. Data were also

collected from males in these classes and this data was used to calculate overall class means

(N=153). Students were enrolled in grade eight with a mean age of 13.19 years (SD=.44). The

students reported their ethnicity as African American (55.7%), Caucasian (24.3%), Multiracial

(12.9%), or Other (7.1%).

Schools were selected based on similarities between the physical education curricula.

Both schools used a traditional sport-based multi-activity curricular model and classes often

began with warm-up activities focused on health and skill related components of fitness. One

school consisted of same-sex PE classes (n=3) that were taught by certified female PE teachers

(n=3). Often, these three classes were team-taught and held simultaneously. The other school

consisted of coeducational classes (n=4) taught by a certified male PE teacher (n=1). Overall,

42.5% of students (n=65) were enrolled in same-sex classes and 57.5% (n=88) were enrolled in

coeducational classes.

Measures

PSDQ-S. The sport self-concept subscale of the Physical Self-Description Questionnaire

– Short (PSDQ-S) was used to measure self-concept in this study (Marsh, Martin, & Jackson,

2010). The sport self-concept subscale focuses on being athletic and good at sports and consists

of the following three items: (a) I am good at most sports; (b) I have good sports skills; and (c) I

play sports well. Each item was measured on a 6-point Likert scale ranging from 1 (not at all

true) to 6 (very true). Sport self-concept represents a domain specific facet of physical self-

concept that is closely related to skill tests such as the standing broad jump and Underkoffler

softball throw for distance. Items from the PSDQ-S have demonstrated good reliability, test-

retest stability over both short and long term, a well-defined factor structure that is invariant over

gender, and both convergent and discriminant validity (Marsh et al., 2010) (See Appendix B).

Standing broad jump. The standing broad jump is a physical ability assessment

designed to measure lower body muscular power and strength. This test has been validated for

children as young as six through college-age and reliability coefficients ranging from .83-.99

have been reported (Miller, 2010; Reiman & Manske, 2009; Ruiz et al., 2011). In addition, the

standing broad jump has demonstrated strong correlations with the 1RM squat (r=.805) (Miller,

2010) and multiple lower body (r= .83-.86) and upper body (r=.69-.85) muscular strength tests

(Castro-Pinero, Artero, & Espana-Romero, 2010). In this assessment, participants are allowed

approximately five minutes of warm-up attempts before completing three trials. Participants are

directed to jump as far forward as possible along a tape measure that is fixed to the floor. The

distance of each jump from the baseline to the heel of the closest foot to the start line is recorded

to the nearest quarter-inch. The criterion score is the average of three trials (Miller, 2010). Norms

for the standing broad jump are reported in Table 3.1.

Underkoffler softball throw for distance. The Underkoffler softball throw for distance

is intended to measure softball ability and upper body muscular power and strength. This test has

demonstrated good reliability (r=.95) and validity (r=.63-.81) among adolescent females (Collins

& Hodges, 1978; Reiman & Manske, 2009). Previous research has verified that the Underkoffler

Softball Throw is correlated with the modified pull up (r=.71) and push up (r=.63) tests of upper

body strength in adults (Negrete, Hanney, Kolber, Davies, & Riemann, 2011). In this assessment,

participants are allowed to complete four sub-maximal-maximal warm-up throws (25%-100%

effort) before attempting three throws for maximum distance. Each participant then completes

three trials consisting of three maximal throws. The criterion score is the distance of the farthest

of the nine throws, measured to the nearest foot from the point where the ball first lands (Collins

& Hodges, 1978; Negrete et al., 2011; Reiman & Manske, 2009). Norms for the Underkoffler

softball throw for distance are not currently available.

Table 3.1. Normative scores on the standing broad jump for females aged 11-14 (Miller, 2010).

Age

Percentile

72”

74”

77”

80”

64”

66”

69”

71”

59”

60”

63”

64”

52”

54”

57”

58”

44”

46”

48”

Procedures

Permission to conduct this study was obtained by the University Institutional Review

Board prior to data collection. In addition, students and parents were provided a full description

of study procedures and child assent/parent consent was collected prior to participation.

Approximately 169 students were recruited to take part in the study and students who did not

provide assent/consent forms were not allowed to participate (10.46 %, n=16). All students were

assigned identification numbers to ensure confidentiality.

Data Collection

Data collection took place over three class periods. Participants first completed

questionnaires containing demographic questions and the PSDQ-S (Marsh et al., 2010). Surveys

took approximately 30 minutes to complete. Second, students completed two physical ability

assessments; the standing broad jump and Underkoffler softball throw for distance. The physical

ability assessments were administered by the primary researcher and the physical education

teachers for each class. Each test was completed in one class period and took approximately 50

minutes to complete.

Data Analysis

Data were initially screened for outliers. Reliability coefficients, descriptive statistics,

and simple correlations were then calculated using SPSS statistical software. Hierarchical

regression analysis was used to test Hypotheses 1 and 2. Specifically, girls‟ individual standing

broad jump and softball throw scores were standardized and entered into the first block

representing individual physical ability predictors. The standing broad jump and softball throw

means for each class were calculated and assigned to individual students within their respective

class. Class mean scores for standing broad jump performance and the softball throw were then

standardized and entered in the second block representing class ability predictors. The outcome

measure was sport self-concept.

A second hierarchical multiple linear regression was conducted to investigate Hypothesis

3 (i.e. class structure will moderate the BFLPE in physical education) (Aguinis, 2004).

Standardized scores were used again for class ability variables, softball throw (CST) and class

standing broad jump (CLJ), and a dichotomous variable was created for the categorical variable

class type (CT) (e.g. coed=1, same sex=0) (Aiken & West, 1991; Aguinis 2004). CST, CLJ, and

CT were first entered as predictors in Block 1, with sport self-concept as the outcome variable.

Two product terms (CST X CT; CLJ X CT) were created and entered into Block 2 (Aiken &

West, 1991; Aguinis, 2004). Significant interaction terms were used to determine if CT

moderated the relationships between the CST, CLJ, and sport self-concept.

Results

Descriptive Statistics and Simple Correlations

The Cronbach‟s reliability coefficient for sport self-concept was .94, representing

excellent inter-item correlation. Means, standard deviations, and simple correlations for all

variables are reported in Table 3.2.

Table 3.2. Descriptive Statistics and Simple Correlations Among Variables.

Variable

SSC

3.98

1.53

0.07

48.62

9.94

0.29**

0.16

682.82

232.79

CLJ

-0.10

0.07

0.03

52.22

4.17

CST

-0.03

0.02

0.25*

0.71**

861.39

85.70

Note. SSC = sport self-concept; LJ = standing long jump; ST = Underkoffler Softball Throw;

CLJ = class-level long jump mean including same sex and coed classes; CST = class-level

Underkoffler Softball Throw mean including same sex and coed classes. * p< .05; ** p<

.01.

The mean score for sport self-concept was above the mid-point of the scale. The mean scores for

the standing broad jump and the class standing broad jump were equivalent to the 5

percentile

for thirteen year-old adolescent females (See Table 3.1). The broad jump and softball throw

mean scores for each of the seven classes are presented in Table 3.3. Classes 1 – 4 were coed and

classes 5 – 7 were all girls (i.e., same-sex). Relationships among variables were initially

investigated by testing bivariate correlations. There was a moderate-to-strong positive correlation

between CST and CLJ. There were small-to-moderate positive correlations between sport self-

concept and individual Underkoffler softball throw and the individual Underkoffler softball

throw and class Underkoffler softball throw.

Table 3.3.Class Average Scores of Same Sex and Coed Classes.

BFLPE Hierarchical Regression Analysis

Results from the BFLPE hierarchical regression analysis are presented in Table 3.4.

Results provided partial support for Hypothesis 1 and 2. Examination of Block 1 revealed that

the Underkoffler softball throw for distance was the only individual ability predictor of sport

self-concept (β= .25). Individual standing broad jump scores did not appear to be related to sport

Class

CLJ M

CST M

Type

Girls

Boys

Total

Class 1

53.88

831.48

Coed

Class 2

53.85

886.45

Coed

Class 3

59.28

966.00

Coed

Class 4

59.01

1052.32

Coed

Class 5

49.72

926.13

Class 6

46.13

746.78

Class 7

53.28

824.31

Note. CLJ = class long jump mean; CST = class Underkoffler Softball Throw;

Coed = coed classes; SS = same sex classes; Girls n= 97; Boys n= 56; total N= 153.

self-concept (β= .03). Examination of Block 2 revealed partial support for a BFLPE.

Specifically, the CST had a negative relationship (β= -.34) with reports of sport self-concept,

suggesting a BFLPE. Although CLJ only bordered on significance (β= -.29, p= .10) as a

predictor of sport self-concept, it is possible that this effect would be more prominent with a

larger sample size. Overall, individual measures of ability accounted for 6% of the variance in

sport self-concept and class measures of ability accounted for 5% of the variance in sport self-

concept.

Table 3.4. Hierarchical Regression Results for Big Fish Little Pond Effect.

Moderation Analysis

The overall regression model evaluating the moderation effects of CT on the relationship

between CST, CLJ, and sport self-concept was not significant (F=.94, p=.45). Furthermore, there

were no significant interaction effects (F=.72, p=.61). Thus, results indicated that CT did not

moderate the relationship between CST, CLJ and sport self-concept in this study.

Discussion

Sport Self Concept

Predictor

R²

Block 1

.06

.03

.16

0.21

.83

.25

.17

2.21

.03

Block 2

.05

CLJ

-.29

.25

-1.67

.10

CST

-.34

.26

-2.04

.04

Quality physical education is designed to educate the whole child, which includes

positively influencing aspects of the affective domain such as self-esteem (NASPE, 2011).

However, according to the BFLPE framework, the structure of the physical education

environment can also negatively influence physical self-concept by facilitating social

comparisons of ability among students (Marsh & Craven, 2002; Marsh et al., 2008a, 2008b;

Thijs et al., 2010). Therefore, this study was designed to investigate a BFLPE for sport self-

concept among adolescent females in physical education. It was hypothesized that: (a) individual

ability would demonstrate a significant positive relationship with sport self-concept; (b) overall

class ability would demonstrate a significant negative relationship with sport self-concept (also

known as a BFLPE); and (c) class structure (e.g. same sex, coed) would moderate the BFLPE in

physical education. It was theorized that same-sex classes would reduce a BFLPE when

compared to coed classes because female students would be less likely to use males as targets for

social comparisons of ability.

A primary theoretical prediction of the BFLPE is that domain-specific student ability will

predict domain-specific self-concept (Marsh et al., 2008a) and previous research has provided

evidence for this effect in the physical domain (β= .40 - .59) (Chanal et al., 2005; Trautwein et

al., 2008). Results from this study are similar and indicate that students‟ performance on the

Underkoffler softball throw for distance positively predicted sport self-concept (β=.25).

However, individual student performance on the standing broad jump did not predict sport self-

concept (β=.03), therefore there was only partial support for Hypothesis 1. It is likely that the

Underkoffler softball throw for distance was a significant predictor while the standing broad

jump was not due to the nature of the skills tests. The softball throw for distance is directly

transferable to a specific sport and measures a very sport-specific skill, and is therefore closely

related to sport self-concept. On the other hand, the standing broad jump assesses a skill-related

component of fitness that is applicable not only to sport, but to the strength, endurance, and

coordination components of physical self-concept. Therefore, future research should examine the

potential for the standing broad jump to positively predict additional components of physical

self-concept.

Results of this study also provide partial support for Hypothesis 2, which stated that class

ability would negatively predict sport self-concept. The class-level performance on the softball

throw did negatively predict sport self-concept (β= -.34), providing some evidence of a BFLPE

in the physical education classes included in this study. These results are comparable to previous

research in the physical domain that has identified similar associations (β= -.39 - -.68) between

physical ability and physical self-concept in gymnastics and fitness-based contexts (Chanal et al.,

2005; Trautwein et al., 2008). Although the class-level performance on the standing broad jump

did not negatively predict sport self-concept, it bordered on significance (β= -.29). It is possible

that class performance on the standing broad jump may negatively influence other components

of physical self-concept (e.g. strength, endurance, coordination), therefore, future research

should include additional components of physical self-concept in order to have a better overall

picture of how class-level ability may influence physical self-concept. Considering the

multidimensional framework of physical self-concept, these results are not necessarily

surprising. For example, performance on a sport-specific skill, overall fitness, and skill-related

components of fitness may contribute to a student‟s overall perception of his or her ability, much

in the same way that perceptions of sporting ability, health, strength, endurance, and flexibility

all contribute to an individual‟s overall physical self-concept. Therefore, future research should

examine how multiple physical ability assessments may be related to the various components of

physical self-concept.

Although results only partially support Hypothesis 1 and 2, these findings do make a

significant contribution to existing literature. These results provide additional support to the

emerging evidence for a BFLPE in physical education. In addition, a considerable strength of

this study is that it uses a sport-specific measure of upper body physical ability and a valid and

reliable measure of lower body physical ability in combination with domain specific physical

self-concept measures to investigate a BFLPE. These findings provide additional support for a

BFLPE in physical education and contribute to the robustness of the BFLPE framework for self-

concept development (Marsh et al., 2008b; Nagengast & Marsh, 2012). Critics of the BFLPE

framework emphasize the importance of examining contextual factors that contribute to the

BFLPE (Dai & Rinn, 2008) yet identifying specific individual or situational moderators of the

BFLPE has been somewhat difficult (Marsh et al., 2008a; Seaton et al., 2010; Seaton et al.,

2011). For example, a preference for a cooperative learning environment has been identified to

enlarge a BFLPE in the academic domain, yet a competitive orientation has not been identified a

statistically significant moderator (Seaton et al., 2010). Follow-up research has been unable to

replicate findings that a BFLPE is enhanced based on cooperative orientations (Seaton et al.,

2011). Similar results have been reported for individual ability, which has been suggested to

influence the size and direction of a BFLPE (Seaton et al., 2010; Seaton et al., 2011). Further

investigation has indicated that individual ability does not moderate the BFLPE (Seaton et al.,

2011). In addition, several motivational constructs (e.g. intrinsic motivation, self-efficacy),

which were originally hypothesized to be moderators of a BFLPE, have consistently

demonstrated no moderation effects (Seaton et al., 2010; Seaton et al., 2011). Seaton and

colleagues have examined 67 different potential moderators and reported that none of the

constructs examined moderate a BFLPE (Seaton et al., 2011). As a result, researchers have

emphasized the importance of examining environmental variables such as forced social

comparisons and frame of reference in the BFLPE (Seaton et al., 2008; Seaton et al., 2010;

Seaton et al., 2011). Therefore, a second major contribution of this research was the investigation

of class type as a potential moderator of a BFPLE for adolescent females in physical education.

