      

       

. ... . . ;. 4%.;

8' ‘tg‘l‘q $355 1:. lug":

I- 53': III? - 51% :
MM 11mm} 3“”)‘3 ,u, a“???

2‘] ;§;§1l‘.
Taw‘rﬂ' 4% ):g .f‘

:4" ‘ 3‘93 ("719"
MHIUMﬁI

I WI}. 1.1—.
' ' nun. P‘IlﬁIfi guwﬁ.

   
   

    
     
   

    
 
 
 
 
  

  

III ' 11K
"Advqba'ﬁzqu I1:&1I‘\'}‘Ih

I Hi 3‘:- hw" nu. '

  
 
    
    

 

    

               

   
  

  
   
 

 

 

       
 
    
 

5.5973“— ‘
'. ' .fz'g'agiﬁ rI rg: ‘24:}: --
' ‘ .I s . ' .
.I, .If... ”I". - 4"“ 'l .
. r Eryn» an: I I l Iu‘L'In'I'E g _.__
' II wz- 12-“! I .
I : |“1I’;;§ :":’11):" 'I ﬂirlﬁit‘” II”; I‘l' I I 3%.. '-;)‘$I"\""0 .
I, ’ k“!!! 11$,“ Vii)”; .. I ‘ . ‘ . l"_‘ I‘.T.Ii| ...
.. ,! , . l3 . '3! {3%. 3-33“ III." «’vl'If-J . i.J~-'I’.g'.‘. ."~%.--.2... . ‘3
L2)” “KW! "if Eva" l“.Ib‘h‘:“'..1;~a::.,‘;. 1 Ig', . . r6 . ‘I "1"?) , v V 3". I. .
lI' h" l' " ‘Iﬁ: " 'u'. ' . I . . .
:Illlll'é‘ {im’i‘n‘ #4311“ 21.5.:i..Ir.I_:I'-II‘II‘: ":3 W§ "> . v 'I ,"
’I‘I’ h 4 “II. HI I; 1W. ”"3'3‘ '5?" {:7 .I. 3 'a'. 1,3." ‘ ”2' 3 5,; " I"'.' ' ..." _ .'
' ."0 min In; '11 .o 'II'Z'y «I‘m-u . I]; '.:‘.f". I ,‘w :r '.. '1. .;' 4 f p 3
I ”.13; |'.;:‘ .‘v' ..II . a. . I 7' ..,..3_
' . 'I‘fl‘I‘iIIIWII? I ”"1? 51m 'I‘I‘.‘ W “WI-:41:- I (v.9; i. .. r- ...I. ' .392: f ‘ .
I ' tﬂ‘, $95.3. “I “l“"ll‘.|"'a‘.n“1;”.v’:"fl": 'II If; ‘T- “:5. $1. . 5"" I. II . . ' f. I"; .. 21». 3' 4.- ’I'III". ;‘. ’ ' I I in. ‘ '30?
I. I, z. !{!¥.1YII.;!'-g: i‘j' '1': “2.4,! ‘ ; ‘I ' 1.0 'I 1.143;. 3I I I 1- .j. ....”- M: ‘I I . H 3' .l I II p I. ' I:
»hﬂHInH¢; Eq~n‘.n j} . ‘- ;w' ”ﬂ"“ ‘ " ‘:. _-- 3 . .1
l Ivl' “VII"‘IE I XII; '. 'lrl'l I ' ‘ ‘I'v‘..13"l.'n€i I, ‘- ‘5 I‘ f“. .,I. . ‘.~,. 3 y “‘ ‘II' I" . ‘» ' V
I :“I|KI:LIIV‘;‘VI.1,I:'I'. £21.. ' 1'" ' ' "I I U" I; I 'I n. |II '.‘, I I '. ' y, ‘ 'I . ' r . I r. . I t‘ ’
' ’ 1",," ‘1! I III" I “'5‘ ‘f' . "5 hr“ "I .z" n d" ' ”‘nl'h ”I 'zlu' V. I ' | ' . ' ‘3 . t 0 . I' A
I . IIJII 'I.‘w| "‘1“. ._f.'ll'."I- II I I I’ . 3‘: M | I I I I I h . g.
l ' ' I - . I . .
Wyn} IIJIJI. I“ ‘ “LIN“ .Ju ;. .y , '3’!” I- I“ r. 91 I ,ll' ' I l I A.
Md Mﬂwwha‘ ?“w%¢~i.lfé' . - w ' .’ ' v. ' .-.;:ww
.‘ ‘m‘I-"h :fI‘g' l"" 'I ""IJV '.'I I I ..‘I-r I '1. I . J ‘,- - - ._:' I?"
‘ >""‘:"'~ ‘ ‘I ' "'l c v ' a ‘ - -. i'I
I" “J It) ”6:19 IInII' I w} .9 .~ {l ”.3.“ I ”I! '. 1 I l I 3 {.I'I I‘ .l: I ,I . . . . I :- I 3“...
I 01%”); Mom I‘" ' ' ' ' | ' : I ’ : . . } . _ - . ‘
I mtﬁhmruﬁm‘ ' w‘ ..- ‘ . . .-L
{If-133.111“ 'I '9‘. ' ‘ ° ' " .. a , - *‘ ’w
. . Miwvw .... ~» .. . . . .. .I. ..
'01]. 'I‘:' II 'IV'I“ ‘-- . ‘I.I I ,' ‘5'",1
l ' ' "H“‘ﬁuﬂhd . ‘ ‘ | . I ' '~ I ‘ I «I I :-‘. I . .
' ¢I1§WIWI$#- ‘. , . . ' ..w..~31v'. .q'h:%N.
. " ‘ =:‘:'I WWI. I. .3 . In: J. . I I
I . lIrjIEIIﬂﬁII .III’T‘, ' “ I'Inl: I.” ”'3'; lo I , I ‘ II I I . II? “JV“ ' I "I M ”‘13.!" ‘I
’9 "1;,4 ' N m‘j', n I11"! III“ I ‘IT'}|'I' I‘m; ”WI " . 'IlIz' H ‘ :I I" h}. I" I'lv " I . | ' “I". I '. l 'I III. I "I". 'III In I' ‘ :ﬁ‘m m,
' ~ ft. '1 IIII ‘ ‘.'~ " I- I .. I‘I II ‘m. '.:I':
“In: 13:3" ‘I".I‘II""I"5W‘,' ”I'll III IMIII‘EII "H": I I. ‘ f. 'I'II‘I" .‘1 ' " .I"" ‘I I' 'IH‘I'I'Iw'
I ‘3”! {.11 In, 'I 'I ‘ I ’ I 3 lo .: ""'I'..I-
TI'II. ling "MI 5‘, f .III‘JuIIp‘qI '.' My” .| . -' .. ' ~ "' ~ 1335 I I
"VII-II .I '1 I' 17"" ,.
3-H.“ .‘qlul'lj- }:I III 1.“ VIII I .|‘ I,‘ I . II ‘ . ." IN I‘TI‘I‘ “II"; .5. :‘
l . I III I.’ _ . I ‘ .1 I l e” I 1.,
‘Lo? .g.’ III‘ "l" .‘t-‘JII ll ' ‘ yIn-II'. 'I' " Lb," H I
. I ‘ , .'. II I
"(I 115‘ "I ”'1”! ' ‘ I , ‘ - - ‘7 '3 ,3. ‘I “Ohm!”
' J" N‘l 'I'i " . I ' III'I'HI .2“ I.‘ I filly." -II 'I I3 I‘ “I
I:I‘. p't‘x’ﬁ‘. I k H V 'p . . - . I l .I I .4134? 1‘; ”I

I
l

III’I'UIIIII'
“I I1 WW5 «bootﬂu vow-- 0"?” "':;':"“ll'd'm ""1"” ...-4“": AW-
4.1.4“ II "‘1‘ ‘ ’

- , 2 :ii
"I ’ II I -, If, 0' ' "'II'MH WI, [IVIIEI'I‘IH .LRIIII
"III.,.IIIII....I.S ,. . ..131'53‘IIIIIWIM

-.‘_.

o

'i' r’. 1: $8

IE SITY LIBRARIES

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII III III

31293 01712 022

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the

dissertation entitled

Understanding Youth Sport Participation
Through Perceived Coaching Behaviors,
Social Support, Anxiety and C0ping

presented by

Susan Walter Hayashi

has been accepted towards fulfillment
of the requirements for

Ph.D. . Physical Education and
degree in

 

 

Exercise Science

Major professor

May 1, 1998

 

MS U is an Afﬁrmative Action/Equal Opportunity Institution O~12771

 

 

 

 

Mlchlgan State

LIBRARY
Unlverslty

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

 

 

 

 

 

 

 

DUEDUE DATEDUE DATEDUE
AUG 0 32002 LIu Is. I: u :. IIIW 0 6 2889
051 3 "2-“ man)

JH ‘UUI
1II4 1 It 05
APSE] 2937
:14 ?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

we animus-m4

 

UNDERSTANDING YOUTH SPORT PARTICIPATION
THROUGH PERCEIVED COACHING BEHAVIORS,‘ SOCIAL SUPPORT,
ANXIETY and COPING
By

Susan Walter Hayashi

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Physical Education and Exercise Science

1998

ABSTRACT

UNDERSTANDING YOUTH SPORT PARTICIPATION THROUGH PERCEIVED
COACHING BEHAVIORS, SOCIAL SUPPORT, ANXIETY, AND COPING

By

Susan Walter Hayashi

Understanding youth sport participation has been the emphasis of many researchers
over the past 20 years. Although insight as to why youth sport athletes continue and
discontinue sport participation has been provided, the past research has not identiﬁed
how variables, particularly social and psychological variables, interact to affect
children's decisions to continue or discontinue their sport participation. Therefore, the
purpose of this study is to build on the past research and to examine if youth sport
athletes who leave sport differ in their perceptions across time of coaching behaviors,
perceptions of social support, level of anxiety and ways of coping from athletes who
continue their sport participation. The participants in this study included 132 female
youth gymnasts. The gymnasts in the study were participants in the sports of rhythmic
and artistic gymnastics. The years of gymnastics participation of the participants
ranged from 2 to 12 years. The athletes were training and/or competing in their sport
for at least 6 months of the year. The data collection took place over a six month
period of time with the gymnasts completing the same set of inventories three times
during those six months. The data was examined using Generalized Estimating
Equations which accounts for correlated observations over time. The results revealed
that the interaction between several variables influenced youth sport participation.

Speciﬁcally, for gymnasts with higher anxiety and low abilities to cope with adversity

are more likely to discontinue participating in gymnastics. Support from family and
friends also played a role in gymnast's participation. That is, gymnasts who perceived
more support from family and friends were more likely to continue in gymnastics.
Finally, gymnasts' participation was also influenced by how they perceived their
coaches' behaviors. Gymnasts who perceived their coaches to provide low amounts of
non-reinforcement/ignoring mistakes feedback and who perceived their coaches to
provide high amounts of punishment oriented feedback, were more likely to
discontinue their gymnastics participation. These results moved the sport participation
research beyond achievement motivation and expanded our understanding of the

relationships between variables which influence sport participation decisions.

Cepyﬁght by
Susan Walter Hayashi
1998

To Carl,

Heinz, and Trudi

ACKNOWLEDGEMENTS

To begin, I would like to thank my advisor, Dr. Martha Ewing, for the
opportunity she gave me to study at Michigan State University, for the quality
education I received and her unconditional support throughout my doctoral program.
She provided me with many opportunities to learn and grow. She also recognized that
every student is unique and she celebrated those differences in her students. My
deepest appreciatation is to her for working with me long distance to complete this
dissertation.

I would also like to thank my committee members: Dr. Deborah Feltz, Dr.
Carole Ames, Dr. Jackie Lerner and Dr. Crystal Branta. Through their guidance and
instruction I received a top rate education. I would particularly like to thank Dr.
Branta for stepping in when I needed some important help.

My appreciation for the friendship of Sue and Terry Duncan is difﬁcult to
express through words. Your support and assistance in conducting the statistical
analyses allowed me to complete my dissertation. Thank you for all that you did for
me! Also, I would like to acknowledge Geoffrey Colon, Shannon Siege], and Lori
Ganno-Overway for helping me collect my data and being willing to repeatedly visit
the gyms.

Finally, I would like to thank my family. To my husband, Carl, who is my

vi

rock and provides me with unconditional love and support. I love you with all my
heart. To my parents, Heinz and Trudi, who have always been a special part of my
adventerous life. Thank you for everything. To my other parents, Howard and Betty,
who are the best "outlaws" in this world. Your support means so much to me. To
Ingrid, Steve, Donna, and Mike who cheered me along every step of the way. All of

you mean the world to me and I thank you for being a special part of my education.

vii

 

TABLE OF CONTENTS

List of Tables ....................................................................................................... v
List if Figures ....................................................................................................... vi
Chapter I: INTRODUCTION .............................................................................. I
Nature of the Problem ............................................................................. 1
Psychological Variables ............................................................... 7
Social Variables ........................................................................... 9

Youth Sport Environment ............................................................ 11

Statement of the Problem ........................................................................ 12
Research Questions .................................................................................. 13

Basic Assumptions ................................................................................... l4
Delimitations ............................................................................................ 14
Chapter II: Review of Literature ......................................................................... 15
Theoretical Frameworks Explaining Sport Participation ........................ 15
Competence Motivation Theory .................................................. 15

Achievement Goal Orientation Theory ....................................... 22

Cognitive Affective Model of Stress ...................................................... 24
Coping ...................................................................................................... 29

Sport Participation ................................................................................... 31
Chapter IH: Method ............................................................................................. 34
Subjects and Design ................................................................................. 34
Instrumentation ......................................................................................... 35
Athletic Coping Skills Inventory - 28 ........................................ 35

Coaching Behavior ....................................................................... 36

Social Support .............................................................................. 38

Sport Anxiety Scale ..................................................................... 39

Procedure .................................................................................................. 40

Pilot Study ................................................................................................ 43
Treatment of the Data .............................................................................. 43

viii

Chapter IV: Results ............................................................................................. 49

Preliminary Analyses ............................................................................... 49
Internal Consistency Analyses ..................................................... 49
Multivariate Analysis of Variance .............................................. 50
Factor Analyses ............................................................................ 50

Generalized Estimating Equations ........................................................... 57

GEE Procedure ................................................................. 58
Research Questions .......................................................... 59
GEE Results Format ........................................................ 60
GEE Analyses Results ................................................................. 60

The Effect of Anxiety, Coping with Adversity, Sport
Support and Family Support

on Gymnastics Participation ............................................ 61
The Effect of Perceived Coaching Behaviors on
Gymnastics Participation ................................................. 64
QCBAS subscales ............................................................ 64
L88 subscales ................................................................... 67
Gymnasts Who Discontinued Participation ............................................ 70
Chapter V: Discussion ......................................................................................... 77
Theoretical Implications .......................................................................... 77
Psychological and Sport Resilience ............................................ 88
Practical Implications ............................................................................... 95
Future Directions ..................................................................................... 98
Conclusion ................................................................................................ 100
Appendices ........................................................................................................... 101
Appendix A: Athletic Coping Skills Inventory-28 ............................... 101
Appendix B: Leadership Scale for Sports ............................................. 102
Appendix C: Coaching Behavior Assessment System
Questionnaire ................................................................... 106
Appendix D: Social Support Inventory ................................................. 108
Appendix E: Sport Anxiety Scale ......................................................... 111
Appendix F: Parent Consent Form ....................................................... 114
Appendix G: Michigan State University Human Subjects Committee
Permission Letter ............................................................. 116
Appendix H: Demographic Questionnaire ............................................ 118
Appendix I: Dropout Questionnaire ..................................................... 119
Appendix J: Means and Standard Deviations ...................................... 120
Appendix K: Quotes of Former Gymnasts ........................................... ' 121
List of References ................................................................................................ 128

ix

LIST OF TABLES

1. Data Collection Schedule ................................................................................ 42
2. QCBAS Factor Analysis for Time 1: Factor Loadings and
Variance Explained .................................................................................. 52
3. QCBAS Factor Analysis for Time 2: Factor Loadings and
Variance Explained .................................................................................. 55
4. QCBAS Factor Analysis for Time 3: Factor Loadings and
Variance Explained .................................................................................. 56
5. Internal Consistency Analyses for QCBAS: Cronbach's
Alpha Coefﬁcients ................................................................................... 57
6. Correlations Among Anxiety, Coping with Adversity, Sport Support and
Family Support Collapsed Across the Three Time Periods ......................... 62
7. Estimated Regression Coefﬁcients and z-Statistics for the Final Models
Following Backwards Elimination of Nonsigniﬁcant Effects ...................... 62
8. Simple Effects in Coping with Adversity by Sport Support for Effects of
Anxiety on Continued Gymnastics Participation Collapsed Across Time 63

9. Correlations Among Positive and Informational Feedback, Punishment-
Oriented Feedback, and Non-Reinforcement/Ignoring Mistakes Feedback

Collapsed Across the Three Time Periods .................................................... 65
10. Estimated Regression Coefﬁcients and z-Statistics for the Final Models
Following Backwards Elimination of Nonsigniﬁcant Effects ...................... 66

11. Simple Effects of Non-Reinforcement/Ignoring Mistakes Feedback for
Effects of Punishment-Oriented Feedback on Gymnastics Participation ..... 67
12. Correlations Among Democratic, Positive Feedback, Social Support and

Training and Instruction Collapsed Across the Three Time Periods ........... 69

13. What Former Gymnasts Liked the Most About Gymnastics: Rank Order

and Frequency Percentages ..................................................................... 71
14. What Former Gymnasts Liked the Least About Gymnastics: Rank Order

and Frequency Percentages ..................................................................... 72
15. Why Former Gymnasts Discontinued Participation: Rank Order and

Frequency Percentages ............................................................................. 73
16. What was the Best Part for Former Gymnasts About Leaving

Gymnastics: Rank Order and Frequency Percentages ............................ 74
17. What was the Hardest Part for Former Gymnasts About Leaving

Gymnastics: Rank Order and Frequency Percentages ............................ 75

LIST OF FIGURES

l. Gould and Petlichkoffs (1988) Integrated Motivational Model of Youth Sport

Participation and Withdrawal .................................................................. 3
2. Harter‘s (1978) Competence Motivation Model ............................................. 17
3. Smith's (1986) Conceptual Model of Stress ................................................... 26

xi

CHAPTER 1

Introduction

Nature of the problem

Understanding youth sport participation has been the emphasis of many
researchers over the past 20 years (e.g., Gould & Horn, 1984; Gould & Petlichkoff,
1988; Orlick, 1974; Petlichkoff, 1996; Sapp & Haubenstricker, 1978; Weiss &
Petlichkoff, 1989). Although insight as to why youth sport athletes continue and
discontinue sport participation has been provided, the past research has not identified
how variables, particularly social and psychological variables, interact to affect
children's decisions to continue or discontinue their sport participation. Therefore, the
purpose of this study is to build on the past research and to examine if youth sport
athletes who leave sport differ in their perceptions across time of coaching behaviors,
perceptions of social support, level of anxiety and ways of coping from athletes who
continue their sport participation.

Several models have been proposed to explain sport participation and
withdrawal including Gould and Petlichkoffs (1988) integrated motivational model of
youth sport participation and withdrawal; Linder, Johns, and Butcher's ( 1990) model of
factors in the voluntary withdrawal from Sport; and Smith's (1986) cognitive-affective
model. These models have provided theoretical frameworks for explaining youth sport

participation.

2

Of particular interest in this study is Gould and Petlichkoffs (1988) integrated
motivational model of youth sport participation and withdrawal (see Figure 1). This
model explains youth sport participation and withdrawal as processes that are
inﬂuenced by a common set of factors (Gould & Petlichkoff, 1988). Furthermore,
Smith's (1986) cognitive-affective model and Linder, Johns, and Butcher's (1990)
model of factors in the voluntary withdrawal from sport are incorporated into Gould
and Petlichkoffs (1988) model. Smith's model is not only one of the underlying
theoretical frameworks in Gould and Petlichkoffs (1988) model but much of the
research on Smith's model has focused on athletes who bumout which is addressed in
the cost-beneﬁt analysis and sport withdrawal components in Gould and Petlichkoffs
model. Linder et al.'s (1990) model addressed voluntary sport withdrawal, which is
also addressed in the sport withdrawal component of Gould and Petlichkoffs model.

This study will focus on the withdrawal components of Gould and Petlichkoffs
model. The integrated motivational model of youth sport withdrawal encompasses
three major components: (a) surface-level motivation for withdrawal and underlying
theoretical motives for withdrawal (e.g., Smith, 1986), (b) a cost-beneﬁt analysis, and
(c) type or level of sport withdrawal. Each of the components will be examined in the
following paragraphs to provide an overview of the model. The ﬁrst component of the
model addresses the descriptive research by identifying the personal and situational
reasons for sport withdrawal along with the theoretical frameworks for explaining
reasons for sport withdrawal. The theoretical frameworks incorporated into the model

are Harter's (1978, 1981) competence motivation theory, Maehr and Nicholls' (I980)

 

 

Component 1

 

MOTIVATION FOR YOUTH SPORT PARTICIPATION AND WITHDRAWAL

 

(la) Surface—Level Motivation for Participation and Withdrawal

....................................................................................................

 

 

 

 

Participation Withdrawal
Persog Situational Persongl Situational
Psychological Physical Psvchological Physical
“ Fun “ Skills * Playing * Other ' “ Skills did “ Program
"‘ Achieving improved, time, etc. interests not im- emphasis
success of etc. * No fun, etc. prove, etc. (too
winning, etc. serious),

 
     
 

  
 
 
 

etc.
(lb)
Underlying Theoretical Motives
for Participation and Withdrawal

 

 

 

 

I Achievement Orientatiorq I Competence Motivation I I Cognitive-Affective Stress I

Component 2 V

Costs - Beneﬁts Analysis

Component 3 V V

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(3a) Sport Involvement (3b) Sport Withdrawal
Type or Level Intensity Type or Level
Sport-Speciﬁc .__, Highly active Sport-Speciﬁc
Domain-general ._._, Slightly active Domain-Gench

 

 

 

 

 

Figure 1: Gould and Petlichkoff’s (1988) Integrated Motivational
Model of Youth Sport Participation and Withdrawal

4

achievement goal orientation theory, and Smith's (1986) cognitive affective model of
stress.

Harter's (1978, 1981) competence motivation theory has been the most
frequently used theoretical framework to understand youth sport participation. In
addition, several researchers have utilized Maehr and Nicholls' ( 1980) as well as
Nieholls (1984) achievement goal orientation theories to examine participation
motivation and attrition in sport. Smith's (1986) cognitive affective stress model has
received only limited attention in the sport participation literature. However, athletes'
responses to stress and their ability to cope with stress within sport have been found to
impact athletes' decisions to continue or leave sport (e.g., Burton & Martens, 1986;
Pooley, 1980). Due to the limited amount of research on the stress/anxiety-coping-
participation relationship, further research is warranted on the influence of
stress/anxiety and youth sport participation. This study will focus on the stress/anxiety
theoretical framework within the model along with the situational reasons for sport
withdrawal. The situational reasons for sport withdrawal have received less attention
in the literature than the personal reasons for withdrawing from sport.

'The second component of the model is a cost-beneﬁt analysis in which athletes
are believed to engage. Speciﬁcally, the cost-beneﬁt analysis suggests athletes'
decision to maintain involvement or to withdraw from sport is based on the perceived
beneﬁts and costs athletes experience in their sport. Limited research has been
conducted on the cost-beneﬁt analysis. Petlichkoff (1988) has conducted the only

study that addressed the cost-beneﬁt analysis in predicting sport persistence. She

5

identiﬁed various types of participants including starters, nonstarters, survivors (i.e.,
athletes who are members of the team but receive little to no playing time), dropouts,
and cuttees (individuals cut from the team during tryouts). Petlichkoff found that
starters and nonstarters had higher levels of satisfaction than the other types of
participants. Speciﬁcally, she found that survivors had lower levels of satisfaction
than did dropouts. Petlichkoff concluded that survivors and dropouts differed in their
cost-beneﬁt analysis. Dropouts found the cost of lack of playing time was greater than
the beneﬁt of being a member of a team, whereas, survivors found the benefit of being
a part of the team was greater than the cost of dissatisfaction with playing time,
subsequently, survivors continued their sport participation.

