H

 

THESIS “' \

LIERARY
Michigan 5mm
University

   
  

 

This is to certify that the

thesis entitled
THE DEVELOPMENTAL RELATIONSHIP BETWEEN PERCEIVED
AND ACTUAL COMPETENCE IN MOTOR ABILITY AND THE
RELATIONSHIP OF EACH TO MOTIVATION TO PARTICIPATE
IN SPORT AND PHYSICAL ACTIVITY

presented by
Beverly Dianne Ulrich

has been accepted towards fulfillment
of the requirements for

___Eh_.JL_dcgree in W Physical

Education

 

Major professor

Datemoxemhew

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

lllllllllllllllllllllllllllllllllllllllllll

 

 

310672 1438
MSU RETURNING MATERIALS:
Place in book drop to
LJBRARJES remove this checkout from
w your record. F_I______NES W‘Ill

 

 

 

be charged if book is
returned after the date
stamped below.

Jim"33375 9062

 

 

l ’ .9 4;! ,1 APlRZNZE’ZBU 13‘

 

FEB 0 6 100:;
JUN t 2‘9”

3E“? :1 5 35325"

 

 

 

 

ABSTRACT
THE DEVELOPMENTAL RELATIONSHIP BETWEEN PERCEIVED AND ACTUAL
COMPETENCE IN MOTOR ABILITY AND THE RELATIONSHIP OF EACH TO
MOTIVATION TO PARTICIPATE IN SPORT AND PHYSICAL ACTIVITY
BY

Beverly Dianne Ulrich

Perceived competence has been theorized as having an important
affect on onehs level of motivation (Griffin & Keogh, 1982; Harter,
19813; Nicholls, 1978). Actual competence purportedly indirectly
affects level of motivation by influencing cumPs perceptions (Harter,
19813; Bandura, 1977). Recently investigators have begun to examine
the interrelationships among these variables within the motor domain.
Critical years in the development of these interrelationships, however,
have been virtually unexplored, (inen the childhood years). This
investigation examined the developmental relationship between perceived
and actual competence in motor ability and the relationship of each to
motivation to participate in sport and physical activity.

Twenty-five males and 25 females at each grade level, kindergarten
through fourth, participated as subjects in this study. Perceived
competence in motor ability was measured via two psychometric scales,

the Perceived Competence Scale for Children (Harter, 1979b) and the

Beverly Dianne Ulrich

Pictoral Scale of Perceived Competence and Social Acceptance for Young
Children (Harter, Pike, Efron, Chao, & Bierer, 1983). Actual motor
ability was measured by performance on a nine—item motor
abilities/sport skills assessment battery; Participation motivation
was reflected by the childfs participation in organized sports.
Univariate and multivariate analyses suggested that for children
in these grade levels, perceptions of competence in motor ability were
not significantly related to participation in organized sports.
Multivariate analysis indicated that actual ability was significantly
related to participation in organized sports. Post hoc univariate E
tests and discriminant function analysis indicated that the difference
in favor of the participants was primarily due to performance on the
sport skill items. ldultivariate analysis and correlational procedures
suggested a significant relationship between perceived competence.and
actual competence in the motor domain. Again, subjects} Scores on the
sport skill items were most important in distinguishing among subjects
with high, average, and low perceptions Of their competence. None of
the major relationships examined changed significantly across these
five grade levels. Fbr young children, perceived competence may not be
an important variable affecting participation in sport. Those who do
participate, however, may develop greater levels of skill which, as age
and accuracy of self perceptions increase may become important to onefis

choice to be involved in sport.

ACKNOWLEDGMENTS

My doctoral program has been a long one but certainly filled with
constant stimulation and growth. I would like to acknowledge several
individuals who, through their perseverance, encouragement, and
assistance, were instrumental in helping me complete this stage of my
career.

Iku Vern Seefeldt, for his unwaivering faith in my capabilities
and encouragement throughout my program. The high standards he sets
for himself and his dedication to research and teaching will be
standards of performance toward which I will strive.

Dr. Deb Feltz, whose research efforts and professional attitude I
highly respect, and whose advice as a statistician and scholar was
particularly helpful throughout my dissertation.

Inn Joe Varro, for his valued feedback and interest in my program
and dissertation.

Dr. Crystal Branta, whom I have known as a committee member,
teacher, mentor, colleague, and friend. In each of these roles she has
demonstrated the wisdom to know when to advise and when to respect my
need to grow and learn on my own. I will always be grateful for her
contribution to my years as a graduate student.

Crystal and Jack, for providing me with a home away from home,
whenever needed. Their unconditional friendship throughout the last
few years has enhanced a relationship which Dale and I will always

treasure .

ii

Joy Kiger, for her willingness to help in any way possible, no
matter how busy she was.

The principals, teachers, and children of Carterville Elementary
School, Carterville, Illinois and Winkler and Parrish Elementary
Schools, Carbondale, Illinois, for their interest and participation in

this study.

And Dale, for sharing this experience with me as a colleague and

husband, for his support, patience, and love.

iii

TABLE OF CONTENTS

Page
LIST OF TABLES .................................................. vi
LIST OF FIGURES..... ............................................ vii
CHAPTER I: INTRODUCTION ........................................ 1
Motivation: Theories and Mediating Constructs
Attribution Theory .................................... 5
Self Perception as a Mediator in Motivation ........... 5
Empirical Evidence for the Relationship Between
Perceived Competence and Motivation.. .............. 10
Actual Ability Relative to Perceived Ability and
Motivation ......... . ....... . ........ . .............. 17
Development of Perceptions of Competence
Descriptive Data ............... ..... .................. 23
Relationships of Cognitive Processes to the Accuracy
of One's Perceptions of Self ....................... 24
Harter's Model of Competence Motivation ............... 29
Gender Differences ......................................... 36

Young Children's Extent of Involvement in Sport and Games.. 41

Overview of the Study ...................................... 43
Research Hypotheses ........................................ 43
Definitions ................................................ 44

CHAPTER II: METHOD

Overview ................................................... 45
Subjects ................................................... 46
Data Collection ............................................ 47
Instrumentation
Perceived Competence Scales ........................... 51
Assessment Battery for Motor Abilities and Sport
Skills ............................................. 57
Questionnaire Regarding Participation in Physical
Activity and Sport ................................. 62

iv

CHAPTER III: RESULTS

Relationship #1: Perceived Competence to Participation
Motivation .............................................. 63

Relationship #2: Motor Abilities to Participation
Motivation .............................................. 70

Relationship #3: Perceived Competence in Motor Ability

to Demonstrated Motor Ability ........................... 78

Summary of Results ......................................... 89
CHAPTER IV: DISCUSSION ......................................... 91
APPENDIX A: PILOT STUDY ........................................ 105
APPENDIX B: PILOT STUDY - SUMMARY STATISTICS ................... 113

APPENDIX C: INFORMATION LETTER FOR PARENTS AND CONSENT FORM.... 118

APPENDIX D: MOTOR ABILITIES/SPORT SKILLS ASSESSMENT BATTERY:
PROTOCOL AND SCORESHEET .................................... 121

APPENDIX E: QUESTIONNAIRE REGARDING PARTICIPATION IN PHYSICAL
ACTIVITY AND SPORT ......................................... 127

APPENDIX F: SAMPLE PICTURE PLATES FROM THE PICTORAL SCALE OF
PERCEIVED COMPETENCE AND SOCIAL ACCEPTANCE FOR YOUNG
CHILDREN, AND FROM PLATES ADDED BY THE INVESTIGATOR........ 133

LIST OF REFERENCES .............................................. 138

LIST OF TABLES
TABLE

1. Perceived Competence in Motor Ability: Cell Means and
Standard Deviations for Participants and Nonparticipants
by Gender and Grade Level ...............................

2. Univariate E Values and Standardized Discriminant
Function Coefficients for the Motor Abilities/Sport
Skills Test Items .......................................

3. Means and Standard Deviations for Main Effects
Participation, Gender, and Grade Level on the Motor
Abilities/Sport Skills Battery Items ....................

4. Intercorrelation Matrix for Items on the Motor
Abilities/Sport Skills Assessment Battery ...............

S. Univariate Flvalues for the Motor Abilities/Sport Skills
Test Items ..............................................

6. Univariate E_Values and Standardized Discriminant
Function Coefficients for the Motor Abilities/Sport
Skills Test Items ........... ........ ............ . .......

72 Means and Standard Deviations for Groups (Top, Middle, and
Bottom‘Third on Perceived Competence) on the Motor
Abilities/Sport Skills Test Items ........ ....... ........

8. Canonical Loadings for Perceived Competence Items and

Motor Ability/Sport Skill Items on the Canonical
Variate ......... . ......................................

vi

Page

66

72

73

75

79

83

84

87

LIST OF FIGURES
FIGURE Page

1‘ Effectance motivation, adapted from White, 1959 (Harter,
19818, p. 6) ooooooooo 000000.00 ooooooooooo ooooooo oooooooo 3O

2. Phase II of the development of effectance motivation
(adapted from Harter, 1981a, p. 7) ............ . ......... 32

3. Phase III of the development of effectance motivation
(adapted from Harter, 19813, p. 10 and Harter & Connell,
in press) ................................ . ....... . ...... 34

4. Number of males and females who were classified as

participants within each grade level..... ...............
48

5. Subsets of children (classified by grade level) who were

not significantly different on perceptions of competence

in motor abilities ..................... . ............ .... 6S
6. Mean values for perceived competence by hours of play ...... 69
7. Performance in motor abilities/sport skills items by hours

per week spent in gross motor activity ........... . ...... 80
8. Possible relationship among three elements of Harter's

Model of Competence Motivation, within the physical

domain, and for young children......... ................. 102

vii

CHAPTER 1

INTRODUCTION

Today more than ever, accountability is an issue in physical
education, as in all eduCational programs. For years public school
administrators, teachers, and coaches have promoted physical education
and sport programs based on a variety of benefits such as physical
fitness, motor skill development, positive self-concept, sportsmanship,
leadership, cooperation, and moral development (Dauer, 1971; Martens &
Seefeldt, 1979; Smoll & Lefebvre, 1979). .Apparently the public has not

been convinced of either the importance or the ability of the schools

 

to produce such benefits within the school setting. In the past few
decadescunnmunity sponsored sport programs for youth and young children
have continually expanded, while funding for physical education is in
constant danger of being reduced or eliminated. In addition, the
"accountability" era in education is forcing physical educators to
produce empirical evidence to support the claims they have made. In
Illinois for example, one of the few states which mandates physical
education five days a week in the elementary schools, a strong lobby
recently attempted to reduce the extent of this mandate. 'This specific
effort was unsuccessful. 'The establishment.of a comufittee to document
justification for physical education, however, by two national
associations (Seefeldt, 1984) accentuates the critical nature of this

issue.

Several studies have demonstrated that significant changes can be
made in the gymnasium, that is, attributes such as fitness (Shephard,
Lavellée, Jequier, Rajic, & LaBarre, 1980), cooperation (Rogers,
Miller, & Hennigan, 1981),1and motor skills (Miller, 1978; Masche,
1970) can be affected. The importance of some of the objectives listed
above and the fact that the gymnasium is the most appropriate place in
which to promote them seems obvious. For others, their value as
specific attributes to be developed via physical activity is not so
clear. Should cooperation, development of positive self-concept, and
leadership_qualities apply more to physical education than to any other
academic discipline, or do conditions unique to the gymnasium hold
special promise for nurturing such attributes?

Of these attributes, the development of a positive self-concept
has recently been a popular source of investigation by researchers in
many educational disciplines (Carlson, 1970; Davis, 1981; Pate, 1978).
ReSults of experimental studies have been equivocal. Studies involving
physical education and physical activity have resulted in as many
supportive as non—supportive conclusions regarding the affectance of a
positive change in self-concept (Hughes, 1973; Martinek, Zaichkowsky, &
Cheffers, 1977; Schempp, Cheffers, & Zaichkowsky, 1983; Smith, 1982).