Contrary to Hypothesis 3, results did not indicate that class structure (e.g. same sex, coed)

moderated the BFLPE in physical education. It was hypothesized that female students in coed

physical education classes would experience forced social comparisons with male classmates,

who may be perceived as more skilled and that these social comparisons would negatively

impact sport self-concept. Females enrolled in same-sex classes are not subjected to this external

social comparison of ability and it was thought that controlling for class type would diminish a

BFLPE for these students, yet results do not support this hypothesis. Previous research

investigating individual-level comparisons in the academic domain has suggested that students

tend to choose classmates with higher ability as individual-level comparison choices and that

these choices sometimes have positive effects on achievement, which could in turn positively

influence self-concept (Blanton, Buunk, Gibbons, & Kuyper 1999; Huguet et al., 2009; Seaton et

al., 2008). However, higher school-average and class-average ability results in a BFLPE and

consistently impacts individual students‟ self-concept negatively (Huguet et al., 2009; Marsh et

al., 2008b; Nagengast & Marsh, 2012;Seaton et al., 2011). Therefore, individual, class-level, and

school-level comparisons of ability can exist simultaneously within a BFLPE (Huguet et al.,

2009; Marsh et al., 2008b; Seaton et al., 2008). It is possible that class type did not moderate a

BFLPE due to the sample size and low power of the statistical analysis. However, it is also

possible that the girls in these coed classes did not choose male classmates as reference points for

ability comparisons. These results are similar to those of previous work examining social

comparison effects on fitness test scores, which indicated that students compare themselves to

others of the same gender (Marsh, 1993). On the other hand, if the girls did chose male students

as targets for comparison, the effects may not have been strictly negative. In other words,

outperforming male targets might have enhanced girls‟ physical self-concept. These results

contribute to previous research suggesting that individual and generalized social comparisons of

ability can exist simultaneously (Huguet et al., 2009; Marsh et al., 2008b).

Results of the current study do not provide evidence that adolescent girls in coed classes

are at an increased risk of experiencing a BFLPE. Instead, it is likely that a BFLPE may be a

result of one‟s individual standing in her class or school (Huguet et al., 2009; Marsh et al.,

2008b; Nagenast & Marsh, 2012; Seaton et al., 2011) and girls in same-sex classes are still at

risk for making social comparisons that result in a BFLPE. Although students are using class-

level comparisons, the class type may not play a significant role as previously hypothesized

because more generalized comparisons are made (Marsh et al., 2008b). Students may take the

overall ability of their classes into consideration over that of individual classmates, making

gender less relevant.

Conversely, generalized comparisons may coexist with individual comparisons (Huguet

et al., 2009; Marsh et al., 2008b). For example, female students in coed classes may not choose

to compare themselves to male students and instead target classmates of the same gender.

Second, students in both same-sex and coed classes may opt to compare themselves to

classmates with higher ability, despite gender. Regardless of comparison-level choice, it is

important for physical educators to be aware that there is potential for a BFLPE in physical

education classes and that all students have an equal chance of having their physical self-concept

negatively influenced, regardless of ability level (Seaton et al., 2011). Future research

investigating a BFLPE in physical education should focus on the comparison process and

comparison-level choices (Huguet et al., 2009; Seaton et al., 2010; Marsh et al., 2008b).

Results of this study also highlight the role social comparison within a BFLPE plays in

physical self-concept development among adolescent females (Huguet et al., 2009; Marsh et al.,

2008b; Seaton et al., 2010). Specifically, these results suggest that generalized social comparison

effects , which are implied within the BFLPE framework, can negatively impact physical self-

concept among girls in physical education, regardless of class type. The physical education

environment is unlike many traditional school subjects and often emphasizes competition,

requires students to demonstrate their skills in front of others, and promotes social comparisons

of physical ability. In the case of a BFLPE where social comparison effects are present, students‟

sport self-concept may suffer. BFLPE researchers have suggested that teachers in the academic

domain should deemphasize competition and focus instead on personal improvement to

counteract possible social comparison effects in the classroom (Liem, Marsh, Martin,

McInerney, & Yeung, 2013). Although competition is often perceived negatively by students in

physical education (Carlson, 1995; Ntoumanis, Pensgard, Martin, & Pipe, 2004), some students

enjoy competition and it is unlikely that the elimination of competitive activities would be

successful. Instead, teachers should structure competitive activities in ways that address a broad

range of ability levels and allow all students to be successful. In addition, teachers need to

consider how students are grouped by ability and avoid creating scenarios where low-skilled

students are constantly competing against high-skilled classmates (Rink, 2013).

In physical education, students are often asked to display their bodies and physical

abilities in front of other students. Students are frequently grouped by ability and social

comparisons are sometimes encouraged. Often, the best “athletes” in class are pointed out for

their achievements and rewarded based on innate ability instead of skill mastery, improvement,

or fair play. Teachers should keep in mind that all students are subject to negative effects from a

BFLPE (Seaton et al., 2011) and create a mastery climate that focuses on personal improvement

instead of innate ability (Solmon, 2003; Rink, 2013). Physical educators can use this information

to create a more positive experience for all students. During adolescence, student enjoyment in

physical education begins to decline (Subramaniam & Silverman, 2007), which is sometimes a

result of feeling unsuccessful in physical education (Carlson, 1995; Ntoumanis et al., 2004). If

students are placed in situations that facilitate social comparison and feelings of incompetence,

their physical self-concepts will be negatively impacted. Thus, teachers should discourage

student engagement in social comparison (Solmon, 2003) to avoid potential negative

consequences of a BFLPE.

In conclusion, this research highlights a BFLPE for students in physical education and

indicates that girls are susceptible to a BFLPE for sport self-concept regardless of class type.

Teachers should keep in mind that all students are susceptible to a BFLPE and avoid

emphasizing social comparison of ability in their classes. Instead, a focus on personal

improvement and individual success can potentially reduce negative effects of social comparison

on students‟ physical self-concept. Despite the contributions of this study, future research should

continue to investigate a BFLPE and potential moderators of a BFLPE related to individual

students, teacher characteristics, and contextual variables in the physical domain. Identifying

possible methods to diminish a BFLPE can lead to interventions that may buffer against negative

influences of social comparisons of ability in physical education (Marsh et al., 2008b; Marsh et

al., 2013).

Researchers should keep in mind that social comparison may not always have negative

effects on achievement (Blanton et al., 1999; Huguet et al., 2009; Seaton et al., 2008) but may

still negatively influence self-concept (Seaton et al., 2008) and continue to investigate whether

class or student-level comparison choices play a greater role in the BFLPE framework (Blanton

et al., 2009; Seaton et al., 2010). This study is not without limitations. The cross-sectional design

and sample size limit the generalizability of these results. Likewise, there may be differences

between class type which were not detected in this study due to low statistical power. This study

also only examined one subdomain of physical self-concept. Future BFLPE research in physical

education should include both male and female students in diverse settings and multiple physical

education environments. In addition, future research would benefit from examining additional

components of physical self-concept and how multiple measures of ability may impact each

domain-specific area of physical self-concept. Finally, research investigating multiple schools

and classes will be better suited to meet the methodological suggestions of multi-level modeling

to investigate a BFLPE (Marsh et al., 2008a).

CHAPTER FOUR: SUMMARY AND CONCLUSIONS

Quality physical education benefits students in several ways and can influence their

emotional development, specifically their perceptions of competence, self-esteem, and self-

concept (Bailey, 2009). Unfortunately, girls often become less engaged and involved in physical

education as they grow older (Subramaniam & Silverman, 2007), therefore they may not be

receiving the same positive emotional, physical, or cognitive benefits as their classmates.

Likewise, girls consistently report lower physical self-concepts than males in physical education

(Caglar, 2009; Schmalz & Davison, 2006). Taken together, these two trends are quite disturbing

and put young girls at risk for detrimental effects of low physical self-concept such as body

dissatisfaction, depression, or disordered eating (O‟Dea & Abraham, 1999; Rosenblum & Lewis,

1999) in addition to negative consequences associated with physical inactivity like increased risk

for obesity, heart disease, and diabetes (Centers for Disease Control and prevention [CDC],

2011). A positive physical self-concept is associated with participation in physical education and

physical activity (Beasley & Garn, in press, Cumming et al., 2011), yet to date there has been

only speculation as to why females report lower physical self-concepts than males in physical

education and many suggest that traditional gender stereotypes are to blame (Caglar, 2009;

Klomsten, Marsh, & Skaalvik, 2005; Schmalz & Davison, 2006). Therefore, the purpose of this

dissertation was to investigate potential explanations for gender discrepancies in physical self-

concept scores among physical education students. Specifically, the roles frame of reference,

social comparison, and the physical education environment play in physical self-concept

development among adolescent females was examined through quantitative and qualitative

research designs.

The results of Chapter Two provide evidence that some girls in same-sex physical

education classes perceive the coed environment as one where they experience pressures

regarding their physical ability and appearance and where there is a perceived risk for emotional

and physical harm. In addition, the role of competition in physical education received significant

attention. Participants conformed to traditional gender norms and viewed males as the more

aggressive and physically dominant gender. However, competition itself was not considered to

be unpleasant and was desirable in appropriate levels and contexts. In addition, results supported

that female physical education students may use multiple sources of information in the process of

physical self-concept development. All of the participants perceived same-sex physical education

as a setting where they were free from the negative experiences associated with coed classes.

Results of Chapter Three provide evidence that female students in both same-sex and

coed physical education can experience a BFLPE and use social comparisons of physical ability

to construct their physical self-concepts. Class type did not increase or decrease the BFLPE for

these students, indicating that the girls in coed classes were not at an increased risk for

experiencing negative consequences to their physical self-concept as a result of a BFLPE.

Instead, results suggest these girls in physical education were susceptible to a BFLPE, regardless

of class type. This study also highlights the importance of considering generalized and

individual-level comparison targets as choices for social comparison for girls in physical

education.

The overarching purpose of these two studies was to provide insight into possible

explanations for gender discrepancies in physical self-concept scores among students in physical

education. The context of physical education provided a rich setting for investigating gender,

physical self-concept, and the physical education environment. Since the implementation of Title

IX, there has been continual debate regarding whether or not same-sex physical education is the

answer to current issues that girls face in physical education, such as participation opportunities,

interaction and engagement with the teacher, and overall enjoyment (Derry & Phillips, 2004;

Hannon & Ratcliffe, 2007; Lirgg, 2006; Hills & Croston, 2012). Despite significant attention

directed toward this issue, research has yet to draw any definitive conclusions regarding which

type of class structure is considered to be more beneficial (Lirgg, 2006). Results of these two

studies were no different, yet they do offer a unique perspective and highlight the importance of

physical self-concept in physical education.

First, results provide evidence that young girls‟ physical self-concept is constructed in

relation to the surrounding environment and that there is support for both the BFLPE and the I/E

frame-of-reference theoretical frameworks for physical self-concept development in a physical

education setting. Specifically, participants in the qualitative study outlined in Chapter Two

identified four sources of information (direct observations of achievement, teachers‟ responses

and comments, responses and comments from classmates, and responses and comments from

others outside the class) for physical self-concept development in physical education. Not only

do these findings support previous research (Skaalvik & Skaalvik, 2002), but they also indicate

that social comparison effects can play a significant role in physical self-concept development.

Evidence for social comparison effects was also supported by the results of the quantitative

study, which identified a BFLPE for girls in both coed and same-sex classes. BFLPE researchers

have suggested deeper investigation of internal and external comparison-level choices (Huguet et

al., 2009; Marsh et al., 2008; Seaton, Marsh, & Craven, 2009), which was addressed in Chapter

Two. Participants not only used internal comparisons of ability and achievement, but they also

identified external sources for comparison such as friends, family, classmates, and other girls in

general. Previous research has suggested that individual and generalized social comparison

effects can exist simultaneously when a BFLPE is present (Huguet et al., 2009; Marsh et al.,

2008; Seaton, Marsh, Yeung, & Craven, 2011) and results of these two studies support this. In

other words, students may choose particular students to use as comparison targets while at the

same time taking the overall ability of their class, grade level, or school into consideration.

Physical educators must consider that female students in their classes are forming their

physical self-concepts based on both individual and generalized comparisons. Girls experience

significant external pressure to conform to feminine body ideals (Azzarito & Solmon, 2009),

which often creates conflict with the activities and tasks that they are asked to perform in

physical education. If teachers reinforce the dominant masculinity that is often present in

physical education (Kirk, 2002, 2003), girls‟ perceived pressure to conform to traditional gender

expectations may in turn be transferred to the physical education class. For example, when

teachers only offer activities that girls feel are gender inappropriate or that put their bodies on

display in conventionally masculine ways, some female students may evade participating.

Likewise, if students are uncomfortable in required uniforms, they may refuse to dress out or

even opt out of physical education altogether. Teachers should consider offering a variety of

activities that students feel are appropriate for both genders and reevaluate providing traditional

sports experiences like basketball, volleyball, or flag football. In addition, teachers may want to

reconsider evaluating student participation based solely on wearing required uniforms and

instead use more authentic forms of assessment that evaluate task mastery and student effort

(Solmon, 2003; Wood, 2003). Finally, it is imperative that teachers deemphasize social

comparison in their classes. Students not only compare with generalized others, but they also use

individual social comparisons as a basis for their own physical self-concept. Emphasizing

personal improvement instead of innate athletic ability will aid in counteracting students‟ ability

comparisons with peers (Moreno-Murcia, Hernandez, Vaillo, & Camacho, 2012). Second,

consistent with previous findings (Azzarito & Solmon, 2006, 2009; Larsson, Fagrell, & Redelius,

2009; McKenzie, Prochaska, Sallis, & Lamaster, 2004), participants in the first study indicated

that in coed physical education they were vulnerable to perceived pressures to conform to a

gendered body, felt susceptible to the male gaze, and were exposed to negative consequences

such as name-calling when participating in activities. Overall, there was the feeling that coed

physical education put girls at risk for emotional or physical harm and that same-sex classes

were a preferable alternative. The internalization of traditional gender ideals and an increased

awareness of gender expectations influenced girls‟ participation, engagement, and enjoyment in

coed physical education.