Finally, the third component of the model is sport withdrawal. Research has
found athletes' withdrawal from sport can include leaving to participate in another
sport or staying with the same sport but changing the intensity of their participation or
completely withdrawing from sport altogether. Research by Gould, Feltz, Horn, and
Weiss (1982) and Klint and Weiss (1986) found that if athletes discontinue their
participation, they will probably reenter sport. Consequently, their research dispelled
the notion that if a child discontinued their participation in a speciﬁc sport, they had
dropped out of sport altogether. This study will focus on the interaction of variables
that lead up to withdrawal from Sport or changes in participation status rather than the
third component of the model. The author recognizes the importance of the different
types and levels of withdrawal; however, studying this component is beyond the scope

of this study.

6

While Gould and Petlichkoffs (1988) model incorporates the variables of
anxiety, coping, social support, and perceived coaching styles as playing a role in
withdrawal from Sport, the interaction of these variables has not been examined in
previous sport participation research. Utilizing their model provides a framework for
studying these variables in relation to youth sport participation. Furthermore, learning
how anxiety, coping and social support variables interact, as well as how perceived
coaching behaviors interact provides an avenue for understanding how some athletes
are able to "bounce back" and persist while others withdraw in light of negative or
challenging sport experiences is of interest. Although variables other than anxiety,
coping, social support, and perceived coaching styles are included in the model, the
variables identiﬁed above appear to be particularly salient in being able to differentiate
between athletes who persist in sport and those who withdraw.

When examining research in psychology it is apparent how researchers have
emphasized the relationship between anxiety, coping and social support and an
individual's ability to persist in light of challenging experiences (e.g., Cowen, Work,
Wyman, Parker, Wannon & Gribble, 1992; Milgram & Palti, 1993; Werner, 1984).
Within sport research some of these variables including anxiety and perceived
coaching styles have received a great deal of attention in the literature while other
variables such as coping and social support received less attention. In the following
sections an overview of the research on youth sport participation and the psychological
variables (coping and anxiety), social variables (family and sport support), and the

youth sport environment (perceived coaching behaviors) is provided.

Psychological Variables

The psychological variables of anxiety and coping have received limited
attention in the youth sport participation literature. Rather, most of the anxiety
research has focused on the relationship between anxiety and performance (e.g.,
Burton, 1988; Gould, Petlichkoff, Simons, & Vevera, 1987) and the coping literature
has focused on the relationship between coping and injuries or performance stress
(e.g., Crocker, & Graham, 1995; Hanson, McCullagh, & Tonymon, 1992; Smith,
Smoll, & Ptacek, 1990). Sport has been characterized as an ideal situation for
studying the antecedents, dynamics, and consequences of anxiety (Hackfort &
Spielberger, 1989). If the Sport arena can serve as an ideal research facility for
studying anxiety, then it seems logical that anxiety could very well affect youth sport
participation. Furthermore, if sport and anxiety go hand in hand, then athletes who
exhibit positive coping strategies would probably be able to persist through difﬁcult
sport circumstances and continue their sport participation more readily than athletes
with poor coping strategies. Speciﬁc relationships between anxiety and sport
participation will be addressed below followed by a similar discussion of coping and
sport participation.

Agility. Research on the relationship between anxiety and youth sport
participation has demonstrated that, just as in adults, stress affects children in different
ways (Compas, 1987). Stress has been shown to affect children's enjoyment of sport
(e.g., Scanlan & Lewthwaite, 1986) as well as anticipated stress has led to children

avoiding sport involvement (Orlick & Botterill, 1975; Pierce, 1980). Although

8

causality has not been found between stress and sport withdrawal, there appears to be
a relationship between a stressful Sport environment and withdrawal. Several
researchers have found that former youth athletes indicated too much pressure as one
reason for withdrawal from their sport (Gould, Feltz, Horn, & Weiss, 1982;
McPherson, Marteniuk, Tihanyi, & Clark, 1980; Pooley, 1980). Walter (1996) found
that gymnasts who left their sport had the highest levels of anxiety during the 3
months prior to withdrawal compared to their peers who continued their gymnastics
participation. Thus, it appears that a better understanding of the role stress and
anxiety play in youth sport participation decisions is warranted. This study proposes
that it is not so much whether athletes encounter stressful situations that affects their
sport participation decisions but the interrelationship between athletes' perception of
situations, athletes' coping strategies and their ability to cope with the stress they
encounter in sport.

_C_gp_ipg. Coping appears to play an integral role in allowing adults and
children to deal with stressful situations (Garmezy, 1983; Shure & Spivak, 1982).
Coping has been deﬁned as "constantly changing cognitive and behavioral efforts to
manage speciﬁc external and/or internal demands that are appraised as taxing or
exceeding the resources of the person" (Lazarus & Folkman, 1984, p.141).
Speciﬁcally, coping responses have been referred to as strategies that mediate between
perceived stressful events and outcomes (Crocker, 1992).

Within the sport domain, researchers have primarily focused on the relationship

between athletes' coping skills and dealing with and preventing injuries as well as how

9

coping skills allow athletes to deal with perfonnance-related stress (e.g., Madden,
Kirkby, & McDonald, 1989; Smith, Smoll, & Ptacek, 1990). The ability to cope with
difﬁcult coaches or competitive situations appears to play an integral role in athletes'
responses to stressors in sport (Gould, Finch, & Jackson, 1993) and, thus, is more than
likely involved in athletes’ Sport participation decisions. None of the research in sport
on coping skills has been related to sport participation. Furthermore, research needs to
consider how coping skills relate to youth sport participation because research in
psychology has demonstrated that an individual's coping ability relates to resilience to
stressful situations (Rutter, 1983). It seems logical that athletes who utilize more
effective coping strategies will be able to deal more effectively with negative sport
situations. Subsequently, athletes will experience more enjoyment in sport and will
probably maintain their sport participation for a longer period of time. Thus, further
research on the relationship between coping skills and youth sport participation is
warranted.
Social Variables

Social variables in Sport can encompass a number of areas including, but not
limited to, group dynamics, team cohesion, and social support. Research on social
support in sport has focused on three areas: (a) the role of feedback from signiﬁcant
others in children's development of perceptions of competence (Black & Weiss, 1992;
Hem & Hasbrook, 1987), (b) the role of family and signiﬁcant others in socializing
children into and through sport (Brustad, 1992; Martin & Dodder, 1991; Weiss &

Glenn, 1992), and (c) the coach-athlete relationships (Home & Carron, 1985; Smith,

10

Zane, Smoll, & Coppel, 1983) and teammate-athlete relationships (Weiss & Duncan,
1992).

Research examining the use of support from within (i.e., coaches, teammates)
and outside (i.e., family, friends) of sport by young athletes and the differences
between these support groups has basically been ignored. Recently, Walter (1996)
found that gymnasts who left their sport identiﬁed more family support during the 3
months prior to leaving gymnastics and less support from within the sport environment
(i.e., coaches, teammates). Thus, differences in the use of family support and support
from within Sport have been found between the gymnasts who continued their sport
participation and those who left gymnastics. Within the psychological resilience
literature, the presence of a supportive family environment has been identiﬁed as being
extremely important characteristics of resilient children (see Gannezy, 1983).

Research has not identiﬁed the process of how parents inﬂuence or protect their
children to help them cope with stressful situations. It appears support from a family
provides an authoritative ﬁgure, who is more knowledgeable than the child and can
provide an outlet from the stressful situation (Maccoby, 1983).

When applying these ﬁndings to sport, not only having support from family is
important to deal with stressful situations but also having support from within sport
from coaches, teammates, and athletic trainers is important for coping with challenging
sport experiences. Having support from within the sport environment may be
important because social support sources from within sport generally have greater

knowledge and experience regarding sport situations and the speciﬁc situations the

ll

athlete is encountering than do family and friends. Further research on understanding
the role of social support in sport is necessary to understand the role social support
plays in participation decisions.

Youth Sport Environment

The youth sport environment can include but is not limited to individual
personal performance, team performance, the type of leadership, and overall team
climate (Horn, 1992). Of central interest to this study is the type of leadership and
more speciﬁcally the perceptions of coaching behaviors by youth sport athletes
because of the interaction coaches have with their athletes, the integral role coaches
play in setting the tone for their team's Sport environment and the fact that research
has demonstrated that coaching styles influence athletes' self-esteem (Smoll, Smith,
Barnett, & Everett, 1993).

In addition, athletes who left sport have reported negative coaching styles
and/or conflict with coaches as reasons for leaving sport (Gould, Feltz, Horn, &
Weiss, 1982; McPherson, Marteniuk, Tihanyi, & Clark, 1980; Pooley, 1980). Barnett,
Smoll, and Smith (1992) found that coaches who were trained to utilize a positive
coaching style had youth sport baseball players who evaluated their coaches,
teammates and the sport of baseball more positively than athletes who played for
untrained coaches. They also found that the players who played for trained coaches
the previous season had signiﬁcantly less attrition at the beginning of the next baseball
season than the players who played for untrained coaches the previous season. Thus,

coaching behaviors have been found to have an effect on youth sport participation but

12

only limited research has been conducted to examine the relationship between
perceived coaching behaviors and youth sport attrition. The current study examines
how the relationship between various coaching behaviors influences youth sport
participation and withdrawal decisions.

In conclusion, if sport researchers were to not only utilize Gould and
Petlichkoffs (1988) integrated motivational model of youth sport participation and
withdrawal as a framework for studying sport withdrawal but also examine the
interaction of variables within the model, a better understanding of the process which
leads to various participation decisions would be provided. The examination of the
interaction of psychological variables and social variables, as well as the interaction of
youth sport experiences of children who continue or discontinue their sport
participation, will provide greater insight into the youth sport participation process.

mtement of the Problem

The purpose of this study was to examine how psychological, social and
environmental variables interact to affect youth sport participation decisions.
Speciﬁcally, this study examined whether youth sport athletes who leave sport differ in
their perceptions of coaching behaviors, perceptions of social support, level of anxiety
and ways of coping from athletes who continue their sport participation. Research in
general psychology has emphasized the importance of considering cognitive,
psychological, social and environmental variables to understand human behavior. Due
to limited research in sport on the interaction between variables that inﬂuence sport

participation decisions, further exploration of psychological, social and environmental

13

variables relative to sport participation is warranted.
Resegrch Questions

Based on the ﬁndings in previous youth sport participation research, research in
general psychology, and logical reasoning, four research questions are presented.
These research questions are developed out of the Gould and Petlichkoffs (1988)
model. Gould and Petlichkoffs model identiﬁes the variables of negative sport
experiences, anxiety, social support and coping and have suggested how these
variables can be either costs or beneﬁts which lead to youth sport participation or
withdrawal. For example, costs of being involved in gymnastics could include
negative gymnastics experiences or the anxiety a gymnast feels. However, a gymnast
also has beneﬁts from being involved in gymnastics such as social support from
teammates, coaches, and family as well as learning new skills. If the beneﬁts
outweigh the costs, the gymnast could continue her/his sport participation. This study
builds on the relationships Gould and Petlichkoff have identiﬁed and examines the
dynamic process of youth sport participation. Speciﬁcally, the dynamic process of
youth sport participation examined in this study utilizes the variables identiﬁed above
and examines how these variables interact to lead to the behaviors of continued sport
participation and withdrawal or the intention to continue or withdraw from sport.

Two research questions are posed in the following paragraphs. The research
questions examine the behavior of gymnasts with regard to their continued
participation or withdrawal. The research questions are as follows:

1. What is the relationship among anxiety, ways of coping, support from

14

within sport and support from family/friends for female gymnasts who continue and
discontinue their sport participation?
2. What is the relationship among perceived coaching behaviors for female
gymnasts who continue and discontinue their sport participation?
Basic Assumptions

There are two assumptions which underlie this investigation.

1. The participants of this study will respond honestly to the survey
instruments.
2. Due to the fact that most of the athletes who participated in this study

have chosen to participate in their sport, it is assumed that the majority of the athletes
will continue participating in their sport. However, some athletes may choose to
discontinue participating in gymnastics. Any differences that emerge on the variables
examined between the athletes who continued and those who discontinued their sport
participation is of particular interest in this study. Thus, if none of the athletes choose
to leave their sport, limitations in the analyses could develop.
Delimitgtions

This study is delimited to female youth sport athletes who participate in
gymnastics, train and compete in their sport for a minimum of 6 months of each year,
who are between the ages of 9 and 14 years of age and who live in either west Texas
or central Michigan. The generalizability to athletes who do not meet these criteria is

questionable.

CHAPTER II

Review of Literature

This chapter will examine the theoretical frameworks incorporated by Gould
and Petlichkoff (1988) in their integrated motivational model of youth sport
participation and withdrawal. Included in the model are Maehr and Nicholl's (1980)
achievement goal orientation theory, Harter's (1978, 1981) competence motivation
theory, and Smith's (1986) cognitive affective model of stress. In addition, the youth
sport participation literature utilizing these theories will be explored. Finally, the
coping literature will be examined. The previously mentioned literatures are examined
as a basis for understanding how variables interact to inﬂuence youth sport
participation.

Theoretical Frameworks Explaining Sport Participation

Gould and Petlichkoffs (1988) integrated motivational model for youth sport
participation and withdrawal identiﬁes three theoretical frameworks that underlie youth
sport participation decisions. These theoretical frameworks include Harter's (1978,
1981) competence motivation theory, Maehr and Nicholls' (1980) achievement goal
orientation theory, and Smith's (1986) cognitive affective model of stress. An
overview of each of these theoretical frameworks is provided in the following sections
along with how these theories have been used to explain withdrawal from sport.

Competence Motivation Theog. Harter's (1978, 1981) competence motivation

15

16

theory has been popular with sport psychology researchers and has been the most
frequently utilized theory to examine youth sport participation. Several features of
Harter's theory, which have had particular appeal in the sport literature, are the theory's
developmental framework, the identiﬁcation of psychological and sociological
variables, and the fact that the model is testable.

Harter (1978) developed competence motivation theory by extending Robert
White's (1959) model of effectance motivation. According to White (1959), people
are born with an innate urge to deal competently with their environment or to "have an
effect". That is, individuals are motivated to engage in mastery attempts to have an
effect on their environment. If the mastery attempts resulted in successful
performances, then intrinsic pleasure is experienced, individual's perceived competence
is enhanced and continued participation in the activity is desired. Harter (1978) was
interested in competence motivation; however, she felt White's model was too global
and untestable. Thus, Harter (1978) attempted to move effectance motivation into a
developmental framework that could be tested empirically (see Figure 2).

Weiss and Chaumeton (1992) summarized Harter's competence motivation
theory as a multidimensional framework that explains both the initiation of mastery
attempts in particular achievement domains and the development of achievement
behaviors such as persistence, approach/avoidance, and striving for higher levels of
competence or challenge. Competence motivation begins with the notion of
individuals engaging in mastery attempts. The participation in mastery attempts is

characterized as being the urges individuals feel to demonstrate their abilities. If those

17

.032 32.3.32 8:83:25 33: meats—w

”N «...—ma

 

 

38830 9:332

 

333x com @5335 8m EoEouuoEBm
+
_m>o._&35 mo mam—eve:
+
39:82 beau—2
235935 8"— 3.5.385

 

 

 

 

BBS“ EoEoBQE§ ..o 363
./ SEE

9.3589233 8:885 3 \ _mmoom
«En—on .2659. .2566 8.. .805 0235on

 

 

39:83. @832

\

 

 

...—80 SC

/.

88830 option

 

 

8:02.303 c8 505628503 .3 x23
1?
132%? ..o mpg—302

+
89882 boﬁﬁz
Engage:— uom 3.652
.8 30802853” 323m

 

 

 

 

 

3.6.2? 3 .ﬁ
“_anﬁm
so 859.869

 

 

 

388m

1..
09.2.30 EEO

 

/ 3:50 ..o

 

- Inga
.9330 .8 \
22.92?— .«Eme \

8:809:00
we :03 338.6; /-

 

 

833$ D382
5 banana

/

goon—

   

 

 

 

 

 

  

:23.»qu
oocaﬁobm

  
 

A

 

833E
22.55

 

 

2.355833.
553.5“. 52.5.
A :53 a was

 

 

28o baa: é
889$ Essay—.23 5
ho sound—«EB:—

 

 

«

 

2238qu BESS
+
8:309:00 329.com

 

 

\

 

  

 

l8

mastery attempts are optimally challenging and success in them is attained, then an
increase in intrinsic pleasure, perceived competence, and internal perceptions of
control is likely to occur. Furthermore, success at these optimally challenging mastery
experiences results in continued motivation to participate. This facilitates the
development of a more intrinsically motivated individual that is represented on the left
side of Harter‘s (1978) competence motivation model. The consequences of repeated
failures at optimally challenging mastery attempts are also addressed in the model.
The results of repeated failures on individuals include increased anxiety, lower
perceived competence, and an external locus of control. This is likely to result in a
more extrinsically motivated individual. The right side of the competence motivation
model depicts the developmental path toward an extrinsic orientation (Harter, 197 8).

In addition, the model depicts the inﬂuence of significant others and the use of
rewards as having either positive or negative effects on the development of
competence motivation. An interaction of the constructs serves to maintain, increase,
or decrease competence motivation, or alternatively stated, inﬂuence the development
of a primarily intrinsically or extrinsically oriented person (Weiss & Chaumeton,
1992). For example, a gymnast engages in a mastery attempt such as practicing
gymnastics routines. Over time, if the gymnast receives positive reinforcement from
signiﬁcant others such as parents, coaches, or teammates, has internal self-rewards and
mastery goals, and perceives the routines as optimally challenging, these will lead to
high levels of perceived competence and intrinsic pleasure, and the gymnast's

competence motivation will increase. If this does not occur, a decrease in competence

l9

motivation occurs. For the purposes of this study, how signiﬁcant others within sport
and outside of sport help athletes to deal with challenging sport experiences such as
negative coaching styles and stressful situations which may influence participation
decisions is of primary interest.

To help explain some of the surface level reasons for youth sport withdrawal,
Gould and Petlichkoff (1988) incorporated Harter's (1978, 1981) competence
motivation theory into their model. The sport literature has utilized perceived
competence from within competence motivation theory to explain sport withdrawal
with dropouts being characterized as having lower levels of perceived competence than
individuals who continue sport participation. In addition to perceived competence
inﬂuencing competence motivation, a child's perception of his or her control over the
environment (perceived control) and his or her motivational orientation or the domain-
speciﬁc nature of perceived competence (physical, social or cognitive competence) are
also said to inﬂuence competence motivation. Thus, with regard to sport participation,
several factors interact to inﬂuence decisions to withdraw from sport. For example,
when Harter's model is incorporated into Gould and Petlichkoffs model, withdrawal
from sport would occur when an athlete's competence motivation decreases. This
decrease in competence motivation results from a combination of failure at mastery
attempts, a lack of or negative social support from signiﬁcant others, low perceptions
of competence, an external perception of control, and, ﬁnally, an increase in anxiety.

Initial research by Roberts, Kleiber, and Duda (I981) utilized Harter's theory to

examine sport participants and nonparticipants. They found that participants had

20

higher perceptions of cognitive and physical competence, general self-worth and
expectations for future success than did nonparticipants. F eltz and Petlichkoff (1983)
built on Roberts et al.'s (1981) ﬁndings by examining children in interscholastic sports
and found that current participants had higher perceived physical competence when
compared with dropouts. Thus, both Roberts et al. (1981) and Feltz and Petlichkoff
(1983) found that sport participants where higher in perceived physical competence
than sport nonparticipants and dropouts.

In contrast, Klint (1985) examined current and former gymnasts on their
perceptions of physical and social competence. She found that former gymnasts were
higher in perceptions of physical and social competence than current gymnasts. These
results were explained as the former gymnasts having had success in gymnastics and
in other sports since leaving gymnastics while the current gymnasts were still in the
midst of striving to attain their success. This led to Klint and Weiss's (1986) study
that examined how perceptions of competence are related to particular motives
children have for sport participation. They found that children high in perceived
physical competence were more motivated to participate for skill development reasons,
and children high in perceived social competence were more motivated by the
afﬁliation aspect of sport when compared to their low perceived competence
counterparts.

In addition, to the examination of perceived competence and sport participation,
Horn (1985) and Black and Weiss (1991) have examined the relationship between

perceived competence and social support. Speciﬁcally, Horn ( 1985) examined how

21

coaches' feedback inﬂuenced female adolescent softball players' perceptions of
competence and performance control. She found that players had higher perceptions
of competence if they received more frequent criticisms in response to performance
errors. In contrast, she found players had lower perceptions of competence if they
received more frequent positive reinforcement and nonreinforcement in response to
desirable performances. Horn (1985) concluded that athletes wanted contingent and
appropriate feedback from their coaches because it suggested to the athletes that the
coach thought they could improve their performances. Black and Weiss's (1991) study
examined athletes' perceptions of coaches' behaviors in relation to perceptions of
ability and motivation in swimmers. Their ﬁndings support Hom's (l985) results.
Aside from these two studies the importance of coaching behaviors on the psycho-
social development of athletes has gone largely unexamined in the sport literature.
Weiss and Chaumeton (I992) emphasized the need for more research that examines
the inﬂuence of coaches on participants continued participation or withdrawal from
sport.

Several researchers have examined the relationship between perceived
competence and anxiety. Weiss, Bredemeier and Brustad (1987) examined the
relationship between negative trait anxiety and perceptions of competence, perceptions
of control, and motivational orientation in youth sports. They found that anxiety was
related to several achievement-related characteristics identiﬁed in Harter's model.

Boys who were identiﬁed as high trait anxious reported lower perceived competence

and more external perceptions of control than low trait anxious boys. Furthermore,

22
girls who were high competitive trait anxious reported lower levels of perceived
competence, a more external perception of control, and lower intrinsic motivation than
girls who were low in competitive trait anxiety.

In addition, Brustad (1988) studied predictors of negative and positive affect in
competitive youth sport. He found that athletes high in competitive trait anxiety had
higher performance-related worries and worries about negative evaluation from
coaches, parents and peers than their low competitive trait anxiety counterparts.

Based on these ﬁndings it is possible to see the important role anxiety plays in the
experiences of youth sport athletes. Unfortunately, this perceived competence and
anxiety research has not been linked back to the youth sport participation research.
Again, additional research is called for by Weiss and Chaumeton (1992) to better
understand the role of anxiety and perceived competence in sport behavior.

When considering the sport participation research that utilized Harter's (I978)
theory within Gould and Petlichkoffs model, competence motivation relates to the
personal surface-level explanations cited for sport withdrawal. For example, athletes
who felt they were not good enough to play (surface-level explanation) a sport such as
softball may have felt that they were not engaging in a mastery attempt (underlying
theoretical motive). The fact that the mastery attempt was not optimally challenging
led the athlete to feel that they did not have any social support, which led to increased
anxiety. Ultimately, this could lead to withdrawal from sport. Additional research is
needed to more fully understand the situational variables identiﬁed within sport

withdrawal.

23

Achievement Goal Orientation Theorv. Researchers including Ames (I984),
Dweck (1986), Maehr and Nicholls (1980), and Nicholls (1984) have developed
several achievement goal orientation theories. Only the work by Maehr and Nicholls
(1980) and Nicholls (1984) has been used to examine the relationship between
motivation and participation in sport. Gould and Petlichkoff (1988) have incorporated
Maehr and Nicholls (I980) achievement goal orientation theory into their integrated
motivational model for youth sport participation and withdrawal to help explain sport
participation patterns. An overview of Maehr and Nicholls (I980) theory will be
provided in the following paragraphs.