The inconsistency in the results of self—perception studies may be
attributed to several sources. Byrne (1983) conducted a thorough
review of the literature and concluded that a universally accepted and
operational definition of self—concept does not exist. lflany terms are
used interchangeably, such as self—esteem, self-confidence, self-image,

body image, self—efficacy, reflected self, and self-concept. In

addition, contradictions frequently exist between theories espoused and
measurements used in such research (Sonstroem, 1982). Wylie (1974), in
her classic review of the state of the art in self-concept research,
recommended either abandoning self-concept theory or making theoretical
and methodological changes to make it more scientifically acceptable.
She went on to support the latter.

One of the major criticisms of the methodology used in self—
concept research relates to the use of a single score for self-concept.
This score is based on the non—weighted sum of items reflecting self—
perceptions in several domains, such as cognitive, physical, and
social. It suggests that people perceive themselves equally in all
areas. Sufficient evidence exists indicating that people may feel
positive about themselves in one or more domains and negative about
their abilities in another. Further, these feelings and the relative
importance of each domain may change with age (Sonstroem, 1982).

The most significant and productive self relationships to be
investigated involve the areas of a specific domain. A curriculum
specialist might investigate the effect of open classrooms versus
traditional settings on self-perceptions in the academic area. In
physical activity and sport one might investigate the effect of a free
play situation versus instruction in motor skills on childrenks
perceptions of themselves in the domain of physical activity.

One might ask why perceptions of onefs abilities are important.

If one feels positive about oneself in the area of physical skills,
will one be more likely to choose to become involved in sport and

physical activity? Are perceptions of ability reflective of actual

ability? Limited data exist from which we might answer these questions.
However, studies involving upper elementary grade level students

suggest that these relationships exist and are positive (Boling & Kirk,
1983; Guyot, Fairchild, & Hill, 1981; Roberts, Kleiber, & Duda, 1981).

A second question which relates to those above is Thaw do
perceptions of competence relate to the primary goals of elementary
physical education, iJL, development of motor skills and improvement
of fitness?" Ostensibly, these are promoted so that children are
competent enough to participate in the common games and sports of
childhood and through these games maintain a healthy level of fitness.
Little is known about why children choose to become involved. Young
children seem to have an incessant desire to be active, but as age
increases many become less active. If the skills and concepts learned
in the gymnasium are only used during that period of required
involvement, the results of physical education would seem to fall short
of the objectives of the educational system in general.

The investigation of the relationship of perceptions of ability to
actual ability and the relationship of each to participation motivation
seems relevant to providing an optimal movement or motor skill
development program for children. By understanding the
interrelationships of these variables we may determine which ones can
be affected within the movement setting. Further, we may better
understand how such variables may affect childrenfls desires to become
involved in and persist in the games and sports of our culture.
Examining theories of motivation seems essential if we are to

understand these interrelationships.

Motivation: Theories and Mediating Constructs

Attribution Theory

 

The most influential approach to motivation theory relative to
sport and physical activity has been attribution theory. In this, the
basic assumption is that actions are based on a search for
understanding (Weiner, 1972). Attribution theory assumes that we judge
our successes and failures based on several causal elements, to
determine why we succeeded or failed. The major causal categories used
are ability, effort, luck, and task difficulty. Weiner (1974) placed
these into a two-dimensional model in which causality (internal or
external) and stability (stable or unstable) were the dimensions. Each
of the four causal elements fell into one of the two levels of each
dimension. .Ability and effort were internal factors, luck and task
difficulty were external. Ability and task difficulty were stable
factors, while luck and effort changed, or were unstable; One may
legitimately desire to include additional factors in the model,
particularly in the sport domain, because the theoretical predictions
derive from the dimensions and not the causal factors per se (Roberts &

Pascuzzi, 1979).

Self-Perception as a Mediator in Motivation

While attribution theory may remain useful in the investigation of
certain aspects of motivation, a growing number of researchers are
producing support for perception of ability as the most important
determinant of achievement behavior (Covington & Omelich, 1979;

Nicholls, 1978; Roberts & Pascuzzi, 1979; Spink & Roberts, 1980).

Several investigators have theorized that attributions of ability and
the self—concept of ability play the central role in mediating
motivation (Bandura, 1977; Griffin & Keogh, 1982; Harter, 1981a; Kukla,
1978; White, 1959).

The concept of competence was first introduced by White, in 1959,
as a psychological construct mediating intrinsically motivated
behavior. In his paper, "Motivation reconsidered: The concept of
competence." he proposed "effectance" motivation to explain why an
individual feels impelled to engage in mastery attempts. .He suggested
that individuals act because they feel a need to have an effect. If
these performance attempts are satisfying, one feels competent, which
provides a feeling of efficacy, inherent pleasure and joy. One is also
likely to want to repeat that performance. However, White's
explanations for such motivational concepts were very general. He
viewed competence motivation as a global motive which directed the
organism in all achievement-oriented tasks. .As such, it did not lend
itself to empirical investigation.

In 1978, Harter proposed a framework which expanded WhiteRs
theory, providing a more specific model of motivation which could be
empirically tested. Harter (1981b) concentrated on identifying the
specific domains in which competence may be measured, and viewed onefis
perceptions of competence as the central mediator of<xu¥s motivation
to achieve in that area. She also has focused on implications of
success and failure, the function of rewards in the cOntrol exerted by
socializing agents, and the relative influence of intrinsic and

extrinsic motivation on the competence motive system.

Harter (1981b) identified three primary competence domains (physi—
cal, cognitive, and social) plus a general area termed self-worth.

This is not surprising, as several theorists in the area of self-
concept had done this previously (Coopersmith, 1967; Martinek &
Zaichkowsky, 1977; Piers, 1969). However, she demonstrated that rather
than assess all domains at once with a collection of specific and
overlapping items, from which a total score is used to estimate self-
worth or general self-concept, each domain must be assessed inde—
pendentlyu General self—concept may or may not be related to any of
the specific domains. Children may feel positive about their physical
and social skills, negative about cognitive ability, but still have a
generally positive feeling about themselves. 'The use of a composite
score may neutralize subjects? strengths and weaknesses, thus, valuable
research as well as practical information may be lost (Harter, 1982;
Sonstroem, 1982).

A second prominent theoretical model which relates perceived
ability to effectance motivation has been proposed by Nicholls (in
press). The basic assumptions of his theory are that peoplefs actions
are purposeful and are motivated by a desire to (3) demonstrate and/or
develop high ability and (b) avoid demonstrating low ability.
Individuals can perceive their own level of ability in two ways.

First, one may view competence relative tocnufs past performances or
gains in knowledge. Second, competence may be viewed relative to one%;
peers.

The use of each of these rationales is related to development.

Young children focus on self and past experience when they must make

choices concerning motor tasks. Adolescents and older children begin
to judge their past performances relative to peer performances. If
this comparison leaves them lower than average, they expect to
demonstrate low ability. If perceptions indicate that their own
abilities are higher than similar others, positive outcomes are
expected from performance. Adults use both rationales in varying
situations. Thus, the individualfs assessment of the probability of
the outcome influences subsequent behavior.

Nicholls (in press) drew on attribution theory to some extent by
further linking ones perception of ability to effort. For low
perceived—ability players, high effort which results in failure,
clearly exposes a basic lack of ability. Therefore, in order to avoid
exhibiting this low ability they avoid high effort situations. In
sport situations, two options are open to these players. They may drop
out, which frequently occurs, or they may look for younger, less
skilled opponents so that some level of success is possible. In
children, an additional type of low perceived-ability individual may
exist. This individual has not given up hope and continues to try
hard. However, repeated experiences of failure eventually lead to
avoidance.

The application of the theories discussed above to sport and game
situations appears logical. However, they have been developed by
psychologists and validated primarily via cognitive tasks. Further
investigation in the motor domain is warranted. Recently some
direction has been offered by specialists in exercise (Sonstroem, 1982)

and motor development (Griffin & Keogh, 1982).

Sonstroem (1982) reviewed selected research in the area of
exercise and self-perceptions and provided some suggestions for
improving research in this domain. Based on his own research and the
research of others, his postulations are without a structured
theoretical framework, but fall closely in line with aspects of
theories of self-perception emanating from developmental psychology.

He stated that in exercise, self—perceptions appear to be situation
specific, i.e., it is more realistic to expect one's level of physical
fitness to be related to physical self-concept than global self-
concept. Further, the possible correlation between specific domains
such as physical activity and global perceptions of self—worth depend
on the value which that domain holds for the individual. In
adolescents, physical attractiveness or peer group status may relate
most highly to perceptions of self-worth. ‘Younger children, especially
boys, may place higher value on physical ability (or fitness) levels.
However, Sonstroem also supported the concept at least in Older
subjects, that perceptions of fitness are more significantly related to
motivation in physical performance situations than are actual fitness
levels.

Griffin and Keogh (1981, 1982) have Offered movement confidence as
a construct reflecting an individualis feeling of adequacy in a
movement situation. They identified movement confidence as a pervasive
and mediating influence in movement behavior. Underlying the
importance of this construct is the notion that children who are
confident about movement will choose to be active, will do so in an

assured manner and will be more likely to persist. ‘The alternate

10

situation implies that children lacking confidence in their abilities
will choose not to be active.

These scientists suggest that in required movement situations a
cycle of decreasing confidence and skill may occur. Children lacking
confidence in motor ability will attend more to feelings of inadequacy,
shifting attention from task relevant information, and falling farther
behind in skill level and thus movement confidence. 'Do break this
cycle they proposed the need to identify such children and structure
activities to promote success in a supportive environment. In essence
they emphasized the need to improve competence to increase confidence
and therefore motivation Egnmggg,

In some movement situations an additional variable which may
affect onefls feelings of confidence and motivation is the anticipation
of a specific form of sensory stimulation during performancee(Griffin &
Keogh, 1982). 'This may be positive and motivating as in the feelings
one perceives while bouncing on a trampoline, or negative, such as the
feeling of being unable to breathe easily when first learning to swim.
Empirical Evidence for the Relationship Between Perceived Competence
and Motivation

Evidence has been growing to support the theoretical relationship
between perceived competence and effectance motivation or motivation to
achieve both in the academic setting and in situations involving physi-
cal activity. Some contradictory evidence has resulted as well. Per-
ceived competence has most frequently been assessed via such scales as
Harter%3(1979b) Perceived Competence Scale for Children, or by having

children rank themselves relative to peers. Effectance motivation is

11

reflected in a childfs choice to participate or persist in a
non—required activity, or by the level of difficulty of a task chosen
within a required situation.

Harter‘s model for the development of mastery motivation has been
verified experimentally at various points in its development. Further
support has resulted from subsequent studies. lflotivation and
perceptions of competence in the cognitive domain have been examined
most frequently.

In two studies, Harter (1981a, 1981b) examined the relationships
among several components of her model as they applied to the cognitive
domain. Her subjects were third through sixth grade level children.
In the first study (1981a), she administered her assessment instruments
to measure the relationship between intrinsic motivation, perceptions
Of competence, perceptions of control, and actual cognitive competence
(as measured by achievement test scores). Higher order factoring
revealed that children who were intrinsically motivated were those who
perceived themselves as more competent than the others. In turn, they
actually were more competent and expressed greater understanding of
what controlled their successes or failures. Conversely, those who
perceived themselves as less competent actually had lower achievement
test scores, were extrinsically motivated, chose to perform easier
tasks, and did not know what controlled their successes and failures.
In the second study (Harter, 1981b), she examined the relationship Of
perceived cognitive competence to subscales of the motivational

orientation scale. Results indicated that the greater oneds sense of

12

cognitive competence the more intrinsically motivated one is to master,
to be curious, and to seek challenges.