Interestingly, despite such negative perceptions of coed physical education, class type did

not influence the BFLPE identified in either same-sex or coed classes. Although these findings

were consistent with BFLPE research (Seaton, Marsh, & Craven, 2010; Seaton et al., 2011), they

do provide a unique perspective in the ongoing debate of coed vs. same-sex physical education.

Social comparison effects were evident in both class types, indicating that the atmosphere of the

physical education was more important than the gender make-up of the class. From discussions

with the girls in the first study, it was clear that their previous experiences in coed physical

education often revolved around participation in traditionally gender-stereotyped activities that

were dominated by the high-skilled males in the class. Many have argued that there is no right or

wrong answer regarding whether students benefit more from coed or same-sex physical

education and that teachers, the environment, and pedagogy are more important than class type

(Kirk, 2003; Lirgg, 2006). Likewise, recent research has suggested that teachers can influence

the gendered culture of physical education by attempting to provide gender equity in their classes

(Azzarito & Katzew, 2010; Azzarito & Solmon, 2006). Therefore, by striving to demonstrate an

awareness of gender and working to provide equitable experiences for all students, teachers can

create more positive experiences for girls in coed classes.

Third, results of the first study highlight how competition can greatly impact student

perceptions of the physical education atmosphere. An acceptable level of competition, concern

for safety, and enjoyment was emphasized. Although participants did ascribe to traditional

gender beliefs and expressed feelings that males were aggressive and rough, they also felt that in

an appropriate setting competition was desirable and enjoyable. It is essential that practitioners

keep in mind that like class type, competition is not necessarily inappropriate for physical

education. Instead, teachers can structure activities so that all students can be successful

regardless of gender, skill level, or athletic ability. Teachers should also carefully consider ability

and skill level when grouping students and avoid placing students in scenarios where they will be

dominated by higher-skilled competitors (Rink, 2013). By doing so, physical educators may also

avoid potential negative consequences associated with ability comparisons and a BFLPE in

physical education.

In conclusion, results of these two studies fill an existing gap in the research examining

the associations between physical self-concept, frame of reference, social comparison, the

physical education environment, and gender in physical education. Future studies should

continue this valuable line of research by investigating the multiple sources of information

students in physical education use to construct their physical self-concepts. Identifying sources

of information and comparison-level choices in the physical domain will contribute to existing

BFLPE research and hopefully lead to intervention strategies to reduce the BFLPE in physical

education classes. Future research should also continue to examine female students‟ perceptions

of the physical education environment and their experiences in physical education. Continued

scholarship in this area will provide additional explanations for gender discrepancies in physical

self-concept scores among students in physical education and aid in identifying methods to

increase girls‟ engagement, enjoyment, and participation in physical education.

REFERENCES

Aguinis, H. (2004). Regression analysis for categorical moderators. New York, NY: The

Guilford Press.

Aiken, L.S., & West, S.G. (1991). Multiple regression: Testing and interpreting interactions.

Newbury Park, CA: Sage Publications.

Asci, F.H. (2002). An investigation of age and gender differences in physical self-concept among

Turkish late adolescents. Adolescence, 37, 365-371.

Azzarito, L. (2010). Future girls, transcendent femininities and new pedagogies: Toward girls‟

hybrid bodies? Sport, Education, and Society, 15, 261-275.

Azzarito, L., & Katzew, A. (2010). Performing identities in physical education (En)gendering

fluid selves. Research Quarterly for Exercise and Sport, 81, 25-37.

Azzarito, L., & Solmon, M.A. (2006). A feminist poststructuralist view on student bodies in

physical education: Sites of compliance, resistance, and transformation. Journal of

Teaching in Physical Education, 25, 200-225.

Azzarito, L., & Solmon, M.A. (2009). An investigation of students‟ embodied discourses in

physical education: A gender project. Journal of Teaching in Physical Education, 28,

173-191.

Bailey, R. (2009). Physical education and sport in schools: A review of benefits and outcomes.

In R. Bailey & D. Kirk (Eds.), The Routledge physical education reader (pp.29-37).New

York, NY: Routledge.

Bain, L.L. (2009). The hidden curriculum re-examined. In R. Bailey & D. Kirk (Eds.), The

Routledge physical education reader (pp.39-47).New York, NY: Routledge.

Beasley, E.K. & Garn, A.C. (in press). An investigation of adolescent girls‟ global self-

concept, physical self-concept, and self-determined motivation in physical education.

Journal of Teaching in Physical Education, accepted for publication May 29, 2013.

Blanton, H., Bluunk, B.P., Gibbons, F.X., & Kuyper, H. (1999). When better-than-others

compare upward: Choice of comparison and comparative evaluations as independent

predictors of academic performance. Journal of Personality and Social Psychology, 76,

420-430.

Brown, D., & Rich, E. (2002). Gender positioning as pedagogical practice in teaching physical

education. In D. Penney (Ed.), Gender and physical education (pp. 13-23). London:

Routledge.

Caglar, E. (2009). Similarities and differences in physical self-concept of males and females

during late adolescence and early adulthood. Adolescence, 44, 407-419.

Carlson, T.B. (1995). We hate gym: Student alienation from physical education. Journal of

Teaching in Physical Education, 14, 467-477.

Castro-Pinero, J., Artero, E.G., & Espana-Romero, V. (2010). Criterion-related validity of field-

based fitness tests in youth: A systematic review. British Journal of Sports Medicine, 44,

934-943.

Centers for Disease Control and Prevention. (2011). Facts about physical activity. Retrieved

from: http://www.cdc.gov/physicalactivity/data/facts.html.

Chalabaev, A., Sarrazin, P., Trouilloud, D., & Jussim, L. (2009). Can sex-undifferentiated

teacher expectations mask an influence of sex stereotypes? Alternative forms of sex bias

in teacher expectations. Journal of Applied Social Psychology, 39, 2469-2498.

Chanal, J.P., Marsh, H.W., Sarrazin, P.G., & Bois, J.E. (2005). Big-fish-little-pond effects on

gymnastics self-concept: Social comparison processes in a physical setting. Journal of

Sport & Exercise Psychology, 27, 53-70.

Chepyator-Thomson, J.R., & Ennis, C.D. (1997). Reproduction and resistance to the culture of

femininity and masculinity in secondary school physical education. Research Quarterly

for Exercise and Sport, 68, 89-99.

Clarke, G. (2002). Difference matters: Sexuality and physical education. In D. Penney (Ed.),

Gender and physical education (pp. 13-23). London: Routledge.

Cockburn, C., & Clarke, G. (2002). “Everybody‟s looking at you!”: Girls negotiating the

“femininity deficit” they incur in physical education. Women’s Studies International

Forum, 25, 651-665.

Collins, D.R., & Hodges, P.B. (1978). A comprehensive guide to sports skills tests and

measurement. Springfield, IL: Charles C. Thomas.

Connell, R. (2008). Masculinity construction and sports in boys‟ education: A framework for

thinking about the issue. Sport, Education and Society, 13, 131-145.

Craven, R.G., & Marsh, H.W. (2008). The centrality of the self-concept construct for

psychological wellbeing and unlocking human potential: Implications for child and

educational psychologists. Educational & Child Psychology, 20, 104-118.

Crocker, P.R.E., Eklund, R.C., & Kowalski, K.C. (2000).Children‟s physical activity and

physical self-perceptions. Journal of Sports Sciences, 18, 383-394.

Crocker, P.R.E., Sabiston, C.M., Forrestor, S., Kowalski, N., Kowalski, K.C., & McDonough,

M. (2003). Predicting change in physical activity, dietary restraint, and physique anxiety

in adolescent girls. Canadian Journal of Public Health, 94, 332-337.

Crocker P.R.E., Sabiston, C.M., Kowalski, K.C., McDonough, M.H., & Kowalski, N. (2006).

Longitudinal assessment of the relationship between physical self-concept and health-

related behavior and emotion in adolescent girls. Journal of Applied Sport Psychology,

18, 185-200.

Cumming, S.P., Sherar, L.B., Hunter-Smart, J.E., Rodrigues, A.M.M., Standage, M., Gillison,

F.B., & Malina, R.M. (2012). Physical activity, physical self-concept, and health-related

quality of life of extreme early and late maturing adolescent girls. The Journal of Early

Adolescence, 32, 269-292.

Cumming, S.P., Standage, M., Loney, T., Gammon, C., Neville, H., Sherar, L.B., & Malina,

R.M. (2011). The mediating role of physical self-concept on relations between biological

maturity status and physical activity in adolescent females. Journal of Adolescence, 34,

465-473.

Dai, D.Y., & Rinn, A.N. (2008). The big-fish-little-pond effect: What do we know and where do

we go from here? Educational Psychology Review, 20, 283-317.

Daley, A.J., & Buchanan, J. (1999). Aerobic dance and physical self-perceptions in female

adolescents: Some implications for physical education. Research Quarterly for Exercise

and Sport, 70, 196-200.

Derry, J.A., & Phillips, D.A. (2004). Comparisons of selected student and teacher variables in

all-girls and coeducational physical education environments. Physical Educator, 61, 23-

34.

Dunton, G.F., Jamner, M.S., & Cooper, D.M. (2003). Physical self-concept in adolescent girls:

Behavioral and physiological correlates. Research Quarterly for Exercise and Sport, 74,

360-365.

Flintoff, A., & Scranton, S. (2005). Gender and physical education. In K. Green & K. Hardman

(Eds.). Physical education: Essential issues (pp.161-175). London: SAGE Publications.

Gheris, J., Kress, J., & Swalm, R. (2010). Students‟ views on physical development and physical

self-concept in adventure-physical education. Journal of Teaching in Physical Education,

29, 146-166.

Gorely, T., Holroyd, R., & Kirk, D. (2003). Muscularity, the habitus, and the social construction

of gender: Towards a gender-relevant physical education. British Journal of Sociology of

Education, 24, 429-448.

Guerin, F.G., Marsh, H.W., & Famose, J.P. (2004). Generalizability of the PSDQ and its

relationship to physical fitness: The European French connection. Journal of Sport &

Exercise Psychology, 26, 19-38.

Hagger, M.S., Biddle, S.J.H., & Wang, C.K. (2005). Physical self-concept in adolescence:

Generalizability of a multidimensional, hierarchical model across gender and grade.

Educational and Psychological Measurement, 65, 297-232.

Hagger, M.S., Stevenson, A., Chatzisarantis, N.L.D., Gaspar, P.M.P., Ferreira, J.P.L., & Rave,

J.M.G. (2010). Physical self-concept and social physique anxiety: Invariance across

culture, gender, and age. Stress and Health, 26, 304-329.

Hannon, J.C., & Ratcliffle, T. (2007). Opportunities to participate and teacher interactions in

coed versus single-gender physical education settings. Physical Educator, 64, 11-20.

Hills, L.A., & Croston, A. (2012). „It should be better all together‟: Exploring strategies for

„undoing‟ gender in coeducational physical education. Sport, Education, and Society, 17,

591-605.

Huguet, P., Dumas, F., Marsh, H.W., Regner, I., Wheeler, L., Suls, J., Seaton, M., & Nezlek, J.

(2009). Clarifying the role of social comparison in the big-fish-little-pond effect

(BFLPE): An integrative study. Journal of Personality and Social Psychology, 97, 156-

170.

Jones, R.J.A., Polman, R.C.J., & Peters, D.M. (2009). Physical self-perceptions of adolescents in

years 8, 9, and 10 in independent schools, state comprehensive schools, and specialist

sport colleges in England. Physical Education and Sport Pedagogy, 14, 109-124.

Kirk, D. (2002). Physical education: A gendered history. In D. Penney (Ed.), Gender and

Physical Education (pp. 13-23). London: Routledge.

Kirk, D. (2003). Student learning and the social construction of gender in sport and physical

education. In S. Silverman & C. Ennis (Eds.), Student learning in physical education:

Applying research to enhance instruction (2

ed.) (pp.67-81). Champaign, IL: Human

Kinetics.

Klomsten, A.T., Marsh, H.W., & Skaalvik, E.M. (2005). Adolescents‟ perceptions of masculine

and feminine values in sport and physical education: A study of gender differences. Sex

Roles, 52, 625-636.

Klomsten, A.T., Skaalvik, E.M., & Espenes, G.A. (2004). Physical self-concept and sports: Do

gender differences still exist?. Sex Roles, 50, 119-127.

Kreft, I.G. & De Leew, J. (1998). Introducing multilevel modeling. London, UK: Sage.

Larsson, H., Fagrell, B., & Redelius, K. (2009). Queering physical education: Between

benevolence towards girls and a tribute to masculinity. Physical Education & Sport

Pedagogy, 14, 1-17.

Lawler, M., & Nixon, E. (2011). Body dissatisfaction among adolescent boys and girls: The

effects of body mass, peer appearance, culture, and internalization of appearance ideals.

Journal of Youth Adolescence, 40, 59-71.

Lee, A.M. (2009). Contributions of research on student thinking in physical education. In R.

Bailey & D. Kirk (Eds.), The Routledge physical education reader (pp. 135-152). New

York, NY: Routledge.

Liem, G.A.D., Marsh, H.W., Martin, A.J., McInerney, D.M., & Yeung, A.S. (2013). The big-

fish-little-pond effect and a national policy of within-school ability streaming: Alternative

frames of reference. American Educational Research Journal, 50, 326-370.

Lirgg, C. (2006). Social psychology and physical education. In D. Kirk, D. Macdonald, & M.

O‟Sullivan (Eds.), The handbook of physical education (pp.141-162). London: Sage

Publications.

Margas, N., Fontayne, P., & Brunel, P.C. (2006). Influences of classmates‟ ability level on

physical self-evaluations. Psychology of Sport and Exercise, 7, 235-247.

Marsh, H.W. (1986). Verbal and math self-concepts: An internal/external frame of reference

model. American Educational Research Journal, 23, 129-149.

Marsh, H.W. (1990). A multidimensional, hierarchical self-concept: Theoretical and empirical

justification. Educational Psychology Review, 2, 77-172.

Marsh, H.W. (1993). The multidimensional structure of physical fitness: Invariance over gender

and age. Research Quarterly for Exercise & Sport, 64, 256-273.

Marsh, H.W. (1996). Physical self-description questionnaire: Stability and discriminant validity.

Research Quarterly for Exercise and Sport, 67, 249-264.

Marsh, H.W. (1997). The measurement of physical self-concept: A construct validation

approach. In K. Fox (Ed.). The physical self: From motivation to well-being (pp.27-58).