Maehr and Nicholls (1980) suggested that success and failure are assigned
different meanings by individuals and that achievement behavior is thought of as
behavior directed toward demonstrating that one possesses desirable qualities. Rather
than deﬁning one achievement behavior, Maehr and Nicholls' (1980) identiﬁed three
forms of achievement motivation: ability, task, and social approval. Ability-oriented
motivation is when the goal of the behavior is to maximize the subjective probability
of attributing high ability to oneself and minimize the chances of attributing low
ability to oneself. Task-oriented motivation is when the goal of behavior is focused on
the process and conducting a task for its own sake rather than to demonstrate
comparative ability. Finally, social approval-oriented motivation refers to behavior
that is directed at maximizing chances of attributing high effort to oneself and
minimizing the chances of attributing low effort to oneself. The goal of the social-

approval behavior is to indicate virtuous intentions rather than ability (Maehr &

24

Nicholls, 1980). These orientations explain the motivated behavior of individuals and
why some individuals participate in an activity and others avoid participating.

According to Gould and Petlichkoffs (1988) model, Maehr and Nicholls'
(1980) achievement goal orientation theory explains youth's decisions to withdraw
from sport by the athlete's achievement goals and by the perception of success in
achieving these goals. Within the model, utilizing achievement goal orientations to
explain sport withdrawal is associated with the psychological and physical
explanations cited for sport withdrawal. Unfortunately, limited research has been
conducted examining the relationship between achievement goal orientations and youth
sport participation. Ewing (1981) utilized Maehr and Nicholls' work as the basis of
examining why children drop out of sport. She found that social approval oriented
individuals were more persistent than ability or task-oriented individuals (Ewing,
1981). In addition, she found that non-participants were more ability oriented than
sport dropouts. Petlichkoff (1988) conducted the other study utilizing Maehr and
Nicholls' (I980) achievement goal orientation theory. She examined the difference in
achievement goal orientations between starters, nonstarters, survivors, cuttees, and
dropouts during preseason, precompetitive, and postseason. She concluded that while
achievement goal orientations were related to athletes' level of satisfaction with their
sport experience, achievement goal orientations were not related to sport persistence
(Weiss & Chaumeton, 1992). Thus, the research utilizing achievement goal
orientations has been limited and has resulted in mixed ﬁndings.

Cognitive Affective Model of Stress. Smith (1986) developed a theoretical

25
model of stress titled the cognitive-affective model of stress (see Figure 3). Smith
used Thibaut and Kelly's (1959) social exchange framework to explain the process of
dropping out where individuals strive to maximize positive experiences while striving
to minimize negative experiences. Speciﬁcally, Smith (1986) suggested that an
athlete's decision to participate and persist in sport is a result of a cost-beneﬁts
analysis where athletes are always trying to maximize the beneﬁts (e.g., learning new
skills, having fun) while minimizing the costs (e.g., being yelled at for performance
errors, missing out on time with friends due to intense training).

The cognitive-affective model of stress addresses both stress and burnout in
sport where burnout is a speciﬁc type of stress response. Smith suggests that both
stress and burnout result from interactions among situational factors, cognitive events,
physiological responses and output or coping behaviors. In Smith's (1986) model the
distinction is made between athletes who burnout and others who discontinue their
sport participation. The withdrawal of athletes who burnout from sport is due to
chronic stress or increased stress-induced costs whereas the reasons others leave sport
are primarily due to a change of interests, an incompatible preferred alternative and/or
value reorientation (Smith, 1986). Smith (1986) also points out that many individuals
discontinue their sport participation for reasons other than burnout. The purpose of
this study does not encompass burnout. Rather, the focus of this study is how
variables including stress/anxiety interact to affect youth sport participation before the
point of burnout. Consequently, the stress portion of the model is of particular interest

in this study due to the interaction of situational, cognitive, physiological and

26

 

 

 

Personality and Motivational Factors

_1 1_

Situation —-> Cognitive appraisal ——-> Physiologic —9 C0ping and
Demands/ * of demands «— responses task behaviors
Resources * of resources e. g., arousal

* of consequences

* “meaning” of
consequences

 

 

 

 

 

Figure 3: Smith’s (1986) Conceptual Model of Stress

27
behavioral components to the model. Both the stress and burnout portions of the
model have been addressed separately in the sport literature (e.g., McCann, 1995;
Smoll & Smith, 1988).

According to Smith (1986) the word stress has typically been used in two
ways. The ﬁrst way refers to situations that tax the physical and/or psychological
capabilities of the individual. The second way the term stress has been used relates to
the individual's response to the situation. The cognitive-affective model takes into
account both the situation and the individual's reactions to the situation. In the
following paragraphs each of the components of the model will be addressed.

The situational component involves interactions between environmental
demands and personal or environmental resources. If an imbalance between demands
and resources occurs, stress will develop. Smith (1986) points out that stress occurs
not only when demands exceed resources but also when resources exceed demands.
Furthermore, the situational component can consist of external and internal situations
which can generate stress (Smith, 1986). External situations can include negative
feedback from coaches, disagreement among teammates, or competing against a strong
opponent. Examples of internal situations include desires to reach certain goals,
personal standards for competing, or conﬂicts in general.

The second component of the model is cognitive appraisal. Cognitive
appraisal is an active process and refers to the psychological reality individuals create
for themselves in the situations they encounter. According to Smith's (I986) model

the response to the situation or the cognitive appraisal contains four elements:

28
appraisal of the demands, appraisal of the resources, appraisal of the nature and the
likelihood of the potential consequences if the demands are not met; and the personal
meaning of those consequences for the person. The appraisal made is not always
accurate.

The physiological responses refer to the third component of the model.
Physiological responses are the bodily responses such as increased heart rate or sweaty
palms. These responses are reciprocally linked to cognitive appraisal. For example,
when a situation is appraised as being dangerous, then arousal occurs as part of the
recruitment of resources to deal with the situation.

The ﬁnal component of the stress model is output behaviors. Output behaviors
refer to an individual's attempt to cope with the situation and include task-oriented,
social and other classes of coping behaviors. The model emphasizes that the output
behaviors are not merely produced to meet the demands of the situation, rather
cognitive appraisals and physiological responses mediate the situation and output
behaviors. Ultimately, the coping behaviors are produced through this process. In
addition to these interactions, the model also addresses how these four components can
be affected by motivational and personality factors

Unfortunately, research utilizing Smith's (I986) cognitive-affective stress model
has focused on burnout within sport. According to Smith (I986), some athletes leave
sport due to burnout which occurs due to stress resulting from a perceived imbalance
between performance demands and their ability to meet those demands and/or because

they lack the coping capabilities. Gould and Petlichkoff (1988) suggest Smith's (I986)

29
model could also be used to explain withdrawal from sport beyond burnout. Gould
and Petlichkoff applied Smith's model to the cost-beneﬁt analysis as the process in
which youth athletes engage to make sport participation decisions. In addition, Gould
and Petlichkoff address athlete's perceptions of the situation and their ability to cope
with stress as contributing to sport withdrawal.

Furthermore, while Gould and Petlichkoff (1988) have included situational
factors such as social support and coaching leadership style in their model as surface
level explanations for withdrawal from sport, the research on sport participation has
largely ignored situational variables. Smith's model (1986) incorporates the appraisal
of the situation stress as well as coping capabilities. Athletes make appraisals of their
situations, encounter stress and have strengths and weaknesses in utilizing coping
skills. How athletes appraise their sport situations, the degree of stress they encounter
and their use of coping skills could all inﬂuence sport participation decisions. Thus,
utilizing Smith's model from within Gould and Petlichkoffs (1988) model to examine
sport participation in general could produce a greater understanding of the process that
leads to withdrawal than previously available. Finally, Gould and Petlichkoffs model
does not address how variables identiﬁed within their model could interact to inﬂuence
participation decisions. In the following section the coping literature will be
examined for explaining how coping skills could affect changes in sport participation
status.

m

By incorporating Smith's (1986) model into Gould and Petlichkoffs (I988)

30

model the concept of coping is addressed as playing a role in the youth sport athlete's
decision to participate and/or withdraw. Unfortunately, not a great deal of emphasis
has been placed on coping within the youth sport participation literature except when
talking about athlete burnout (McCann, I995). The majority of athletes do not suffer
from burnout, and understanding how athletes who do burnout use or do not use
coping skills has been ignored. By further examining the coping literature we can see
the value of incorporating the coping research into the youth sport participation
research.

According to the work of Lazarus and Folkman (1984) and Lazarus and
Launier (1978), stress results from a transaction between the environment and person
factors. Lazarus and Folkman (I984) and Folkman and Lazarus (I980, 1985) stated
that a stressful situation is only so if it is perceived as stressful by the individual and
an individuals' response to the situation perceived as stressful is coping. According to
Folkman and Lazarus (I980), coping strategies have been classiﬁed as problem-
focused coping where individuals utilize cognitive and behavioral efforts to change the
problem causing the distress and emotion-focused coping where individuals regulate
emotional responses to the stressor.

When considering coping in children, children typically are dependent on adults
for survival and generally have limited personal control over stressors (Compas,
Malcame, & Fondacaro, I988; Liederman, 1983). Furthermore, the ability to
cognitively appraise stressors and employ coping resources is often a function of

cognitive development and functions which children are only in the process of

31

developing (Shaie, 1977-78). In addition, children have been shown to differ in their
sensitivity to the environment. Speciﬁcally, it has been found that children who
perceive more situations as stressful may need to cope with a greater number of
situations than their less responsive counterparts (Compas, I987).

Relating the coping research to a child's sport experiences, if a child is enrolled
in a sport program they may not be aware of the potential stresses involved with
practices, interacting with coaches, teammates, and competing. Thus, they have
limited control over stressors. On top of not having control over stressors, certain
athletes may be more susceptible to perceiving situations in sport as stressful and may
lack also an understanding of ways to cope. If a youth athlete has this combination of
circumstances and encounters stressful situations in sport such as negative coaching
behaviors, they may discontinue their sport participation. The psychological resilience
literature has examined coping in relation to resilience or invulnerability to stress (e.g.,
Gannezy, 1983). This work has shown that having positive coping skills enables
individuals to deal more effectively with stressful situations. Children with
maladaptive coping skills are not able to deal effectively with the stressful situations in
their lives. Thus, understanding how coping relates to children's abilities to deal with
challenging sport experiences could provide insight into factors affecting youth sport
participation.

Sport Participation
In the previous section an overview of the theories incorporated into Gould and

Petlichkoffs (I988) integrated motivational model of youth sport participation and

32

withdrawal has been provided. This section will focus on utilizing Gould and
Petlichkoffs (1988) model to explain the relationships between variables inﬂuencing
youth sport participation.

Gould and Petlichkoff‘s (1988) model incorporates a cost-beneﬁt analysis,
which is based on Smith's theoretical model, to address how athletes decide to
continue or leave sport. Athletes' decisions to remain involved in or to withdraw from
sport is based on the perceived beneﬁts and cost he or she experiences in the situation
(Gould & Petlichkoff, 1988). While Gould and Petlichkoff‘s (1988) model identiﬁes
important variables which lead to this cost-beneﬁt analysis, the situational variables
identiﬁed in the model including social support, coaching styles, anxiety and coping
skills, have gone largely unexamined in the sport participation literature.

Furthermore, although Gould and Petlichkoffs (I988) model does not suggest
how variables interact to lead to changes in sport participation status beyond the cost-
beneﬁt analysis, their model does provide a framework for studying variables which
interact to inﬂuence sport participation. Speciﬁcally, Gould and Petlichkoffs (1988)
model ﬁrst identiﬁes surface-level reasons along with the underlying theoretical
motives for youth sport participation and withdrawal. These reasons for participating
and withdrawing from sport are also supported within the sport participation literature
as reasons for participation and withdrawal. In addition, the cost-beneﬁt analysis
included in Gould and Petlichkoffs (I988) model suggests that an interaction between
the reasons for participation and withdrawal occurs. Unfortunately, their model does

not suggest how the various reasons for participation and withdrawal interact to lead to

33

continued participation or withdrawal. Speciﬁcally, what has not been examined is
how these reasons relate to psychological variables such as anxiety and coping, social
variables such as social support and environmental variables such as coaching
behaviors to inﬂuence sport participation decisions.

Finally, Weiss and Petlichkoff (1989) identiﬁed several missing links which
needed to be addressed to obtain a better understanding of the youth sport participation
process including further exploring the particular reasons for attrition, asking the
athletes in more depth about their sport experiences and reasons for withdrawal, the
development of longitudinal studies, and gaining an understanding of the social
structure of sport. By incorporating the recommendations from Weiss and Petlichkoff
(I989), understanding how variables interact to inﬂuence youth sport participation
decisions can be achieved. Speciﬁcally, by addressing some of the missing links in
sport participation, research studying the interaction of variables inﬂuencing youth
sport participation can be used to build on the work of Gould and Petlichkoff (I988).
By examining youth sport participation in this manner, a better understanding of how
several variables interact to inﬂuence youth athletes' decision to continue or withdraw

from sport can be gained.

CHAPTER III

Method

Subiects and Design

The participants in this study included 132 female youth sport athletes of which
62.6% were from Texas and 37.4% from Michigan. The participants ranged between
the ages of 9 and 14 (M=11.61, §_l_)_=1.40 ). The gymnasts had been with their current
coach an average of 3.66 years (SL351 .95). Of the 132 gymnasts who volunteered to
participate in the study, 131 gymnasts participated at Time 1 of the data collection,
126 gymnasts participated at Time 2 and 120 gymnasts participated at Time 3. One
gymnast who agreed to participate in the study was repeatedly absent during the data
collection times and was subsequently dropped from the study. Of the remaining 131
participants who began in the study, 17.5% of the participants or 23 gymnasts
discontinued their gymnastics participation during the course of the study. A more
detailed discussion of these gymnasts will be provided later in the results section. A
total of nine clubs participated in the study while more than 20 were contacted for
participation. Due to exaggeration by some coaches of the number of gymnasts who
met the criteria for participation and to unwillingness of some clubs to participate, the
total number of participants was slightly lower than had been anticipated. However,
this lower number did not affect the subsequent analyses.

The gymnasts in the study were participants in the sports of rhythmic and

34

35

artistic gymnastics. The years of gymnastics participation of the participants ranged
from 2 to 12 years (M=6.31, SQ=2.31). The athletes were training and/or competing
in their sport for at least 6 months of the year. The ethnic background of the
gymnasts was predominantly European-American (E=IO3). In addition, several other
ethnic backgrounds were represented among the gymnasts including African-American
031:1), Mexican-American (ﬂ=2), Asian-American (ﬂ=1), and Other (11:24).
Permission to participate in this study was obtained from the coaches, parents and
athletes. The athletes were given the opportunity to discontinue their participation in
the study at any time without penalty.

Instrumentation

A total of ﬁve instruments were administered during this study. These
included the athlete's coping skills in sport, two measures of perceptions of coaching
behaviors, social support, and anxiety in sport. All inventories were modiﬁed to be
sport speciﬁc. For example, if a questionnaire utilized the words "competitive event",
for the gymnasts the words "competitive event" were changed to the word "meet".
The psychometric properties, including internal consistency analyses, of the
instruments used in this study are provided in the following paragraphs.

Athletic Coping Skills Inventory-28. Smith, Schutz, Smoll and Ptacek (1995)
developed this 28 item coping inventory (see Appendix A). The Athletic Coping
Skills Inventory-28 (ACSI-28) has seven speciﬁc subscales: (a) Coping With
Adversity, (b) Peaking Under Pressure, (c) Goal Setting/Mental Preparation,

(d) Concentration, (e) Freedom From Worry, (0 Confidence and (g) Achievement

36

Motivation and Coachability. Statements are rated on a 4-point Likert scale with
response choices ranging from almost never to almost always. A Personal Coping
Resources score can be obtained by summing the subscales. Only the Coping with
Adversity subscale will be used because it most closely addresses the area of coping
that is of interest for this study. In addition, the Personal Coping Resources score.
which can be obtained with the ACSI-28, is fairly strongly correlated with the Sport
Anxiety Scale (r = -.43) which will be used in this study to assess gymnasts' anxiety.
The Coping with Adversity subscale had a lower correlation with the Sport Anxiety
Scale (r = -.29 ).

The coping with adversity subscale has demonstrated adequate validity and
reliability for both team and individual sports (Smith, Schutz, Smoll & Ptacek, 1995).
The coping with adversity subscale demonstrated adequate reliability (r = .75).
Convergent and discriminant validity was demonstrated when the relationship between
coping with adversity subscale and other coping scales including Rosenbaum's (1980)
Self-Control Schedule, the Ways' of Coping Checklist (Vitaliano, Russo, Carr, Maiuro,
& Becker, 1985), the Mental Health Inventory (Veit & Ware, 1983), the Washington
Self-Description Questionnaire (Smoll, Smith, Barnett, & Everett, 1993), and Coppel's
(1980) Self-Efﬁcacy Scale were examined.

_C_oa¥ching Bethvior. To assess athlete's perceptions of their coaches' behaviors,
two inventories were selected including the Leadership Scale for Sports (LSS)
(Chelladurai & Saleh, 1978; 1980) (see Appendix B) and the questionnaire version of

the Coaching Behavior Assessment System (QCBAS) (Amorose & Horn, in press;

37

Hour & Glenn, 1988) (see Appendix C). These two questionnaires were used because
the LSS provides a general measure of leadership style whereas the QCBAS provides
a more speciﬁc measure of coach's behavior with regard to feedback patterns.

The LSS contains 40 items representing ﬁve dimensions of leader behavior in
sport. Each item is scored on a ﬁve-point Likert scale (always to never) and the
athlete is asked to indicate to what extent her/his coach exhibits that particular type of
behavior. The ﬁve subscales included in the LSS are Democratic and Autocratic
Styles which measure the coach's decision making style, Social Support and Positive
Feedback which measure the coach's motivational tendencies, and one subscale which
measures the coach's instructional behavior (Training and Instruction).

The stability of the ﬁve subscale structure has been supported through
independent factor analyses. In addition, admissible levels of internal consistency and
test-retest reliability were obtained across several samples of students and athletes.
Cronbach's alpha was used to assess intemal consistency and the following scores
were obtained for each of the subscales: Training and Instruction: r = .87; Democratic
Behavior: r = .84; Social Support: r = .82; Positive Feedback: r = .87; and Autocratic
Behavior: r = .37. All the coefﬁcients except the Autocratic Behavior subscale for the
LSS exceeded the .65 criterion alpha level advocated by Nunnally (1978). For the
Autocratic subscale, item contributions to each subscale's alphas were examined to
understand each item's contribution to the internal consistency of the subscales. While
deleting one of the items would raise the alpha coefﬁcients, deleting the item would

not raise the alpha enough to reach the acceptable level. Thus, the autocratic subscale

38
will not be used in the subsequent analyses.

The QCBAS is made up of 16 items that represent different types of coaches'
feedback responses. Athletes are asked to rate how typical it is for their coach to give
them that kind of feedback in practices and games. The questionnaire is a ﬁve-point
scale where athletes rate their responses from Very Typical to Not All that Typical.
Eight different feedback categories are represented by the 16 items. Seven of the
categories correspond to those identiﬁed in the original CBAS (Smith, Smoll, & Hunt,
1977). An additional category (reinforcement combined with technical instruction) is
included in the QCBAS. This category was identiﬁed in an observational study by
Horn (1985). Therefore, of the eight categories, three of these categories relate to the
types of feedback coaches provide to players' performance successes
(praise/reinforcement, non-reinforcement, reinforcement combined with technical
instruction). The other ﬁve categories pertain to the feedback coaches provide during
players' performance errors (mistake contingent encouragement, ignoring mistakes,
corrective instruction, punishment, and corrective instruction combined with
punishment). The reliabilities are reported in the results section for this inventory.

Socizi Support. Gymnasts' perceptions of the social support from within the
sport environment and from outside the sport environment were assessed. This
questionnaire, designed by the author, asks the gymnasts to identify who provides
them with support in gymnastics (see Appendix D). The objective is to assess
differences between support utilized from within the sport environment and from

outside the sport environment. There are seven questions that refer to family/friends

39

providing support and seven questions that refer to people in gymnastics providing
support. The responses are measured on a four-point Likert scale and the participants
chose from responses ranging from "not important support" to "extremely important
support." The questionnaire was pilot tested prior to administration. The questionnaire
is based on the work of Antonucci (1986), Levitt, Weber, and Clark (1986), and
Sternberg and Grajek (1984). Internal consistency analyses were conducted on the
subscales. The following reliabilities were obtained for family support r = .69. The
reliabilities for sport support were r = .71.

Sport AnxieﬂLmﬁ. Smith, Smoll, and Schutz (1990) developed the Sport
Anxiety Scale (SAS) (see Appendix E). This scale contains 21 items and responses
are measured on a four-point scale (Not At All to Very Much So). The questionnaire
contains three subscales: Somatic Anxiety and two classes of cognitive anxiety: Worry
and Concentration Disruption. The subscales can be totaled for an overall anxiety
score. For the purposes of this study the total anxiety score was used because sound
reasoning within the sport participation and psychology literature does not exist for
examining anxiety based on the subscales and because individuals respond differently
to anxiety.

For the participants in this study, internal consistency analyses were examined
and a reliability of .92 was obtained. Exploratory and conﬁrmatory factor analyses,
performed by the authors, demonstrated a stable factor structure. Convergent,
discriminant and construct validity were also assessed. SAS demonstrated convergent

validity due to its high correlation with the Sport Competition Anxiety Scale (Martens,

40

1977), r =.81, and the somewhat lower correlation with the STAl Trait scale
(Spielberger, Gorsuch, & Lushene, 1970), r = .48. Furthermore, when examining the
SAS subscales with the SCAT (Somatic Anxiety, r = .80; Worry, r = .66;
Concentration Disruption, r =.47), the correlations suggest that the SCAT is primarily
a measure of somatic anxiety and not an adequate measure of the cognitive dimensions
of sport-speciﬁc anxiety (Smith, Smoll, & Schutz, 1990). Also, the SAS demonstrated
construct validity when it successfully predicted tensions and confusion responses
within a stressful precompetitive setting as well as when it discriminated between
groups of athletes who differed in performance level.
Procedure

Permission to conduct this study was obtained from all the coaches, parents and
athletes prior to the beginning of data collection. A letter was sent to the coaches of
all prospective clubs to explain the purpose of the study. A follow-up phone call was
placed to the coaches to answer any questions and to request if they would be willing
to have their gymnasts participate in the study. After obtaining permission from the
coaches to have their athletes participate in the study, the parents were contacted to
obtain permission for their child to participate in this study. The nature of the study
was explained to the parents and they were asked to complete a permission form (see
Appendix F) which allowed their child to participate in the study. Those athletes who
participated in the study were guaranteed conﬁdentiality of their responses through the
use of a code number for each subject. Furthermore, permission to conduct the study

was obtained from Michigan State University's Human Subjects Committee prior to

41

any data collection (see Appendix G). All athletes were given the option to not
participate in the study and were allowed to discontinue their participation at any time
during data collection without penalty. The author administered the inventories in
Texas while three trained assistants administered the inventories in Michigan.

The data collection took place at the training facilities of the youth sport
athletes. Arrangements were made with the coaches to have the athletes complete the
inventories. All of the inventories were administered three times. The ﬁrst two times
the inventories were administered, they were administered at approximately one month
intervals. Initially, the gymnasts were given the inventories during their pre-
competitive and early competitive gymnastics season (see Table l). The second time
the inventories were administered to the gymnasts was during their early competitive
and competitive seasons. Finally, the inventories were administered to the gymnasts
who were still participating near the end of the competitive season. Each data
collection session took approximately 15-20 minutes and all the inventories were
administered at one time. The instruments were counterbalanced each time they were
given to the participants. The athletes also completed a demographic questionnaire
(see Appendix H). During the three months following completion of the data
collection, each of the clubs was contacted to see if any gymnasts had discontinued
participating in gymnastics.