In two behavioral studies, Harter (1979a, 1982) examined the
relationship of sixth graders} scores on the perceived competence scale
to an academic task situation and to athletic involvement. In the
first study, children were divided into high (top third) and low
(bottom third) subgroups based on their perceived cognitive competence
scores. ldhen given their choice of anagrams varying in difficulty
level, the high perceived competency group chose considerably more
difficult anagrams than those of the low perceived competency group.
The second study involved sixth graders from a school in which athletic
participation was a prominent school value. ‘Those children selected
for sport teams scored significantly higher on the physical and social
perceived competence scales than did non-team members. In this
situation Harter utilized sport participation as the measure of
intrinsic motivation in the physical domain but only examined those who
made the team versus all others. Of equal interest would have been an
analysis of the scores for children who tried out for but did not make
the team.

External studies in the physical domain have provided support for
a relationship between perceived competence and motivation to
participate. The first published report of a study which directly
tested Harter’s theories relative to perceived competence and
participation motivation in sport was conducted by Roberts et al.
(1981). These researchers utilized Harterfis Perceived Competence Scale

and interview questions to compare fourth and fifth grade participants

13

and nonparticipants in organized athletics. They found that
participants demonstrated higher levels of perceived competence, were
more.persistent, and had higher expectations of future success. Causal
attributions for involvement and non-involvement were ability oriented.
They concluded support for that portion of Harterksrnodel which
suggests that perceived competence in physical skills has an important
relationship to participation motivation in sport contexts.

Guyot, Fairchild, and Hill (1981) assessed fourth, fifth, and
sixth grade level children on some of the same variables as did Roberts
et a1. Their subjects were also classified as participants and non-
participants in sport. For both males and females, participation in
sport correlated significantly with physical self-concept. Physical
self-concept was measured via a subscale of the Piers-Harris Self
Concept Scale. In addition, they examined the relationship between
physical fitness/motor ability levels and participation. -Results
suggested a significant relationship for males but not for females.

Not all studies investigating sport participation and perceptions
of ability in children have supported a significant relationship.
Lewko and Ewing (1980) studied 9 to 11 year old boys and girls. They
determined children's perceptions of competence by asking them to rate
their ability on a scale from one to ten. Results suggested that boys
perceived their ability to be high, regardless of their level of
involvement. Girls who were participants perceived themselves as
significantly more competent in motor skills than non-participant
girls. 'The method of assessment used in this study may have affected

the results. Being good in sports is socially desirable for boys at

14

this age; therefore, the tendency to overrate oneself on a single item
measure may have been great. In Harter's (1982) and Roberts et al.s
(1981) studies a more complex and psychometrically sound scale was used
(Perceived Competence Scale for Children, Harter, 1979b). It is worded
specifically to avoid the suggestion that one response is more socially
desirable than another.

Maul and Thomas (1975) compared third grade girls who were
involved in a gymnastics program to ones not involved in any organized
sport program. They found no significant differences between the
groups on concepts of ability to perform gross motor movements. In
this study an unpublished scale from a doctoral dissertation was used.
In their discussion the authors questioned the validity of the test.

The very nature of sport encourages social comparison. Nicholls
(in press) identified this process as important to determining oneHs
perceptions of ability which subsequently affects one's choice to
participate in achievement situations. Social comparison seems
particularly important within the sport setting. 'The outcome of a
contest, win or lose, is relative to the caliber of the participants.
Several researchers have examined the relationship between the social
comparison process in the conception of ability and outcome in sport
situations.

College age subjects have demonstrated that in sport situations
they perceive ability level to be the most common cause of outcome, and
that participants judge their performance in a contest relative to
their opponentls ability. University students were asked to imagine

themselves involved in a variety of competitive situations in sport,

15

having won or lost the contest (Roberts & Pascuzzi, 1979). They were
asked to associate one of the four attributes to specific sets of
circumstances. By far the most frequently chosen attribute was
ability. Effort was also indicated by some, but not as a major
determinant.

When involved in actual competitive situations, college age
subjects have also attributed success or failure to ability (Spink &
Roberts, 1980). Following participation in a racketball tournament,
subjects were asked to indicate to what they attributed the outcome of
their match. Responses related to perceptions of their ability as it
compared to their opponents' level of ability. Satisfaction wilfli
outcome was also related to perceptions of ability relative to that of
their opponents, regardless of outcome.

Some evidence exists for the use of social comparison by younger
subjects in their perception of ability. Orlick and Botterill (1975)
interviewed children aged 7 to 19 years who had never played in
organized sport programs. 'They found that 75% said they wanted to play
but did not try out because they felt they were not sufficiently
skillful to make the team. Perhaps as Nicholls (in press) suggested,
they chose to avoid demonstrating their inability. Fourth and fifth
grade sport participants were given a test of perceptions Of athletic
ability and were asked to rate their own motor ability relative to that
of their teammates (Roberts et al., 1980). A low but significant
correlation was obtained between the two scores, suggesting that
perceived competence is related to oneds perception of ability relative

to peers.

16

Two assumptions underly Sonstroem’s (1982) postulations regarding
the study of self-perceptions in sport. First, that physical skills
represent a separate domain in self-perceptions and second, that
self-perceptions contain motivational properties. Several studies
utilizing adults and children as subjects supported the concept that
perceptions of self in specific areas such as fitness or body image may
be affected experimentally (Brown, Morrow, & Livingston, 1982;
Rohrbacker, 1979). They demonstrated further that global perceptions
of self were not affected.

A study of boys at the middle-school level suggested that their
perceived level of physical ability related to their motivation to
participate in organized school sport programs. Three hundred ninety—
three boys were given a test of perceived competence in physical
ability at the start of the school year; These scores were compared
later to subjects} choices to play or not play intramural sports. Two
separate discriminant function analyses yielded canonical correlations
of .35 and..40 with 64 and 68% of the subjects correctly classified as
participants or nonparticipants, respectively.

Griffin and Keogh%;(1981, 1982) theory of movement confidence has
much heuristic appeal. Based on similarities to the work of others in
achievement motivation such as Harter (1981b) and Bandura (1977), some
support is anticipated. Thus far, the only reported empirical support
for their model demonstrated the reliability and consistency of one
aspect. They determined that independent observers could reliably and

consistently identify behaviors Of young children as being confident or

17

non-confident in situations perceived as "high risk" (Keogh, Griffin, &

Spector, 1981).

Actual Ability Relative to Perceived Ability and Motivation

The preceeding discussion emphasized the central role of
perceptions of ability in choice to participate, and the desire to be
affective and to persist in achievement tasks, especially in sport and
games. Considering this hypothesis and the support it has generated,
one might ask how important the child's actual skill level is in
motivation to participate and persist in motor activity. In physical
education and sport situations, facilitators generally focus on the
improvement of true skill level. Pangrazzi (1982) stated that to avoid
developing a feeling of incompetence, young children especially need
skilled instructors who can bring about true improvement in skill
level. Ostensibly, by raising skill level one increases the childfis
desire to use those skills in games and sport. Does subsequent

motivation occur because the child is truly able to perform at a higher

 

level of skill or because the child thinks improvement has occurred?
Further, does one (ability) automatically result in the other (thinking
one is able)?

Certainly successful participation requires some basic level of
skill. To play soccer one must be able to dribble and pass the ball;
the ability to kick a rolling ball is a prerequisite to playing
kickball. Experience and research tell us that children at the lower
end of the ability continuum learn quickly that they are not as skilled
as their peers, are not welcome in games and sport, and learn to avoid

displaying their lack of ability. Building confidence without skill

18

would not be likely to provide a lasting effect on childrenfls desire to
participate and persist in sport.

Bandura (1977) identified four sources of self—confidence
information: mastery experiences, modeling, verbal persuasion, and
emotional arousal. He considered mastery experiences or competence to
be the most durable and strongest in developing confidence in oneEs
self. Bressan and Weiss (1982) and Griffin and Keogh (1982) emphasized
the importance of the psychosocial construct (ifih, movement confidence)
in motivating children to move. 'They expanded on this concept by
suggesting that the best way to increase confidence is by developing
competence. They concluded that the development of competence in
movement, in a supportive atmosphere, with the necessary incentives,
provides the necessary ingredients for developing achievement
motivation.

A few investigators have examined the relationship between
perceived competence and actual competence in children of elementary
school age, within the academic or physical education setting.
Generally, a moderate relationship has been observed. Several studies
in the academic setting demonstrated that for third through sixth grade
children, perceptions of ability clearly differentiated between
children of actual high and low ability (Boersma & Chapman, 1978;
Chapman & Boersma, in press; Harter, 1981b). While children in this
age group demonstrated a relatively accurate concept of their ability,
level of achievement correlated more highly with perceptions of ability
than with actual ability (correlations were .61 and .31, respectively).

In the physical activity setting, Boling and Kirk (1982) studied the

19

relationship of perceived competence of fifth grade boys in motor
skills and actual competence. ‘Their results suggested that those who
scored in the upper and lower third on the perceived physical ability
scale could accurately be distinguished by their performance on a
series of gross motor tasks.

One researcher examined the relationship between perceived and
actual competence in physical skills across the ages of 4 to 12 years,
in two separate studies. Harter (1982) compared the scores of third
through sixth grade children on perceived competence in physical
ability to ratings of their motor ability made by their physical
education teacher. Correlations were in thee.60s for all four grade
levels, suggesting that these children were moderately accurate in
judging their own motor ability. Correlations between these variables
for younger children were very low; an £_of .26 for the four- and five—
year-old children and an £.of .32 for the six- and seven-year-Old
subjects (Harter & Pike, in press).

Two points should be noted regarding the studies just cited.
First, ability ratings for the younger children were made by classroom
teachers, who may have been less accurate in their assessments of motor
ability than physical educators. This factor may have contributed to
the lower correlations. Second, the pattern of correlations for the
third through sixth grade subjects was quite different within the
domain of physical activity, than the pattern of correlations for the
academic area. Harter found lower, but linearly increasing,

relationships between perceived and actual competence in academics.

20

However, the correlations were similar at each grade level for the
physical activity domain.

The relationship between perceived ability and actual ability
appears to be affected by immediate and long-term feedback received
from significant others. Although perceptions Of onefs ability appear
to be based on years of self-evaluation and comparison, several studies
(Dweck & Repucci, 1973; Haas & Maehr, 1965; Maehr, Mensing, & Nafzger,
1962) have demonstrated that the accuracy of self—perceptions of
children of the fifth grade level and older may be significantly
affected by feedback from others, at least temporarily. In these
studies subjects received positive or negative feedback from
significant others on a short term basis. Results indicated that a
significant change in perceptions of ability could be affected, both
positively and negatively, depending on the type of feedback given.
This occurred regardless of actual ability in academic settings as well
as in physical activity situations. Both types of feedback altered
children's opinions, but the effects of negative feedback faded more
rapidly than did the effects of praise.

One effect of feedback from a physical education instructor may be
increased accuracy in perceptions of physical ability. Yeatts and
Gordon (1968) studied seventh graders who had or did not have physical
education during their elementary school years. Correlations between
their fitness scores and self-image of physical abilities suggested
that those who had elementary physical education were significantly

more accurate in their assessment of personal fitness.