Champaign, IL: Human Kinetics.

Marsh, H.W. (1999). Physical self-description questionnaire outline. SELF Research Centre.

Marsh, H.W. (2002). A multidimensional physical self-concept: A construct validity approach to

theory, measurement, & research. Psychology: The Journal of Hellenic Psychological

Society, 9, 459-493.

Marsh, H.W. (2006). Self-concept theory, measurement, and research into practice: The role of

self-concept & educational psychology. Leichester, UK. British Psychological Society.

Marsh, H.W. & Craven, R.G. (2002). The pivotal role of frames of reference in academic self-

concept formation: The “big fish-little pond” effect. In F. Pajares & T. Urdan (Eds.),

Academic Motivation of Adolescents (pp. 83-124). Greenwich, CT: Information Age

Publishing, Inc.

Marsh, H.W.,& Craven, R.G. (2006). Reciprocal effects of self-concept and performance from a

multidimensional perspective. Perspectives on Psychological Science, 1, 133-163.

Marsh, H.W., Debus, R., & Bornholt, L. (2005). Validating young children‟s self-concept

responses: Methodological ways and means to understand their responses. In D.M. Teti

(Ed.), Handbook of research methods in developmental science (pp. 138-151).Oxford:

Blackwell Publishing.

Marsh, H.W., Hau, K.T., Sung, R.Y.T., & Yu, C.W. (2007). Childhood obesity, gender, actual-

ideal body image discrepancies, and physical self-concept in Hong King children:

Cultural differences in the value of moderation. Developmental Psychology, 43, 647-662.

Marsh, H.W., Hey, J., Roche, L.A., & Perry, C. (1997). Structure of physical self-concept: Elite

athletes and physical education students. Journal of Educational Psychology, 89, 369-

380.

Marsh, H.W., Kong, C.K., & Hau, K.T. (2001). Extension of the internal/external frame of

reference model of self-concept formation: Importance of native and nonnative languages

for Chinese students. Journal of Educational Psychology, 93, 543-553.

Marsh, H.W., Martin, A.J., & Jackson, S. (2010). Introducing a short version of the physical self-

description questionnaire: New strategies, short-form evaluative criteria, and applications

of factor analyses. Journal of Sport & Exercise Psychology, 32, 348-482.

Marsh, H.W., & O‟Mara, A. (2008). Self-concept is as multidisciplinary as it is

multidimensional: A review of theory, measurement, and practice in self-concept

research. In H. W. Marsh, R.G. Craven, & D.M. McInerney (Eds.), Self-processes,

learning, and enabling human potential (pp.87-118). Information Age Publishing Inc.

& The Montana Council of Teachers of Mathematics.

Marsh, H.W., Papaioannou, A., & Theodorakis, Y. (2006a). Causal ordering of physical self-

concept and exercise behavior: Reciprocal effects model and the influence of physical

education teachers. Health Psychology, 25, 316-328.

Marsh, H.W., Richards, G., Johnson, S., Roche, L., & Treymayne, P. (1994). Physical self-

description questionnaire: Psychometric properties and a multitrait-multimethod analysis

of relations to existing instruments. Journal of Sport and Exercise Psychology, 16, 270-

305.

Marsh, H.W., Seaton, M., Trautwein, U., Ludtke, O., Hau, K.T., O‟Mara, A.J., & Craven, R.G.

(2008a). The big-fish-little-pond effect stands up to critical scrutiny: Implications for

theory, methodology, and future research. Educational Psychology Review, 20, 319-350.

Marsh, H.W., & Shalveson, R.J. (1985). Self-concept: Its multifaceted, hierarchical structure.

Educational Psychologist, 20, 107-125.

Marsh, H.W., Trautwein, U., Ludtke, O., Koller, U. (2008b). Social comparison and big-fish-

little-pond effects on self-concept and other self-belief constructs: Role of generalized

and specific others. Journal of Educational Psychology, 100, 510-524.

Marsh, H.W., Trautwein, U., Ludtke, O., Koller, U., & Baumert, J. (2006b). Integration of

multidimensional self-concept and core personality constructs: Construct validation and

relations to well-being and achievement. Journal of Personality, 74, 403-456.

Marsh, H.W., & Yeung, A.S. (1998). An extension of the internal/external frame of reference

model: A response to Bong. Multivariate Behavioral Research, 36, 389-420.

McKenzie, T.L., Prochaska, J.J., Sallis, J.F., & Lamaster, K.J. (2004). Coeducational and single-

sex physical education in middle schools: Impact on physical activity. Research

Quarterly for Exercise and Sport, 4, 446-449.

Miller, D.K. (2010). Measurement by the physical educator: Why and how (6

ed.). New York,

NY: McGraw Hill.

Moller, J., Pohlmann, B., Koller, O., & Marsh, H.W. (2009). A meta-analytic path analysis of the

internal/external frame of reference model of academic achievement and academic self-

concept. Review of Educational Research, 79, 1129-1167.

Moreno-Murcia, J.A., Hernandez, E.h., Vaillo, R.R., Camacho, A.S. (2012). Motivation and

physical self-concept in physical education: Differences by gender. The Open Education

Journal, 5, 9-17.

Nagengast, B., & Marsh, H.W. (2012). Big fish in little ponds aspire more: Mediation and cross-

cultural generalizability of school-average ability effects on self-concept and career

aspirations in science. Journal of Educational Psychology, 104, 1033-1053.

National Association for Sport and Physical Education. (2011). Physical education is critical to

education the whole child [position statement].Reston: VA: Heidorn, B.

National Association for Sport and Physical Education & American Heart Association. (2010).

2010 Shape of the nation report: Status of physical education in the USA. Reston, VA:

National Association for Sport and Physical Education.

Negrete, R.J., Hanney, W.J., Kolber, M.J., Davies, G.J., & Riemann, B. (2010). Can upper body

extremity functional tests predict the softball throw for distance: A predictive validity

investigation. The International Journal of Sports Physical Therapy, 6, 104-111.

Nigg, C.R., Norman, G.L., Rossi, J.S., & Benisovich, S.V. (2001). Examining the structure of

physical self-description using an American university sample. Research Quarterly for

Exercise and Sport, 72, 78-83.

Ntoumanis, N., Pensgard, A.M., Martin, C.,& Pipe, K. (2004). An idiographic analysis of

amotivation in compulsory school physical education. Journal of Sport & Exercise

Psychology, 26, 197-214.

O‟Dea, J.A., & Abraham, S. (1999). Association between self-concept and body weight, gender,

and pubertal development among male and female adolescents. Adolescence, 34, 69-79.

Patton, M. (2002). Qualitative research and evaluation methods (2

ed.). Thousand Oaks, CA:

Sage Publications.

Peart, N.D., Marsh, H.W., & Richards, G.E. (2005). The physical self-description questionnaire:

Furthering research linking physical self-concept, physical activity and physical

education. Retrieved from: http://www.aare.edu.au/05pap/abs05.htm#P.

Penney, D., & Evans, J. (2002). Talking gender. In D. Penney (Ed.), Gender and physical

education (pp. 13-23). London: Routledge.

Reiman, M.P., & Manske, R.C. (2009). Functional testing in human performance. Champaign,

IL: Human Kinetics.

Rink, J. E. (2013). Teaching physical education for learning (7

ed.). New York, NY: McGraw-

Hill.

Rosenblum, G.D., & Lewis, M. (1999). The relations among body image, physical attractiveness,

and body mass in adolescence. Child Development, 70, 550-64.

Ruiz, J.R., Castro-Pinero, J., Espana-Romero, V., Artero, E.G., Ortega, F.B., Cuenca, M.M.,

Jimenes-Pavon, D., Chillon, P., Girela-Rejon, M.J., Mora, J., Gutierrez, A., Suni, J.,

Sjostrom, M., & Castillio, M.J. (2011). Field-based fitness assessment in young people:

the ALPHA health-related fitness test battery for children and adolescents. British

Journal of Sports Medicine, 45, 518-524.

Schmalz, D.L., & Davison, K.K. (2006). Differences in physical self-concept among pre-

adolescents who participate in gender-typed and cross gendered sports. Journal of Sport

Behavior, 29, 335-352.

Schneider, M., Dunton, G.F., & Cooper, D.M. (2008). Physical activity and physical self-concept

among sedentary adolescent females: An intervention study. Psychology of Sport and

Exercise, 9, 1-14.

Seaton, M., Marsh, H.W., & Craven, R.G.(2009). Earning its place as a pan-human theory:

Universality of the big-fish-little-pond effect across 41 culturally and economically

diverse countries. Journal of Educational Psychology, 101, 403-419.

Seaton, M., Marsh, H.W., & Craven, R.G. (2010). Big-fish-little-pond effect: Generalizability

and moderation: Two sides of the same coin. American Educational Research Journal,

47, 390-433.

Seaton, M., Marsh, H.W., Yeung, A.S., & Craven, R.G. (2011). The big fish down under:

examining moderators of the „big-fish-little-pond‟ effect for Australia‟s high achievers.

Australian Journal of Education, 55, 93-114.

Skaalvik, E.M., & Skaalvik, S. (2002). Internal and external frames of reference for academic

self-concept. Educational Psychologist, 37, 233-224.

Solmon, M.A. (2003). Student issues in physical education classes: Attitudes, cognition, and

motivation. In S. Silverman & C. Ennis (Eds.). Student learning in physical

education: Applying research to enhance instruction (2nd ed.) (pp.67-81). Champaign,

IL: Human Kinetics.

Subramaniam, P.R., & Silverman, S. (2007). Middle school students‟ attitudes toward physical

education. Teaching and Teacher Education, 23, 602-611.

Thijs, J., Verkuyten, M., & Helmond, P. (2010). A further examination of the big-fish-little-pond

effect: Perceived position in class, class size, and gender comparisons. Sociology of

Education, 83, 333-345

Trautwein, U., Gerlach, E., & Ludtke, O. (2008). Athletic classmates, physical self-concept, and

free-time physical activity: A longitudinal study of frame of reference effects. Journal of

Educational Psychology, 100, 988-1001.

United States Department of Health and Human Services. (2011). Healthy People 2020.

Retrieved from http://HealthyPeople.gov.

Van Manen, M. (1990). Researching lived experience: Humans science for an action sensitive

pedagogy (2

ed.). London, ON: The Althouse Press.

Van Manen, M. (2011). Phenomenology online: A resource for phenomenological inquiry.

Retrieved from http://www.phenomenologyonline.com/.

Webb, L., McCaughtry, N., & Macdonald, D. (2004). Surveillance as a technique of power in

physical education. Sport, Education, and Society, 9, 207-222.

Welk, G.J., & Joens-Matre, R. (2007). The effect of weight on self-concept and psychosocial

correlates of physical activity in youths. Journal of Physical Education, Recreation, and

Dance, 78, 43-46.

Whitehead, J.R., & Corbin, C.B. (1997). Self-esteem in children and youth: The role of sport and

physical education. In K.R. Fox (Ed.), The physical self: From motivation to well-being

(pp. 175-204). Champaign, IL: Human Kinetics.

Wood, T.M. (2003). Assessment in physical education: The future is now!. In S. Silverman & C.

Ennis (Eds.). Student learning in physical education: Applying research to enhance

instruction (2nd ed.) (pp.187-203). Champaign, IL: Human Kinetics.

APPENDIX A: EXTENDED REVIEW OF LITERATURE

Physical Self-Concept and Gender in Physical Education:

A Review of Past Research, Current Trends, and Future Directions

Dramatic declines in physical activity levels and participation in physical education have

contributed to an increase in overweight and obesity for children and adolescents (Cumming et

al., 2011; United States Department of Health and Human Services [USDHHS], 2011). Regular

physical activity engagement in exercise settings such as physical education can benefit students

not only physically, but cognitively, emotionally, and socially as well (Bailey, 2009; USDHHS,

2011). Despite the benefits of physical activity, there is still a pressing need to motivate students

to participate in physical education classes. Physical self-concept, an individual‟s perception of

himself or herself within the physical domain (Marsh, 1990), is an intrinsic factor that can help

explain student engagement in physical activity both in and out of physical education (Carlson,

1995; Welk & Joens-Matre, 2007). Positive physical self-concept is related to physical activity

participation and both adoption and adherence (Crocker et al., 2003; Cumming et al., 2011;

Dunton, Jamner, & Cooper, 2003; Marsh, Papaioannou, & Theodorakis, 2006a) and a multitude

of other important outcomes such as increased happiness, intrinsic motivation, and physical

performance and decreased depression and anxiety (Craven & Marsh, 2008; Crocker, Sabiston,

Kowalski, McDonough, & Kowlaski, 2006; Schneider, Dunton, & Cooper, 2008).

Physical self-concept research has consistently identified gender differences across

multiple physical activity settings including physical education, with females reporting lower

physical self-concept than males (Asci, 2002; Caglar, 2009; Hagger, Biddle, & Wang, 2005;

Marsh, Hey, Roche, & Perry, 1997; O‟Dea & Abraham, 1999; Schmalz & Davison, 2006;

Whitehead & Corbin, 1997). Although gender differences in physical self-concept are well

documented, there is currently only speculation as to why these differences are so distinct within

physical education. Most researchers place the blame for these differences on Western society‟s

gender stereotypes (Asci, 2002; Caglar, 2009; Crocker, Eklund, & Kowalski, 2000; Crocker et

al,, 2003; Klomsten, Marsh, & Skaalvik, 2005; Klomsten, Skaalvik, & Espenes, 2004; Schmalz

& Davison, 2006). This is a reasonable conclusion considering that schools are complex social

environments where gender stereotypes are reinforced and that traditionally masculine subjects

such as physical education magnify differences between males and females (Bain, 2009;

Connell, 2008; Kirk, 2003; Penney & Evans, 2002). Individuals define gender and evaluate

themselves in relation to the social context (Ridgeway & Correll, 2004). Settings like physical

education, which highlight gender and reinforce traditional societal beliefs, create an

environment that consequently requires students to emphasize their own expressions of gender

and behaviors (Penney & Evans, 2002).

While physical education should provide equal opportunities for all, regrettably “gender

segregation has the tendency to re-emerge within the gym” (Larrson, Fagrell, & Redelius, 2009,

p.14). As a result, physical education can influence physical self-concept in both positive and

negative ways (Marsh, Asci, & Tomas, 2006a; Whitehead & Corbin, 1997). Simply accrediting

gender differences in physical self-concept to gender stereotypes in physical education is likely a

naive, inadequate explanation. The complete understanding of this phenomenon is far more

complex. Investigating the relationships between gender and physical self-concept construction

within the gendered environment of physical education can, however, provide new insight into

methods to increase participation and physical education enjoyment for children and adolescents.