The gymnasts who discontinued their gymnastics participation were contacted

by phone shortly after they left gymnastics. They were asked to complete the set of

 

Table 1

Data Collection Schedule

— ——— ——_—.—— _—_——— _

August through November - Skill Acquisition / Prepare Routines

Januaﬂ - Pre-competitive Season/Early Competitive Season

Time I ---
Data Collection

February - Early Competitive Season

Time 2 -----
Data Collection _l\_/I__arch through May - Competitive Season

Time 3 ----- April - Almost the End of the Competitive Season
Data Collection

End of May - Competitive Season Ends

3 Month Follow-up June, July aid August - Gyms contacted for Dropouts

 

43

inventories one ﬁnal time and to complete an open-ended set of questions (see
Appendix I) which asked about their decision to leave gymnastics. The questionnaires
were sent to the gymnasts with a return envelope. If the former gymnasts did not
return the questionnaires within two weeks, a follow-up phone call was placed to make
sure they had received the packet. They were asked if they would return the
completed questionnaires. Eight of the gymnasts answered the open-ended questions
directly over the phone. In total, 19 of the 23 former gymnasts, or 83%, completed
the open-ended questionnaires.
Pilot Study

A pilot study was conducted to examine whether the questionnaires and
instructions were clear and understandable. The participants in the pilot study were 1 1
female gymnasts who ranged between the ages of 10 and 14 and attended clubs that
did not participate in the rest of the study. Overall, the gymnasts did not have any
difﬁculty understanding the questionnaires’ items or following the instructions. Based
on the results it was determined that the questionnaires to be used in this study were
appropriate for the population being studied.

Merit of the Data

The data of the present study were analyzed via the Generalized Estimating
Equations (GEE) procedure. In the following paragraphs an overview of GEE is
provided. Finally, the rationale for the implementation of the GEE analysis within this
study is presented.

Investigations where the participants are studied over a period of time represent

44

an important but often overlooked strategy in understanding youth sport participation
(Weiss & Petlichkoff, 1989). One of the advantages of conducting investigations with
repeated observations is that researchers are able to examine individual change over
time (Duncan, McAuley, Stoolmiller, & Duncan, 1993). Unfortunately, standard
multivariate statistical methods often perform poorly in settings that incorporate
correlated observations, non-normally distributed data, time-varying covariates, missing
observations and other characteristics common in longitudinal data (Duncan et al.,
1993). To deal with the limitations often encountered when using standard
multivariate statistical methods with repeated measures data, more widely applicable
approaches to parameter estimation or modeling are required (Ware, I985). Liang and
Zeger (1986) developed generalized estimating equations (GEE) which provides an
approach for parameter estimation.

GEE is based on the generalized linear models (GLM) framework of
McCullagh and Nelder (1989). Speciﬁcally, GEE extends GLM to handle correlated
observations. Most of the techniques typically used for multivariate statistical analyses
are special applications of the GLM framework (Tabachnick & F idell, 1996).
Furthermore, the GLM framework has already been established as a powerful and
ﬂexible analytic format (Duncan et al., 1993). Examples of methods included in GLM
are multiple regression, ANOVA, ANCOVA, discriminant function analysis, logistic
regression, canonical correlation, MANOVA, MANCOVA, proﬁle analysis, and
structural equation model. GLM produces consistent parameter estimates as long as

several criteria are met, including that the model is correctly speciﬁed in terms of the

45

covariates included and their functional relation to the outcome, and the relation
between the mean and variance of the residuals is correctly speciﬁed (Duncan et al.,
1993)
Duncan et al. (1993) provided the following example for the relationship
among classic linear regression, GLM and GEE:
"For example, in classic linear regression, statistical inference proceeds under
the assumption that the residuals are normally distributed with constant
variance; but in GLM only the assumption that the variance of the residuals is
constant and independent of the mean is necessary. GLM, however, assumes
independent observations. If the level of correlation between observations is
appreciable as it usually is for repeated measures, inferences drawn from
statistical procedures that ignore the correlation can be quite misleading.
Liang and Zeger (1986) have shown that GEE produces consistent estimates
and robust test statistics even when the exact form of the correlation among
observations is misspeciﬁed. Test statistics are most efﬁcient, however, when
the postulated correlation structure is close to the true structure."
Through this example it is possible to see how GEE has developed and the strength of
using GEE in repeated measures analyses.
One of the primary reasons for using GEE in this study was because of the
correlated observations that emerge through repeated measures over time. Through
GEE, the GLM framework is extended to handle correlated observations (Duncan,

McAuley, Stoolmiller, & Duncan, 1993). As stated in the procedure, gymnasts will

46

complete the questionnaires three times over the course of the gymnastics season.
Thus, it is necessary to have a procedure that can handle correlated observations. By
utilizing GEE, it is possible to account for the fact that there is dependence among the
dependent variables.

Furthermore, GEE is useful for analyzing either discrete or continuous
longitudinal data and can be used to analyze non-normal longitudinal data (Duncan,
Duncan, Hopes, & Stoolmiller, 1995). In this particular study both discrete and
continuous dependent variables were used. For example, whether or not gymnasts
continue or discontinue participating in gymnastics is the discrete dependent variable
and the intended participation as measured by the future participation question is the
continuous dependent variable.

The fact that GEE can examine non-normal longitudinal data is a particular
beneﬁt for the current study because of the limitations in other statistical techniques in
handling repeated measures and dichotomous outcomes. Furthermore, there is no
constraint within GEE regarding number of variables or number of observations.
Incomplete data can be analyzed and the examination of unequal time cells is possible.
According to Zeger and Liang (1986), GEE‘s approach to modeling longitudinal data
requires the following: (a) speciﬁcation of the link formation - an expression relating
the expected value of the dependent variable to a linear function of the covariates; (b)
description of the outcome variance as a function of the outcome mean; and (c)
designation of a "working correlation matrix" for the set of responses from each

subject.

47

The analyses of the predicted relationships would be conducted by running
each of the research questions through GEE. The iterative operations that would be
performed by GEE on the predicted comparisons include:

1) Obtaining initial estimates for regression coefﬁcients. These are unweighted
least squares regression coefﬁcients. These estimates assume
independence of the data.

2) Given the current estimate of regression coefficients, the working
correlation matrix for the dependent variables is estimated.

3) Given (1) and (2) the regression coefﬁcients are updated taking into account
the data are not independent. The coefﬁcients are weighted based on
correlations among the dependent variables. For example, in this study
anxiety is measured three times over a 6 month period. The procedure
would take these highly correlated data into account when
calculating the regression coefﬁcients.

4) Finally, a continuous iterative process (3) occurs until the solution
converges and/or maximum iterations have been reached.

In conclusion, GEE is believed to be the most appropriate method to answer
the research questions identiﬁed in this study. Speciﬁcally, by utilizing GEE analyses
correlated observations which emerge in repeated measures analyses will be accounted
for and the richness of repeated measures data will be appropriately analyzed.
Through GEE analyses the researcher will be able to examine how and which

variables are able to predict changes in the participation status of the gymnasts. The

48

interpretation of the ﬁndings would be based on which variables or combination of
variables are better able to identify those gymnasts who continue or discontinue their

gymnastics participation.

CHAPTER IV
Results

The results for this study are presented in three sections. In the ﬁrst section, results
from the preliminary analyses are presented. The preliminary analyses of the data included
examining internal consistency of the measures, Multivariate Analysis of Variance to examine
for differences in responses by state of residence, and factor analyses on the QCBAS. In the
second section of this chapter the results from the Generalized Estimating Equations (GEE)
analyses are presented. The GEE analyses examined the previously stated research questions.
The ﬁnal section will address the ﬁndings from the open-ended questions from the gymnasts
who discontinued their participation.

Prelimiryary Analvses

Internal Cons_ist_enc1Anst_e§

For this population, the reliability of the inventories was examined. Internal
consistency measures of reliability were computed for all the inventories and subscales except
the QCBAS inventory by calculating Cronbach's (1951) alpha coefﬁcients. The results are
presented in the method section along with where the inventories are identiﬁed. The internal
consistencies were not reported for the QCBAS in this section. A factor analysis was
conducted on the QCBAS to ﬁrst determine the underlying structure of the gymnasts'
perceptions of their coaches' feedback. The internal consistencies for the factors are reported

in the section reporting the ﬁndings of the factor analysis.

49

50
Multivariate Analysis of Variance

A preliminary multivariate analysis was conducted to examine if gymnasts'
responses differed according to their state of residence. A one-way multivariate
analysis of variance (MANOVA) was performed on 12 dependent variables: anxiety,
coping, family support, sport support, the LSS subscales (democratic, social support,
positive feedback, training and instruction) and the QCBAS factors which were
identiﬁed in the factor analysis (see Factor Analysis section). The independent
variable was state (Texas, Michigan). The multivariate analysis was not signiﬁcant,
Wilks, Lambda=.88, F(l 1,1 19) = 1.47, p < .15. Consequently, all subsequent analyses
were combined for gymnasts from Texas and Michigan. See Appendix K for means
and standard deviations.

Factor Analyses

According to Smoll, Smith, Curtis and Hunt (1978) and Amorose and Horn (in
press), a principal axis factor analysis should be conducted on the QCBAS to
determine the structure underlying gymnasts' perceptions of their coaches' feedback.
Although the QCBAS contains items which represent eight different feedback
categories, according to Smoll et al. (1978) and Amorose and Horn (in press) the
items should not be merely totaled for each of the feedback categories. Rather, a
factor analysis is done to extract the combination of these feedback categories to
identify how athletes' perceive their coaches' feedback and to weight each item's
contribution to the resulting construct. Thus, principal axis factor analyses were

conducted on the scores obtained from the administration of the QCBAS for each data

51

collection time.

A principal axis factor analysis was conducted on the data from Time 1, Time
2, and Time 3. For each factor analysis, factor scores were calculated for use in the
GEE analyses. The three different factor analyses were conducted due to the fact that
we were primarily concerned with how changes over time in the gymnasts' perceptions
of coaching behaviors contributes to gymnastics participation. Using the factor scores
in the GEE, which is different than the scalar scores we typically use, allows for the
different weighting of the items to be reﬂected by the factor loadings. We were not
concerned with a single measure of coaching behaviors. The subsequent GEE
analyses allowed us to examine these changes over time. Following the identiﬁcation
of the factor structure, internal consistency of the factors were assessed.

Three interpretable factors were extracted at Time 1, Time 2, and Time 3
accounting for 49.7%, 50.6%, and 54.5% of the total variance, respectively (see Table
3 for factor loadings and variance explained by each factor for Time I). The criteria
for extracting factors were a minimum eigenvalue of 1.0, and varimax rotation resulted
in the identiﬁcation of three conceptually distinct factors. The criteria for item
inclusion on a particular factor were a minimum loading of .40 on that factor to label
the factor. There were two uses of the data: (a) to label the factor to describe the
items which load in a similar manner and (b) to compute factor scores for use in the
subsequent GEE analyses. At Time 1 and Time 3 a couple of the items cross-loaded.
Typically, items which cross load would be eliminated from the factor analyses and

the data would be reanalyzed. However, for our purpose items which cross load were

Table 2

52

QCBAS Factor Aggvsis for Time 1: Factor Loadings and Variance Explained

 

Time 1
Factors and Items:

"Good pass."
"Way to go! You really extended
your legs that time."

Positive and
Informational
Feedback
0.64

0.72

"Great routine! Now you're keeping

your toes pointed."
"Excellent practice today."
"That's OK, Keep working at it."
"Hang in there! You'll do better
next time."

"That was a really stupid mistake."

"Your technique looks lousy!
Keep your head up."

"That performance sucked."

"No, that's not right, you need

to work on a faster release"

"How many times have I told
you to extend your legs?"
"You dropped your elbow.
Next time keep it up."

"Coach ignores good performance"

"Coach doesn't say anything
about your performance."

"Coach ignores your error or poor
performance."

"Coach doesn't say anything to
you about your error or
poor performance."

Eigenvalue
Percent Variance
Reliability

0.78
0.58
0.21

0.36
-.10

-.21
-.26

.08

-.08

-.31
-.44

-.55

-.02

-.11

4.81
30.1
0.82

Factor Loadings

Punishment-
Oriented
Feedback
0.01

-. l9

-. l 7

-.08

-. 12

-. 16
0.92

0.49
0.57

0.21
0.53

-.09
.20

.19

.01

.12

2.01
12.6
0.88

Non-Reinforcement/
lgaoring Mistakes

-.08
-.20
-.04
-.05
-.ll

-.09
0.11

0.13
.17

0.09
0.02

-.07
0.50

0.52

0.69
0.72

1.12
7.0
0.73

53

retained because their factor loadings are primarily being used to compute factor
scores and only generally used to label a factor. When examining the items which
cross load, depending on how the athlete interprets the statement (for example, "coach
doesn't say anything about your performance"), it makes intuitive sense that it could be
interpreted as positive and informational feedback or non-reinforcement/ignoring
mistakes feedback. These criteria and procedures are consistent with Amorose and
Horn (in press). The total variance each factor accounted for and an explanation of
how the factors were labeled is provided below.

For Time 1, factor 1 accounted for 30.1% of the total variance and was labeled
Positive and Informational Feedback. Items that related to feedback coaches provide
athletes following performance successes included statements such as "Way to go!
You really extended your legs that time.", "Excellent practice today", and "Good
Pass!". Factor 2 accounted for 12.6% of the total variance and was labeled
Punishment-Oriented Feedback. The items on this factor included the following
statements "That was a really stupid mistake." and "Your technique looks lousy! Keep
your head up." These items related to feedback coaches provide athletes following
player performance errors. The ﬁnal factor related to feedback which coaches provide
following player performance successes and errors and accounted for 7.0% of the total
variance. This factor was labeled Non-reinforcement/Ignoring Mistakes. Items that
were included on this factor were statements such as "Coach ignores good
performance." and "Coach doesn't say anything to you about your error or poor

performance." The items that loaded highest were used to label the construct.

54

For Time 2, factor I accounted for 31.3% of the total variance, factor 2
accounted for 12.5% of the total variance and factor 3 accounted for 7.4% of the total
variance. See Table 3 for factor loadings and variance explained by each factor for
Time 2. For Time 3, factor 1 accounted for 31.4% of the total variance, factor 2
accounted for 15.2% of the total variance and factor 3 accounted for 7.7% of the total
variance (see Table 4 for factor loadings and variance explained by each factor for
Time 3). The factor structures from Time I to Time 2 to Time 3 included almost
identical items. Items with factor loadings below .40 were included on the factors for
Time 1, Time 2, and Time 3. This was done to provide consistency in reporting
because factor scores were calculated which uses the factor loadings as weights for the
item's contribution to the factor score. Due to the fact most of the items loaded
consistently across time, the same labels were kept for each of the factors.

Internal consistency measures of reliability were computed for the
QCBAS subscales by calculating Cronbach's (1951) alpha coefficients (see Table 5).
The alpha coefﬁcients for Time 1, Time 2, and Time 3 were .82, .80, and .90 for
Positive and Informational Feedback, .88, .83, and .83 for Punishment-Oriented
Feedback, and .73, .74, and .81 for Non-reinforcement/Ignoring Mistakes, respectively.
All of the reliabilities met acceptable levels advocated by Nunnally (1978). All of the
alpha coefficients changed slightly over time. These changes may be due to the
decrease in the number of gymnasts during each time period. Consequently, there was
a slightly different population on which the internal consistency analyses were

conducted. As stated previously, these slight changes have been found acceptable by

Table 3

55

QCBAS Factor Analvsis for Time 2: Fagtor Loadings and Variance Explained

 

Time 2
Factors and Items:

"Good pass."
"Way to go! You really extended
your legs that time."

Positive and Punishment-

F actpr Lpadings

Informational Oriented

Feedback

0.36

0.74

"Great routine! Now you're keeping

your toes pointed."
"Excellent practice today."
"That's OK, Keep working at it."
"Hang in there! You'll do better
next time."

"That was a really stupid mistake."

"Your technique looks lousy!
Keep your head up."

"That performance sucked."

"No, that's not’ right, you need

to work on a faster release"

"How many times have I told
you to extend your legs?"
"You dropped your elbow.
Next time keep it up."

"Coach ignores good performance"

"Coach doesn't say anything
about your performance."

"Coach ignores your error or poor
performance."

"Coach doesn't say anything to
you about your error or
poor performance."

Eigenvalue
Percent Variance
Reliability

0.68
0.53
0.71

0.65
-. I9

-.15
-.31

0.17

-.13

0.29
-.11

-.33

-.18

-.08

5.02
31.3
0.80

Feedback
-.03
0.07
0.03
-.19
-.10

-.22
0.67

0.76
0.64

0.53
0.60

0.36
0.34

0.37

.08

-.02

1.99
12.5
0.83

Non-Reinforcement/
Igporing Mistakes

-.62
-.49
-.31
-.37
-.06

-.08
0.23

0.16
0.29

0.06
0.03

-.16
0.71

0.62

-.02
0.25

1.19
7.4
0.74

56

Table 4
QCBAS Factor Analysis for Time 3: F aLctor Loadings and Variance Explained

 

 

Tim; 3 Factor Laagings

Factors and Items:

Positive and Punishment:
Informational Oriented Non-Reinforcement]
Feedback Feedback Igppring Mistakes

"Good pass." 0.74 -.15 -.24
"Way to go! You really extended

your legs that time." 0.85 0.02 -.13
"Great routine! Now you're keeping

your toes pointed." 0.72 0.06 -.09
"Excellent practice today." 0.54 -.07 -.17
"That's OK, Keep working at it." 0.75 -.08 0.01
"Hang in there! You'll do better

next time." 0.69 -.15 0.09
"That was a really stupid mistake." -.22 0.65 0.26
"Your technique looks lousy!

Keep your head up." -.31 0.74 0.10
"That performance sucked." -.31 0.59 0.31
"No, that's not right, you need

to work on a faster release" 0.26 0.55 0.01
"How many times have I told

you to extend your legs?" -.08 0.71 0.10
"You dropped your elbow.

Next time keep it up." 0.40 0.49 -.03
"Coach ignores good performance" -.46 .33 0.43
"Coach doesn't say anything

about your performance." -.41 0.38 0.52
"Coach ignores your error or poor

performance." -.05 0.05 0.83

"Coach doesn't say anything to
you about your error or

. poor performance." -.01 0.16 0.76
Eigenvalue 4.81 2.01 1.12
Percent Variance 31.4 15.2 7 7

Reliability 0.89 0.83 0.81

57

 

Table 5

Internal Consistencv Analyses for OCBAS: Cronbach's Alpha Coefﬁcients

 

Cronbach's Alpha Coefﬁcients

£11111 11.11122 Lima
Positive and Informational Feedback .82 .80 .90
Punishment-Oriented Feedback .88 .83 .83
Non-Reinforcement/lgnoring Mistakes .73 .74 .81

 

researchers (Bohmstedt, 1970; Carmines & Zeller, 1989).

Finally, the factor pattern weights obtained from the factor analyses were used
to calculate three factor scores for each time period for each gymnast in the study.
These factor scores were then used as a measure of the gymnast's perceptions
concerning their coaches' feedback style in the subsequent analyses.

Generalized Estimating Equations

In this section the results of the Generalized Estimating Equations (GEE)
analyses are presented. To build on the treatment of the data section presented in
Chapter 3, an overview of the procedure for running the GEE analyses is provided.
This will be followed by the format in which the GEE results are presented. In
addition, the research questions for this study are re-stated. Finally, the GEE analysis

results for each of the research questions are presented.

58

GEE Procedure. The procedure for conducting GEE analyses is described
below. First, the correlations of the independent variables are computed. These
correlations collapsed across time represent interindividual associations of the
measures. Then the correlation among the dependent variables collapsed over time is
calculated. This correlation is used in calculating and updating the beta weights
between the independent or predictor variables and the dependent measures.

Then, the initial multivariate model, which is based on the variables identiﬁed
in the research questions, is speciﬁed. The initial multivariate model includes all
possible interactions and main eﬂ‘ects of the variables identiﬁed in the model. The
model is estimated and the signiﬁcance of the model is examined. If the entire model
is not signiﬁcant, then a backwards elimination procedure is employed. The
backwards elimination is used to eliminate the least signiﬁcant interaction term and the
model is then re-estimated. A full factorial GEE analysis with a backwards
elimination procedure is necessary for examining both main and interaction effects of
the variables which inﬂuence gymnasts' actual participation or intended participation.

For example, in research question number one, the model incorporates four
independent variables. The model is estimated and if the 4-way interaction is not
signiﬁcant, backwards elimination is employed and the 4-way interaction is eliminated.
Then, the model is re-estimated with all the 3-way interactions, 2-way interactions and
main effects and the 3-way interactions are examined for signiﬁcance. If all the 3-
way interactions are not signiﬁcant, then the least signiﬁcant 3-way interaction is

eliminated and model is re-estimated. If a signiﬁcant interaction emerges then that

59

signiﬁcant interaction is kept but the backwards elimination procedure continues for
the 2-way interactions and main effects not involved in the 3-way interactions.

The backwards elimination procedure is continued until only the signiﬁcant
interactions (if there are any) are kept in the model and any lower order interactions
and/or main effects which are not signiﬁcant and do not contribute to the higher order
interaction are removed from the model. If there are not any signiﬁcant interactions,
then the backwards elimination procedure is conducted for the main effects where the
least signiﬁcant main effect is removed until only signiﬁcant main effects remain in
the model. The estimated regression coefﬁcients and associated z-statistics for the
ﬁnal model are reported.

Finally, if there are signiﬁcant interactions, the simple effects are computed for
the signiﬁcant interactions. To determine the nature of the interaction effects, using
the ﬁnal model that produces a regression equation, simple effects are calculated using
standard techniques (Aiken & West, 1991). The procedure explained by Aiken and
West (1991) serves as the basis for the interpretation of the interaction. Calculating
the simple effects makes it possible to interpret the signiﬁcant interactions.

Research Questions. Two research questions were previously identiﬁed for this
study. Based on these research questions, which are once again provided below,
several models were tested using GEE analyses.

1. What is the relationship among anxiety, ways of coping, support from
within sport and support from family/friends for female gymnasts who continue and

discontinue their sport participation?

60

2. What is the relationship among perceived coaching behaviors for female
gymnasts who continue and discontinue their sport participation?

GEE Results F ormat. The results of the GEE analyses will be presented in the
following format. First, the correlation matrix for the variables in the model collapsed
across time will be reported. Second, the correlations among the dependent variables
will be reported. The correlation among the dependent variables is reported because
the regression weights, which will be reported next, are based on the correlations
among the dependent variables. Third, the results from the backwards elimination and
the subsequent speciﬁed model will be reported. The results from the speciﬁed model
are composed of the predictors for the dependent variable. Finally, the simple effects
will be reported which allow the signiﬁcant interactions to be examined. The results
for the research questions are provided in the following paragraphs.

GEE Analyses Results

In this section the results from the GEE analyses for each of the research
questions are provided. Each research question is identiﬁed for the section where the
results are presented. The results are reported in the following order. First, the effect
of anxiety, the coping with adversity subscale from the SAS, sport support and family
support subscales from the social support inventory on gymnastics participation are
reported. Then, the results from the effect of perceived coaching behaviors, from the
QCBAS subscales and LSS subscales, on gymnastics participation is provided. The

means and standard deviations for each inventory are provided in Appendix J.

61

The Effect of Anxieg, Coping with Adversity, Sport Support and Family
Support on Gymnastics Participation. Gymnasts were monitored during the study and

for the 3 months following the completion of the questionnaire phase of the study to
see if any gymnasts discontinued participating in gymnastics. The dichotomous
dependent variable was the gymnasts who actually continued or discontinued their
gymnastics participation. The main effects for the analyses were anxiety, c0ping with
adversity, sport support, and family support. The original speciﬁed model was the
interaction between anxiety, coping with adversity, sport support, and family support.
The correlations between anxiety, Coping with adversity, sport support and family
support collapsed across the three time periods for all of the participants, are presented
in Table 6. The correlation among the dependent variables (continued, discontinued
participation) collapsed over time was r = .96.

GEE analysis was ﬁrst used to estimate the following model: the interaction
between anxiety, coping with adversity, sport support, and family support on continued
gymnastics participation. The backwards elimination procedure was employed because
the original model was not signiﬁcant. Following backwards elimination, the ﬁnal
model which emerged included the two-way interaction between anxiety and coping
with adversity, and the main effect for family support (see Table 7). The signiﬁcant
main effect for family support had a negative direction. Speciﬁcally, higher family
support is associated with continuing in gymnastics and low family support is

associated with discontinuing in gymnastics.