21

Only a few researchers have examined the relationship between
actual motor ability and participation motivation. The results of
these studies have been equivocal. Magill and Ash (1979) compared the
actual motor ability of first through fifth grade boys and girls who
were participants or nonparticipants in organized sport programs. They
found no relationship between motor ability and choice participation.
However, their choice of tasks used to represent motor ability did not
appear to be representative. The items were: (a) the Minnesota Manual
Dexterity Test, (b) a finger tapping task, and (c) a stabilometer task.
Physical fitness scores were also available for fourth and fifth grade
subjects (Texas AHPER test items). Scores on these tasks differen-
tiated fourth but not fifth grade participants from nonparticipants.
King (1982) studied the task choices of preschool and primary grade
level children in a free—play gymnasium situation and reported that
motor ability was not a significant factor in their task choices.
Although she used a representative battery of items to test motor
ability, her task choices did not appear to provide a sufficient range
of difficulty levels to necessitate a decision about perceived ability
on the part of the child. Guyot et al. (1981) studied boys and girls
in Grades 4 through 6 who were participants and nonparticipants in
sport. They found a significant relationship between participation and
motor ability for boys, but not for girls.

Some support for the role of actual ability in achievement
motivation/expectancy resulted from a study of 11 and 12 year-Old youth
soccer players by Scanlon and Passer (1981). The relative importance

of actual soccer ability, general self-esteem, overall win-loss record,

22

opponents"win—1oss record, anxiety, and prior record against the same
opponent in predicting players' pre—game expectancies was examined.
Soccer ability was most predictive, followed by self-esteem. The
additional factors did not add significantly to the prediction
equation. However, self-perception measures were general. Had a
specific measure of perceived competence for physical or soccer ability
been used, it may have been more predictive.

On the basis of the studies reviewed in the last two sections, it
appears that a significant relationship may exist between perceived
competence and motivation to participate for children in the upper
elementary grades. Further, these children demonstrate moderate
accuracy in identifying their own levels of ability. Evidence that
these relationships exist in the movement setting is beginning to
accumulate, although not without some controversy; However, the
relative importance of motor ability to the perceived competence-
motivation paradigm suggested by motivation theorists is not clear. In
addition, the question of these relationships existing for younger
subjects is unanswered.

The need to investigate the relationships among performance,
psychological, and behavioral variables goes beyond the expansion of
our body of knowledge. In the applied setting such information may
help us understand children with low motivation, and eventually
discover better ways to help them. By the time children reach third
grade, the youngest age group usually studied, they are reporting
perceptions of themselves that have been developing for several years.

If we want to understand why some children are not motivated to be

23

active or have a low perception of their ability, so that we may affect
them in the most positive way, we must understand the critical

developmental years, iJL, the preschool and early elementary years.

Development of Perceptions of Competence

Descriptive Data

Assessment of the concept of ability, as well as of most "selF'
attributes, indicates that children report generally positive
perceptions of themselves. liean scores across the age range of
preschool to Grade 6 are negatively skewed. Statistically significant
differences are not usually found between adjacent grade levels. A
trend is suggested in competence scores, across the preschool and
elementary ages, which indicates a shift to a somewhat less positive
concept of ability; ‘While mean scores decrease, variation within age
groups increases with age (Harter, 1978; Harter & Pike, in press;
Martinek & Zaichkowsky, 1977; Nicholls, 1978). However, all of the
data described above are cross-sectional in nature and based on means.
Thus, changes within the group may be masked. Those who were positive
about their abilities at six years of age may not be the ones who
believe they are the best performers at age eight or nine. Also,
differential rates of development of accurate self-perceptions may be
occurring but would not be detectable in studies that are cross-

sectional in design.

24
Relationships of Cognitive Processes to the Accuracy of One's
Perceptions of Self

Perceptions of onefls self are considered to be relatively stable
and various domains are hierarchically ordered by approximately seven
or eight years of age (Felker, 1974; Harter, 1982; Piers, 1969;
Zaichkowsky, Zaichkowsky, & Martinek, 1980). 'The importance of events
occurring prior to this time seems obvious to a developmentalist.
However, until recent years little effort had been made to provide
empirical evidence for the processes by which changes in self-
perceptions occur, or the variables which affect those changes.

Recently, the identification of the various cognitive processes
necessary for making accurate assessments of ability and the onset of
these processes have shown promise in explaining the development of
accuracy in making judgments about ability; One such process is called
social comparison procedures. This process involves evaluation of
(nufs own ability via comparison with peers (Festinger, 1954). Veroff
(1969) believes that between 5 and 10 years of age social comparison is
the most important means Of perceiving one's ability. Only beyond this
age do children use more objective normative information to evaluate
their own abilities.

Observation of children younger than four or five years of age
reveals a lack of interest in comparing their abilities with others
(Veroff, 1969; Suls & Sanders, 1979). By age four or five children
begin interacting with peers and according to some researchers, making
comparisons. lchlintock and associates (McClintock & Moskowitz, 1976;
McClintock, Moskowitz, & McClintock, 1977) found that between ages four

and one half and five, if children were given a choice, they began to

25

choose situations in which a competitive advantage over peers was
provided, rather than one in which they were concerned only with their
own outcome. Such tendencies were found to increase until about age
seven and then leveled off. Ruble, Feldman, and Boggiano (1976)
observed that kindergarten, first, and second grade level children in a
timed task situation with a peer co-actor, glanced more frequently at
the peer as age increased. Masters (1971) reviewed a large number of
studies on social comparison in children, and concluded that by age
four children are highly involved in social comparison. A
representative study involved self-rewarding behavior, in which
children who had previously received fewer tokens than a peer,
dispensed additional tokens to themselves.

Ruble, Boggiano, Feldman, and Loebl (1980) contended that social
comparison ability develops in stages. They suggested that studies
such as those reviewed by Masters (1971) represent lower stages of
development and that mature, useful functioning does not occur in the
preschool years. Young children's observations of their peers may have
been motivated by curiosity, a desire to seek information about Others,
or to be like others, but not by evaluative motives. Ruble et a1.
(1980) developed a complex design to study children%s(kindergarten,
second, and fourth grade levels) ability to assess accurately their own
behavior compared to others on a novel gross motor task; Expectations
of rewards in the study were manipulated to maximize the childls effort
at giving honest predictions (even if the child honestly expected his
or her own skill level to be lower than others) of future chance of

beating opponents of various skill levels. Not until fourth grade were

26

accurate assessments made. Kindergarten children actually rated their
abilities more highly than did the second and fourth grade children.

Stipek (1981) suggested that the highest and lowest achievers may
be able to judge their own ability and compare themselves to peers
better than average achievers because they are more likely to receive
clear and consistent feedback about abilities. She found that
kindergarten and first graders, trichotomized into high, middle, and
low ability groups were very inaccurate. Even the low achievers
overestimated their level of ability. By second and third grade,
children were more accurate about their own abilities, and more
accurate still about the abilities of peers. She suggested several
explanations, including feelings of omnipotence, self—enhancement bias,
and faulty information processing due to pre-operational thinking.

On the basis of the studies cited above, apparently social
comparison procedures do not begin to function in an evaluative and
somewhat accurate manner until approximately eight or nine years of
age. Pre-school and primary elementary age children are aware of
others} performances, but this awareness stems from a curiosity.
Outcome or rewards give more salient information than processes or
abilities. Further, younger children tend to be unrealistically
optimistic about their abilities.

Additional cognitive capacities which are developing during the
preschool and primary years, and which may relate to the development of
accurate self-concepts of ability, were elaborated by Nicholls (1978).

1. In order to rank order or compare onehs ability to peers, the

child must be able to:

27

a. seriate,
b. decenter (view oneself with relationship to others), and
c. relate temporally separated outcomes to each other.

Children in the pre-operational stage do not have the cognitive
skills for such processes. By age five or six, as children enter the
concrete operational stage these cognitive skills begin to emerge.

2. Understanding of the relationshipgbetweengperformance norms
and ability required for success.

Older children and adults (in the formal operations stage) reason
that tasks at which most people succeed require little ability; those
which only few can do require high ability.

3. Ability to determine the cause of outcome - why one won or
as.

Experiences CLe., repeated failures or successes) may give one a
feeling of being a loser or winner, good or bad. Until one can
separate ability from effort (around age 11) accurate ratings of
ability relationships are diminished. Nicholls characterized many
young children as persisting in motor and other skills under the
assumption that their inherent ability is not low, only additional
effort is needed to improve their unsuccessful attempts.

The ability to accurately assess oneds level of ability has been
shown to change with age. Harter (Hatter, 1982; Harter & Pike, in
press) and Nicholls (1978) compared childrenks self-ratings of academic
competence with ratings done by the teacher; Harter found correlations

of .32 (preschool/kindergarten), .43 (Grades land 2), .28 (Grade 3),

.32 (Grade 4), .50 (Grade 5), and .55 (Grade 6). Nicholls reported

 

28

correlations of .21, .27, .58, .71, .57, .80, and .78 for ages 7
through 13, respectively. The change, while not linear, shows a
positive relationship to age.

Some support exists for the possibility that the accurate
assessment of one's competence may develop at different rates for
different domains. Correlations between teacher and student ratings of
physical competence (Harter, 1982; Harter & Pike, in press) increased
during the preschool and primary grade school years and plateaued at a
moderately high value during the upper elementary level years. 'The
correlations obtained for this domain were .26 and .32 for ages four to
five and six to seven, respectively, while across Grades 3 through 6 a
consistent value of .62 was obtained. During the early years physical
competence ratings were slightly less accurate than cognitive, both
being much higher than the accuracy of social skill ratings. From the
third through sixth grade levels, children were much more accurate
concerning their perceptions of ability in physical skills than in
cognitive or social skills.

Competitive motor activity situations may provide more concrete
and obvious feedback than academic situations. Because of this,
children may develop accurate perceptions of ability at an earlier age
than in other domains (Kleiber, 1981; Pascuzzi, 1981). Pascuzzi (1981)
examined the effect of finishing first, second, or third in a foot race
on perceptions of competence in preschool and primary elementary grade
level children. Place of finish did not affect significantly the
perceptions of ability of preschool girls. It did affect the

subsequent ability, affect, and expectancy ratings made by preschool

29

boys and by boys and girls of school age. In addition to the type of
feedback provided, the social value of sport, especially for boys, may
suggest greater attention to movement than to academic performances.

In summary, based on the concepts related to cognitive functioning
discussed above, one may expect preschool age children to be inaccurate
raters of their own ability. .As cognitive processes mature and social
comparison situations increase with age, children become more
realistic. .Accuracy may develop earlier in motor skill than in other
domains due to the many opportunities for comparison during the early
years, more frequent contingent feedback and the importance of motor

skills to young children.

Harter's Model of Competence Motivation

In order to investigate the relationships among variables related
to achievement motivation in physical activity, as is proposed in this
study, they should be viewed as part of a larger theoretical model
which may provide some evidence (theoretical or empirical) for their
relationship. Harter (1981a, 1981b; Harter & Connell, in press) seems
to provide the best developmental model for understanding factors that
cause children to choose to achieve (to be effective, to be active)
and to persist in achievement situations. iHer three-phase model
concentrates on the early childhood and elementary years and suggests
that perceptions of competence are a key component in development of
motivation. Following is a summary of Harter's model.

Phase I (See Figure 1). In this phase, infants are intrinsically
motivated (designated in the model as effectance motivation) to engage

in achievement tasks (labelled mastery behaviors). By producing an

30

 

Mastery Behaviors

Components of the Mastery
Process:

Independent Mastery Attempts
Challenge Seeking

 

 

 

 

 

 

 

 

 

 

Curiosity
_E_f_f_e_ct on the
Effectance Environment
Motivation Product
Success/Failure

 

 

 

 

l

Positive Affect

Feelings of Efficacy
Joy in being a Cause
Pleasure from
Exercising Schemes

 

 

 

 

 

Figure l: Effectance motivation, adapted from White, 1959
(Harter, 19810, p. 6)

31

effect on the environment, they experience pleasure and joy. Awareness
of the effects that their actions have on the environment produces a
positive effect; often smiling or laughing. Piaget described this
process as infants'lattempts at effectance and their joy in being a
cause. He evoked no special motive to account for this behavior,
assuming as Harter does, that it is innate to the organism.