Therefore, the purposes of this review are to document the emergence of physical self-concept as

a stand-alone dimension of self-concept and examine the traditionally masculine environment of

physical education as an explanation for gender differences in physical self-concept.

The first section of this paper provides an overview of the development of a

multidimensional model of self-concept and the evolution of physical self-concept as a distinct

construct worthy of investigation. In the second section, the development of physical self-

concept research in physical education is examined. Also, similarities between gender

construction and physical self-concept are outlined. In the third section, the traditionally

masculine culture of physical education is analyzed in relation to how it may contribute to

gender differences in physical self-concept. Finally, this paper concludes with a summary of the

limitations of existing research, implications for practitioners, and suggestions for future

research.

Overview of Self-Concept and Physical Self-Concept

There has been controversy concerning the definition, structure, and measurement of self-

concept (Marsh, 1990; Marsh, 2006; Shalveson, Hubner, & Stanton, 1976). Early research rarely

utilized a consistent definition of self concept, which created inconsistent findings across the

literature. There was also debate concerning whether self-concept had a unidimensional or

multidimensional structure. Shalveson and Marsh (1976) argued that unidimensional self-

concept was overly simplistic and did not capture the complex nature of the construct. There

was also little empirical support or consistent measurement of unidimensional self-concept. In

1976, Shalveson and colleagues synthesized previous self-concept research in order to create a

working definition and theoretical framework to unify future investigations (Shalveson et al.,

1976). Marsh and Shalveson (1985) later revised this model, which is currently used in a variety

of academic, physical activity, and social settings (Craven & Marsh, 2008; Marsh & Craven,

2006). The Marsh/Shalveson (1985) theoretical model broadly defines self-concept as a person‟s

perception of himself/herself that is formed through interactions with and interpretations of the

surrounding environment (Marsh, 1990; Marsh & Shalveson, 1985; Shalveson et al., 1976). The

Marsh/Shalveson model suggests that self-concept: (a) can be differentiated from other

constructs (b) has a multifaceted and hierarchical structure; and (c) undergoes developmental

changes (Marsh, 1990; Marsh & Shalveson, 1985; Shalveson et al., 1976).

Definition

The Marsh/Shalveson (1985) definition of self-concept distinguishes between self-

concept and self-esteem. Self-concept has often been described as the descriptive aspect of self-

perception while self-esteem was deemed the evaluative aspect (Marsh & Shalveson, 1985).

According to Shalveson et al.‟s (1976) definition, self-concept is both evaluative and descriptive

by relying on an individual‟s self-perceptions (reflective statements about the self) (Fox, 1997) in

addition to the frame of reference with which he or she evaluates these perceptions. Therefore, in

order to distinguish between self-concept and self-esteem in the Marsh/Shalveson (1985) model,

self-esteem is considered the global component of self-concept (Marsh, 1990; Marsh &

Shalveson, 1985; Shalveson et al., 1976).

Although closely related, self-concept should also be distinguished from competence,

perceived competence, perceived ability, and self-efficacy. Competence refers to an individual‟s

actual abilities/inabilities, effectiveness/ineffectiveness, and successes/failures (Elliot & Dweck,

2005; Schultheiss & Brunstein, 2005). Perceived competence can then be thought of as

generalized self-statements regarding a person‟s competence. Accordingly, perceptions of ability

are self-statements of ability applied to specific contexts (Fox, 1997). Self-efficacy, on the other

hand, is an individual‟s personal belief about his or her ability and capability to succeed in a

given task (Bandura, 1986). Based on these definitions, perceptions of competence and ability

can be characterized as self-descriptive and cognitive evaluations of competence while self-

efficacy can be distinguished as self-evaluative and affective evaluations of competence.

Perceptions and feelings of competence have been pinpointed as a “cornerstone of our self-

worth” (Rhodewalt & Vohs, 2005, p.559) and self-efficacy is considered to act as a precursor to

self-concept development (Bong & Skaalvik, 2003). Therefore, although each is a stand-alone

component of the self, competence, perceived competence, perceived ability, and self-efficacy

are all considered to be integral components of self-concept.

Prior to the development of the Shalveson et.al (1976) definition, terminology related to

self-concept was inconsistent. Multiple definitions and a variety of terms related to the central

constructs of the self resulted in confusion and conflict in related literature (Shalveson et al.,

1976). Recent research has been effective at distinguishing between constructs; however, it

should be noted that the terms self-worth and self-esteem are often used interchangeably (Fox,

1997).

Structure

According to the Marsh/Shalveson (1985) theoretical model, self-concept has a

multidimensional and hierarchal structure. In this model self-esteem (global self-concept) is

divided into academic and nonacademic self-concepts. In the nonacademic domain, self-concept

is further separated into social, emotional, and physical self-concept sub-domains (See Figure 1).

Consequently, each sub-domain is considered to also have a multidimensional structure (Marsh,

1990; Marsh & Craven, 2006; Marsh & Shalveson, 1985). Later revisions of the original

Shalveson et al. (1976) model determined that less emphasis should be placed on the hierarchal

structure and more focus directed toward the multidimensionality and developmental aspects of

self-concept (Marsh, 1990; Marsh, 2002; Marsh & O‟Mara, 2008a; Marsh & Shalveson, 1985).

Currently, there is a substantial amount of support for the multifaceted nature of self-

concept and it is inferred that global measures are not clear indicators of self-concept within a

specific domain (Craven & Marsh, 2008; Marsh, 1990; Marsh & Craven, 2006; Marsh &

O‟Mara, 2008b; Marsh & Shalveson, 1985; Marsh, Trautwein, Ludtke, Koller, & Baumert,

2006b). Although global self-concept is not insignificant, it is no longer considered a useful tool

for understanding the distinct dimensions of self-concept. Therefore, it is strongly suggested that

researchers focus on the dimensions of self-concept most pertinent to their field (Craven &

Marsh, 2008; Marsh, 1990; Marsh, 2002; Marsh & O‟Mara, 2008a; Marsh & Craven, 2006;

Marsh & Shalveson, 1985; Marsh, Trautwein, et al., 2006). For researchers in physical

education, physical self-concept is the component of self-concept most applicable to

understanding this construct within the complex physical and social atmosphere of physical

education (Craven & Marsh, 2008; Marsh, 1990; Marsh & Craven, 2006; Marsh & O‟Mara,

2008a; Marsh, Trautwein, et al., 2006; Marsh & Shalveson, 1985).

Measurement of physical self-concept. Although the majority of initial research

utilizing the Marsh/Shalveson (1985) model of self-concept was related to the academic domain,

there has been progress in the application of this model to various areas in the physical domain,

such as physical education (Craven & Marsh, 2008; Marsh, 1999; Marsh, 2002; Marsh, 2006;

Marsh & Craven, 2006; Shalveson et al., 1976). Physical self-concept refers to an individual‟s

perception of himself or herself within the physical domain and is separated into two specific

categories: physical ability and physical appearance (Marsh, 1990; Marsh, 1996; Marsh, 1999;

Marsh; 2002; Marsh & Craven, 2006; Marsh & Shalveson, 1985; Peart, Marsh, & Richards,

2005; Shalveson et al., 1976).

Early exploration of the multidimensional structure of physical self-concept began with

studies that used global multidimensional measures such as the Self-Description Questionnaire

(SDQ) series of assessments (Jackson & Marsh, 1986; Marsh & Jackson, 1986; Marsh & Peart,

1988; Marsh & Redmayne, 1994). These assessments identify physical ability and physical

appearance as components of physical self-concept (Marsh, 1997). The first studies to apply the

Marsh/Shalveson (1985) model to a physical setting used the SDQ to examine self-concept in

relation to athletic participation among female high school students and young women (Jackson

& Marsh, 1986; Marsh & Jackson, 1986). Results indicated that for young women, athletic

participation was related to physical ability self-concept (Jackson & Marsh, 1986; Marsh &

Jackson, 1986).

The SDQ was then used to examine self-concept and its relationship to physical

fitness (Marsh, 1993a; Marsh, 1993b; Marsh & Peart, 1988). It was determined that physical

ability was moderately correlated with fitness (Marsh & Peart, 1988) and that multiple fitness

indicators displayed moderate correlations with physical fitness self-concept (Marsh, 1993b).

These studies concluded that specific components of physical fitness had unique relationships

with physical fitness self-concept. In addition, physical fitness was not related to academic self-

concept, indicating that physical self-concept is distinct from other self-concept dimensions

(Marsh, 1993b). These early studies provided a basis for the multidimensionality and

distinctiveness of physical self-concept (Marsh, 2002; Marsh, 2006). Although these

investigations made a significant contribution to self-concept literature by providing evidence

that physical self-concept is a distinct facet of self-concept, the use of global self-concept

measures in combination with samples that consisted primarily of young female participants

were major limitations.

As physical self-concept began to emerge as a separate and multidimensional component

of self-concept, the need for measures to assess this construct apart from global self-concept

became evident. Originally, investigations used global measures that often had only one or two

physical scales, research was inconsistent, physical self-concept measures were unreliable, and

there was a lack of a theoretical basis (Fox, 2000; Marsh, 1997). As a result, measures were

developed using a construct validity approach and multidimensional theoretical basis in order to

examine physical self-concept in more detail. These initial instruments were Richards‟ Physical

Self-Concept Scales (RPSC) (As cited in Marsh, 1997), the Physical Self-Perception Profile

(PSPP) (Fox, 1990; Fox & Corbin, 1989), and the Physical Self-Description Questionnaire

(PSDQ) (Marsh, Richards, Johnson, Roche, & Tremayne, 1994).

The first of these new instruments, the RPSC, was developed based on the

Marsh/Shalveson (1985) model of self-concept, previous self-concept research, and data from

the SDQ series (As cited in Marsh, 1997). This instrument identified seven facets of physical

self-concept: body build, appearance, health, physical competence, strength, action orientation,

and overall physical satisfaction (As cited in Marsh, 1997). The RPSC was one of the first

multidimensional physical self-concept instruments and is considered to be valid and reliable

across gender and age groups (Richards & Marsh, 2005). In addition, this measure provided a

basis for the development of alternative measures such as the PSPP and PSDQ.

Currently, the PSPP and PSDQ are considered to be valid and reliable and are the most

widely used instruments in physical self-concept research (Fox, 2000; Marsh, Asci, et al., 2002 ;

Sonstroem, 1998). The PSPP was developed using the Shalveson et al. (1976) model of self-

concept, previous self-concept research, and content validation from a sample of university

students (Fox & Corbin, 1989). This measure identified four multidimensional physical self-

concept scales and one global scale: sport, condition, body, strength, and global physical self-

worth (Fox, 1990; Fox & Corbin, 1989). When the PSPP was initially developed, it was

identified as the most valid and reliable physical self-concept instrument up to that point (Marsh,

1997). The PSPP was first used with university students and subsequent testing and revisions led

to the creation of children‟s versions of the PSPP (PSPP-C) and The Children and Youth

Physical Self-Perception Profile (CY-PSPP), both of which have been validated among children

as young as eight years old and up to high school age (Fox, 2000; Hagger, Biddle, Chow,

Stambulova, & Kavussanu, 2003; Hagger, et al., 2005; Jones, Polman, & Peters, 2009; Welk,

Corbin, Dowell, & Harris, 1997; Welk & Eklund, 2005; Whitehead & Corbin, 1997). Since its

original development, the PSPP and successive versions have been validated among a variety of

cultures and populations (Fox, 2000; Hagger et al., 2003). The PSDQ and PSPP are highly

correlated; however some researchers have reported that the PSPP has high internal correlations

while the PSDQ has smaller measurement error (Marsh, 1997; Marsh, Asci, et al., 2002 ; Marsh

et al., 1994). Therefore, Marsh and colleagues suggest that the PSDQ is the instrument best

suited to examine physical self-concept (Marsh, 1997; Marsh, Asci, et al., 2002 ).

The PSDQ was developed in response to previous research that identified physical self-

concept as a separate and multidimensional construct of self-concept (Jackson & Marsh, 1986;

Marsh & Jackson, 1986; Marsh & Peart, 1988). The preliminary version of the PSDQ combined

items from the SDQ II and Sutherland and Marsh‟s (1987) physical ability and self-description

scale (Marsh & Redmayne, 1994). In contrast to previous investigations that only examined two

sub-components of physical self-concept, this study examined six facets: endurance, balance,

flexibility, strength, appearance, and general ability. This initial instrument examined the

multidimensionality of physical self-concept using the Marsh/Shalveson model and in turn

provided support for each of the six components of physical self-concept (Marsh, 2002; Marsh,

2006; Marsh & Redmayne, 1994; Marsh & Sonstroem, 1995).

Following Marsh & Redmayne‟s (1994) identification of six components of physical self-

concept, the initial version of the PSDQ was revised based on the Marsh/Shalveson theoretical

model, the SDQ series of measures, and Marsh‟s (1993) study of physical fitness indicators

(Marsh et al., 1994). The original instrument was reconstructed to identify nine physical self-

concept scales and two global scales (Marsh et al., 1994). The final version of the PSDQ divided

physical self-concept into two specific categories: physical ability and physical appearance,

which were further separated into nine sub-categories (Marsh, 1996; Marsh, 1999; Marsh, 2002;

Marsh & Craven, 2006; Marsh et al., 1994). Specifically, Marsh and colleagues suggested an

individual‟s perception of his or her physical strength (e.g. “I am a physically strong person”),

body fat (e.g. “I have too much weight on my body”), physical activity levels (e.g. “I do sports,

exercise, dance, or other physical activities almost every day”), endurance (e.g.“I can run a long

way without stopping”), sporting ability (e.g. “I am good at most sports”), physical coordination

(e.g. “I am good at coordinated movements”), physical health (e.g. “I get sick a lot”), physical

appearance (e.g. “I have a nice looking face”), and physical flexibility (e.g. “My body is

flexible”) all contribute to his or her overall physical self-concept (e.g. “Physically, I am happy

with myself”) (See Figure 1)(Marsh et al., 1994).