62

 

Table 6

Correlations Among Anxiety, Coping with Adversig, Sport Support and Family

Support Collapsed Across the Three Time Periods

 

 

Anxiety Coping Sport Family
Support Support
Anxiety 1.00
Coping -.51 1.00
Sport Support -.20 0.30 1.00
Family Support -.07 0.08 -.45 1.00

 

 

 

Table 7

Estimated Regtession Coefﬁcients and z-Statistics for the Final Models Following
Backwards Elimination of Nonsigpiﬁcant Effects

 

 

B z
intercept -l .562 -6.825
Anxiety .002 .137
Coping -.005 -.470
Family Support -.042 -2.357‘
Anxiety x Coping -.017 -2.3I4"'

 

* Note. z-values of 1.96 or greater are signiﬁcant at p < .05.

 

63

The results for the simple effects are presented in Table 8. The results from
the simple effects revealed that there was a signiﬁcant effect of anxiety on leaving
gymnastics for those with low coping with adversity. The direction of the results is
positive. Thus, these results suggest that for those with low coping with adversity and
higher anxiety are more likely to discontinue participating in gymnastics. In contrast,
for those with low coping with adversity and lower anxiety, the more likely they are to
continue to participate in gymnastics. Furthermore, for those with moderate coping
with adversity and high coping with adversity the effect does not emerge. Thus,
moderate and high coping skills may serve as protective mechanisms and allow

gymnasts with higher anxiety to continue participating in gymnastics.

 

Table 8

Simple Effects in Coping with Adversig by Sport Support for Effects of Anxietv on
Continued Gymnastics Participation Collapsed Across Time

 

 

 

B 2
Effect of Anxiety on:
Low Coping .054 2.017 *
Moderate Coping .002 .137
High Coping -.050 -1.795

 

Note. " z 3 1.96, p < .05.

 

One limitation of these ﬁndings is that when analyzing the simple effects for
the interactions, the signiﬁcant results did not emerge until plus or minus three
standard deviations were utilized. Thus, the signiﬁcant simple effects are not very

representative of this group of gymnasts.

64

The Effect Perceived Coaching Behaviors on Gymnastics Participatign. The

 

effect of perceived coaching behaviors on gymnastics participation was examined.
This was the second research question. Speciﬁcally, perceived coaching behaviors
were assessed using the subscales for the QCBAS and the LSS. Separate analyses
were conducted on the QCBAS subscales on the gymnastics participation and the LSS
subscales on gymnastics participation. The results of the GEE analyses with the
QCBAS subscales are presented ﬁrst and then the results of the GEE analyses with the
LSS subscales are presented second.

QCBAS Subscal_§. The QCBAS subscales were calculated through the
principle axis factor analyses reported earlier in the results. The three QCBAS
subscales that emerged following the factor analyses were (1) positive and
informational feedback, (2) punishment-oriented feedback, and (3) non-
reinforcement/ignoring mistakes. The three subscales served as the independent
variables. Gymnasts were monitored during the study and for the 3 months following
the completion of the questionnaire phase of the study to see if any gymnasts
discontinued participating in gymnastics. The dichotomous dependent variable was the
gymnasts who actually continued or discontinued their gymnastics participation. The
original model that was speciﬁed was the 3-way interaction between positive and
informational feedback, punishment-oriented feedback, and non—reinforcement/ignoring
mistakes. The correlations between positive and informational feedback, punishment-
oriented feedback, and non-reinforcement/ignoring mistakes feedback collapsed across

the three time periods for all of the participants, are presented in Table 9. The

65

correlation among the dependent variables (continued,discontinued participation)

collapsed over time was r = .96.

 

Table 9

Correlations Among Positive and Informational Feedback, Punishment-Oriented
Feedback and Non-Reinforcement/Igporing Mistakes Collapsed Across the Three

Time Periods

 

 

 

Positive and Punishment- Non-reinforcement/
Informational Oriented Ignoring
Feedback Feedback Mistakes

Positive and

Informational

Feedback 1.00

Punishment-

Oriented

Feedback -.18 l .00

Non-Reinforcement/

Ignoring Mistakes -.42 -.11 1.00

 

 

GEE analysis was used to estimate the following model: the interaction
between positive and informational feedback, punishment-oriented feedback, and non-
reinforcement/ignoring mistakes feedback on continued gymnastics participation. The
results of the ﬁrst model estimation revealed that the three-way interaction was not
signiﬁcant. Thus, the backwards elimination technique was employed. The two-way

interaction between punishment-oriented feedback and non-reinforcement/ignoring

66

mistakes feedback was signiﬁcant. The results from the speciﬁed model are presented

in Table 10.

 

Table 10

Estimated Regtession Coefﬁcients and z-Statistics for the Final Models Following
Backwards Elimination of Nonsigpiﬁcant Effects

 

 

B z
intercept -I .5 50 -6.751
Punishm ent-Ori ented Feedback 0.002 0.272
Non-Reinforcement/ Ignoring Mistakes 0.000 -.071
Punishment- Non-Reinforcem ent/
Oriented x Ignoring Mistakes
Feedback -.01 1 -2 .774’

 

’ Note. z-values of 2.58 or greater are signiﬁcant at p < .01.

 

The results for the simple effects are presented in Table 11. The results from
the simple effects revealed several things including that there was a signiﬁcant effect
of punishment-oriented feedback on continued gymnastics participation for those
gymnasts with perceptions of coaches providing low non-reinforcement/ignoring
mistakes feedback. The direction of the effect is positive thus gymnasts who perceive
their coaches to provide low amounts of non-reinforcement/ignoring mistakes feedback
and who perceive their coaches to provide high amounts of punishment oriented

feedback, will be more likely to discontinue their gymnastics participation. In

67

contrast, if these same gymnasts perceive their coaches to be low in providing
punishment-oriented feedback, they will be more likely to continue their gymnastics

participation.

 

Table l 1

Simple Effects of Non-Reinforcement/Igroring Mistakes for Effects of Punishment-
Oriented Feedback on vanastics Participation

 

 

B 2
Effects of Punishment-Oriented Feedback on:
Low Non-Reinforcement/Ignoring Mistakes 0.024 2354*
Moderate Non-Reinforcement/Ignoring Mistakes 0.002 0.272
High Non-Reinforcement/Ignoring Mistakes -.020 —l .641

 

Note. "‘ z 3 1.96, p < .05.

 

A limitation with these ﬁndings is that when analyzing the simple effects for
the interactions, the signiﬁcant results did not emerge until plus or minus two standard
deviations were utilized. Thus, the signiﬁcant simple effects are not very
representative of this group of gymnasts.

LSS Subscales. The LSS assesses athlete's perceptions of their coaches'
behaviors and contains 40 items that represent ﬁve subscales. Only the following four
LSS subscales were used in the GEE analyses: (a) democratic coach's decision making

style, (b) positive feedback a coach provides, (c) social support a coach provides, and

68

(d) coach's training and instructional behavior. The ﬁfth subscale, an autocratic
coach's decision making style, was eliminated due to low reliabilities (see internal
consistencies section). The four subscales served as the independent variables.
Gymnasts were monitored during the study and for the 3 months following the
completion of the questionnaire phase of the study to see if any gymnasts discontinued
participating in gymnastics. The dichotomous dependent variable was the gymnasts
who actually continued or discontinued their gymnastics participation. The main
effects for the analyses were a democratic coach's decision making style (democratic),
the positive feedback a coach provides (positive feedback), the social support a coach
provides (social support), and a coach's training and instructional behavior (training
and instruction). The original model that was speciﬁed was the 4-way interaction
between democratic, positive feedback, social support, and training and instruction. In
this section the results of the GEE analysis will be presented. The correlations
between democratic, positive feedback, social support, and training and instruction
collapsed across the three time periods for all of the participants, are presented in
Table 12. The correlation among the dependent variables (continued,discontinued
participation) collapsed over time was r = .96.

GEE analysis was used to estimate the following model: the interaction
between democratic, positive feedback, social support, and training and instruction on
gymnastics participation. The results of the ﬁrst model estimation did not reveal a
signiﬁcant four-way interaction. Thus, the backwards elimination technique was

employed. During the backwards elimination, none of the three-way or two-way

69

interactions emerged as signiﬁcant.

 

Table 12

Correlations Among Democratic. Positive Feedlgck. Social SupporLand Training and
Instruction Collapsed Across the Three Time Periods

 

 

Democratic Positive Social Training and
Feedback Support Instruction
Democratic l .00
Positive
Feedback .03 l .00
Social Support .06 .03 1.00
Training and
Instruction -.55 -.25 -.33 1.00

 

 

 

Finally, backwards elimination was employed on the main effects. None of the
main effects emerged as signiﬁcant. Thus, no model was speciﬁed for the LSS
subscales of democratic, positive feedback, social support and training and instruction
on the gymnasts who actually continued or discontinued their gymnastics participation.
No further analyses were conducted because no ﬁnal model emerged. These results
suggest that we are not able to predict gymnasts' participation from these particular
coaching behaviors.

In conclusion, the results for this study revealed several ﬁndings. When

gymnasts' decision to continue or discontinue participating in gymnasts was examined,

70

several variables were identiﬁed which predicted that behavior. For example, the
interaction between anxiety and coping with adversity as well as support from family
and friends inﬂuenced gymnasts participation decisions. In addition, the interaction of
two perceived coaching behaviors (punishment-oriented feedback and non-
reinforcement/ignoring mistakes) inﬂuenced gymnasts decisions to continue or
discontinue gymnastics participation.
Gymnasts Who Discontinued Participation

The 23 gymnasts who discontinued their gymnastics participation were between
the ages of 10 and 13 (M=II.7, S_D = 1.19). The former gymnasts had been with their
current coach an average of 3.65 years (SI;= 2.55). Eighty-seven percent of the
dropouts were Anglo-American and their years of gymnastics participation ranged
from two to ten years (M=5.44, SD = 2.15). The gymnasts who discontinued their
gymnastics participation were contacted to answer a series of questions regarding their
decision to leave gymnastics. Of the 23 gymnasts who discontinued their gymnastics
participation, 19 answered the questionnaires. The former gymnasts were asked how
long it took them to decide to leave gymnastics. Their answers reﬂected that for most
of the gymnasts deciding to leave gymnastics was a process that extended over a
period of time. Speciﬁcally, six of the gymnasts took more than 6 months to decide,
ﬁve of the gymnasts took 3-4 months to decide, ﬁve of the gymnasts took 1-2 months
to decide and two gymnasts decided in 1-2 weeks. Finally, one gymnast had no
choice as the decision to leave gymnastics was made for her due to ﬁnances. She was

told that she must leave gymnastics because the family could no longer afford for her

71

to attend. These results indicate that for most gymnasts, the thought process of
leaving gymnastics is a long one and begins either at the end of the previous
competitive season or around the beginning of the gymnastics season in which they
ultimately leave gymnastics.

The former gymnasts responded to ﬁve open-ended questions. A transcription
of the former gymnasts' responses are presented in Appendix K. The former gymnasts
were asked what they like most about being in gymnastics. The rank order and
frequency percentages of what the former gymnasts liked most about being in

gymnastics are presented in Table 13. The ﬁve most popular areas identiﬁed by the

 

Table 13
What Former Gymnasts Liked the Most About Gymnastics: Rank Order and
Frequency Percentages

 

 

Frequency

Rank Order Percentagas
1) Learning New Skills 53%

2) Teammates/Friends 32%

3) Coaches 21%

3) Competing 21%

5) Traveling to Competitions 16%

6) Challenge 11%

7) Being in Top Physical Condition 5%

7) Performing Something I'm Good At 5%

7) Winning 5%

7) Having fun 5%

7) Support from Family 5%

 

72

former gymnasts included: (a) Learning New Skills, (b) Teammates/Friends, (c)
Coaches and Competing, (tie), and (e) Traveling to Competitions. More than half of
the gymnasts identiﬁed Learning New Skills as something they liked the most about
gymnastics. Thus, the former gymnasts appear to have valued the learning component
in their sport.

The second question asked the gymnasts what they liked least about being in
gymnastics. The former gymnasts identiﬁed several things including: (a) Time
‘ Demands, (b) Coaches, (c) Fear of Event/Injury and Pressure on Self, (tie), (e)
Teammates and Practicing when Tired (tie). The rank order and frequency
percentages are presented in Table 14. Time demands of gymnastics emerged as

something the former gymnasts liked the least about gymnastics.

 

Table 14

What Former Gymnasts Liked the Least About Gymnastics Rank Orer and
Frequency Percentages

 

 

Frequency
Rank Order Percentages
1) Time Demands 53%
2) Coaches 26%
3) Fear of Event/Injury 16%
3) Pressure on Self 16%
5) Teammates 11%
5) Practicing when Tired 11%

7) People with Bad Attitudes 5%

 

73

Gymnasts who discontinued their participation then responded to the question
asking why they left gymnastics. The majority of former gymnasts indicated the time
demands placed on them was at least one of the reasons they Ieﬁ gymnastics (see
Table 15). This is consistent with what the former gymnasts liked least about
gymnastics. It appears that what these gymnasts liked the most about gymnastics was
unable to override the time constraints involved with participating in their sport. In
addition, the second most popular reason given for leaving gymnastics was that they
were tired of gymnastics. Considering the average number of years the former
gymnasts had participated in gymnastics was slightly over 5 years and considering the
average age of the former gymnasts was approximately 11.5 years of age, it seems
logical that they could have grown tired of gymnastics by then. The other most cited

responses were the Coaches, Injury, and Family Could Not Afford It.

 

Table 15

Why Former Gymnasts Discontinued Participation: Rank Order and Frequency
Percentages

 

Frequency

Rank Order Percentages
1) Time Demands 53%

2) Tired of Gymnastics 26%

3) Coaches 21%

4) Injury 16%

5) Family Could Not Afford It 11%

6) Too Stressful 5%

7) Summer Vacation 5%

 

74

The former gymnasts were also asked what the best and the hardest parts were
about leaving gymnastics. The former gymnasts identiﬁed three things they liked best
about their decision to leave gymnastics. The most popular responses were (a) Free
Time to Pursue Other Activities, (b) Reduced Stress, and (c) Time for Homework.
The rank order and frequency percentages are presented in Table 16. It is interesting
that all of the gymnasts identiﬁed only three things that they liked best about leaving
gymnastics because these same gymnasts had identiﬁed several reasons for leaving

gymnastics.

 

Table 16

Wh w the Bes Part for Former G nast A L avin G ti 5: r

gag Frequency Percentages

 

Frequency
sweet MP r ma 8
1) Free Time to Pursue Other Activities 74%
2) Reduced Stress 32%
3) Time for Homework 21%

 

In contrast, the four most popular responses by former gymnasts on the hardest part
about leaving gymnastics were (a) Leaving Teammates, (b) Leaving Coaches, (c)
Losing Ability to do Skills, and ((1) Missing Gymnastics and the Great Times (see
Table 17).

Finally, the gymnasts were asked whether they had joined any activities since

leaving gymnastics. Eight of the former gymnasts indicated that they had not joined

75

 

Table 17

What was the Hardest Part for Former Gymnasts About Leaving Gymnastics: Rank
Order and Frequency Percentages

 

Frequency

Rank Order Percentages
1) Leaving Teammates 42%

2) Leaving Coaches 37%

3) Losing Ability to do Skills 26%

4) Missing Gymnastics and the Great Times 21%

5) Telling Coaches 5%

6) Giving Up Years of Hard Work 5%

 

any other activities since leaving gymnastics. Seven of those eight said they
alreadywere in other activities before leaving gymnastics and planned to continue
those activities. Furthermore, eleven of the former gymnasts indicated they joined
other activities since leaving gymnastics but in examining their answers it appears they
already were involved in most of these activities. The activities they are involved in
are predominantly school activities such as sports and clubs. Thus, the gymnasts are
not discontinuing their activities, rather they appear to be switching to other activities
or focusing on activities to which they were previously committed.

In conclusion, the age of the dropouts (Range=10-l3; M=11.7, _SQ= 1.19) and
current gymnasts was almost the same (Range= 9-14; M=l 1.61, §D=I.40). In
addition, the dropouts (M=3.65, _S_12=2.55) had been coached by their current coach for
almost the same period of time as the current gymnasts (M=3.66, S_D_=l.95). The

former gymnasts indicated the decision to leave gymnastics generally took place over

76

several months. Thus, it was not a quick decision that was made. Few gymnasts were
forced to leave the sport. The dropouts indicated learning new skills was what they
liked the best about being in gymnastics and the time required to participate in the
sport was what they liked the least about gymnastics. The time demands of
participating in gymnastics was also the most frequently identiﬁed reason for leaving
gymnastics. Finally, the former gymnasts identiﬁed that leaving their teammates and
coaches was the hardest part about leaving gymnastics while the best part about
leaving gymnastics was having the free time to pursue other activities. Thus, the time
demands of the sport of gymnastics played an integral part in gymnasts decisions to
leave their sport.

In conclusion, the results for this study revealed several ﬁndings. When
gymnasts actual decision to continue or discontinue participating in gymnastics was
examined, several variables were identiﬁed which predicted that behavior. For
example, the interaction between anxiety and coping with adversity as well as support
from family and friends inﬂuenced gymnasts participation decisions. In addition, the
interaction of two perceived coaching behaviors (punishment-oriented feedback and
non-reinforcement/ignoring mistakes) inﬂuenced gymnasts decisions to continue or

discontinue gymnastics participation.

CHAPTER V

Discussion

The purpose of this study was to examine how psychological variables, social
variables, and environmental variables interact to affect youth sport participation
decisions. Speciﬁcally, this study examined whether female gymnasts who leave
gymnastics differ in their perceptions of coaching behaviors, perceptions of social
support, level of anxiety and ways of coping from gymnasts who continue their sport
participation. The discussion is presented in the following manner. First, the research
questions and results are discussed in light of theoretical implications. Then, practical
implications are offered. And ﬁnally, future research directions and conclusions are
presented.

Theoreticﬁal Implications

For this study, two research questions were identiﬁed. The ﬁrst research
question addressed the relationship between anxiety, coping with adversity, support
from within sport and support from family/friends among female gymnasts who
continue and discontinue their sport participation. The second research question
concerned the relationship between perceived coaching behaviors in female gymnasts
who continue and discontinue their sport participation. Several interesting ﬁndings
related to gymnastics participation emerged from these research questions. The
research questions are discussed in light of theoretical implications. In addition, a

section is included to examine psychological and sport resilience relative to the

77

78

ﬁndings from this study.

Gould and Petlichkoffs (1988) integrated motivational model of youth sport
participation and withdrawal was used as the conceptual framework for this study.
This model explains youth sport participation and withdrawal as processes that are
inﬂuenced by a common set of factors (Gould & Petlichkoff, 1988). Incorporated into
Gould and Petlichkoffs (1988) model are Harter's (1978, 1981) competence motivation
theory, Maehr and Nicholls' (1980) achievement goal orientation theory and Smith's
(1986) cognitive affective model of stress which serve as the underlying theoretical
motives for participation and withdrawal.

Based on these underlying theories, the following constructs are included in
Gould and Petlichkoffs (1988) model: anxiety, coping, social support, and coaching
behaviors. What is not addressed in the model is how the variables of anxiety, coping
and social support interact to inﬂuence withdrawal from sport or how speciﬁc
coaching behaviors interact to inﬂuence withdrawal from sport. Furthermore, prior to
this study the interaction between anxiety, coping and social support had not been
examined relative to sport participation.

The results from the present study provide insight into how anxiety, coping and
social support interact as well as how speciﬁc coaching behaviors interact to inﬂuence
decisions to participate in and withdraw from gymnastics. The ﬁndings were that
anxiety and coping interact to inﬂuence gymnasts' actual participation. More
speciﬁcally, gymnasts who cannot cope effectively with adversity or difﬁcult

situations, which can easily arise in sport in general and in this case in gymnastics,

79

and the higher these gymnasts' anxiety, the more likely they will discontinue
participating in gymnastics. Thus, a relationship emerged between gymnasts' abilities
to cope, their level of anxiety and their decision to participate in gymnastics.

Furthermore, these results are consistent with what has been found in research
in psychology where an individual's coping ability relates to resilience in stressful
situations (Garmezy, 1983; Rutter, 1983). That is, if a gymnast can cope effectively
with the stressful situations in their sport environment, they will be more likely to
continue participating in gymnastics. It may be beneﬁcial for the sport participation
research to futher examine the psychology literature to gain greater insight into the
process of sport participation.

In addition, these ﬁndings identify the need for gymnasts to develop adequate
coping skills or strategies to deal with the stressful situations they may encounter in
gymnastics which would allow them to continue their sports participation. The coping
research in sport has primarily focused on athlete burnout (Gould, Tuffey, Udry, &
Loehr, 1996a, 1996b, 1997; McCann, 1995) and athletes' coping with injuries (Smith,
Smoll, & Ptacek, 1990). In addition, many youth athletes have not fully developed
their coping skills (Shaie, 1977-7 8) and sport can be a stressful environment (Brustad,
1988), thus, the current ﬁndings provide a greater understanding of the role of coping
with adversity in gymnasts' participation decisions.

An example of how the ﬁndings from this study relate to gymnastics is
provided. Kim is a gymnast who, under her coaches' orders, has been working to

learn several difﬁcult skills. Kim has been having difﬁculty learning the new skills

 

80

and her coach has become frustrated with her. Kim has become increasingly more
nervous before practice because she knows she needs to perform the skills and she
does not want to disappoint her coach. As a result of this, Kim is feeling a great deal
of pressure but does not know how to deal with the pressure. According to the current
ﬁndings, if Kim does not know how to deal with the situation she is in (low coping
skills) and she is very stressed about Ieaming the new skills and dealing with her
coach (high anxiety) then Kim will be more likely to discontinue participating in
gymnastics if the situation does not change. In contrast, a gymnast who encounters
the same situation of needing to learn several new skills and is having difficulty
learning the new skills but is able to cope effectively with the pressure she is feeling,
this gymnast will be more likely to continue participating in gymnastics.

Speciﬁcally, the results from the current study expand our knowledge of which
psychological variables (i.e., anxiety and coping) inﬂuence gymnastics participation as
well as how anxiety and coping interact to inﬂuence gymnastics participation. That is,
if gymnasts have low abilities to cope with adversity, and experience higher anxiety,
the more likely they are to leave gymnastics. Furthermore, when we consider the
reasons the gymnasts identiﬁed for discontinuing their gymnastics participation in the
follow-up questionnaires, we recognize only part of the story is being told regarding
youth athletes' participation decisions. We now know that in the months prior to
leaving gymnastics, anxiety and coping are interacting and are related to the decision
to leave gymnastics. However, a better understanding of the psychological variables

that inﬂuence youth sport participation is necessary.

 

81

The analyses also identiﬁed that support from family and friends played a role
in decisions to continue participating in gymnastics while support from individuals
from within the sport domain such as coaches or teammates were not related to actual
participation decisions. This particular ﬁnding does not support the work of Walter
(1996) who found that gymnasts who continued their gymnastics participation utilized
more support from within gymnastics (i.e., support from coaches, teammates) than
gymnasts who discontinued their participation. Walter (1996) also found that
gymnasts who discontinued participating in gymnastics utilized more and more family
support as they got closer to leaving gymnastics. For the group of gymnasts who
participated in this study, support from family and friends played a signiﬁcantly more
prominent role in gymnastics participation than did support from within gymnastics.

These conﬂicting results between past research and the present study suggest
the need to further examine whether differences exist between where athletes obtain
support (i.e., support from within sport and support from outside of sport) and whether
that relates to sport participation decisions. These differences may have occurred
because the majority of the gymnasts in the current study were not involved in as
intense a gymnastics program as were the gymnasts in the study by Walter (1996).
The gymnasts who are involved in highly competitive gymnastics programs, such as in
the present study, may see their family and friends outside of gymnastics as integral
parts of their gymnastics participation. In contrast, gymnasts who participate in
extremely demanding gymnastics programs, such as in the Walter (1996) study, may

perceive individuals outside of their sport domain as being distracting and not sources

82

of support.