Phase II (See Figure 2). Harter contends that actually from birth
on, socializing agents have an effect on childrenfis motivation
behaviors. Parents and significant others react to children's attempts
to master tasks in two ways, (a) by evaluating the product, and (b) by
projecting a level of acceptance or rejection of the attempt.

Reactions to the product project right or wrong, success or failure.
Through social learning processes such as modeling and reinforcement
these responses feed into children's sense of competence and affect the
development of their orientation about intrinsic motivation. For
example, if parents react to their childls successful attempts at
batting a ball with smiles and verbal praise the child may desire to
continue the activity and feel competent at batting balls. Reactions
to the process project a level of acceptance. Sharing childrenfls joy
in being affective has a positive impact on their emerging sense of
personal worth. Ignoring or conveying a sense of little value for such
efforts may temper childrenfls responses, leading to less positive
feelings of worth.

While affect is the central correlate to motivation for mastery
behaviors, perceptions of competency and feelings of self-worth become

related to motivation, as well. Thus, in Phase II the parents or

32

C. d 652 cote: Go... 30302:
c2323»: 02.2330 .0 2.0.5.033“. on: .o H enact “a 0.52“.

      
       

20.
6.532.; .3350
-0352: .i
Auuocvoa .0897 c2333.: Imam
-o>:ouo£ .o gauge—co ”:11 .n

            

 

            

 

.- -- 5:02.05 I _-l
3c2onEoU .o 3.0.0.:8 _
.o oacom

 

         

5.53 .2393;
.o 02.3

_o>ocaaon_o
co _o>ocaa<

 

   

oucoioooxx .8050

 

 

£050 >4 8:32..
.o co_.o:.o>w

  
  

 

 

m0 —O>Od =030>=02

 

 

 

eocofiozw

 

 

053=O&\u00093m
8:39...

    
   

.coEco.._>cw

 

  

 

£3.50 9.303 093.225
29:32 fotoi .cotcoaevc.
8390....

30.32 2: .o 2cecoafiou

2.. S .85

 

 

9.030300 30:02

 

 

 

33

caregivers lay the foundation for the development of childrenfis sense
of competence during infancy and early childhood. Presumably climbing
fences, running down hills, and chasing a friend still bring inherent
pleasure, but the purity of the perceptions begins to be tempered by
the reactions of significant others.

Phase III (See Figure 3). The hypothesized relationship of the
major correlates affecting competence/effectance motivation in Phase
III has been characterized by Harter (Harter & Connell, in press) as a
chain. Four primary correlates are now identified (actual competence,
perception of competence, competence affect, and understanding what
controls performance outcome). ‘The first three of these appeared in
Phase II; however, their relationship to each other and to motivation
becomes attenuated somewhat.

In this phase, the effect of children's actions on their
environment may be interpreted as their actual level of competence. As
in Phase II, significant others respond with both evaluative and
acceptance/rejection information. Each continues to have an effect on
perceptions of competence and affect, respectively; However, older
children begin to perceive themselves in a more complex manner as the
capacity for logical thought and appreciation of the relationship
between cause and effect emerges. Harter referred to this new aspect
of information processing as the internalization of cognitive-
informational structures (labelled in Figure 3 as Internalized Set of
Mastery Goals and Criteria for Success). Internalized mastery goals
and criteria for success begin to be formed when children adopt the

performance standards of the significant Others in their world. This

34

 

 

Unknown Control

 

 

 

 

 

 

 

 

 

Actual Competence

 

 

Effect on the environment:
If Known Control Success or failure
(

internal or external) T; ‘

 

 

 

 

 

 

 

 

 

 

 

t' ---------------- .1
: Evaluation by whenJi
1 I

b.---------------J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Domain Specific *
Competence/Effectance r----"-----’-"-""fi'
Motivation , Internalized set of Mastery '
: Goals and Criteria Lo: Success :
L ................ J
Eerceived M l l ‘
l' -------------- H
: Acceptance or Reiection l:
: by others :
L ............ _.|

 

 

(Signifies most important FCompetence Affect ]/

relationships in the model)

(signifies correlates to motivation)

 

 

 

 

 

 

[:1
l l
Liz—.11
(signifies antecedents to motivation)

 

 

Figure 3: Phase 1]]: of thedevelopment of effectance motivation (adapted from Harter.
19810. p. 10 and Harter 8. Connell, in press). _

35

process allows children to judge how much they value a specific domain
and what level of performance constitutes success or competence.

The consistency and relevancy of the evaluative feedback children
receive from others influences the degree to which goals and criteria
for success are internalized. This, in turn, affects perceptions of
who or what controls performance outcome. If children are given clear,
consistent, and relevant evaluations about their performances they will
develop consistent and realistic internalization structures and
understand who controls performance outcomes, iJL, internal or
external/powerful others. Inconsistent evaluations lead to ambiguities
about the source of control (unknown control).

Harter (Harter & Connell, in press) contended that'the driving
force for the "motivation chain" is the child's perception of who
controls the outcome of performance situations. Those who understand
who controls it are the high level performers; those without a clear
understanding perform less competently. Competence affect results from
cxuys perceptions of competence and the acceptance or rejection of that
performance level by others. ‘It also affects motivation level but not
as directly as perceived competence.

Support for the directionality of the variables in Harter’s model
comes from a study involving upperelementary and junior high school
level subjects (Connell, 1981). First, subjects were tested on all
elements in her model. Second, structural equation modeling procedures
(a type of path analysis) were used to test four possible models or

directional relationships among the motivational correlates proposed.

36

The model presented was the one determined to best fit the data
collected.

Most of the resultant relationships seem plausible; However, her
reference to the unknown control element as being directly related to
performance seems overzealous, based on her cross-sectional data and
correlational procedures. Further, Connell (1981) from whose
dissertation these results were derived, refers to the understanding of
control element as "setting the stage" for the cognitive and affective
processing of performance information. Intuitively this seems to be a
more reasonable interpretation.

In summary, Harter (Harter & Connell, in press) proposed that a
clear understanding of who controls performance outcomes (one's self or
powerful others) leads to demonstration of higher levels of actual
competence, which leads to higher perceptions of competence and to a
strong motivation to have an effect on the environment or demonstrate
mastery of tasks in that domain. Conversely, children who do not know
why they are successful or unsuccessful perform less competently,
perceive themselves as less competent and lack motivation to achieve in

that domain.

Gender Differences
Evidence is accumulating that there are gender differences in the
domain of physical activity for the three central variables to be
investigated in this study, iJL, perceived competence, actual motor
ability, and motivation to participate. Based on her own research,
Harter (1978) noted a trend toward gender differences in motivation

levels for tasks in which a gender difference was noted in skill level.

37

For tasks in which males demonstrated an advantage over females, either
in measures of true ability or those which are stereotyped as
masculine, males reported higher levels of motivation. However, for
tasks in which females demonstrated equal or higher skill levels
compared to males, level of motivation appeared to be equal. .Although
Harter's investigations have dealt primarily with cognitive skills, a
similar trend may be expected in motor skills.

Males appear to have an.advantage in level of skill when compared
to females both in true ability and as stereotyped by society. Prior
to adolescence males have a slight, although in most practical
situations a non—significant advantage in actual motor ability. ldhile
minor anatomical differences may account for this disparity in skill
level, strong cultural expectations are almost certain to play a part
in fostering this difference (Herkowitz, 1978).

The stereotype that sport and game participation is more
appropriate for males than for females, while changing, certainly still
exists. Parents demonstrate a greater response to the involvement in
sport of their sons than to their daughters' involvement (Lamb, 1976).
Children also display gender differences regarding their opinion of the
appropriateness of participation by females in sport. In a study
involving third through sixth grade boys and girls, Selby and Lewko
(1976) found males less favorable toward female participation in sport
than were females, regardless of their own level of participation.

Motivation level in sport and physical activity may be measured by
participation versus nonparticipation or persistence in a task. When

motivation is measured via percent of each gender participating in

38

sport, males continue to dominate, even though the number of female
participants is increasing (Kleiber & Roberts, 1983; Lever, 1976; Youth
Sports Study, Phase I, 1976). ldhen persistence in a task was analyzed,
a gender difference was not found (Roberts et ed., 1981). IHowever,
this measure was of a childls anticipated persistence, not demonstrated
persistence. Phase one of a study of sport participation by youth
(State of Michigan, 1976) found that one out of four boys participated
in and completed a season in baseball while only one in ten girls did
ex» Additional research is necessary to determine if females who
choose to participate will try just as hard to succeed as their male
counterparts.

Males typically perform sport skills better than females, and they
also perceive their general level of athletic ability to be higher than
do females (Duquin, 1978; Wiggins, 1973). 'This difference in predicted
ability may not be true for all tasks, however, Evidence suggests that
for those tasks categorized as having "masculine" components such as
speed or strength, males perceive their ability to be higher than
females, but for "neutral" tasks a gender difference is not found
(Corbin, Landers, Feltz, & Senior, 1983; Corbin & Nix, 1979; Corbin,
Stewart, & Blair, 1981).

Gender differences in self perceptions may be masked by the use of
global self-concept scores. The assessment of global self traits or
attributes usually fails to reveal gender differences (Magill & Ash,
1979; Martinek & Zaichkowsky, 1977; Piers, 1969). However, such
measurements tend to mask differences which may be occurring within

separate domains (e49, physical, social, and cognitive) by utilizing a

39

composite score for all assessment items which weights each domain
equally. In reality, children are likely to value some competencies
more than others, or feel significantly more competent in one area than
in others (Rosenberg, 1979). Ability in physical skills is generally
considered to be an important source of status for children (Smoll,
1974), however, athletic inVolvement may be a more salient source of
pride and status for males than for females (Kleiber, 1979; Kleiber &
Hemmer, 1981). Harter (1981) did not find a gender difference for the
cognitive, social, and global subscales across Grades 3 through 6.
Only on the physical subscale did males report significantly higher
perceptions of ability than did females.

The reactions and evaluations of significant others to young
childrenksattempts to perform movement tasks are important to the
development of perceptions of competenceu Gender differences in
perceived competence, particularly in the physical activity domain, may
begin as a result of differential responses by parents. There are
several ways in which significant others may have a differential
effect. Stereotyped attitudes are likely to be reflected in parents'
reactions to and evaluations of the movement performances of their
offspring, ioe" more encouraging for boys, less encouraging for girls.
This may influence developing perceptions of competence and motivate
some to repeat such movement‘s more often than others. The task
experiences one has are further expected to influence onefis attitudes
(Breer & Locke, 1965). If young boys are experiencing movement
activities more frequently, §2d_if they are positive, a more positive

perception of competence should develop. Further, the increased

4O

participation would provide more opportunities for comparison of
abilities whereby a more accurate conception of motor ability may
result. With accuracy may come the realization for some boys that they
are not as good as they thought they were, thereby discouraging
participation. 'The actual point at which this may begin to occur, and
whether it does occur earlier for males than for females is unknown,
but warrants further investigation.

Nicholls (1978) hypothesized that an understanding that ability
and effort are distinct causal factors should lead to more accurate
assessment of one's own competence and thus affect motivation. He
studied children ages 5 through 13, and developed four stages of
reasoning which he believes children go through regarding reasoning
about their achievements in tasks. He found that boys were more
advanced in their reasoning in an academic task situation, and the
older males (ages 9 to 13) chose more difficult tasks to perform than
did females of the same age. However, they were no more accurate in
estimating their own level of competence. Accuracy improved with age
for both genders.