The PSDQ is a 70 item instrument that requires individuals to respond to declarative

statements on true or false Likert scale. PSDQ construct validation has demonstrated that the

items are closely related to the constructs being measured (Marsh, 1996; Marsh, 1997; Nigg,

Norman, Rossi, & Benisovich, 2001). The PSDQ specifically has good reliability, test-retest

stability over both short and long term, and a well-defined factor structure that is invariant over

gender (Marsh, 1997; Marsh, 2002; Peart et al., 2005). In addition, it has both convergent and

discriminant validity when compared to three other physical self-concept instruments and

external criteria (Marsh, 1997; Marsh, 2002; Peart et al., 2005).

Since the original development of the PSDQ, a short form of the instrument has been

created (Marsh, Martin, & Jackson, 2010; Peart et al., 2005). The PSDQ-S condensed the 70-

item PSDQ to 40 items and is considered to be psychometrically sound (Marsh et al., 2010; Peart

et al., 2005). Since the development of the PSDQ and PSDQ-S, support for these physical self-

concept measures and theoretical structure has since been established for a variety of

populations, cultures, and age groups in various sport, exercise, and physical education settings

(Dunton et al.,2006; Guerin, Marsh, & Famose, 2004; Marsh, 1996; Marsh, 1997; Marsh, 2002;

Marsh, 2006;Marsh, Gerlach, Trautwein, Ludtke, & Brettschneider, 2007a; Marsh, Perry,

Horsely, & Roche, 1995; Marsh & Sonstroem, 1995; Nigg et al., 2001).

The RPSC, PSPP, and PSDQ are all considered to have good construct validity. Although

both the PSPP and PSDQ are widely used in physical self-concept research, it has been

suggested that the PSDQ and PSDQ-S are the more reliable and valid instruments (Marsh, 1997;

Marsh, Asci, et al., 2002 ; Marsh et al., 2010; Richards & Marsh, 2005). The PSPP is often

reported to have high correlations between factors and this is considered to be a major limitation

of the instrument (Marsh, Asci, et al., 2002 ; Richards & Marsh, 2005). The RPSC and PSDQ are

psychometrically similar; however, the RPSC does not address all of the components identified

in the PSDQ (Richards & Marsh, 2005). Therefore, the PSDQ has many advantages when

compared to the RPSC and PSPP. Nonetheless, the PSDQ is not frequently used in physical

education and researchers instead tend to choose shorter versions of the PSPP to examine

physical self-concept among children (Hagger et al., 2005; Welk & Eklund, 2005). Although the

RPSC is a shorter instrument that addresses more components than the PSPP, it is still not as

comprehensive as the PSDQ and PSDQ-S (Marsh et al., 2010; Richards & Marsh, 2005).

Considering recent support for the PSDQ-S (Marsh et al., 2010) and the suggestion that the

PSDQ be used instead of the PSPP (Marsh, Asci, et al., 2002 ), it is likely that the shorter and

more “child-friendly” PSDQ-S will become the instrument of choice in physical education

research.

In summary, the RPSC, PSPP and PSQD are considered valid, reliable, and established

instruments in physical self-concept research. However, a major strength of the PSDQ is its

ability to examine a wider range of physical self-concept dimensions while maintaining smaller

correlations between factors. With emerging support for the PSDQ-S, researchers can now

investigate a wider range of physical self-concept factors in a shorter amount of time and provide

new and valuable insight into the physical self-concepts of children in physical education.

Development

According to the Marsh/Shalveson (1985) model, self-concept should be examined from

a developmental perspective (Marsh, 1990; Marsh & Shalveson, 1985; Shalveson et al., 1976),

meaning that self-concept evolves as children grow and develop. Children exhibit a

multidimensional self-concept at a young age (Marsh & Craven, 2006; Marsh, Ellis, & Craven,

2002b; Marsh, Debus, & Bornholt, 2008); however, the multifaceted nature of self-concept

becomes more distinct as children grow and mature cognitively (Marsh 1990; Marsh, 2002;

Marsh & Craven, 2006; Marsh & Shalveson, 1985). As self-concept becomes more

multidimensional, the hierarchy weakens and self-concept becomes increasingly more domain

specific (Marsh, 1985; Marsh & Shalveson, 1985). There is strong support that as children age,

physical self-concept also becomes more differentiated and multidimensional (Marsh 1993b;

Marsh et al., 2005; Marsh, Hau, et al., 2007); indicating that physical self-concept is also best

studied from a developmental perspective (Marsh, 1993b; Marsh, 2002b; Marsh, 2006; Marsh, et

al., 2007a; Marsh, Hau, Sung, & Yu, 2007b).

Not only is there more distinction among the specific domains of self-concept as children

mature but self-concept also becomes more realistic. For instance, young children often have

very high self-concepts that do not reflect actual ability. As children transition from

preadolescence to adolescence, their self-concepts often decrease and become more accurate.

Following adolescence, self-concept then begins to stabilize (Marsh & Craven, 2006; Marsh,

Craven, & Debus, 1998; Marsh, Ellis, et al., 2002 ; Marsh & Shalveson, 1985). The same

developmental shift occurs with physical self-concept, which undergoes a decrease during

adolescence and then becomes increasingly more established with age (Marsh et al., 1998). In

physical education, physical self-concept maintains the multidimensional structure that has been

recognized in other physical settings (Dunton et al., 2006; Gheris, Kress, & Swalm, 2010; Marsh,

Hau, et al., 2007) and it is suggested that in this context physical self-concept should also be

examined from a developmental perspective (Marsh, Hau, et al., 2007 ).

Frame of reference. An individual‟s frame of reference plays an important role in the

development of global and physical self-concept (Marsh & Craven, 2002). There is support for

frame of reference effects and the development of physical self-concept in physical education

(Gheris et al., 2010; Trautwein, Gerlach, & Ludtke, 2008). The Internal/External Frame-of-

Reference (I/E) Model provides a framework for understanding the development of self-concept

in relation to internal and external comparison (Craven & Marsh, 2006; Marsh, 1990; Marsh &

Craven, 2002). First of all, internal comparisons of ability within each domain can impact self-

concept in other domains. For example, a student may have average math ability and above

average physical ability. The student will in turn use this internal comparison of ability in two

separate domains to construct self-concepts in each area (Marsh & Craven, 2006; Marsh et al.,

1997). In a physical setting, students may compare current performance on a specific task with

their own previous performances (Gheris et al., 2010). As a result, the construction of a positive

physical self-concept is part of an internal self-evaluation process.

In combination with internal evaluations, external comparisons influence the

development of physical self-concept. External comparison refers to when individuals use social

comparison to form self-concepts. As children grow older, external comparison increasingly

plays a larger role in self-concept development (Marsh, 1993b; Marsh, 2006; Marsh & Craven,

2002; Marsh, et al., 2008b) and the context of physical education provides a complex setting to

negotiate the development of physical self-concept. For instance, physical education classrooms

often have children exhibiting skills in a social setting. Such a public display of ability provides

an optimal scenario for external social comparisons (Gheris et al., 2010; Trautwein et al., 2008).

In order to explain social comparison effects on self-concept development Marsh and

colleagues developed the Big-Fish-Little-Pond Effect (BFLPE) theory (Marsh & Craven, 2002;

Marsh, Seaton, Trautwein, Ludtke, Hau, O‟Mara, & Craven, 2008b). The basic premise of the

BFLPE is that students with high ability will have a lower self-concept when surrounded by

others with high average ability. Conversely, students with average ability will have a higher

self-concept when surrounded by others with low average ability (Chanal, Marsh, Sarrazin, &

Bois, 2005; Marsh, Seaton, et al., 2008). There is support for the BFLPE for physical self-

concept in physical education (Chanal et al., 2005; Marsh, 1993a; Marsh, 2002; Marsh, 2006;

Marsh, et al., 1997; Marsh & Peart, 1988; Trautwein et al., 2008). For instance, in the Chanal et

al. (2005) study, gymnastics self-concept was influenced by the performance of others. When

students were surrounded by classmates with superior gymnastics skills, they reported a lower

self-concept for this area (Chanal et al., 2005). In addition, Marsh (1993a) has also reported that

social comparisons of running ability among students impacted their own self-perceptions in the

physical education classroom. While support for the I/E Frame-of-Reference Model and the

BFLPE in physical education does exist, it is sparse and conclusions from this research are

limited. Physical education is an ideal setting for internal comparisons of performance as well as

social comparisons of ability. Therefore, there is a substantial need for more study of how a

child‟s frame of reference and social context influence the development of a positive physical

self-concept in physical education.

Reciprocal effects model. In combination with frame of reference effects, there is a

connection between outcomes and physical self-concept development. Marsh and colleagues

have described a Reciprocal Effects Model (REM) to explain this relationship (Craven & Marsh,

2008; Guay, Marsh, & Bovin, 2003; Marsh & Craven, 2002; Marsh & Craven, 2006; Marsh &

Perry, 2005). For example, Marsh & O‟Mara (2008b) provided evidence to support a positive

reciprocal relationship between academic self-concept, school grades, and educational

attainment. Academic self-concept was demonstrated to be consistently related to school grades

and educational attainment, yet evidence supporting a REM with global self-esteem was weak

and inconsistent. Therefore, this investigation also indicates that a REM is also domain specific

(Marsh & O‟Mara, 2008b).

In physical education, there is support for the REM between physical self-concept and

physical activity levels, exercise behaviors, skill development, and performance. For example,

students with higher physical self-concepts are more active and exhibit positive exercise

behaviors (Marsh, Papaioannou, et al., 2006; Trautwein et al., 2008; Welk & Eklund, 2005). As a

result, when students are more physically active their physical self-concepts will likely be higher.

The REM is also associated with skill development in physical education. When physical self-

concept development is supported, gains in skill performance are more likely to be maintained.

As a result, favorable skill performances positively impact physical self-concept (Peart et al.,

2005). In accordance with the REM, when students are satisfied with their skills and personal

performance their physical self-concept will be more positive- which becomes a tool used to

improve performance (Craven & Marsh, 2008; Marsh & Craven, 2002; Marsh et al., 2005).

A major goal of physical education is to increase children‟s physical activity levels while

providing them opportunities to develop skills that will enable them to be physically active for

the rest of their lives (Ennis, 2011). If teachers support the development of a positive physical

self-concept, they can in turn promote skill development and physical activity. Likewise, by

fostering skill development and providing opportunities for physical activity, teachers can

encourage positive physical self-concept development among their students. An adequate

amount of support for the REM in physical education exists, yet there is still a need for studies to

explore this model and justify the importance of positive physical self-concept development in

physical education.

Physical Self-Concept, Physical Education, and Gender Construction

Research on Physical Self-Concept in Physical Education

Past research has provided a body of evidence to support physical self-concept as a stand-

alone and multidimensional construct of self-concept. As a result, investigations in a variety of

physical activity and sport settings, including physical education, have utilized this framework in

order to understand self-concept within the physical domain. Unfortunately, inquiry in physical

education based upon Marsh‟ model of physical self-concept has been limited, perhaps due to the

length of the PSDQ. The majority of investigation in this area has used various versions of the

PSPP and has been based on Fox‟s physical self-perception framework (Goni & Zulaika, 2000;

Hagger et al., 2005; Whitehead & Corbin, 1997). Findings from this research are however,

consistent with studies utilizing the PSDQ and the Marsh/Shalveson (1985) model of physical

self-concept. In general, physical self-concept research in physical education has been related to

issues of curriculum, physical self-concept interventions, and gender differences in physical self-

concept.

For instance, investigations based on both theoretical frameworks have concluded that the

physical education curriculum can influence physical self-concept. Specifically, when the

curriculum attempts to improve physical self-concept, it can positively influence the physical

self-concepts of students ranging from elementary to high school (Daley & Buchanan, 1999;

Gheris et al., 2010; Goni & Zulaika, 2000). In addition, programs that place emphasis on

participation in cooperative activities instead of competition are more likely to improve physical

self-concept sub-domains for females (Daley & Buchanan 1999; Marsh & Peart, 1988). Research

with the PSDQ has demonstrated that curriculums such as adventure or fitness education that

focus on improving self-concept and fitness simultaneously may positively affect physical self-

concept (Gheris et al.,2010; Marsh & Peart, 1988).

There has also been investigation of the ability of school-based fitness interventions to

improve physical self-concept of sedentary high school females. These interventions were not

however, as successful as hypothesized. Physical self-concept, measured with the PSDQ, was

only affected when there were increases in physical activity levels in conjunction with

improvements in fitness that resulted in a reduction of body fat (Dunton et al., 2006; Schneider et

al., 2008). Despite issues with research design, the major contribution of past research related to

curriculum and physical self-concept interventions with children is the notion that a positive

physical self-concept can be developed and maintained as a result of external influences

(Whitehead & Corbin, 1997).

Physical self-concept research utilizing the Marsh/Shalveson (1985) framework that

focuses solely on physical education is sparse. The studies that have been done either use only

global measures of self-concept (Goni & Zulaika, 2000) or include only female participants

(Dunton et al., 2006; Marsh & Peart, 1988; Schneider, et al., 2008). As a result, there are major

gaps in the literature examining physical self-concept among students enrolled in physical

education. Regardless of instrumentation or theoretical framework, past physical self-concept

research does provide support for the restructuring of the traditionally masculine curriculum of

physical education in order to positively influence physical self-concept (Azzarito & Solmon,

2009; Azzarito & Katzew, 2010).

Gender Effects in Physical Education

Although physical self-concept research limited to physical education utilizing the

Marsh/Shalveson (1985) model is incomplete, there have been substantial investigations related

to the presence of gender differences in physical self-concept. Specifically, females tend to

report a lower physical self-concept than males and these distinct gender differences are

consistently demonstrated in studies based on both the Marsh/Shalveson (1985) and Fox (1990)

models of physical self-concept (Asci, 2002; Caglar, 2009; Hagger et al., 2005; Hagger,

Stevenson, Chatzisarantis, Gaspar, Ferreira, & Rave, 2009; Hayes, Crocker, & Kowalski, 1999;

Jones et al., 2009; Marius, Claudia, Florina, & Densia, 2011; Marsh et al., 1997; Martin, Liem,

Coffey, Martinez, Parker, Marsh, & Jackson, 2010; O‟Dea & Abraham, 1999; Schmalz &

Davison, 2006; Welk & Eklund, 1995; Whitehead & Corbin, 1997).