Although research on support from family and friends generally has not been
tied back to participation, the results are consistent with and build on research that has
examined parental support in sport in general. Speciﬁcally, previous research has
found that athletes who perceived their parents to be supportive reported more
enjoyment while participating in their sport (Ommundsen & Vaglum, 1991; Scanlan &
Lewthwaite, 1988). Vaaneren (1995) reported that parental support provided a
buffering effect for when problems emerged such as coaches' criticism of an athlete's
performance or problems with teammates. Thus, parental support appears to play a
role in youth athletes' sport experiences and ultimately sport participation decisions.

In addition, these ﬁndings provide further insight into the work of Burton and
Martens (1986) who were unable to clearly delineate the role of signiﬁcant others in
the decision to drop out of wrestling. The current results indicate that while support
from family and friends may not directly contribute to the decision to drop out of
gymnastics, the support family and friends provide plays a role in allowing gymnasts
to continue their sport participation.

Limited research has been conducted on the relationship between athletes'
perceived coaching behaviors and sport participation decisions (e.g., Barnett, Smoll &
Smith, 1992). Rather, the coaching behavior research has focused on the relationship
between perceived coaching behaviors and three major areas including: (a) perceptions
of ability and motivation (Black & Weiss, 1992), (b) coaching effectiveness training

and youth sport attrition (e.g., Barnett, Smoll, & Smith, 1992; Smoll, Smith, Barnett,

83

& Everett, 1993), and (c) perceived coaching behaviors and team cohesion/play
satisfaction (e.g., Gardner, Sheilds, Bredemeier, & Bostrom, 1996; Home & Carron,
1985; Weiss & Friedrichs, 1986). Based on the results from the current study we now
know that when gymnasts perceive coaches to provide speciﬁc coaching behaviors
(i.e., punishment-oriented feedback and non-reinforcement and ignoring mistakes) that
they do interact to inﬂuence participation decisions.

Relating the work of Barnett, Smoll, and Smith (1992) to the present study is
particularly relevant because they found that athletes who played for coaches who
were trained to be more encouraging in their interactions with their Little League
baseball players had a signiﬁcantly lower attrition rate from baseball the following
season. If coaches can be trained to provide less punishment-oriented feedback and to
ignore performance errors when they cannot provide positive and informational
feedback, then, in the case of the current study, the gymnasts may perceive their
coaches to be more encouraging and would not leave gymnastics.

Furthermore, Weiss and Petlichkoff (1989) speciﬁcally identiﬁed the
importance of understanding coaching styles and how they relate to sport participation
as one of the missing links in sport participation research. The results from this study
have, for the ﬁrst time, examined the social structure of gymnastics by examining the
relationship between perceived coaching behaviors and how they interact to inﬂuence
gymnastics participation. Speciﬁcally, when athletes perceive the coach ignores
performance errors while providing high amounts of negative feedback, these athletes

will be more likely to discontinue participating in gymnastics. The results from this

84

study have not only provided greater insight on the types of coaching behaviors which
play a role in youth sport participation decisions but insight has also been gained on
how speciﬁc coaching behaviors interact to inﬂuence youth sport participation.

For example, if gymnasts perceive their coach to seldom ignore mistakes they
make and the gymnasts perceive the coach to provide a great deal of feedback that
contains punishment, then these gymnasts may feel they never get a break from their
coach. Thus, gymnasts perceive their coach never misses a mistake they make and is
always yelling at them, even if the feedback they are getting contains a great deal of
information for how to improve. As a result of feeling like they are being continually
watched and criticized, these gymnasts will be more likely to discontinue participating
in gymnastics.

Gymnasts processed the type of feedback provided to them and made
participation decisions based on how they interpret the information from their coaches.
Having open communication channels between coaches and their gymnasts would
provide opportunities to discuss any miscommunication and thus the athlete would
avoid beginning the process of leaving gymnastics. We know from the gymnasts who
left gymnastics that their decision to leave generally took place over a period of
several months. If coaches' behaviors do not become an issue to the gymnast then the
gymnast does not begin to think about leaving. While coaching behaviors may be
factors in gymnasts' participation decisions, it is necessary to also consider that other
factors such as athletes' coping ability and sources of support which may limit the

inﬂuence of certain coaching behaviors on gymnasts' participation decisions. Further

 

 

85

research is necessary to explore these relationships.

These ﬁndings also make intuitive sense. Gymnasts may perceive the coach
will not ignore some mistakes they make and provides feedback that includes mostly
negative language rather than instruction. These gymnasts may feel that they are

never good enough because they do not get the moves correctly, never get a break

 

1'
from the coach, and, consequently, decide to leave gymnastics. In contrast, if
gymnasts perceive the coach does ignore mistakes at least some of the time then the
amount of punishment-oriented feedback does not affect decisions to participate. .

While the gymnasts want to learn the skills, the learning environment the coach has
created with the type of feedback provided does not allow some of the gymnasts to
continue participating.

For example, Horn (1985) examined the relationship between coaching
behaviors and junior high female athletes' perceived competence over the course of the
season. She found that perceived competence increased with criticism-type feedback

from their coaches. Horn (1985) explained these ﬁndings by stating athletes identiﬁed

 

the importance of appropriate and performance-contingent feedback. Thus, if the
junior high athletes received praise but it was not appropriate to the effort or ability
demonstrated, then the athletes interpreted this as they were so bad the coach could
not ﬁnd anything to say to help them improve their performance.

It is possible to relate Horn's (1985) work to the ﬁndings from the current
study. That is, gymnasts may perceive that, coaches' feedback was not appropriate and

contingent on their performance. If gymnasts perceive this to happen repeatedly over

86

the course of the year, then the gymnasts may feel they have little control over
changing the coaches' behaviors, the environment in the gym, or their ability to learn
the skills necessary to be successful. Thus, the gymnasts may have felt that their only
recourse was to leave gymnastics.

Finally, the results of the open-ended questions the former gymnasts completed
are discussed. The reasons the former gymnasts gave for withdrawing from
gymnastics are consistent with previous research (e.g., Gill, Gross, & Huddleston,
I985; Gould, Feltz, & Weiss, 1985; Longhurst & Spink, 1987). Speciﬁcally, the
results of the present study revealed that time demands of the sport were the most
frequently cited reason for leaving gymnastics. Other frequently cited reasons for
leaving gymnastics were the coaches, fear of the events or injury, and pressure on
themselves. The results from the present study were consistent with ﬁndings of why
gymnasts (Klint & Weiss, 1986) and swimmers (McPhereson, Martenick, Tihanyi, &
Clark's, 1980) discontinued their sport participation.

The results also revealed similarities between the things gymnasts liked most
about participating in gymnastics and the ﬁndings of past research regarding the
reasons for sport participation identiﬁed by athletes. For example, numerous studies
found that athletes participated in sport to learn new skills/improving skills (Gill,
Gross & Huddleston, I985; Gould, Feltz & Weiss, 1985; Longhurst & Spink, I987;
Sapp & Haubenstricker, I97 8). These ﬁndings are consistent with the results from
this study. What the gymnasts liked the most about being in gymnastics was Ieaming

new skills. For the gymnasts in this study, additional frequently cited things they

 

 

87

liked the most about being in gymnastics were their teammates/friends in gymnastics,
the coaches and competing. In general, these results support the previous research on
reasons for sport participation (Gill et al., 1985; Gould et al., 1985; Longhurst &
Spink, I987). The ﬁndings differ in that their teammates/friends in a sport have not
been identiﬁed as motives for participating.

Weiss and Petlichkoff (1989) examined the sport participation literature which
has shown that children's attrition from a sport is oﬁen temporary (Gould, Feltz, Horn
& Weiss, 1982; Klint & Weiss, 1986; White & Coakley, 1986). The research has
shown that children participate in a phenomenon of "dropping in and dropping out" of
sport (Weiss & Petlichkoff, 1989). That is, when youth sport athletes discontinue
participating in a sport, they often do not leave sport altogether. Rather, the
individuals either choose to participate in other sports or to participate in the same
sport at a lower intensity. The ﬁndings from the open-ended questions in this study
supported the earlier research that when a child leaves gymnastics, they are not
permanently disengaging from sport or other activities. Rather, they have chosen to
discontinue participating in gymnastics and would rather spend their time in other
activities. For the gymnasts in this study 95% of the gymnasts who left gymnastics
indicated they were either previously involved in other activities and would continue
those activities or had joined other activities since leaving gymnastics.

It has been suggested that identifying the reasons for withdrawal alone does not
provide an adequate picture of the process children may go through before they leave

their sport (Gould, 1987; Gould & Petlichkoff, 1988; Petlichkoff, 1993; Weiss &

 

 

88
Petlichkoff, I989). The results from this study began to identify the process gymnasts

may go through before they withdraw from gymnastics. We know that anxiety and
coping are interacting in the months prior to actual withdrawal to inﬂuence
participation decisions. Thus, while the reasons for withdrawal are similar to ﬁndings
from previous research, the ﬁndings from this study provide insight to the
psychological variables that are inﬂuencing gymnastics participation.

In conclusion, from the current study we have gained insight into how variables
interact to inﬂuence gymnasts' participation decisions. We have also learned the
importance of following gymnasts over an extended period of time to observe their

behaviors. Finally, we now understand more about the complex nature of youth sport

participation.

Psychological and Sport Resilience. When looking beyond the sport literature,

a line of research in psychology can provide guidance for understanding the interaction
between the variables examined in this study and how these variables could affect
Sport participation. Resilience has been deﬁned as when individuals encounter
negative or challenging experiences yet develop normally (Rutter, 1990). Although
youth sport athletes generally are not at risk for abnormal development, they do
encounter negative or challenging situations that could affect decisions to continue
Sport participation. Thus the concept of resilience, where individuals modify their
response to a risk or stressful situation, is applicable to sport. That is, in sport most
youth athletes will encounter negative or challenging situations, such as having

difﬁculty learning a skill or being yelled at by a coach. Some athletes are unaffected

89

by these negative experiences whereas others are greatly affected by these negative
experiences. More speciﬁcally, based on their sport experiences some athletes will
continue their sport participation while others will discontinue their sport participation.
The essential deﬁning feature of resiliency is that there is a modiﬁcation of the
person's response to a risk situation. Resilient individuals have been found to have
lower levels of anxiety (Milgram & Palti, 1993), more positive coping skills (Rutter,
1983), secure attachments (Fongay, Steele, Steele, Higgit, & Target, 1994) and good
peer relationships (Cowen & Work, 1988). The resilience literature has focused on
what allows some children's development to be severely affected by major life events
or daily hassles (non-resilient children) while other children's development is not
affected by these events (resilient children) (Luthar & Zigler, 1991). Researchers have
found that resilient individuals have certain qualities such as positive coping skills,
lower anxiety, and higher social support whereas non-resilient children lack these
qualities (Felner, Primavera, & Cauce, I981; Milgram & Palti, 1993; Rutter, 1983;

Werner & Smith, 1982).
The resilience literature has identiﬁed high-risk children and factors that appear

to be protective mechanisms for children in light of negative experiences. In the sport
Context a similar relationship between factors could be supported. That is, if a child
has stressful events in sport but maintains positive coping skills, has lower anxiety and
has social support from within sport, this child would be resilient and probably
cOtrtinue sport participation. The resilience literature attempts to identify social,

Situational, and individual difference variables that predispose individuals to either

90

negative or positive reactions to stressful events (Garmezy, 1981; Rutter, 1979).

Within the sport literature, only one study (Smith, Smoll, & Ptacek, 1990) has utilized

the resilience literature for examining the interaction of variables.

Smith et al. (1990) framed their study in terms of the research on vulnerability
and resiliency and examined how life stress, social support, and coping skills interact
to determine if certain children were more vulnerable to sport injury. Smith et al.
(1990) indicated that in life event research, situational and individual difference
variables such as social support and coping have been identiﬁed as factors that
increase the vulnerability of people to the impact of negative life events. Smith et al.
( l 990) found that a lack of social support and coping skills and the presence of major

life stress left youth sport athletes vulnerable for injury. They concluded by
emphasizing the need to consider the joint inﬂuence of social support and coping skills
in life event research.

Youth sport athletes have identiﬁed injuries as a reason for leaving sport
(Petlichkoff, 1993) thus it is possible to link Smith et al's. (1990) ﬁndings to youth
Sport participation research. That is, Smith et al. (1990) found that athletes who are
1 ow in social support and coping skills, and encounter major negative life stress, will
be predisposed to injury. For these athletes who are predisposed to injury, they may
8.1 so be vulnerable to withdrawing from their sport due to having low social support
and c0ping skills, and encountering major negative life stress. In addition, it is
itliteresting to consider if youth sport athletes, who encounter negative situations yet

Q<>t1tinue their sport participation, have qualities similar to resilient children and

91

athletes who encounter negative situations yet leave sport, have qualities similar to
non-resilient children. If these scenarios are supported, then the youth sport
participation process may be better understood through examining the interaction of

variables including coping, anxiety, social support and negative experiences as have

been supported in the psychological resilience literature.

Based on the resiliency research in general psychology and the work of Smith,
et al. (1990), Walter (1996) introduced the concept of sport resilience. A pilot study
was conducted on college students examining the concept of sport resilience utilizing a
retrospective approach. Participants were identiﬁed as sport resilient and sport non-
resilient by whether they had continued participating in the same sport for more than
four years during junior high and high school. Interviews were conducted on the
subjects identiﬁed as resilient and non-resilient. Differences between resilient and
non-resilient individuals emerged in their support from within the sport environment.
Sport resilient athletes reported receiving more support from within the sport setting
than did sport non-resilient athletes. These ﬁndings warranted further examination of
the concept of sport resilience on youth sport athletes currently involved in sport.

Based on the psychological resilience literature, Walter (1996) operationally
deﬁned sport resilience as a child who participates in at least one sport for a minimum
of four years, encounters negative sport experiences, has high perceptions of control,
has high self-conﬁdence, has low levels of anxiety, develops supportive relationships
in sport, and continues his/her sport participation. The operational deﬁnition for non-

SI>ort resilience was a child who participates in at least one sport for a minimum of

92

four years, encounters negative sport experiences, has negative coping skills, fails to

develop supportive relationships in sport, has high levels of anxiety, has low self-
conﬁdence, and discontinues her/his sport participation. The minimum of four years

of sport participation in one sport was selected because if an athlete has participated

for at least four years it was assumed that they have committed to the sport and

withdrawal is not a result of sampling various sports.

Walter (1996) found support for the operational deﬁnition of sport resilience
when gymnasts who continued their sport participation had encountered negative sport
experiences, had high self-conﬁdence, had lower levels of anxiety, developed more
supportive relationships in sport than gymnasts who discontinued their sport
participation. The gymnasts who were identiﬁed as sport non-resilient were found to
have during the three months prior to leaving gymnastics lower self-conﬁdence, higher
anxiety, reported fewer supportive relationships in sport, and encountered more
negative situations in sport when compared to the sport-resilient gymnasts (Walter,

l 996). Thus, support for all of the components Walter proposed in the operational
deﬁnition of sport resilience was found except perceptions of control. No differences

Were found in perceptions of control for sport resilient and sport non-resilient

gymnasts (Walter, 1996).
Partial support has been found for the operational deﬁnition of sport resilience.

The results from this study which support the operational deﬁnition of sport resilience
are the interaction between anxiety and coping with adversity. These ﬁndings are

cOnsistent with Walter's (1996) ﬁndings that gymnasts' who were identiﬁed as non-

93

resilient had higher levels of anxiety and lower coping abilities. The results from this
study provide additional insight because the speciﬁc nature of the relationship between
these two variables was identiﬁed. That is, gymnasts who were identiﬁed as being
low in their ability to cope with adversity, and had higher anxiety, were more likely
to discontinue participating in gymnastics. These ﬁndings are consistent with the
psychological resilience literature where resilient individuals have been found to have
lower levels of anxiety (Milgram & Palti, 1993) and more positive coping skills
(Rutter, 1983).

The results from this study did not ﬁnd that support from within the sport
domain was more important than support from family and friends. Rather, the results
indicated that support from family and friends played a role in the continued sport
participation of gymnasts. These ﬁndings, as stated previously, are not consistent with
the work of Walter (1996) who found that support from within sport was important for
gymnasts to continue their sport participation. However, the ﬁndings from the present
study are consistent with the psychological resilience literature that developing strong
social support is important for dealing with difﬁcult situations (Fongay, Steele, Steele,

Higgit, & Target, 1994). Based on these ﬁndings, there appears to be some support
for the concept of sport resilience, however, further work is needed to fully explore
the nature of the variables involved in sport resilience. Speciﬁcally, work is needed
on the deﬁnition of sport resilience, the variables involved in sport resilience, and how
those variables interact.

What the concept of sport resilience and the psychological resilience research

94

add to our understanding of sport participation is an avenue for incorporating the
interaction of variables as well as individual variables identiﬁed in this study as
playing a role in gymnasts participation decisions into Gould and Petlichkoffs (1988)
model. Speciﬁcally, we now know that anxiety and coping interact to inﬂuence
gymnasts' participation as well as support from family and friends playing a role in
gymnastics participation. Gould and Petlichkoffs (1988) model provided insight into
the decisions youth athletes make in participating and withdrawing from their sport.
However, we now know there are speciﬁc variables (anxiety, coping, family/ﬁiend
support, and coaching behaviors) which are involved in participation decisions and
how these variables interact to inﬂuence gymnasts participation decisions. Gould and
Petlichkoffs model fails to identify how different variables interact to inﬂuence youth
athletes participation, the length of time involved in the decision making process for
whether or not to continue participating, and the complexity of youth sport
participation.
The underlying theoretical frameworks identiﬁed in Gould and Petlichkoffs
(1988) model include Harter's (1978, 1981) competence motivation theory, Maehr and
Nicholls' (1980) achievement goal orientation theory, and Smith's (1986) cognitive
affective model of stress. While these theoretical frameworks are depicted in Gould
and Petlichkoffs (1988) model as being directly related to the surface-level
In otivations for participation and withdrawal, nowhere in the model is the interaction
between the variables discussed. In the present results we know that variables are

i nteracting to inﬂuence sport participation. For example, from the results of this study

95

we know that while the gymnasts identiﬁed similar reasons for withdrawal such as
other interests, when examining the time period leading up to their withdrawal from
gymnastics we see that anxiety and c0ping were interacting to inﬂuence those
participation decisions.

In addition, family and friends played an integral role in providing support for
the gymnasts which led to continued participation and certain coaching behaviors were
identiﬁed by the gymnasts to interact to inﬂuence participation decisions. Thus, while
the athletes were able to provide reasons for withdrawal they may not have been
willing to or able to articulate the psychological processes that were occurring to
inﬂuence those participation decisions. Thus, the model is not complete in being able
to explain the sport participation process.

By utilizing the psychological resilience literature and the preliminary work
conducted on sport resilience a more complete understanding of gymnasts sport
participation and withdrawal is provided. For example, the psychological resilience
literature has identiﬁed the same variables as have been identiﬁed in this study,
anxiety, coping and support from family and friends, as being important in helping

individuals to deal with stressful situations. Furthermore, the concept of sport
resilience has been partially supported from the current ﬁndings. Additional research
i S needed to see if the current ﬁndings can be replicated across different sports and for
bOth genders. If future research does support the current ﬁndings, Gould and

Petlichkoffs (1988) model would need to be rethought.

96

Practical Implications

Several practical implications can be offered in light of the results from this
study. First, it is apparent that gymnast's ability to cope with the situations they
encounter in gymnastics plays a role in their participation decisions. Thus, if coaches
are aware of this they can provide networks within their gyms to help the youth
gymnasts cope with the various experiences they encounter in their gyms. For
example, group discussions could be held by older gymnasts with the younger
gymnasts to tell them how they dealt with the stresses of competition, fear of new
skills and overcoming injuries.

Second, coaches need to be aware that many young athletes have not fully
developed their coping skills so they can deal effectively with the various situations
they may encounter. So, when gymnasts face difﬁculty in learning skills, coaches and
athletes may view the same situation very differently. For example, a gymnast may be
having difﬁculty Ieaming a new skill and both the coach and athlete are becoming
frustrated. The coach may view the situation as the athlete merely requiring more
time to learn the skill. In contrast, the gymnast may not know how to cope with not

learning the skill, may view not Ieaming the skill as a failure, their anxiety may

illcrease and they begin to consider leaving gymnastics. A coach needs to understand

that providing sources of support from individuals such as parents, other coaches,
teammates, or sport psychologists to the athlete are important for the athlete to
develop their coping skills. If the coach can provide some resources for support to

help an athlete cope more effectively with the situations they encounter, less anxiety

CC

CC

p2

pa
fer
frc
DOS
is n
that
P101:

hat,“

97

will develop and they will be more likely to continue participating in gymnastics.
Third, if coaches know that parents provide important social support for the
gymnasts which inﬂuences participation decisions, coaches could work on the
communication they have with parents. Oﬁen several problems exist regarding
communication between coaches and parents. For example, parents do not feel they
are kept informed of their child's progress or situations in the gym, no avenue exists
for regular communication between parents and coaches, and/or parents are extremely
critical of coaches' decisions and consequently poor communication exists between
coaches and parents. By alleviating some of the problems that oﬁen exist with coach-
parent communication, coaches can work to cultivate a positive relationship with
parents which includes regular communication. Then, parents and coaches may be
able to work more closely to provide the child with the support necessary to deal with
difﬁcult or challenging situations that may arise in gymnastics.

Finally, if coaches know the speciﬁc behaviors, which can inﬂuence gymnasts'
participation, they may be willing to learn more about how to provide appropriate
feedback to help create an environment where gymnasts want to continue participating.
From previous research we know that athletes do not want feedback that is always
positive and containing no criticism. Rather, athletes want feedback from coaches that

i s rich with information on how they can improve their skills. Athletes understand
that sometimes criticism is necessary. Coaches' education on the importance of
providing feedback that contains information for improvement and awareness that a

balance between positive feedback, negative feedback, and ignoring mistakes feedback

 

98

is necessary for athletes to continue participating in their respective sports.
Future Directions

The present study contributed to the sport participation literature in several
ways. First, the gymnasts were studied over the course of the season. In addition,
how psychological variables including anxiety and coping interact to inﬂuence
gymnastics participation was examined. Social variables were also examined in
relation to gymnasts' participation. Speciﬁcally, support from family and friends
emerged to inﬂuence sport participation decisions. Finally, the interactions between
coaching behaviors and how those behaviors inﬂuence gymnastics participation was
explored. These ﬁndings have provided a new approach for examining the sport
participation process that is based on the missing links identiﬁed by Weiss and
Petlichkoff (1989). Despite these contributions, additional research is needed to better
understand youth sport participation. Thus, future research directions are offered.

First, most of the sport participation research has been grounded in
participation motivation. That is, reasons for sport participation and withdrawal have

been understood through achievement motivation theories such as Harter's (197 8,

l 981) competence motivation theory and Maehr and Nicholls' (I980) achievement goal
Orientation theory. These theories are the underlying motivational theories
incorporated into Gould and Petlichkoff‘s (1988) integrated motivational model of

youth sport participation and withdrawal along with Smith's (1986) cognitive-affective
tflodel of stress.

The ﬁndings from this study are that gymnasts' decisions to continue

 

99

participation and withdrawal are inﬂuenced by the interaction of variables such as
anxiety and coping, various coaching behaviors, as well as social support. While
Gould and Petlichkoffs (1988) model has provided guidance for studying and
understanding youth sport participation, the results from the present study suggest the
need for additional research to see if the ﬁndings from the present study can be
generalized across different sports. If the results can be generalized across sport, then
the parts of the model addressed by this line of research needs to be revised to
incorporate the speciﬁc interaction of variables which lead to sport participation and
withdrawal.

Second, while this study employed a repeated measure technique over
approximately a 6 month period, it seems to be important to follow athletes over
longer periods of time. As Petlichkoff (1996) suggested, if athletes are followed
through phases of participation it would then be possible to examine the role of other
variables such as whether previous experiences inﬂuenced participation withdrawal or
was the decision to withdraw due to experiences during that season.