Some support for Nicholls} theories was produced in a study of
fourth and fifth grade participants and nonparticipants in sport
(Roberts et al., 1981). Participants believed ability was more
important than effort as compared to nonparticipants, but a gender
difference in perception of the role of effort and ability was not
found. Participants demonstrated higher perceptions of competence,
reported that they would persist longer in difficult tasks, and had

higher expectations of future success. ldhile admitting the topic is

41

debatable, these researchers suggested that "

."sport selects out those
who perceive themselves as more competent to begin with rather than
contributing in a substantial way to the development of perceptions of
competence" (p. 213). However, this study was cross-sectional in
design and only identified participants and nonparticipants.
Developmental variables may be examined with greater scientific
validity by utilizing a longitudinal design. By following the same
subjects over several years, one may find that participation does not
increase perceptions of ability for those who find success. For those
who are eliminated from teams, are not given an opportunity to
participate in games, or are unsuccessful in other ways, organized
sport participation may decrease perceptions of ability. Therefore,
some of Roberts et al.s nonparticipants may have been drop-outs whose

perceptions of competence were higher at an earlier time in their lives

but were negatively affected by participation in organized sport.

Younnghildren's Extent of Involvement in Sport and Games

The number of children participating in organized sport and games
has been rising in the last few decades (Kleiber & Roberts, 1983;
Martens & Seefeldt, 1979). In addition to the probable trickle-down
effect of increased societal interest in fitness and activity, Seefeldt
(Seefeldt & Gould, 1980) suggested several reasons for the increase in
numbers: (a) lowering of the age at which children may enter organized
sport programs, (b) increase in the number of sport situations available
to children, (c) greater participation by females, and (d) improved

public transportation, providing greater access to sport facilities.

42

The number of children participating in sport rises during the
elementary school years and peaks during the middle school years.
Magill and Ash (1979) studied first through fifth grade children in
Texas and found an increase in participation by grade level. By third
grade over half of their subjects had participated on at least one
organized team during the past year. The second graders reported close
to 50% participation, while many first grade children were already
involved. Results of Phase I of the Michigan Youth Sports Study (State
of Michigan, 1976) revealed that most participants entered organized
sport at approximately eight or nine although some began as early as
age three or four. By 11 or 12 years of age a peak and subsequent
drop-off in participation was observed for males and females.

In addition to time spent in organized sport activities,
participation in non-agency sponsored physical games is frequent.
Kleiber and Roberts (1983) studied the free-time activity patterns of
children between the ages of 9 and 11. They reported that the most
frequently chosen activity for both boys and girls was physical games,
such as ping pong, basketball, and tag, with percent of participation
greater for males than females. Other categories into which childrenks
after-school activities were classified included physical free-form
activities (playing catch, frisbee, climbing), watching TV, lessons
(music, dance, gymnastics), reading, doing homework, etc.

We must confront the facts that (a)<xnnmunity sponsored youth sport
programs are continuing to grow, with most children reporting that
their initial contact with sport is non—school related, and (b)

reductions in the budgets of elementary level physical education

43

programs continue to be a threat to the provision of adequate programs
for all children. In order to regain community support for and
strengthen our physical education programs, a more scientific
examination of what we can accomplish and why these goals are important
is necessary; Many children become involved in sport prior to the
third grade. We cannot continue to begin our investigation of the
interrelationships of variables such as perceived competence, actual
motor ability, and motivation to participate at the upper elementary
grade levels. By third grade the foundations of motor abilities and
perceptions of those abilities are well established. We must first
understand the normal course of development if we are to facilitate
development in any area.
Overview of the Study

The purposes of this study were to investigate the relationship
between perceived and actual competence in motor ability and the
relationship of each to motivation to participate in sport and physical
activity. The population of interest was young children, specifically
children in kindergarten through the fourth grade levels. Finally,
possible developmental changes among the relationships between these

variables were of concern.

Research Hypotheses

 

It was hypothesized that:

1. There is a significant relationship between perceptions of
competence in motor ability and participation motivation.

2. There is a significant relationship between motor ability and

participation motivation.

44

3. There is a significant relationship between motor ability and
perceptions of competence in motor ability.

4. The strength of each relationship identified above increases with
age.

Definitions

 

Definitions of the following variables will aid in understanding

the design of and analyses of this study:

1.

Perceived Competence in Motor Ability — This variable denotes
children's perceptions of their ability to perform gross motor
tasks, as reflected in their responses to test items.

Motor Ability - This variable is defined by childrenfls performance
on 9 selected gross motor tasks. Four of these items may be
classified primarily as sport skill items; five of them reflect
motor ability items.

Participation Motivation - For this variable,cfluldren were
classified as participants or nonparticipants in organized sport
based on their response to item #2 in the Questionnaire Regarding

Participation in Sport and Physical Activity.

Participants = children who were involved in and completed the
season in at least one organized sport during the
past year.

Nonparticipants = children who did not complete at least one
season of involvement in an organized sport in the

past year.

CHAPTER II

METHOD

Overview

To assess the interrelationships among perceived competence in
physical activity (motor skills), actual competence in motor activity,
and participation motivation, a field-based descriptive study was
conducted. Children in kindergarten through the grade four level were
assessed, in school, regarding their perceived competence in physical
activity. Motor ability was determined by a battery of motor
performance items, also assessed at school. .A take-home questionnaire
was used to determine childrenfls patterns of participation in sport and
the reasons for such choices. Children were asked to complete their
questionnaires with parental assistance.

The selection of thisage range was based on literature pertaining
to the development of perceptions of self and the onset of
participation in organized sport. Such research suggests that by
approximately age seven or eight, children have developed sufficient
cognitive skills and have had the necessary experiences to be forming
stable and hierarchical perceptions about themselves (Harter, 1981b;
Rosenberg, 1979). By third or fourth grade some level of accuracy in
perceptions of self appears to be emerging (Boersma & Chapman, 1978;
Harter, 1981b) Some of the environmental variables which affect the
development of accuracy of perceptions of ability may be more prevalent
at an earlier age in the physical domain (e49, immediate knowledge of
results, opportunities for social comparison). 'Thus, it is possible

that perceptions of physical ability may become established and stable

45

46

earlier than perceptions of ability in other domains. By age eight or
nine many children choose to become involved in organized sports
(Magill & Ash, 1979; State of Michigan, 1976). In order to understand
the relationship between perceived competence and participation in
motor activity at this age and beyond, the antecedent developmental
changes must also be determined. Research which has examined one or
more of the variables of interest for this study has typically focused
on only the early, middle, or late childhood years. This study
involved a more comprehensive age-related analysis, encompassing the
time period when critical developmental events may be occurring.

The purposes of this study were (a) to examine the relationship
between children's self-concept of physical ability and their
motivation to participate in sport and games, (b) to examine the
relationship between childrenfls actual motor competence and their
motivation to participate in sport and games,(kfl to examine the
relationship between children's perceptions of competence and their
actual competence in motor ability, and (d) to examine the nature of

the age-related and gender-related differences in these relationships.

Subjects

Boys and girls in Grades K through 4 served as subjects for this
study. Twenty-five males and 25 females were tested within each grade
level. The mean ages of subjects in Grades K, 1, 2, 3, and 4 were
70.10, 83.08, 95.34, 108.12, and 119.16 months, respectively.

Subjects were drawn randomly from a pool of volunteers from two
elementary schools in the Carbondale, Illinois area. Only school

districts which employed physical education specialists were considered

47

for this study. 'This decision was based on some evidence that

children provided with direct instruction in physical education classes
had significantly more accurate perceptions of their physical ability
than children who did not have physical education instruction (Yeatts

& Gordon, 1968).

Children in the elementary schools in this district are bussed to
achieve racial integration. Within this study, 62% of the subjects
randomly chosen from the pool of volunteers were white. 'Ehirty percent
were black while 8% represented other racial backgrounds. 'The racial
distribution of this sub-population was proportionately similar to that
of the entire Carbondale Elementary School District population for
Grades K through 4.

Children who were contacted in the five grade levels returned
their questionnaires at an average of 70%. The range of values for
percent of response by grade level was 67% to 77%.

Based on their responses to survey questions, of the 250 subjects
in this study, 122 were classified as participants and 128 as
nonparticipants. The percent of participation increased as grade
level increased. Values of 28%, 48%, 52%, 50%, and 66% were obtained
for Grades K through 4, respectively. Actual numbers of male and

female participants at each grade level are presented in Figure 4.

Data Collection
Administrators of schools in the Carbondale, Illinois school
district were contacted for participation, since they represented a
large, suburban district with racial integration throughout all

schools. 'The socio—economic status of students within this district

48

C“)

 

 

 

SlNVdIDllHVd :lO HBSWDN

GENDER AND GRADE LEVEL
Figure 4. Number of males and females who were classified

as participants within each grade level.

49

district was considered to be primarily lower middle class to middle
class.

Contact with school personnel regarding the study was initiated by
a telephone call to the district superintendent. Subsequently, a
meeting was arranged to discuss the nature of a schools' involvement
and purposes of the study. Following the visit, the superintendent
agreed to provide his cooperation, to choose the schools to be
utilized, and to contact the principal at each school to inform them
that this study met with district approval. Subsequent personal
contact was made with each principal, printed materials describing the
nature of the study were presented, and time and facility commitments
were discussed. Each principal agreed to cooperate and chose to inform
the teachers about the study.

Following this a letter explaining the study, a questionnaire
regarding participation in physical activity and sport, and a parental
permission form were sent to the parents of each child in Grades K
through 4 in each participating school. Approximately two weeks
following the date questionnaires were sent home, children's names were
selected at random from the pool of respondents, to fill the quotas as
designated for gender and grade level. Children were asked
individually, prior to testing at school, if they would participate in
the remainder of the study. .All children chosen from the pool of
respondents agreed to be tested. Testing was conducted during the
months of March, April, and May of 1984.

Two different sets of procedures were used to test perceptions of

competence in motor ability. Two procedures were necessary due to the

50

range in ages of the subjects and the protocol designated in the
manuals for the companion assessment instruments used. 'The instrument
used with younger children (The Pictoral Scale of Perceived Competence
and Social Acceptance for Young Children, Harter, Pike, Efron, Chao, &
Bierer, 1983) required individual testing in a quiet, non-distracting
atmosphere. Each child in Grade K through 2 was accompanied from his
or her classroom by this investigator to a quiet conference room within
the school. 'The subject was seated opposite the tester while testing
took place. The tester read directions, descriptions of each picture
plate, and perceived competence questions to subjects from a manual
prepared by the test developers. .Administration time was approximately
15 minutes per child. Following completion of the test children were
accompanied back to their rooms and thanked for their participation.
Children in Grades 3 and 4 received the Perceived Competence Scale
for Children.(Harter, 1979b), by classroom. Printed test forms were
distributed and general directions given by this investigator. Sample
questions and subsequent test questions were read aloud to the
students. In addition to this investigator, classroom teachers
remained in the room to answer students' questions throughout the test.
Administration time was approximately 30 minutes per classroom.
Following the perceived competence testing, and within
approximately two to three weeks, each child was given the motor
ability/sport skills assessment battery; Children were tested in pairs
of the same gender and grade level. .As in the perceived competence
testing for younger children, subjects were accompanied from their

classroom to the gymnasium for testing, then back again by the test

51

administrator. ‘Upon arrival at the gym subjects were asked to remove
their shoes and socks. Test items were presented in a counterbalanced
order, randomized by pairs of subjects. Four sequences of task
presentation were used. Subjects in each test pair took turns being
the first one to perform test items.

Motor ability/sport skill assessments were accomplished with the
assistance of a graduate student from Southern Illinois University.
Inter-rater reliabilities were measured for the nine items and were
found to be high. A mean of the Pearson product correlations based on

10 subjects was .98.

Instrumentation

Perceived Competence Scales

Two testing instruments were deemed necessary to assess perceived
competence in motor ability. Recent evidence suggested that
preschoolers and primary grade level children do not think of
themselves in as many separate domains as do children in the upper
elementary grade levels. In addition, some items appropriate for older
children are not meaningful for younger ones (Harter & Pike, in press).
The assessment instruments chosen were developed by the same principal
researcher with consideration for the points made above (Harter, 1979b;
Harter et alu,l983). Each test contained the same question and
response format.