Studies utilizing the PSDQ have found that males typically score higher on overall

measures of self-concept, domain specific areas such as physical self-concept, and specific sub-

domains within physical self-concept. There is however, conflicting research regarding

specifically in which sub-domains males score higher. For example, it has been reported that

males in both high school and elementary school often score higher on the specific PSDQ

subscales of health, coordination, physical activity, body fat, sports competence, global physical,

appearance, strength, endurance, and global self-esteem (Klomsten et al., 2004). Others have

reported that males score higher on all measures except appearance and self-esteem and the

greatest differences are in sport competence, endurance, and strength (Caglar, 2009).

Furthermore, some studies using the PSDQ state that females score higher on global physical

self-concept items and males score higher on health self-concept items (Martin et al., 2010).

There are also discrepancies among PSDQ studies with regard to flexibility. At least one study

has documented that females report higher flexibility scores, although males tended to still rate

themselves highly on this measure (Klomsten et al., 2004). Finally, some researchers have stated

that when using a measure of global self-concept, males report a more positive overall self-

concept related to physical appearance in general (O‟Dea & Abraham, 1999). Investigations

using the PSPP have reported similar findings (Asci, 2002; Biddle, Page, Ashford, & Jennings,

1993; Jones et al., 2009; O‟Dea & Abraham, 1999; Welk & Eklund, 1997).

Similar to other physical settings, there are gender differences in physical self-concept

scores in physical education. Typically, females in physical education report a lower physical

self-concept and males tend to rate themselves higher in almost all subdomains on both the

PSDQ and PSPP (Caglar, 2009;Biddle et al., 1993; Marsh, Papaioannou, et al., 2006; Welk &

Eklund, 1997; Whitehead & Corbin, 1997). Consistent with conclusions of previous research in

sport and physical activity settings, it is generally accepted that distinct gender differences exist

between students in physical education. However, where the differences lie within the construct

of physical self-concept is currently unclear. In addition, there is little insight as to why these

gender differences are so stable and limited research controlling for physical activity or skill

levels. Many researchers have hypothesized about the impact of gender stereotypes (Asci, 2002;

Caglar, 2009; Crocker et al., 2000; Crocker et al,, 2003; Klomsten et al., 2004; Klomsten et al.,

2005; Schmalz & Davison, 2006), yet there has been little to no investigation in this area.

Schools, specifically physical education classes, are often places where Western society‟s

prominent gender stereotypes developed and reinforced (Bramham, 1999; Connell, 2008; Marsh

& Byrne, 1991). While gender stereotypes are likely a contributing factor to physical self-

concept differences between males and females in physical education, other factors such as the

social construction of gender, social comparisons of ability, or how the body is put on display

may also play a part. Considering the consistency of these gender differences documented in

research and the gender-laden atmosphere of physical education, it is surprising that there has not

been more exploration of gender differences in physical self-concept focused on physical

education settings.

Gender Construction

Gender is considered a central aspect of one‟s personal identity and encompasses

attribute self-perceptions, gender identity, and gender stereotypes (Martin & Ruble, 2004;

Martin, Ruble, & Szkrybalo, 2002, Tobin, Menon, Menon, Spatta, Hodges, & Perry, 2010).

Attribute self-perceptions refer to a child‟s self-assignment of gender-related characteristics. In

other words, it is the degree to which he or she describes himself or herself as more masculine,

feminine, or androgynous. Historically, the self-assignment of gender characteristics has been

labeled as gender identity, but current research has deemed gender identity to be more complex

(Bem, 1974; Burke & Tully, 1977, Zosuls, Miller, Ruble, Martin, & Fabes, 2011).

The current definition of gender identity is “the quality and strength of the cognitive

connections that a person makes between the self and a gender category” (Tobin et al., 2010,

p.607). Gender identity is more self-evaluative than the simple assignment of gender

characteristics to oneself. It encompasses knowledge of one‟s membership to a gender category,

feelings of compatibility with this category, pressures felt to conform to this category, and

feelings of superiority or inferiority associated with this group membership (Egan & Perry, 2001;

Tobin et al., 2010).

Gender stereotypes are cultural beliefs of gender based on how a particular culture views

the typical male or female and are socially-constructed beliefs that each will display certain

characteristics and demonstrate particular behaviors assumed to be acceptable or unacceptable

for each gender (Lips, 2005; Ridgeway, 2009; Ridgeway & Correll, 2004; Tobin et al., 2010).

For example, in Western societies a “typical” male should exhibit more masculine qualities that

tend to focus on action and violence. In contrast, a “typical” female should display more

feminine qualities that focus on emotionality, properness, and supportiveness. Gender

stereotypes are usually mutually exclusive standards that become guidelines for self-evaluation

of gender, or the construction of a gender identity (Basow, 1992; Ridgeway, 2009). In summary,

gender is not simply whether or not someone is male or female and is instead a multidimensional

construct made up of attribute self-perceptions, gender identity, and gender stereotypes.

Gender is considered to be a socially constructed expression of self and there is a

consensus among scholars that apart from biological factors a variety of internal and external

forces are at work as children attempt to make meaning of gender (Bem, 1974; Evans &Penney,

2002; Huffaker & Calvert, 2006; Leaper, 2011; Leman & Tenenbaum, 2011; Marsh & Byrne,

1991; Martin & Ruble, 2004; Martin et al., 2002; Ridgeway & Correll, 2004; Zosuls et al.,

2011). The self-socialization process of gender construction is how boys learn to “act like boys”

and girls learn to “act like girls” from interactions with significant others (i.e. parents, teachers),

peers, and society (Leaper, 2011; Zosuls et al., 2010). Therefore, gender stereotypes and the

social environment work in combination to play a significant role in gender development.

In a similar manner, physical self-concept is also developed through internal and external

frames of reference (Craven & Marsh, 2008). The presence of gender stereotypes in physical

education is a reasonable explanation for physical self-concept gender differences, but the

relationships are likely more complex. The processes of gender construction and physical self-

concept development are similar and involve self-description, self-evaluation, and frame of

reference. While gender stereotypes create an environment that is governed by hegemonic

masculinity (Connell, 2008; Connell & Messerschmidt, 2005; Ridgeway & Correll, 2004), they

do not solely account for the multiple factors that can provide an explanation for physical self-

concept gender differences. It is more likely that gender stereotypes contribute to the masculine

environment of physical education, which influences the self-socialization process of gender

development. Gender construction and physical self-concept development share many parallels

and it is possible the climate created by hegemonic masculinity and the social environment of

physical education play a part in students‟ self-description, self-evaluation, and frame of

reference for both gender construction and physical self-concept development.

Physical Education: A Gendered Environment

Physical education has an extremely gendered history, but over the past forty years an

awareness of gender issues in physical education has been evolving (Bain, 2009; Kirk, 2002;

Kirk, 2003; O‟Sullivan, Bush, & Gehrig, 2002). Initially, the most pressing gender issue was

whether or not girls were receiving equal opportunities in physical education. In the United

States, Title IX (1972) has impacted equal opportunities for girls and women in education, sport,

and physical education. This piece of legislation mandated that physical education could no

longer be segregated by gender and both girls and boys must have equal opportunities in physical

education classes (Brown & Rich, 2002; Kirk, 2003; O‟Sullivan, et. al, 2002). Although Title IX

(1972) has led to some changes-primarily increases in girls‟ participation in sport- many contend

that it has contributed to the dichotomous view of gender and the proliferation of a masculine

model of physical education (Kirk, 2003; O‟Sullivan, et al., 2002). For example, Title IX (1972)

allows the separation of boys and girls for certain activities (i.e. contact sports) and teachers

often continue to separate students during sex-typed activities (Flintoff & Scranton, 2005; Kirk,

2003). Today, many researchers argue that this piece of legislation encourages a bipolar view of

gender and that the greatest lesson from Title IX (1972) is that there is a difference between

access and equity (Brown & Rich, 2002). As a result, physical education classes are now

considered to be places that emphasize and foster gender disparities (Connell, 2008; Kirk, 2003;

Penney & Evans, 2002) and research evidence supports the conclusion that a coeducational

environment may actually contribute to exclusion of girls in physical education while promoting

hegemonic masculinity (Kirk, 2003; O‟Sullivan et al., 2002).

The hegemonic gender order in physical education influences the types of activities

students participate in, their actual and perceived skillfulness, how they view and use their

bodies, and their interactions and relationships with the teacher (Azzarito & Solmon, 2006;

Azzarito & Solmon, 2009; Larrson et al., 2009; Lee, 2009). Surprisingly, physical education

teachers are often aware of the gender order present in their classrooms, yet they continue to

reinforce a hegemonic masculinity (Brown & Rich, 2002; Evans & Penney, 2002; Larrson et al.,

2009). For example, the activities teachers include in their curriculum are often stereotyped by

gender (i.e. boys play football and girls jump rope). Likewise, when teachers have students

demonstrate activities or skills, boys are often chosen to demonstrate male stereotyped activities

while girls are chosen to demonstrate female stereotyped activities (Constantinou, 2008).

Students also reinforce the masculine atmosphere of physical education. For boys and girls,

activity choices, participation, and nonparticipation are all influenced by what is considered to be

gender appropriate (Azzarito & Solmon, 2009; Tischler & McCaughtry, 2011). Both boys and

girls have stereotypes regarding what activities are appropriate for males and females (Klomsten

et al., 2005). However, girls often seem to reinforce these stereotypes more than boys, which can

influence their performance and perceptions of ability (Chalabaev, Sarrazin, & Fontayne, 2009;

Klomsten et al., 2005; Solmon, Lee, Belcher, Harrison, & Wells, 2003).

Many activities are deemed either gender appropriate or inappropriate in physical

education, and research has consistently demonstrated gender differences among perceptions of

ability in gender-typed activities (Lee, 2009). Boys tend to have higher perceptions of ability

when the activities are stereotypically masculine and girls have higher perceptions of ability

when the activities are stereotypically feminine (Caglar, 2009; Klomsten et al., 2006; Lee, 2009).

These findings are similar to those documenting gender differences in physical self-concept,

which is reasonable considering the link between perceptions of ability and self-concept.

In addition, females in particular feel more pressure to conform and avoid participating in

masculine typed activities (Solmon et al., 2003). However, both girls and boys may avoid in

participating in activities that are deemed gender inappropriate in order to avoid the risk

associated with transgressing gender boundaries (Azzarito & Solmon, 2009; Cockburn & Clarke,

2002; Gorely, Holroyd, & Kirk, 2003; Paechter, 2003). As teachers and students reinforce

traditional stereotypes through activity choices and opportunities, students‟ construction of

gender identity and physical self-concept will be affected (Welk & Joens-Matre, 2007). When

children are separated during sex-typed activities, teachers are inadvertently teaching their

students what is appropriate for each gender and children‟s gender identities may be affected by

their feelings of compatibility with each category. As a result, students are receiving a clear

message that that they clearly belong to one of two groups. Similarly, physical self-concept may

be influenced since students typically rate themselves higher in domains that are deemed gender

appropriate (Caglar, 2009; Klomsten et al., 2006). Thus, cultural expectations of gender

appropriateness may influence students‟ self-perceptions and the opportunity to participate only

in gender typed activities could impact physical self-concept in both positive and negative ways.

The hegemonic masculinity present in physical education not only affects teachers‟

activity choices and student participation, it can also influence students‟ perceptions of

skillfulness. Actual skillfulness and perceived skillfulness in gender typed activities are issues

for not only girls but low-skilled males as well (Davison, 2000; Flintoff &Scranton, 2005;

Millington & Wilson, 2010; Solmon et al., 2003; Tischler & McCaughtry, 2011). For males, the

ability to display skillfulness within the gym or on the field can contribute to their ability to

move to the top of the hegemonic masculine pyramid (Bramham, 2003; Hickey & Fitzclarenge,

1999; Tischler & McCaughtry, 2011). This is likely to influence their physical self-concepts. If

boys are either not skilled or do not perceive themselves to be skilled, their physical self-

concepts can suffer as a consequence. Although it is speculated that boys tend to value sport and

activity performance more than girls, girls also feel the need to be skillful in physical education

(Azzarito & Solmon, 2006; Gorely et al., 2003; Flintoff & Scranton, 2001; Klomsten et al.,

100

2005) and perceptions of ability are a central component of physical self-concept (Marsh et al.,

2010). Unfortunately, both teachers and female students view males as more skilled in physical

education (Chalabaev et al., 2009; Larrson et al., 2009). When female students view males as

more skillful, it can decrease their participation in class (Constantinou, Manson, & Silverman,

2009).

Students‟ skill development and performance are associated with their perceptions of

ability and activity levels and when children have higher perceptions of competence they are

more likely to be physically active (Barnett, Morgan, van Beurden, & Beard, 2008; Welk &

Schaben, 2004; Welk & Joens-Matre, 2007). According to the REM, participation in physical

activity and physical self-concept are related; therefore if girls choose not to participate because

they perceive males as more skillful, their lack of participation will hinder the development of a

positive physical self-concept. Likewise, the same can be assumed for low-skilled males in PE.

In addition, research examining the REM in physical education has concluded that skill

development and physical self-concept are closely related (Peart et al., 1988).Therefore, gender

differences related to skillfulness, whether actual or perceived, are certainly contributing factors

to physical self-concept differences.

The physical body also plays an important role in the formation of gender identity and

physical self-concept within the context of physical education (Azzarito & Solmon, 2009;

Flintoff & Scranton, 2005; Kirk, 2002; Kirk, 2003; Marsh, Hau, et al., 2007 ). Physical education

is a classroom where the body is seen as an instrument and tool used to demonstrate knowledge

(Paechter, 2003)-much like a pencil and paper in a mathematics classroom. Whether it is in the

locker room, gym, or outdoors, the body is often used in gendered ways during physical

education and is put on display in front of others (Larrson et al., 2009; Paechter, 2003).

101

Regardless of gender, adolescents unfortunately often report dissatisfaction with their own

bodies (Lawler & Nixon, 2011). For males, the body functions as a tool to display their

masculinity and secure a place in the gender hierarchy of the class. Boys value an outward

appearance of strength and strive to have a muscular physique that rivals those of professional

athletes and meets a masculine ideal (Azzarito & Solmon, 2006; Brown, 2005; Gard, 2006;

Klomsten et al., 2005; Rosenblum & Lewis, 1999). Sites such as the locker room often provide

an opportunity for social comparisons of the body. As males compare their bodies to one

another, their bodies become an outward expression of masculinity. Failing to meet this

masculine ideal may result in the questioning of their masculinity or even their sexuality (Clarke,

2002; Davison, 2000; Gard, 2006; Gorely et al., 2003; Lawler & Nixon, 2011; Paechter, 2003).