Finally, the use of recent statistical advances such as GEE to analyze the
results provides an avenue for studying repeated measures data without the typical

limitations one encountered in longitudinal analyses. This statistical tool is
particularly appropriate for examining youth sport participation because of the need to
Study athletes over extended periods of time. Thus, researchers are encouraged to
e>rplore this statistical technique for examining questions where they may encounter

correlated observations, non-normal distribution of data, or continuous and/or discrete

1 00
dependent variables.

Conclusion

In conclusion, the purpose of this study was to examine how psychological
variables, social variables, and environmental variables interact to affect youth sport
participation decisions. Speciﬁcally, this study examined whether female gymnasts
who leave gymnastics differ in their perceptions of coaching behaviors, perceptions of
social support, level of anxiety and ways of coping from gymnasts who continue their
sport participation. The ﬁndings from this study identiﬁed the interaction of
psychological variables related to gymnastics withdrawal, the importance of social
support from family and friends in continued gymnastics participation, and the
interaction of coaching behaviors to inﬂuence withdrawal from gymnastics. In
addition, this study employed the statistical technique of Generalized Estimating
Equations to analyze the repeated measures data which allowed the researcher to avoid
some of the problems inherent in repeated measures analyses.

Understanding the variables that inﬂuence youth sport participation and how
they interact is important for helping more athletes maintain their sport participation.
While previous sport participation research has been grounded in achievement

motivation and has provided a great deal of insight for understanding sport
participation, the results from this study identify that other variables inﬂuence sport

participation decisions. Thus, the need exists to more completely explore the variables

t1flat inﬂuence youth sport participation.

 

 

APPENDICES

101
APPENDIX A

Athletic Coping Skills Inventory-28

DIRECTIONS A number of statements that athletes have used to describe their
experiences are given below. Please read each statement carefully and then recall as

accurately as possible how often you experiences the same thing. There are no right

or wrong answers. Do not spend too much time on any one statement. Please check
the line which indicates how often you have these experiences when playing sports.

Almost Almost
Never Sometimes Often Always

l. I remain positive and enthusiastic during competition,
no matter how badly things are going.

2. When things are going badly, I tell myself to
keep calm, and this works for me.

3. When I feel myself getting too intense, I
can qurckly relax my body and calm myself.

4. I maintain emotional control regardless of
how things are going for me.

 

102
APPENDIX B

Leadership Scale for Sports

Each of the following statements describe a speciﬁc behavior that a coach may exhibit.
For each statement there are ﬁve choices. Please indicate your perceptions of your
coaches' behavior by circling the appropriate number. Please answer all items.

Always Often Occasionally Seldom Never

(75% of (50% of (25% of
the time the time) the time)
5 4 3 2 I

My coach .....
l) sees to it that gymnasts
work to their capacity. 5 4 3 2 l
2) asks for the opinion of the
gymnasts on strategies for
speciﬁc meets. 5 4 3 2 l
3) helps gymnasts with their
personal problems. 5 4 3 2 l
4) compliments gymnasts for good
performance in front of others. 5 4 3 2 1

5) explains to each athlete the techniques
and tactics of the sport. 5 4 3 2 1

6) plans relatively independent
of the gymnasts. 5 4 3 2 1

7) helps members of the group
settle their conﬂicts. 5 4 3 2 1

8) pay special attention to
correcting gymnasts' mistakes. 5 4 3 2 1

103

9) gets group approval on important
matters before going ahead. 5 4 3

10) tells a gymnast when the gymnast
does a particularly good job. 5 4 3

11) makes sure that the coach's function in the team
is understood by all gymnasts. 5 4 3

12) does not explain his/her
actions. . 5 4 3

13) looks out for the personal
welfare of the gymnasts. 5 4 3

l4) instructs every gymnast individually
in the skills of the sport. 5 4 3

15) lets the gymnasts share in
decision making. 5 4 3

16) sees that a gymnast is rewarded
for a good performance. 5 4 3

17) ﬁgures ahead on what should
be done. 5 4 3

18) encourages gymnasts to make suggestions
for ways to improve practices. 5 4 3

19) does personal favors for the
gymnasts. 5 4 3

20) explains to every gymnast what should be done and
what should not be done. 5 4 3

21) lets the gymnasts set their
own goals. 5 4 3

22) expresses any affection felt
for the gymnasts. 5 4 3

23) expects every gymnast to carry out one's
assignment to the last detail. 5 4 3

104

24) lets the gymnasts try their own way even
if they make mistakes. 5 4

25) encourages the gymnasts
to conﬁde in the coach. 5 4

26) points out each gymnasts'
strengths and weaknesses. 5 4

27) refuses to compromise on
a point. 5 4

28) expresses appreciation when a
gymnast performs well. 5 4

29) gives speciﬁc instructions to
each gymnast on what should be done
in every situation. 5 4

30) asks for the opinion of the
gymnasts on important
coaching matters. 5 4

31) encourages close and informal
relations with gymnasts. 5 4

32) sees to it that the gymnasts efforts
are coordinated. 5 4

33) lets the gymnasts work at their
own speed. 5 4

34) keeps aloof from the
gymnasts. 5 4

35) explains how each gymnast's contribution

ﬁts into the total picture. 5 4
36) invites the gymnasts home. 5 4
37) gives credit where it is due. 5 4

38) speciﬁes in detail what
is expected of gymnasts. 5 4

105

39) lets the gymnasts decide
on skills or routines to be used
in competition. 5 4

40) speaks in a manner which
discourages questions. 5 4

106
APPENDIX C

Coaching Behavior Assessment System Questionnaire

As you perhaps already know, coaches really differ from each other in the type
of feedback they give in response to their athletes' performances.

This questionnaire is designed to ﬁnd out what type of coaching feedback your
coach typically gives you in practices and meets.

Coaching Responsesto Athlete's Successes

Listed below are six examples of the feedback your coach might have given to
you after you had a successful performance in a meet or practice. PLEASE RATE
EACH STATEMENT IN TERMS OF HOW TYPICAL IT WAS TO THE KIND OF
FEEDBACK YOUR COLLEGE COACH GAVE YOU AFTER YOU HAD A
SUCCESSFUL PERFORMANCE.

Very Not at all

Typical Typical
1. "Good Play!" 5 4 3 2 1
2. Coach ignores your good
performance 5 4 3 2 1
3. "Way to go! You really
extended your legs that time" 5 4 3 2 1
4. "Great routine. Now you're
keeping your toes pointed." 5 4 3 2 1
5. "Excellent work in practice
today." 5 4 3 2 1

6. Coach doesn't say anything
to you about your good
performance. 5 4 3 2 1

107
Coaching Response to PlaLers' Errors

Listed below are 10 examples of the type of feedback your coach might have given
you if you had made a mistake or committed an error in a game or practice. PLEASE
RATE EACH STATEMENT IN TERMS OF HOW TYPICAL IT WAS OF THE
KIND OF FEEDBACK YOUR COLLEGE COACH GAVE YOU AFTER A
PERFORMANCE ERROR OR POOR PLAY.

Very Not at all

Typical Typical
1. "That's OK. Keep
working at it!" 5 4 3 2 l
2. Coach ignores your error
or poor performance. 5 4 3 2 1
3. "That was a really
stupid mistake." 5 4 3 2 1
4. "You dropped your elbow.
Next time keep it up." 5 4 3 2 1
5. "How many times have I
told you to extend your
elbow?" 5 4 3 2 1
6. "Hang in there! You'll do
better next time." 5 4 3 2 l

‘7. Coach doesn't say anything
to you about your error or poor

performance. 5 4 3 2 1
8. "Your technique looks lousy!

Keep your head up." 5 4 3 2 1
9. "That performance sucked." 5 4 3 2 1

10. "No, that's not right, you
need to work on a faster
release." 5 4 3 2 l

108
APPENDIX D

Social Support Scale

Who Is Your Support for Gymnastics?

In gymnastics you may have good and bad experiences. We would like to know who
gives you support in gymnastics and the people with whom you share your gymnastics
experiences. Please circle the number which most closely describes how much support
these people provide for you. There are no right or wrong answers.

1) How much support does your mother provide for you in gymnastics?

l 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

2) How much support does your coach provide for you in gymnastics?

 

l 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

3) How much support does your father provide for you in gymnastics?

l 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

4) How much support does your assistant coach provide for you in gymnastics?

l 2 3 4
No Somewhat Important Extremely
Important Important Support Important

Support Support Support

109

5) How much support does your brotherls) provide for you in gymnastics?

l 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

6) How much support does a teammate provide for you in gymnastics?

l 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

7) How much support does your sisterls) provide for you in gymnastics?

1 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

8) How much support does your team provide for you in gymnastics?

1 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

9) How much support do your grandparents provide for you in gymnastics?

l 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

10) How much support does your athletic trainer provide for you in gymnastics?

l 2 3 4
No ‘ Somewhat Important Extremely
Important Important Support Important
Support Support Support

110

11) How much support do other relatives provide for you in gymnastics?

l 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

12) How much support does your doctor provide for you in gymnastics?

 

1 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

13) How much support do friends OLSide of gymnastics provide for you in
gymnastics?

l 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

14) How much support do other people in gymnastics provide for you in gymnastics?

l 2 3 4
No Somewhat Important Extremely
Important Important Support Important
Support Support Support

1 11
APPENDIX E

Sport Anxiety Scale

REACTIONS TO COMPETITION

A number of statements which athletes have used to describe their thoughts and
feelings before or during competition are listed below. Read each statement and then
circle the appropriate number to the right of the statement to indicate how you usually
feel prior to or during competition. Some athletes feel they should not admit to
feelings or nervousness or worry, but such reactions are actually quite common, even
among professional athletes. To help us better understand reactions to competition, we
ask you to share your true reactions with us. There are, therefore, no right or wrong
answers. Do not spend too much time on any one statement, but choose the answer
which describes how you commonly react.

Not At Moderately Very
A_ll Somewhat SQ Much So
I. I feel nervous ........................ l 2 3 4

2. During competition, I ﬁnd
myself thinking about

unrelated things ...................... l 2 3 4
3. I have self-doubts ............... l 2 3 4
4. My body feels tense ........... l 2 3 4

S. I am concerned that I may
not do as well in competition
as I could ............................. 1 2 3 4

6. My mind wanders during
Sport competition .................... l 2 3 4

'7. While performing, I often
do not pay attention to
Vvhat's going on.... 1 2 3 4

112

8. I feel tense in my stomach... I
9. Thoughts of doing poorly
interfere with my concentration

during competition .................... 1

10. I am concerned about

choking under pressure ........... l
11. My heart races .................... l
12. I feel my stomach sinking. l

13. I am concerned about
performing poorly ..................... 1

14. I have lapses in concentration
during competition because
of nervousness ......................... 1

15. I sometimes ﬁnd myself
trembling before or during a
competitive event ..................... 1

16. I'm worried about reaching
my goal ...................................... 1

17. My body feels tight ........... l

18. I'm concerned that others
will be disappointed with my
performance.............................. 1

19. My stomach gets upset
before or during competition. 1

20. I'm concerned I won't be
able to concentrate ................... l

21. My heart pounds before
competition ................................ 1

 

113
APPENDIX F

Parent Consent Form

Dear Parent,

My name is Susan Walter and I am currently a doctoral student a Michigan
State University. I have recently moved to Texas and will complete my degree while
living here in Texas. In order to complete my degree, I am conducting a research
Study to examine how gymnasts deal with positive and frustrating sport experiences.
Speciﬁcally, I am interested in the experiences youth sport gymnasts have while they
are participating in gymnastics and what allows some gymnasts to continue their sport
participation while others leave gymnastics.

I have recently met with your child's coach and they have agreed that the study
I would like to conduct is a worthwhile project to support and would help them gain
insight into better understanding gymnasts. I would like to ask that your child be a
participant in the study.

The purpose of this study is to understand how gymnasts interpret their sport
experiences and what leads some gymnasts to continue participating in gymnastics
while others choose to leave gymnastics. The study would involve your child
completing several questionnaires which explore their feelings toward gymnastics over
the next several months. I will be in the gym three times and your child will be asked
to complete the questionnaires a total of three times. Completing the questionnaires
will be done during gymnastics practice. If your child decides to stop participating in
gymnastics, I will contact them and ask them to complete the questionnaires one ﬁnal
time. All of your child's responses will be kept to completely conﬁdential (i.e., they
will not be shared with any of the coaches or parents) and each gymnast will be
assigned a code so anonymity is assured. There are no risks to your child
participating in this study. Participation is voluntary and your child can withdraw
from the study at any time without penalty.

I feel this is a worthwhile project for your child to participate in because of the
impact if could have on youth sports. If we can better meet the needs of athletes in
their sport, they will enjoy their sport experiences even more and continue
participation for a lifetime.

If you would like your child to participate in this study, please complete the
bottom portion of the letter and return it to your child's coach. If you have any

questions, do not hesitate to contact me, (806) 742-3361. I will be happy to speak
with you.

Sincerely,

 

114

Susan M. Walter
Michigan State University

(cut here)
(keep top portion, send bottom)

I agree to have my child participate in the study described above:

Print:

parent or guardian's name

Signature:

 

 

parent or guardian's name
I understand my rights as a subject in this study and I want to be a participant:

Print:

gymnast's name

Signature:

 

 

gymnast's name

 

 

115
APPENDIX G

Michigan State University's Human Subjects Committee
Permission Letter

 

 

 

OFFICE OF
RESEARCH
AND
GRADUATE
STUDIES

University Committee 0
Research Involving
Human Subiects
(UCRIHS)

Michigan State University
246 Administration Building
East Lansing. Micmgan
48824-1046

517/355-2180
FAX: 517/432-1171

[h‘mﬂamSDwLMMmW
imxsnwwmmuDMNv
MW in Action

M50 as an Mum-um
WWW”? mm“

   
 
 
 
  

MICHIGAN STATE

UNIVERSITY 116

May 16 , 1997

 

'ro: Martha E. Ewing.
201 IM Sports ircle

RE: IRB#: 95-174
TITLE: CONSIDERING RESILIENCY- IN SPORT: A RETROSPECTIVE
APPROACH
REVISION REQUESTED: N/A
CATEGORY: 1 -C
APPROVAL DATE: 05/13/97

The.University Committee on Research Involving Human Subjects'(UCRIHS)
reView of this project is complete. I am pleased to adVise that the
rights and welfare of the human subjects appear to be adequately
protected and methods to obtain informed consent are appropriate.
Berefore, the UCRIHS approved this project and any reVisions listed
a ove.

RENEWAL: UCRIHS approval is valid for one calendar year, beginning with
the approval date shown above. Investigators planning to_
continue a preject be ond one year must use the green renewal
form (enclosed with t e original a proval letter or when a
prOJect is renewed) to seek u date certification. There.is a
maximum of four such expedite renewals ossible. Investigators
wishin to continue a prOJECt beyond tha time need to submit it
again or complete reView.

REVISIONS: UCRIHS must review any changes in procedures involving human
subjects, rior to initiation of t e change. If this is done at
the time o renewal, please use the green renewal form. To
revise an approved protocol at any other time during the year,
send your written request to the. CRIBS Chair, requesting reVised
approval and referenCing the progect's IRE # and title.. Include
in our request a description of theochange and any reVised
ins ruments, consent forms or advertisements that are applicable.

PROBLEMS / '

CHANGES: Shouldoeither of the followin arise during the course of the
work, investigators must noti UCRIHS promptly: I1) roblems
(unexpected Side effects, comp aints, e c.) involVing uman
subjects or (2) changes in the research environment or new
information indicating greater risk to the human sub'ects than
existed when the protocol was previously reviewed an approved.

If we can be of any future hel , lease do not hesitate to contact us
at (517)355-2180 or FAX (51714 2- 171.

Sincerely,

vid E. Wright, Ph.D
CRIHS Chair

DEW:bed

cc: Susan M. Walter

 

 

l 17
APPENDIX H

Demographic Questionnaire

1) What is your birthday:

 

month day year

2) How old are you?

3) Are you a (check the correct response):
Male Female

4) With which ethnic group do you identify yourself with (check only one response)? :
European-American African-American
Mexican-American Asian-American

Other

5) For how many years have you been participating in gymnastics?

6) For how many years have you been working with your current coach? __

1 18
APPENDIX I

Dropout Questionnaire

1) What did you like the most about being in gymnastics?

2) What did you like the least about being in gymnastics?

3) Why did you decide to leave gymnastics? (please explain in detail)

4) HOW long did it take you to decide to leave gymnastics?

5) What was the best thing about leaving gymnastics?

6) what was the hardest part about leaving gymnastics?

7) Have you joined any other activities since you left gymnastics? Yes No

(circle one)
If yes, in what activities are you involved?

 

 

1 19
APPENDIX J

Means and Standard Deviations

 

 

m Mighigan

Standard Standard
Variable Mean Deviation Mean Deviation
Anxiety 45.50 11.54 45.80 10.72
Coping with. Adv. 6.04 2.58 6.31 2.56
Family Support 18.13 4.19 18.86 4.08
Sport Support 20.40 3.51 20.08 3.90
Democratic 29.44 6.86 26.37 7.07
Social Support 24.26 6.60 22.16 5.74
Positive Feedback 20.52 4.18 18.45 4.25
Training & Inst. 51.13 8.30 48.88 8.93
QBAS - Factor 1 0.11 0.94 -.18 0.82
QBAS - Factor 2 -.14 0.92 0.24 0.93
QBAS - Factor 3 -.09 0.88 0.15 0.81

 

 

120
APPENDIX K

Means and Standard Deviations

Sport Anxiety Scale

AC 81-2 8, Coping subscale

Social Support
Family Support
Sport Support

Leadership Scale for Sports
Democratic
Autocratic
Social Support
Positive Feedback

Training and Inst.

Time 1

SD

Mean

45.61

6.14

18.41
20.28

28.29
13.36
23.47
19.75
50.29

11.20

2.57

4.15
3.65

7.08
2.82
6.35
4.31
8.58

Time 2

Mean SD

47.48

6.17

18.90
20.54

28.97
13.26
23.14
20.41
51.30

12.82

6.62

4.01
4.00

7.24
3.22
7.02
3.98
8.17

Time 3

SD

Mean

44.50 13.06

6.27

19.27
20.59

28.06
12.96
21.66
19.92
49.19

2.90

3.99
4.06

7.07
3.16
6.15
4.26
8.83

 

 

 

121
APPENDIX L

Quotes of Former Gymnasts

 

 

 

 

122

Question 1: What did you like most about being in gymnastics?

Raw ontes:

Learning new tricks

I liked going to meets and competing even though 1 got

nervous. I also liked being there with my friends.

My teammates and coach

1 like the people there and Ieaming new tricks.

The support from my family and the competition. I also like winning.
I liked to go to double-mini.

Performing something I'm good at.

I liked the ﬂoor as an event and the challenge of each event.

My coaches and teammates. Learning new skills. Going to nationals and having a
blast. Most of all just having fun when doing gymnastics.

Always being in top physical condition and being able to perform something I‘m good
at.

I liked seeing my teammates (also my friends) and as for skills
I liked to do a lot of dance moves on beam and floor.

Learning new skills. Challenging things in gymnastics - goals & tricks couldn’t do yet
& trying to do them.

The trampoline and to bounce up and down and do ﬂips.

Going to meets. Doing tumbling and dance moves.

The competition.

Go to be with my friends.

Loved Jane" (coach). She made it really fun. My best friend was there.
The (gymnastics) events. actually doing them in practice.

Fun to tumble and flip.

" Note: Name changed for confidentiality.

 

 

123
Question 2: What did you like the least about being in gymnastics?

Raw Quotes:

Teammates

The time it took up. Sometimes it seemed like I just
didn’t want to go at all. The coaches got on my nerves too.

People with bad attitudes.
When I was tired and had to workout

The time-consuming process that you had to commit
to in order to achieve high places.

Going two days a week and staying there for about
four or ﬁve hours.

 

Those days when you just could not be yourself because 1
was having an off day.

Missing out on what my friends are doing beeause I had to
go to gymnastics.

Ididn’t like thebeamandldidn’tlikehavingtogotogym
almost everyday.

Fear of being injured Disappointment I feel about myself
when I don’t perform as I know I am really able to.

I didn’t like it when our coaches seemed mad or disappointed
in us and I didn't like it when they yelled at us.

Not having time with friends. Not being able to study as much.
Didn’t like double-mini.

Long hours of workout. Pressure on self and from coaches. Coaches pressure from
them.

All the time it took-up, so much homework and wasn't getting it done in all.
Conditioning

It was really hard to go, took up most of my time and wanted to do other things.

All the conditioning Didn‘t like all the pressure of always doing tricks right. Pressure ‘
from coaches and other girls how they could do it I compared myself to other girls a

lot.

The coaches, they were mean.

124
Question 3: Why did you leave gymnastics?

Raw Quotes:
Because I wanted to do other sports besides gymnastics.

Because I was sick of the coaches always griping at me. They had favoritism Every
time me and Sally‘ would talk we would get separated but no one else ever would
Also, it took up too much time. I didn’t have time to get my homework done and my

grades were dropping

1haddonethesportforalongtimeandlwastiredofit lwas
always complaining about going 1 was also hurt And the drive!

My mom couldn’t afford cheer and gym. I had to choose. I chose
cheer because I get my tumbling at cheer.

[decidedtoleaveymnasticsbeeauseaﬁerbrealdngmyfoot-llikedthebreakandthe

extra time with my family and friends. I also was going to Junior High School, so I
needed more time for homework.

I was tired of going to (city) and getting home at 10 or 11 o’clock.

This year 1 have had no time to go to practice. Therefore, it has been hard to devote all
the way to it I had been struggling with time so 1 ﬁnally decided to quit and take a
break. One of these days I might decide to go back. It was beginning to become
stressful.

l was just too busy with homework and other school stuff, so there just wasn't any time.

I quit gym beeause I wanted to have a life outside of gym. do other sports
and I wasjust getting tired ofgym and my heart wasn’t in it anymore.

It became too expensive for our family to manage, with four kids in the family.

I wanted to do more things with my friends and family. I wanted to get back into
things 1 did before I got serious about gymnastics.

Made cheerleading at school. Wanted to spend time with friends. School work become
more important.

Getting tired of it Wanted to try something new - cheerleading, basketball.

Was going to be gone most of the summer and mom didn’t want to pay if not there.

Was going to go back in fall. A lot of times didn't get along with coaches. Didn't agree
with coaches a lot of work on corrections.

Homework was more important Wanted to do Future Problem Solving Group.

Getting tired of doing it.

Over the summer, wasn’t going to be in town and didn't want mom to pay fee while
gone. Will wait for Kate‘ (coach) to have baby.

 

 

M'pn 3 ppntinued: 125

Momdidn’twanttodrive insummerandlwantedabreak.

Had broken your ankle and had lmocked you out of Nationals. Had planned to quit after
National beeause getting kind of tiring and do cheerleading Then broke ankle and quit
sooner.

‘ Note: Name changed for conﬁdentiality.

 

 

126

Question 4: What was the best thing about leaving gymnastics?

Raw Quotes:

Do other sports.

Havinga lot more timetodofunstuﬂ‘.

More time on my hands.

I wasn’t so tired or burned out.

The extra time I had for other things important to me.

I get to spend more time playing and talldng to my friends.

 

1 have more time and I’m less stressed now.
1 have time to be with my friends now.
Ididn't have to go. 1 had a free life.

It gave me so much more time to spend with friends and other activities. 1 also have a
lot less stress in my life!

I‘m getting the chance to get back into things 1 really enjoyed doing before I had to go to
gymnastics 4 days a week

More quality time to myself. Being able to study.
More time to do other things like read and stuff. Now have time to do my homework

Don't have to worry about it any more (doing good at meets). No pressure from coaches
and myself to do good at meets.

More time to do whatever. Time for homework.

Have a little bit more free time.

More time to do homework and started making better grades.

Don’t feel all the pressure of always going and always doing the tricks right.

Don’t have something every single night and get to be home.

127

Question 5: What was the hardest part about leaving gymnastics?
Raw ontes:

My coaches

Well, I’ve been in it for six years so it was kind of hard to just give up six years of hard
work. Also, I thought my coach would be mad beeause he coaches me at school too.

Leaving my coach and teammates.
Not seeing my friends or the coaches.