Perceived Competence Scale for Children (Harteri 1982). The
Perceived Competence Scale for Children was designed to measure
childrenfls perceptions of their own ability in each of three competency

domains and their feeling of general self—worth or esteem. The

52

competency domains included; (a) cognitive competence, which was
oriented toward academic achievement, (b) social competence which
related to social skills, and (c) physical competence, which reflected
ability to play sports and games. The general self-worth subscale was
considered and designed to be independent of any specific competency
domain.

The format of each test item represented an improvement over
previous measures of perceptions of oneself in its attempt to reduce
the social desirability effects. A "structured alternative format" was
used, and both responses were worded so that they would be perceived as
socially legitimate. The subject first.selected the statement which
was most like him or her, then discriminated further by indicating if
the answer chosen was really true or just gggg pf true for him or her.

Each of the four subscales has seven items, scored on a 4-point
scale, with 1 indicating low perceived competence and 4 high
perceptions of competence. The scores are summed and then averaged for
each subscale, resulting in four separate subscale means representing
perceptions of competence for each domain and general self-worth.

The validity of the scale is based on factor analytic procedures
which involved an initial sample of over 2,000 third through ninth
grade children from four states. Results indicated four distinct
factors, with only one item cross-loading consistently on a second
factor. Average factor loadings of .57, .45, .53, and .38 were
obtained for items on the cognitive, social, physical, and general
subscales, respectively. The internal consistency of each subscale was

assessed using the Kuder-Richardson formula. These reliability

53

estimates ranged from .73 to .83. Test-retest reliability data
involving a 3-month interval ranged from .70 for the general self-
esteem tol.87 for the physical competence subscales.

Since the main purpose for utilizing this scale was to assess
perceptions of competence in the area of physical activity, and only
seven items are utilized to assess this domain, seven items were added.
An identical format was used for the items, bringing the total item
number for this subscale to 14. The purposes for doing this were (a)
to increase the reliability of the subtest and (b) to provide items
related to tasks assessed in the physical skills test battery.

The addition of items was on advice from personal communication
with the test developer, (Harter, August, 1983). Following data
collection, the total subscale of perceived competence in motor ability
was subjected to reliability analysis using the coefficient alpha, to
test for internal consistency among the items. A standardized item
alpha (reliability coefficient) of .85 was obtained which approximates
the value the test developer found for internal consistency among the
original items. Such a value is considered acceptable for measures
such as this. The range of alphas obtained by item, if the item was

deleted from the test, was from .83 to .85.

Items ip_Harter's subscale Items added:

for physical activity:

 

1. do well at all sports 1. do well at games involving
2. better at sports kicking balls
3. do well at new activity 2. do well at games involving

4. good enough at sports catching balls

54

5. first chosen for games 3. able to run fast
6. play rather than watch 4. do well at games involving
7. good at new games throwing

5. can jump far
6. am strong

7. good at dribbling balls

Example of questions included in Harter's scale:

 

 

 

 

 

 

 

 

 

 

 

 

 

Really Sort of Sort of Really
true true for true for true
for me me me for me

Some kids do Others don't

well at all feel that they

kinds of but are very good

sports when it comes

to sports

Example of questions added:

 

 

 

 

 

 

 

 

 

 

 

 

 

Some kids do Other kids don't
well at games feel that they
that involve but are very good
kicking balls at kicking

balls

The Pictoral Scale of Perceived Competence and Social Acceptance
for Youpg_Children (Harter & Pike, in press). This scale was developed
for children ages four through seven years, and is designed to measure
two general constructs of childrenfis perceptions: (a) perceptions of
competence and (b) perceptions of social acceptance. The competence
domain is further divided into items measuring perceptions of cognitive

and physical competency while the social acceptance domain consists of

55

and physical competency while the social acceptance domain consists of
items reflecting perceptions of peer and maternal acceptances ‘Unlike
the companion scale for older children, a general self—worth subscale
is not included.

The test item format.is a "structured alternative format," as is
used with older children, however, picture plates are used instead of
printed test items. 'The child points to the picture which depicts the
child which is most like him or her, then to a large or small circle
beneath the picture, indicating whether the child in the picture is
really like him or her, or just Egg; pf_like him or her. A set of
plates is available for pre-school through kindergarten level children
while a separate set is available for first and second grade level
children. In order to provide competency items that would be
discriminatory, Harter and Pike (the test developers) determined that
not all items were appropriate for both developmental levels. Some
items do, however, overlap.

Each of the four subscales has six items, scored as in the
Perceived Competence Scale for Children. However, factor analysis
revealed only two domains, suggesting that at least for-the competency
domain, children do not distinguish between their cognitive and motor
abilities. 'The factor analysis was based on a sample of over 250, 4-
through 7-year—old boys and girls. Items had moderate loadings on
their designated factor, and with only two exceptions in the sample of
4- and 5—year-old children, did not cross-load. Average factor
loadings for the sample of 4- and 5—year-old children were .47, .29,

.41, and..58 for cognitive, physical, peer acceptance, and maternal

56

acceptance subscales, respectively. In the same order, mean factor
loadings for the sample of 6 and 7 year-old children were Sit .39,
.60, and .55. Additional support for the validity of the subscales
comes from the demonstrated ability of scores from at least three of
them to discriminate between children who were identified as having
problems related to that competency area and children who were not
experiencing problems.

The internal consistency of each subscale, as measured by the
coefficient alpha, ranged from .50 to .85. When subscales were
combined according to their respective factors, the range was from .75
tx>.89. Total scale reliability for the preschool/kindergarten items
was .88; for first/second grade items, .87.

As with the other perceived competence instrument utilized in this
study, additional items were added to the subscale for physical
ability. Each subscale currently contains six items, therefore six
additional items were added bringing the total to 12. The internal
consistency of the items in the expanded subscale for perceived
competence in physical ability was examined with reliability analysis
using coefficient alpha procedures. An acceptable standardized alpha
of .80 was obtained which approximated that obtained by the test
developer. 'The range of alpha values obtained by item, if deleted from

the test, was from .76 to .81.

 

 

 

Items included ip_Harter.§. Items added £2 the scale:
Pike's physical activity

subscale:

1. good at swinging items for 1. good at jumping

2. good at climbing pre-school 2. good at bouncing a ball

57

 

 

 

3. can tie shoes & kinder— 3. good at kicking a ball
4. good at skipping garten level 4. good at throwing a ball
5. good at running children 5. good at catching a ball
6. good at hopping 6. am strong

Items included _i_rl Harter _&_ Items added _t_:_g_ Elle_§§§1_e_:
Pike's physical activipy

subscale:

1. good at swinging items for 1. good at jumping

2. good at climbing first and 2. good at batting a ball
3. good-at bouncing ball second grade 3. good at kicking a ball
4. good at skipping level 4. good at throwing a ball
5. good at running children 5. good at catching

6. good at jumping a rope 6. am strong

Examples of the picture plates used to assess perceived competence in

motor domain tasks may be found in Appendix F.

Assessment Batterypfor Motor Abilities and Sport Skills

The items included in the motor assessment battery were chosen to
represent (a) specific motor abilities as identified by factor analysis
studies and (b) motor skills common to the most popular organized
sports of children in the community being studied. Test items had to
meet several criteria, as listed below. Each item had to:

1. be appropriate for the age range of 5 to 10 years (appropriate
to children's cognitive as well as motor abilities),

2. be reliable,

3. be valid,

58

4. in conjunction with other items in this assessment battery, not
require a total administration time in excess of 30 minutes,

5. be administerable in a typical elementary school gymnasium or
multi—purpose room, and

6. not require elaborate or non—portable equipment.

Preliminary items were chosen by this investigator and reviewed by
three content experts. ‘Following minor revisions, nine items were
agreed upon as representative of the two content areas specified, motor
abilities and sport skills. Subsequently, a pilot study was conducted
to determine if all items could meet criteria one through four as
indicated above. One hundred ninety-four children in Grades K through
4 were assessed with the battery of task items. (Fbr details of the
pilot study see Appendix AJ Based on the results of this preliminary
study all of the original items were maintained, four of which required
revisions in protocol. Test-retest reliability coefficients for a
sample of 10 first grade subjects ranged from .57 tol.95 for the nine
items. Specific item reliabilities may be found in Appendix B.

Face validity of the items was based on the judgment of three
content experts, as noted previously. The validity of the battery of
items as a measure of general motor ability was examined in the pilot
study. All children who were assessed in the pilot study were also
rank ordered within grade level on the basis of demonstrated motor
performance by their physical education specialist. Rank order scores
were converted to z scores and used in multiple regression and
discriminant function analyses. Results of the multiple regression

analyses suggested that the amount of variance that could be accounted

59

for by the nine items, within each grade level ranged from 56% to 77%
(M_ = 66.4%). Results of the discriminant function analyses suggested
that when used in combination, subjects' scores on the nine items could
successfully classify them as belonging in the top, middle, or lower
third of their grade level in motor performance, as indicated by rank
order scores assigned to them. Percent of correct classification
ranged from .69 to .88 (ll = .79).

A brief description of each item included in the battery, how it
was scored, and the ability or skill which it measured follows.

Complete protocol information is located in Appendix D.

Item Motor ability 3; skill tested

 

1. Standing Broad Jump (power of legs)

The broad jump test required subjects to stand with toes behind a
take—off line on the floor and jump as far as possible onto a tumbling
mat. Take-off and landing were on two feet. Score was measured in
inches, calculated to the nearest 1/2 inch.

2. Flexed Arm Hang (muscular endurance of upper arms
and shoulder girdle)

Subjects grasped the 1 1/2 inch diameter chinning bar with a
pronated grip and maintained as long as possible, a position in which
the chin was above the bar and arms fully flexed. Score was number of
seconds position was maintained, scored to the nearest 1/10th of a
second.

3. Side-step Test (agility without running)
From a standing position astride a center line, subjects were

required to slide 6 feet to the right, touching the right foot to the

60

floor outside the right side line. Without turning, subjects had to
slide in the opposite direction to touch left foot outside the left
side line. Sliding continued to alternate side lines until 10 seconds
had passed. IPoints were scored for each line crossed within the 10
seconds.
4. Sixty Yard Shuttle Run (speed and agility while running)
Two cones were placed 15 yards apart with a starting line on the
floor, even with the outer edge of one cone. Subjects ran two laps
around the cones. Score was the number of seconds which elapsed 'from
start to finish, scored to the nearest 1/10th of a second.

5. Sit-up Test (strength and endurance of
abdominal muscles)

Bent-knee sit-ups were performed, on a tumbling mat. Feet were
held by the subject's testing partner. Subject's score was the number
of sit-ups completed within 30 seconds.

6. Playground Ball Dribble Test (eye-hand coordination and a
sport skill)

In this test subjects were asked to dribble a 9 inch red
playground ball in and out of a row of six traffic cones placed 8 feet
apart. Score was the number of cones the subject dribbled past within
30 seconds.

7. Soccer Ball Dribble Test (eye-foot coordination and a
sport skill)

Subjects were asked to dribble CLe., move the ball with their
feet only) a size five indoor soccer ball in and out of a row of
traffic cones. The row consisted of five cones placed 10 feet apart.
Subjects received one point each time their ball passed between two

cones within 30 seconds.

61
8. Soccer Ball Throw Test (power of upper body and a
sport skill)

From behind a restraining line, subjects used a two-handed
overhead throwing motion to throw a size five outdoor soccer ball for
distance. A step forward was taken as the subject threw the ball.
Score was the distance the ball traveled in the air, measured to the
nearest inch.

9. Softball Repeated Throws Test (eye-hand coordination and a
sport skill)

In this test subjects were asked to throw an indoor softball at a
large target on the wall as often as possible within 30 seconds.
Subjects were required to catch (or retrieve) each rebound. The ball
had to be thrown from within a designated throwing area; catching was
not restricted to that area. Score was the number of times the subject
hit the target within 30 seconds.