Girls, on the other hand, are expected to be thin, smaller, and have a less muscular

physique that projects femininity (Azzarito & Solmon, 2006; Lawler & Nixon, 2011; Molloy &

Herzberger, 1998; Rosenblum & Lewis, 1999). It has been documented that girls value having an

attractive face and slender, more feminine body (Gorely et al., 2003; Klomsten et al., 2005).

When students‟ bodies do not meet the desired ideal it may become an obstacle to their outward

expressions of gender. For instance, if a female has a muscular body it can hinder her outward

projection of femininity and also bring her sexuality in to question since adolescents often view

muscular and athletic females as being “butch” or “lesbian” (Clarke, 2002; Gorely et al.,2003).

Specifically in a physical education setting, the body becomes a tool for performance and

displaying one‟s own gender within a social environment and many students form opinions of

their bodies in relation to the social context (Azzarito & Katzew, 2010; Lawler & Nixon, 2011).

Feelings related to physical appearance are central to physical self-concept (Crocker et al., 2006)

102

and conformity to gender expectations in combination with displays of the body in PE can

influence how students evaluate their own bodies and appearance.

Finally, since both gender and physical self-concept are influenced by external factors

(Ridgeway, 2009; Shalveson et al., 1976), social relationships in PE may potentially impact both.

Teacher-student interactions and relationships are influenced by gender in physical education.

For example, some researchers have stated that teachers either do not provide positive feedback

or overlook when students excel in activities that are deemed gender inappropriate (Brown,

2005; Lee, 2009; Larrson et al., 2009). In addition, there tend to be gender differences in the

frequency and type of feedback teachers provide to students, as well as student perceptions of

teacher feedback (Nicaise, Cogerino, Bois, & Amorose, 2006). By either overlooking positive

performances in gender inappropriate activities or providing more feedback to high skilled-

males, teachers are unknowingly reinforcing gender stereotypes and perhaps hindering physical

self-concept development of both boys and girls (Larrson et al., 2009). Even more disturbing,

teachers may publicly humiliate students during physical education classes when they do not

conform to traditional stereotypes or display adequate skills (Chepyator-Thomas & Ennis, 1997;

Davison, 2000). These interactions could negatively impact students‟ feelings of physical self-

concept.

Teacher-student relationships can also promote gender stereotypes in physical education.

Males who play sport and are coached by the physical education teacher often develop a

relationship above and beyond that of other students. These boys are often praised for their

abilities, given more responsibilities than other students, and have a closer personal relationship

with the teacher. This often places boys at the top of the hegemonic masculinity hierarchy

(Brown, 2005). Unfortunately, teachers often acknowledge male dominance in their classrooms

103

as something that is expected and may avoid challenging traditional stereotypes or favor boys

over girls (Chepyator-Thomas & Ennis, 1997; Larrson et al., 2009; Lirgg, 1993; Brown, 2005;

Chalabaev, Trouilloud, & Jussim, 2009). By either ignoring or condoning the hegemonic gender

order present in physical education, teachers play an important role in the construction of gender

and physical self-concepts of their students.

In summary, students and teachers are both aware of the gender order in physical

education and generally enforce dominant gender stereotypes associated with hegemonic

masculinity (Brown & Rich, 2002; Evans & Penney, 2002; Larrson et al., 2009). These

stereotypes create an environment that regulates opportunities and access to activities, actual and

perceived skillfulness, use and self-evaluation of the body, and teacher-student interactions and

relationships (Azzarito & Solmon, 2009; Chepyator-Thompson & Ennis, 1997; Klomsten et al.,

2005; Lee, 2009). Researchers typically attribute physical self-concept gender differences in

physical education to traditional stereotypes. It is reasonable to assume however, that these

stereotypes simply create an environment that influences the experiences students have, the role

of social comparison, and interactions with significant others which work in combination with

the social construction of gender in order to influence physical self-concept. Thus, the masculine

environment of physical education is likely a contributing factor to students‟ self-descriptions,

self-evaluations, and frame of reference for both gender construction and physical self-concept

development.

Conclusions, Implications, and Directions for Future Research

A positive physical self-concept is a valuable outcome in physical education and is

associated with increased physical activity participation, increased skill development, and

improved performance among students (Crocker et al., 2006; Hagger et al., 2005; Marsh,

104

Papaipannou, et al., 2006; Peart et al., 1988). Although young people‟s physical activity levels

and participation in physical education are declining (USDHHS, 2011), research indicates that

developing a positive physical self-concept can produce favorable outcomes among students

(Chanal et al., 2005; Marsh, Trautwein et al., 2006; Trautwein et al., 2008). The literature also

indicates that current physical education programs may not be the optimal environment to

promote physical self-concept development among all students. Currently, there is a body of

evidence to support gender differences favoring males in physical self-concept scores among

students in physical education (Asci, 2002; Caglar, 2009; Hagger et al., 2005; Hayes et al., 1999;

Schmalz & Davison, 2006). Although many point to gender stereotypes in physical education as

the most significant influence on students‟ physical self-concepts, the relationship between

gender stereotypes and physical self-concept is likely much more complex. While evidence

supports the notion that prominent gender stereotypes in Western society contribute to a

hegemonic masculine environment in schools, there is a need for more examination of how this

context is related to gender construction and physical self-concept among students.

Perhaps the most significant contribution of physical self-concept research in physical

education is the idea that a positive physical self-concept can be developed and maintained

(Daley & Buchanan, 1999; Dunton et al., 2006; Gheris et al., 2010; Marsh & Peart, 1988;

Schneider et al., 2008). This finding holds many implications for physical education teachers. If

teachers provide students with cooperative activity choices, create opportunities to improve

fitness levels, and promote skill development in their classes they can positively impact students‟

physical self-concepts (Craven & Marsh, 2008; Gheris et al., 2010; Schneider et al., 2008).

Unfortunately, these conclusions are limited in generalizability since the majority of studies have

included only female physical education students. Therefore, future research should examine

105

how teachers can foster the development of a positive physical self-concept among both boys

and girls in their classes.

In addition, existing physical self-concept literature has made a substantial contribution

by providing considerable evidence that males have higher physical self-concept across a variety

of settings, cultures, and age groups (Caglar, 2009; Crocker et al., 2000; Marsh et al., 1997;

Martin et al., 2010; Hagger et al., 2005; Hagger et al., 2009). There are conflicting results,

however, with regard to which specific sub-domains males score higher in when compared to

females (Caglar, 2009; Klomsten et al., 2004; O‟Dea & Abraham, 1999). These inconsistencies

are likely due to the use of global self-concept measures and multiple physical self-concept

instruments. Marsh‟s model and measures of physical self-concept are believed to be the most

valid and reliable currently available to researchers (Marsh et al., 1994; Marsh, Asci, et al., 2002;

Marsh et al., 2010). Future investigations examining physical self-concept in physical education

should employ his framework and instrumentation in order to contribute to a deeper

understanding of physical self-concept.

There is overwhelming support indicating that physical education often promotes a

bipolar view of gender and conforms to hegemonic masculinity (Connell, 2008; Kirk, 2002;

Larrson et al., 2009; Penney & Evans, 2002). The masculine environment of physical education

is reinforced by the opportunities teachers provide for students, the activities students choose to

participate in, students‟ actual and perceived skillfulness, the role of the physical body, and

teacher-student interactions (Azzarito & Solmon, 2009; Azzarito & Katzew, 2010; Chepyator-

Thomson & Ennis, 1997; Flintoff & Scranton, 2005; Larrson et al., 2009; Lee, 2009). From

existing research it is reasonable to conclude that physical education continues to reinforce

hegemonic masculinity and a prominent gender order. It is however, premature to assume that

106

gender stereotypes are the sole reason for gender differences in physical self-concept scores as

there is no existing research to support this notion. Instead, future research should investigate the

hypothesis that gender stereotypes indirectly influence physical self-concept by contributing to

hegemonic masculinity in physical education while also controlling for physical activity and skill

levels of students.

Finally, in addition to support for a traditionally masculine physical education culture,

research indicates that there are many parallels between gender construction and physical self-

concept development. For example, both constructs should be viewed from a developmental

perspective and evolve as children develop cognitively (Marsh, Hau, et al., 2007; Martin &

Ruble, 2004; Tobin et al., 2010). In a similar manner, gender construction and physical self-

concept development both involve aspects of self-description, self-evaluation, and frame of

reference (Craven & Marsh, 2008; Martin et al., 2010; Tobin et al., 2010). Therefore, it is likely

that there are existing relationships between the two within a physical education context.

Specifically, future investigations should examine the relationships among the masculine

environment of physical education, gender construction, and physical self-concept development.

Future research should also examine how the traditionally masculine culture of physical

education is related to students‟ frame of reference with regard to gender construction and

physical self-concept development.

In conclusion, current physical self-concept literature makes a substantial contribution by

establishing gender differences in physical education and suggesting that a positive physical self-

concept can be developed and maintained. There are however, no existing investigations

examining potential relationships between gender construction and physical self-concept

development. Moreover, there is no investigation of the potential effects a masculine culture of

107

physical education can have on both gender construction and physical self-concept development.

Therefore, from this review of literature, it is clear there is need to examine the relationships

between gender construction and physical self-concept development in physical education. In

addition, researchers should investigate how hegemonic masculinity in physical education

creates an environment that contributes to gender differences in physical self-concept.

Figure 1. The Hierarchal Structure of Self-Concept and Physical Self-Concept.

Global Self-

Concept

Academic

Nonacademic

Social Emotional Physical

Physical

Ability

Strength

Activity

Levels

Endurance

Sporting

Ability

Physical

Coordination

Flexibility

Physical

Appearance

Body Fat

Health

108

References

Asci, F.H. (2002). An investigation of age and gender differences in physical self-concept among

Turkish late adolescents. Adolescence, 37, 365-371.

Azzarito, L., & Katzew, A. (2010). Performing identities in physical education (En)gendering

fluid selves. Research Quarterly for Exercise and Sport, 81, 25-37.

Azzarito, L., & Solmon, M.A. (2006). A feminist poststructuralist view on student bodies in

physical education: Sites of compliance, resistance, and transformation. Journal of

Teaching in Physical Education, 25, 200-225.

Azzarito, L., & Solmon, M.A. (2009). An investigation of students‟ embodied discourses in

physical education: A gender project. Journal of Teaching in Physical Education, 28,

173-191.

Bailey, R. (2009). Physical education and sport in schools: A review of benefits and outcomes.

In R. Bailey & D. Kirk (Eds.), The Routledge physical education reader (pp.29-37).New

York, NY: Routledge.

Bain, L.L. (2009). The hidden curriculum re-examined. In R. Bailey & D. Kirk (Eds.), The

Routledge physical education reader (pp.39-47).New York, NY: Routledge.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.

Englewood Cliffs, NJ: Prentice-Hall.

Barnett, L.M., Morgan, P.J., van Beurden, E., & Beard, J.R. (2008). Perceived sports competence

mediates the relationship between childhood motor skill proficiency and adolescent

physical activity and fitness: A longitudinal assessment. International Journal of

Behavioral Nutrition and Physical Activity, 5.

Basow, S.A. (1992). Gender stereotypes and roles. Pacific Grove, CA: Brooks/Cole.

Bem, S. (1974). The measurement of psychological androgyny. Journal of Consulting and

Clinical Psychology, 42, 155-162.

Biddle, S., Page, A., Ashford, B., & Jennings, D. (1993). Assessment of children‟s physical self-

perceptions. International Journal of Adolescence and Youth, 4, 93-109.

Bong, M., & Skaalvik, E.M. (2003). Academic self-concept and self-efficacy: How different are

they really?. Educational Psychology Review, 15, 1-40.

Bramham, P. (2003). Boys, masculinities, and PE. Sport, Education, & Society, 8, 57-71.

Brown, D. (2005).An economy of gendered practices? Learning to teach physical education from

the perspective of Pierre Bourdieu‟s embodied sociology. Sport, Education, and Society,

10, 3-23.

109

Brown, D., & Rich, E. (2002). Gender positioning as pedagogical practice in teaching physical

education. In D. Penney (Ed.), Gender and physical education (pp. 13-23). London:

Routledge.

Burke, P.J., & Tully, J.C. (1977). The measurement of role identity. Social Forces, 55, 881-897.

Caglar, E. (2009). Similarities and differences in physical self-concept of males and females

during late adolescence and early adulthood. Adolescence, 44, 407-419.

Carlson, T.B. (1995). We hate gym: Student alienation from physical education. Journal of

Teaching in Physical Education, 14, 467-477.

Chalabaev, A., Sarrazin, P., & Fontayne, P. (2009). Stereotype endorsement and perceived

ability as mediators of the girls‟ gender-orientation-soccer performance relationship.

Psychology of Sport and Exercise, 10, 297-299.

Chalabaev, A., Sarrazin, P., Trouilloud, D., & Jussim, L. (2009). Can sex-undifferentiated

teacher expectations mask an influence of sex stereotypes? Alternative forms of sex bias

in teacher expectations. Journal of Applied Social Psychology, 39, 2469-2498.

Chanal, J.P., Marsh, H.W., Sarrazin, P.G., & Bois, J.E. (2005). Big-fish-little-pond effects on

gymnastics self-concept: Social comparison processes in a physical setting. Journal of

Sport & Exercise Psychology, 27, 53-70.

Chepyator-Thomson, J.R., & Ennis, C.D. (1997). Reproduction and resistance to the culture of

femininity and masculinity in secondary school physical education. Research Quarterly

for Exercise and Sport, 68, 89-99.

Clarke, G. (2002). Difference matters: Sexuality and physical education. In D. Penney (Ed.),

Gender and physical education (pp. 13-23). London: Routledge.

Cockburn, C., & Clarke, G. (2002). “Everybody‟s looking at you!”: Girls negotiating the

“femininity deficit” they incur in physical education. Women’s Studies International

Forum, 25, 651-665.

Connell, R. (2008). Masculinity construction and sports in boys‟ education: A framework for

thinking about the issue. Sport, Education and Society, 13, 131-145.

Connell, R.W. & Messerschmidt, J.W. (2005). Hegemonic masculinity: Rethinking the concept.

Gender & Society, 19, 829-859.

Constantinou, P. (2008). Heightening our awareness of gender stereotypes. Strategies, 21, 28-34.

Constantinou, P., Manson, M., & Silverman, S. (2009). Female students‟ perceptions about

gender-role stereotypes and their influence on attitude toward physical education.

Physical Educator, 66, 85-96.

110