Telling my coaches, especially Bill‘ - he was my favorite - and I felt I had disappointed
him - but he was really decent about it

 

The things I’ve learned, I might not be able to do them anymore.

The sport itself beeause I still love to do it My wonderful coaches and teammates. All
of the great times 1 have had there, and all of the nationals I’ve been to. I miss it.

Leaving my coaches. After a while they beeame my friends.
Leaving my teammates and coaches.

Losing the availability of the gym for working out and not having motivation of the
coach and other gymnasts to help me stay in the best physical condition

Leaving my friends.

Losing my strength and tricks

Leaving the trampoline and all the medals.

Afraid l was going to miss it too much.

Missing some of my best friends at gymnastics. I don’t get to see them much anymore.
Miss all my friends 1 went with

Was going to miss Beth’ (coach).

Miss doing all the events.

Not going to Nationals.

* Note: Name changed for conﬁdentiality.

 

LIST OF REFERENCES

128

LIST OF REFERENCES

Aiken, L.S., & West, SO. (1991). Multiple regress_ionz Testing and interpreting
interactions. Newbury Park, CA: Sage Publications.

Ames, C. (1984). Competitive, cooperative, and individualistic goal structures:
A motivational analysis. In R. Ames & C. Ames (Eds), Research on motivation in
education: Student motivation (pp. 177-207). New York: Academic Press.

Amorose, A.J., & Horn, TS. (in press). Intrinsic motivation: Relationships with
collegiate athletes' gender, scholarship status and perceptions of their coaches'

behavior. Journal of Sport & Exercise Psychology.

Antonucci, TC. (1986 summer). Measuring social support networks:
Hierarchical mapping technique. Generations 10-12.

 

Barnett, N.P., Smoll, F.L., & Smith. RE. (1992). Effects of enhancing coach-
athlete relationships on youth sport attrition. The Spprt Psychologist, 6, 111-127.

Black, S.J., & Weiss, MR. (1991). The relationship among perceived coaching
behaviors, perceptions of ability, and motivation in competitive age-group swimmers.

Journal of Sport and Exercig Psychology, 14, 309-325.

Bohrnstedt, G.W. (1970). Reliability and validity assessment in attitude
measurement. In G. F. Summer (Ed), Attitude measurement (pp. 80-99). Chicago:
Rand McNally.

 

Brustad, R.J. (1988). Affective outcomes in competitive youth sport: The
inﬂuence of intrapersonal and socialization factors. Journal of Sport & Exercise

Psychology, 10, 307-321.

Brustad, R.J. (1992). Integrating socialization inﬂuences into the study of
children's motivation in sport. Journal of Sport & Exercise Psychology, 14, 59-77.

Burton, D. (1988). The dropout dilemma in youth sport: Documenting the
problem and identifying solutions. In RM. Malina (Ed), Youth athletes: Biological,

psychological, and educational perspectives (pp. 245-266). Champaign, IL: Human
Kinetics.

 

129

Button, D., & Martens, R. (1986). Pinned by their own goals: An exploratory

investigation into why kids drop out of wrestling. Journal of Sport Psychology, 8,
183-197.

Carmines,E.G., & Zeller,R.A. (1989). Reliability and validig assessment.
Newbury Park, CA: Sage.

Chelladurai, P., & Saleh, S. (1978). Preferred leadership in sports. Canadian
Journal of Applied Sport Sciences, 3, 85-92.

Chelladurai, P., & Saleh, S. (1980). Dimensions of leader behavior in sports:
Development of a leadership scale. Journal of Smrt Psychology, 2, 34-45.

Compas, BE. (1987). Coping with stress during childhood and adolescence.
Psychological Bulletin, 101, 393-403.

Compas, B.E., Malcame, V.L., & Fondacaro, KM. (1988). Coping with
stressful events in older children and young adolescents. Journal of Consulting ppd

Clinical Psychology, 56, 405-411.

Coppel, DB. (1980). The relationshi of rceiv social su rt and self-
efﬁcacy to major and minor stressors Unpublished doctoral dissertation, University of
Washington, Seattle.

Cowen, E.L., & Work, WC. (1988). Resilient children, psychological wellness,
and primary prevention. American Journﬂ of Communig Psycholpgy, 16, 591-607.

Cowen, E.L., Work, W.C., Wyman, P.A., Parker, G.R., Wannon, M. & Gribble,
P. (1992). Test comparisons among stress-affected, stress-resilient, and nonclassiﬁed

fourth- through sixth-grade urban children. Journal pf Commppig Psychology, 20,
200-214.

Crocker, P.R.E. (1992). Managing stress by competitive athletes: Ways of
coping. International Journal of Sport Pachologicﬂ, 23, 161-175.

Crocker, P.R.E., & Graham, TR. (1995). Coping by competitive athletes with
performance stress: Gender differences and relationships with affect. The Sport

Psyghologist, 9, 325-338.

Cronbach, L.J. (1951). Coefficient alpha and internal structure of tests.
Psychometrika, 16, 297-333.

‘1

2“

 

 

130

Duncan, T.E., Duncan, SC, Hopes, H., & Stoolmiller, M. (1995). An analysis
of the relationship between parent and adolescent marijuana use via generalized
estimating equation methodology. Multivariate Behavioral Research 30 317-339.

 

Duncan, T.E., McAuley, E., Stoolmiller, M., & Duncan, SC. (1993). Serial
ﬂuctuations in exercise behavior as a function of social support and efﬁcacy
cognitions. Journal of Applied Social Psychology. 23. 1498-1522.

Dweck, CS. (1986). Motivational processes affecting learning. American
Psychologist, 41, 1040-1048.

I
Ewing, ME. (1981). Achievement orientations and sport behavipr of males "7""
and females. Unpublished doctoral dissertation, University of Illinois, Urbana-

Champaign.

 

Felner, R.D., Primavera, J., & Cauce, AM. (1981). The impact of social
transitions: A focus for preventive efforts. American Journal of Community

Psychology, 9, 449-459.

Feltz, D.L., & Petlichkoff, L.M. (1983). Perceived competence among
interscholastic sport participants and dropouts. Canadian Journal of Applied Sport
Sciences 8 231-235.

 

Folkman, S., & Lazarus, RS, (1980). An analysis of coping in a middle-aged
community sample. Journal of Health and Socigl Behaviors, 22, 457-459.

Folkman, S., & Lazarus, RS, (1985). Coping as a mediator of emotion.
Journal of Personalig and Social Psychology, 48, 150-170.

Fongay, P., Steele, M., Steele, H., Higgit, A., & Target, M. (1994). The theory
and practice of resilience. Jogpal of Child Psychology and Psychiat_ry, 35, 231-257.

Gardner, D.E., Sheilds, D.L.L., Bredemeier, B.J.L., & Bostrom, A. (1996). The
relationship between perceived coaching behaviors and team cohesion among baseball
and softball players. The Spprt Psychologist, 10, 367-381.

Garmezy, N. (1981). Children under stress: Perspectives on antecedents and
correlates of vulnerability and resistance to psychopathology. In A.I. Rabin, J.
Aronoff, A.M. Barclay, & RA. Zucker (Eds), Further explorations in personalig (pp.
196-269).

Garmezy, N. (1983). Stressors in childhood. In N. Garmezy and M. Rutter

(Eds), Stress, coping, and devslopmsnt in children (pp. 43-84). New York: McGraw-
Hill.

131

Gill, D.L., Gross, J .B., & Huddleston, S. (1985). Participation motivation in
youth sport. International Journal of Sport Psychology, 14, l-14.

Gould, D. (1987). Understanding attrition in children's sport. In D. Gould &

M. Weiss (Eds), Advances in pediatg's sport sciences - Behavioral issues (pp. 61-85).
Champaign, IL: Human Kinetics.

Gould, D., & Horn, T. (1984). Participation motivation in young athletes. In
J.M. Silva, III & R.S. Weinberg (Eds), Psychological foundations of spurt (pp. 359-
370). Champaign, IL: Human Kinetics.

Gould, D., Feltz, D.L., Horn, T., & Weiss, MR. (1982). Reasons for attrition in
competitive youth swimming. Journal of Smrt Behavior, 6, 126-140.

Gould, D., Feltz, D.L. & Weiss, MR. (1985). Motives for participating in
competitive youth swimming. International Journal of Spppt Pacholpgy, 16, 126-140.

Gould, D., Finch, L., & Jackson, S. (1993). Sources of stress in national
champion ﬁgure skaters. Journal of Sport & Exercise Pyghology, 15, 134-159.

Gould, D., & Petlichkoff, L. M. (1988). Participation motivation and attrition in
young athletes. In F .L. Smoll, R.A. Magill, and M.J. Ash (Eds), Childrsn in sport
(3rd ed., pp. 161-178). Champaign, IL: Human Kinetics.

‘ Gould, D., Petlichkoff, L., Simons, J ., & Vevera, M. (1987). Antecedents of,
temporal changes in, and relationships between CSAI-2 subcomponents. Journal of

Sport Psychology, 6, 289-304.

Gould, D., Tuffey, 3., Udry, E., & Loehr, J. (1996a). Burnout in competitive
junior tennis players: I. A qualitative psychological assessment. The Smrt

Psychologist, 10, 322-340.

Gould, D., Tuffey, S., Udry, E., & Loehr, J. (1996b). Burnout in competitive
junior tennis players: II. A qualitative analysis. The Smrt Psychologist, 10, 341-366.

Gould, D., Tuffey, S., Udry, E., & Loehr, J. (1997). Burnout in competitive
junior tennis players: III. Individual differences in the burnout experience. Tbs Sppg

Psychologist, 11, 257-276.

' Hackfort, D., & Spielberger, CD. (1989). Sport-related anxiety: Current trends
in theory and research. In D. Hackfort & S.D. Speilberger (Eds), Anxigg in spgrts:
An intsrnational perspsctive (pp. 261-267). New York: Hemisphere.

 

 

132

Hanson, S.J., McCullagh, P., & Tonymon, P. (1992). The relationship of
personality characteristics, life stress, and coping resources to athletic injury. Journal

of Sport & Exercise Psychology, 14, 262-272.

Harter, S. (1978). Effectance motivation reconsidered. Human stelopment,
2_1, 34-64.

Hatter, S. (1981). A model of intrinsic mastery motivation in children:
Individual differences and developmental change. In W.A. Collins (Ed), Minnesota
Symposium on Child Psychology (Vol. 14, pp. 215-255). Hillsdale, NJ: Erlbaum.

Horn, TS. (1985). Coaches' feedback and changes in children's perceptions of
their physical competence. Journal of Educational Psychology, 77, 174-186.

Horn, TS. (1992). Motivational orientations in sport. In T.S. Horn (Ed),
Advances in sport psychology. Champaign, IL: Human Kinetics.

Horn, T.S., & Glenn, S. (1988, June). The relationship between athletes'

psychological characteristics and their preference for particular coaching behaviors.
Paper presented at the meeting of the North American Society for the Psychology of

Sport and Physical Activity, Knoxville, TN.

Horn, T.S., & Hasbrook, CA. (1987). Psychological characteristics and the
criteria children use for self-evaluation. Journal of Spprt Psychology, 9, 208-221.

Horne, T., & Carron, A. (1985). Compatibility in coach-athlete relationships.
Journal of Sport Psychology, 7(2), 137-149.

Klint, K.A. (1985). Participation motives and self-mrceptions of current Qd
former athletes in youth gmnastics. Unpublished master's thesis, University of
Oregon, Eugene.

Klint, K.A., & Weiss, MR. (1986) Perceived competence and motives for

participating in youth sports: A test of Harter's competence motivation theory. Journﬂ

of Sport Psychology, 9, 55-65.

Lazarus, R.S., & Folkman, S. (1984). Stress, appraisal, and coping. New York:

Springer.

Lazarus, R.S., & Launier, R. (1978). Stress-related transactions between person

and environment. In LA. Pervin & M. Lewis (Eds), Persppctives in interactional
psychology (pp. 287-327). New York: Plenum.

 

\

 

133

Levitt, M.J., Weber, R.A., & Clark, MC. (1986). Social network
characteristics as sources of maternal support and well-being. Developmenﬂ

Psychology, 22, 310-316.

Liang, K.Y., & Zeger, S.L. (1986). Longitudinal data analysis using
generalized linear models. Biometrika 73 13-22.

 

Liederman, PH. (1983). Social ecology and childbirth: The new nursery as
environmental stressor. In N. Garmezy & M. Rutter (Eds), Stress, coping and
development in children (pp. 135-157). New York: McGraw-Hill.

Linder, K.J., Johns, D.P., & Butcher, J. (1990). Factors in withdrawal from
youth sport: A proposed model. Journal of Sport Behavior, 14, 3-18.

Longhurst, K., & Spink, KS. (1987). Participation motivation of Australian
children involved in organized sport. Canadian Journal of Ssisnce, 12, 24-30.

Luthar, S., & Zigler, E. (1991). Vulnerability and competence: A review of
research on resilience in childhood. American Journal of Orthopsychiaﬂ, 61, 6-22.

Maccoby, BE. (1983). Socio-emotional development and response to stressors.

In N. Garmezy and M. Rutter (Eds), Stress, coping, spd dsvelopment in children (pp.
43-84). New York: McGraw-Hill.

Madden, C.C., Kirkby, R.J., & McDonald, D. (1989). Coping styles of

competitive middle distance runners. International Journal of Spprt Psychology, 20,
287-296.

Maehr, M.L., & Nicholls, JG. (1980). Culture and achievement motivation: A

second look. In N. Warren (Ed), Studies in cross;sultura1 psyshplogy (Vol. 3, pp.
221-267). New York: Academic.

Martens, R. (1977). The Sport Competition Anxieg Test. Champaign, IL:

Human Kinetics.

Martin, D.E., & Dodder, RA. (1991). Socialization experiences and level of
terminating participation in sports. Journal of Sport Behavior, 14, 113-128.

McCann, S. (1995). Overtraining and burnout. In S.M. Murphy (Ed), Sport
psychology interventions. Champaign, IL: Human Kinetics.

McCullagh, P., & Nelder, J.A. (1989). Generalized linear model (2nd ed). New
York: Chapman and Hall.

 

 

 

134

McPherson, B., Marteniuk, R., Tihanyi, J ., & Clark, W. (1980). The social
system of age group swimmers: The perception of swimmers, parents and coaches.

Canadian Journal of Applied Sport Sciences, 4, 142-145.

Milgram, N.A., & Palti, G. (1993). Psychosocial characteristics of resilient
children. Joumsl of Research in Personalig, 27, 207-221.

Nicholls, J .G. (1984). Achievement motivation: conceptions of ability,
subjective experience, task choice, and performance. Psychological Rsview, 91, 328-
346.

Nunnally, J .C. (1978). Psychometric theog (2nd. ed). New York: McGraw
Hill.

Ommundsen, Y., & Vaglum, P. (1991). Soccer competition anxiety and
enjoyment in young boys players: The inﬂuence of perceived competence and
signiﬁcant others' emotional involvement. International Journal of Sport Psychology,
2_2_, 35-49.

Orlick, TD. (1974, November/December). The athletic dropout: A high price

for inefﬁciency. Canadian Association for Health, Physical Education, and Recreation
Journal 21-27.

 

Orlick, T. D., & Botterill, C. (1975). Eveg kid can win. Chicago: Nelson-Hall.

Petlichkoff, L.M. (1988). Motivzgion for sport pgrsistence: An empirical
examination of underlying theoreticg cpnﬂsts. Unpublished doctoral dissertation,

University of Illinois at Urbana-Champaign.

Petlichkoff, L.M. (1993). Relationship of player status and time of season to
achievement goals and perceived ability in interscholastic athletes. Pediatric Exsrcise
Science 5 242-252.

 

Petlichkoff, L.M. (1996). The dropout dilemma in youth sports. In Oded Bar-

Or (Ed), Encyclomdia of sports medicine: The child and the sdolesgent athletes (Vol.
5). Oxford, England: Blackwell Scientiﬁc.

Pierce, W.J. (1980). Psychplogical perspecgives of youth smrt participants and
nonparticipants. Unpublished doctoral dissertation, Virginia Polytechnic Institute and

State University, Blacksburg.

Pooley, J .C. (1980). Dropouts. Coaching Review, 3, 36-38.

 

 

 

135

Roberts, G.C., Kleiber, D.A., & Duda, J.L. (1981). An analysis of motivation in
children's sport: The role of perceived competence in participation. Joumsl of Sport

Psychology, 3, 206-216.

Rosenbaum, M. (1980). A schedule for assessing self-control behaviors:
Preliminary ﬁndings. Behéavior Therapy, 11, 109-121.

Rutter, M. (1979). Protective factors in children's responses to stress and
disadvantage. In M.W. Kent, & J.E. Rolf (Eds), Primpary prevention of
psychopathology: Social competence in children (Vol. 3, pp. 49-74). Hanover, NH:
University Press of New England.

Rutter, M. (1983). Stress, coping and development: Some issues and some
questions. In N. Garmezy & M. Rutter (Eds), Stress, coping, and development in
children (pp. 1-41). New York: McGraw-Hill.

Rutter, M. (1990). Psychosocial resilience and protective mechanisms. In J.
Rolf, A.S. Masten, D. Cicchetti, K.H. Neuchterlein & S. Weintraub (Eds), Risk and
protective factors in the development of psychopathology (pp. 79-101). New York:
Cambridge University Press.

Sapp, M., & Haubenstricker, J. (1978). Motivation for ioining and repsons for
not continuing in youth sport programs in Michigan. Paper presented at the meeting of
the American Alliance for Health, Physical Education, Recreation, and Dance, Kansas
City, MO.

Scanlan, T.K., & Lewthwaite, R. (1986). Social psychological aspects of
competition for male youth sport participants: Part 4. Predictors of enjoyment. Journal
of Sport grid Exercise Psychology. 8. 25-35.

Scanlan, T.K., & Lewthwaite, R. (1988). From stress to enjoyment: Parental
and coach inﬂuences on youth participants. In E.W. Brown & C.F. Brown (Eds),
Competitive sports for children And youth: An overview of rese_arch pnd issues (pp.
41-48). Champaign, IL: Human Kinetics.

Shaie, K.W. (1977-78). Toward a stage theory of adult cognitive development.
Journal of Aging and Human Development. 8. 129-138.

Shure, M.B., & Spivak, G. (1982). Interpersonal problem solving in young
children: A cognitive approach to prevention. Americap Joumsl of Occupational
Beg/ion 3. 215-232.

Smith, RE. (1986). Toward a cognitive-affective model of athletic burnout.
Joumgl of Sport Psychology. 8, 36-50.

 

 

136

Smith, R.E., Smoll, F.L., & Schutz, R.W. (1990). Measurement and correlates
of sport-speciﬁc cognitive and somatic trait anxiety: The Sport Anxiety Scale.
Anxieg Researph, 2, 263-280.

Smith, R.E., Schutz, R.W., Smoll, F.L., & Ptacek, J.T. (1995). Development
and validation of a multidimensional measure of sport-speciﬁc psychological skills:

The Athletic Coping Skill Inventory-28. Journal of Sport and Exercise Psychology,
11, 379-398.

Smith, R.E., Smoll, F ., & Hunt, E. (1977). A system for the behavioral
assessment of athletic coaches. Resegrch Quarterly. 48, 401-407.

Smith, R.E., Smoll, F.L., & Ptacek, J.T. (1990). Conjunctive moderator
variables in vulnerability and resiliency research: Life stress, social support and coping

skills, and adolescent sport injuries. Journal of Personalig and Social Psychology, 58,
360-370.

Smith, R.E., Zane, N.W.S., Smoll, F .L., & Coppel, DB. (1983). Behavioral
assessment in youth sports: Coaching behaviors and children's attitudes. Medicine and
Science in Sports and Exercise 15 208-214.

 

Smoll, F .L., & Smith, RE. (1988). Reducing stress in youth sport: Theory and
application. In F.L. Smoll, R.A. Magill, and M.J. Ash (Eds), Children in smrt (3rd
ed, pp. 161-178). Champaign, IL: Human Kinetics.

Smoll, PL, Smith, R.E., Barnett, N.P., & Everett, J.J. (1993). Enhancement of
children's self-esteem through social support training for youth sport coaches. M

of Applied Psyshology, 78, 602-610.

Smoll, F., Smith, R., Curtis 8., & Hunt, E. (1978) Toward a mediational model
of coach-player relationships. Researsh Qwerly for Exgrcise spd Spprt, 49, 528-541.

Spielberger, C.D., Gorsuch, R.L., & Lushene, RE. (1970). STAI. Mgpal for
the Ststs-Tg't-Anxieg Inventot'y. Palo Alto, CA: Consulting Psychologists Press.

Sternberg, R.J., & Grajek, S. (1984). The nature of love. Journal of Personality
2_md Socgl Psvchology, 47, 312-329.

, Tabachnick, B.G., & Fidell, LS. (1996). Using multivariate statistics (3rd ed).
New York: Harper Collins.

Thibaut, J.W., & Kelly, H.H. (1959). Th; social pgchology of gsoups. New
York: Wiley.

 

 

 

137

Vaaneren, NW. (1995). Interpersonal stress, performance level, and parental
support: A longitudinal study among highly skilled young soccer players. Th; Smtt
Psychologist, 9, 225-241.

Veit, C.T., & Ware, J.E. (1983). The structure of psychological distress and

well-being in general populations. Journal of Consulting and Clinical Psychology, 51,
730-742.

Vitaliano, P.P., Russo, J., Carr, J.E., Maiuro, R.D., & Becker, J. (1985). The
Ways of Coping Checklist: Revision and psychometric properties. Multivm'ate
Behavioral Research 20 3-26.

 

Walter, SM. (1996, June). A preliminm exploration of smrt resiliency. Paper
presented at the meeting of the North American Society for the Psychology of Sport

and Physical Activity Conference, Minett, Ontario, Canada.

Ware, JR (1985). Linear models for the analysis of longitudinal studies. M
American Statistician. 39. 95-101.

Weiss, M.R., Bredemeier, B.J. & Brustad, R.J. (1987, June). Comxtip'vs trait

anxieg in children's sport: The relationship to perceived commtence, pgrceivsd
control, and motivational orientations. Paper presented at the meeting of the North

American Society for the Psychology of Sport and Physical Activity, Vancouver, BC.

Weiss, M.R., & Chaumeton, N. (1992). Motivational orientations in sport. In
T.S. Horn (Ed), Advances in smrt psycholpgy. Champaign, IL: Human Kinetics.

Weiss, M.R., & Duncan, SC. (1992). The relationship between physical
competence and peer acceptance in the context of children's sports participation.

Journal of Sport & Exercise Psychology, 14, 177-191.

Weiss, M.R., & Friedrichs,W. (1986). The inﬂuence of leader behaviors, coach
attritions, and institutional variables on performance and satisfaction of collegiate
basketball teams. Journal of Smrt Psyshology, 8, 332-346.

Weiss, M.R., & Glenn, SD, (1992). Psychological deveIOpment and females'
sport participation: An interactional perspective. Qpest, 44, 138-157.

Weiss, M.R., & Petlichkoff, L.M. (1989). Children's motivation for
participation in and withdrawal from sport: Identifying the missing links. Pediatric
Exercise Science 1 195-211.

 

Werner, E.E. (November, 1984). Resilient children. Young Children, 68-72.

 

 

I38

Werner, E.E., & Smith, RS. (1977). Kauai's children come of age. Honolulu:
University of Hawaii Press.

Werner, E. E., & Smith, RS. (1982). Vulnerpble. but invincible: s longitudinal
study of resilient children and youth. New York: McGraw-Hill.

White, A., & Coakley, J. (1986). Mgking decisions: The response of young
people in the Medwav Towns to the "Ever thought about sport?" Campaign. London;
West Sussex Institute of Higher Education.

White, R. (1959). Motivation reconsidered: The concept of competence.
Psychological Review. 66. 297-333.

Zeger, S.K., & Liang, K-Y. (1986). Longitudinal data analysis for discrete and
continuous outcomes. Biometrics. 42, 121-130.

 

 

nIcrtIan smTE UNIV. LIBRARIES
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHIIIIIIIII
31293017120225