Test-retest reliability of the items as revised from the pilot
study was assessed during this study for 10 subjects in each of three
grade levels. Five male and five female kindergarten, second, and
fourth grade level subjects were tested a second time on the motor
ability/sport skill assessment battery within approximately two weeks
of initial testing. The Pearson product correlations for the
kindergarten grade level subjects ranged from .53 to .89 (ll of .72).
Values for second grade level subjects ranged from .57 to .98 (M_of
.79) while the range of values for fourth grade level subjects was from

.65 to .91 (ll of .77).

62

Questionnaire Regarding Participation in Physical Activity and Sport

 

In order to examine the level of participation in sport by
subjects and to classify each child as a participant or nonparticipant
in organized sport, a questionnaire was developed. The questionnaire
was sent home with each child with the request that parents or
guardians assist their child in its completion.

Of primary interest was whether or not the child had participated
in and completed the season in at least one organized sport during the
past year. Those who answered this question with a yes were classified
as participants for the purposes of this study. Those who did not
complete at least one season of participation in an organized sport
program were classified as nonparticipants.

Additional information was requested of the subjects which related
to reasons for participating or not participating, amount of time spent
at practice, the sport in which the subject participated, how important
being good at sports was to them, their choices of activities when
given free time, and the amount of time spent in gross motor activity
during a week. Such information was used to describe this sample and
to understand better why these children were or were not participants
in sport.

A copy of the Questionnaire Regarding Participation in Physical

Activity and Sport is located in Appendix E.

CHAPTER III

RESULTS

Three major relationships were examined, via univariate as well as
multivariate procedures. First, analysis of variance (ANOVA) and
multivariate analysis of variance (MANOVA) were used to examine the
relationship between childrenfls perceptions of their competence in
motor ability and their participation motivation. Second, a MANOVA was
used to examine the relationship between motor abilities and
participation motivation. “Third, the relationship between perceptions
of competence in motor abilities and actual motor performance was
analyzed via a MANOVA, as well as correlational techniques. The
analyses and results obtained will be discussed separately for each

major relationship.

Relationship #1: Perceived Competence to Participation Motivation
AEQXA,‘ a 2 X 2 X 5 (Participation by Gender by Grade Level)
ANOVA was performed to compare group differences on perceptions of
competence in motor ability. Due to unequal numbers of participants
and nonparticipants (participation main effect) the general linear
model for regression analysis technique was utilized. This method
allowed for the examination of each effect holding it orthogonal to all
other effects in the model.
The main effect of participation (participant versus nonpartici-

pant in organized sport) was of primary concern. lhawever, sufficient

63

64

research suggests that age and gender are related to perceptions of
ability. Therefore, these variables and all possible interactions were
included in the analysis to investigate the possibility that the
relationship between perceptions of competence and involvement may vary
in relation to gender and or age (identified as grade level in this
study).

Results revealed that the participation main effect, FIl, 230) =
2.69, p < .10 was not significant. The main effect for gender,

[(1, 230) =- 11.16, p< .001, and the grade level main effect, 11(4, 230)
= 9.94, p < .0001 were significant. None of the interactions
(participation by gender, participation by grade, gender by grade,
participation by gender by grade) were found to be signiicant.

Follow-up for determining how the genders differed on perceptions
of competence was accomplished by inspection of mean values. The mean
for males was 3.36 (SD 2 .48) while females had a mean of 3.13 (SD =
.57). ZPerceived competence mean scores could range from one to four
with one indicating low perceived competence and four reflecting high
perceptions of competence. 'The mean values suggested that males
perceived themselves to be more competent in motor skills than did
females.

To investigate the differences among grade levels in perceptions
of competence in motor abilities, mean scores were subjected to tests
for linear and quadratic trends and Scheffé tests of significant
differences between and among groups. A significant linear effect was
found [(1, 245) = 43.34, p < .0001, with mean values decreasing as

grade level increased. The test for a quadratic effect was not

65

significant. Results of the Scheffé procedures suggested three

homogeneous subsets of groups (Figure 5). Third and fourth grade level

Grade Level 4 3 2 1 K
Subset One

Subset Two

 

Subset Three

 

Figure 5. Subsets of children (classified by grade level) who were not
significantly different on perceptions of competence in

motor abilities.

subjects were not significantly different; first, second, and third
grade level subjects did not differ significantly; and kindergarten,
first, and second grade level subjects were not significantly
different. Table 1 contains all cell means and standard deviations.
MANOVA — A second analysis examined the relationship between
perceived competence in motor ability and participation, gender, and
grade level. In this MANOVA procedure subjects' item scores on the
motor ability subscale of the perceived competence scale were used as
the dependent variables. .Although such a multivariate analysis would
conceptually provide a more powerful analysis of the relationship
between the independent and dependent variables than the original
ANOVA, it was not chosen as the primary model for analysis. 'Fhis
decision was made for two reasons. First, only half of the total
number of items used in the motor ability subscale were similar across

all grade levels and therefore half of the available information would

66

NHum
was
a no

oHum
an:
m mu

“HHoo mom muaoaHOHuuoaao: Ho sonaazss

mHum
HHuz
N no

oHum

H

nu:

no

oHnm
NH":
3 HQ

mHnm
ON":
a ow

oum
0H":
m uu

NHum cum cum
«Ha: wHuz mu:
N ha H no M no

"HHou pom muamaHuHuuma Ho poneazs

 

 

Ho>oa ocouu

Ho. oo.m mm. mo.m ow. om.m no. Hm.m om. mm.m Hooch
No. OH.m mo. Hm.m oc. No.m 0e. wo.m mm. HN.m Hm. mm.m moHoEom
we. mm.m Hm. wo.m em. mo.m mm. m¢.m om. mm.m mm. N©.m mOHmz
mm. HN.m Ho. mc.m mm. oo.~ 0o. em.m om. cm.m Ne. mm.m *smuamQHOHuumacoz
me. 0H.m mm. ow.m oo. cm.m om. em.m He. mH.m mm. ha.m mOHmsom
mm. mm.m ac. m~.m mm. cH.m He. mq.m ma. eq.m em. oo.m mOHmz
om. mm.m mm. oH.m on. ON.m He. em.m as. om.m mm. mm.m smucoaHOHuumm
mm .2. NW H... mm .2. mm H... mm n W m 3838
Hmuoa N cOHuoaHowuumm

 

maOHumH>on cumcamuw new name: HHoo

H OHan

Ho>oq Dunno can nouaoo A; muaoaHOHuuoaaoz pom mucoaHOHuumm How
"auHHHn< uouoz aH ooaouoqeoo oo>Hoouom

67

be ignored. Second, most of the items that were similar across the
grades were added to the subscale by this investigator and were
therefore not original items on the scale.

Results of this analysis were similar to those obtained in the
ANOVA analysis. The main effect of participation was not significant,
Wilks' Lambda 3 .95, [(7, 224) = 1.65, p < .12. The main effects for
gender, Wilks' Lambda a .91, [(7, 224) = 3.26, p_ < .003 and grade,
Wilks' Lambda 8 .69, {(28, 809) =- 3.13, p< .0001 were significant. In
addition, one interaction effect was found to be significant, gender
by grade, Wilks' Lambda =- .83, F_(28, 809) = 1.49, p< .05. This
suggested that for one or more items the difference between males'anul
females' perceptions of ability in motor skills was not the same across
all grade levels.

As a follow-up to the significant gender by grade level
interaction effect the univariate E_values were examined for each
dependent variable. At the univariate level only one of the perceived
competence items, ability to dribble a ball, demonstrated a significant
interaction effect, _F_(4, 230) = 3.25, p < .01. Newman—Keuls'
procedures were used to examine differences between all pairs of cell
means involved in the interaction. ‘The primary source of this
interaction appeared to be the fact that males and females did not
differ significantly in their perceptions of competence on this item at
any grade level except fourth. At this grade level the mean for
females (2.68) was significantly lower than for males (3.24). Further,
the mean value for females was significantly lower than for all other

groups except the third grade level males.

68

Because many of the subjects in this study were younger than the
age at which most children become participants in organized sport, an
additional variable, which may also reflect participation motivation,
hours per week spent in gross motor activity, was examined. The
general linear model for regression analysis was used with perceived
competence in motor ability as the dependent variable. Independent
variables were hours of play (a continuous variable) and gender and
grade level (each identified as a categorical variable).

Results suggested that the hours of play main effect was
significant, F_(1, 230) a 3.78, p < .05. A plot of mean values for this
main effect (Figure 6) suggested that as the number of hours in a week
spent in gross motor activity increased, perceptions of motor ability
also increased. The amount of variance in perceptions of competence
for which the full model could account was .26. Hours of play
specifically accounted for only .014 of the total variance. This
appears to be a very small proportion when compared to that accounted
for by the other significant main effects in the model, gender and
grade level, which accounted for .043 and .174, respectively.

In Figure 6, the mean for 20 hours per week conspicuously departs
from the positive relationship between perceived competence and hours
per week spent in gross motor activity. The severity of the drop may
result from the fact that 1 of the 11 subjects had a mean score for
perceived competence which was almost three standard deviations below
the mean for the subject's grade level. A specific reason why this

subject had such a low score could not be identified. lVithout this

3.554
3.50-
3.45-
3.40-
3.354
3.30-
3.25-
3.204
3.15-

69

 

3.10:

PERCEIVED COMPETENCE IN MOTOR ABILITY

cf

I I I I T r
05 IO 15 20 25 30

N: (121) (62) (34) (ll) (8) (14)

Figure 6. Mean values

HOURS PER WEEK SPENT IN
GROSS MOTOR ACTIVITY

for perceived competence by hours of play.

70

score the mean value for 20 hours per week would have been 3JH.
providing a less severe drOp in the graph.

As noted above, the main effects gender F(1, 230) = 13.44, p <
.0003 and grade level, _F_(4, 230) = 13.57, p < .0001 were also
significant. The relationships between these variables and perceived
competence in motor ability were examined previously in a model with
participation as the additional independent variable. Results were
similar. 'Further interpretation of the results relative to these main
effects would therefore be redundant (see pages 64-65). None of the

interactions included in this model were significant.

Relationship #2: Motor Abilities to Participation Motivation

MANOVA - A 2 X 2 X 5 (Participation by Gender by Grade Level)
MANOVA was applied to examine the relationship between childrens'
choices to be participants or non-participants in organized sport and
their performance on nine selected motor ability and sport skill items.
As in the first relationship examined, participation was the effect of
primary interest. Gender and grade level were also included because
ample research indicates that each has a significant relationship to
motor performance. ZFurther, the relationship between participation and
motor ability may change with different levels on one or both of these
variables.

The results revealed a significant main effect for participation,
Wilks' Lambda = .89, [(9, 222) = 3.01, p< .002; gender, Wilks' Lambda =
.71, _F_(9, 222)= 9.91, p_< .0001; and grade, Wilks' Lambda = .26,
£136, 833) a ILLOZ, p_< .0001. None of the interactions (participation

by gender, participation by grade, gender by grade, participation by

71

gender by grade) were significant. Univariate Fltests, as well as
discriminant function analyses were used as follow-up procedures when
significant multivariate results were obtained to determine which
dependent variables contributed most to differentiating the groups.

Results of the univariate E tests for the involvement main effects
showed significant differences between participants and nonparticipants
on four of the nine items in the motor assessment battery (Table 2).
These items were the soccer ball dribble, basketball dribble, softball
repeated throws, and the soccer ball throw items, ith, the sport skill
items. Inspection of the means for these two groups (Table 3)
indicated that participants performed better than nonparticipants on
all of these items.

The discriminant function analysis suggested that five items could
be used to discriminate significantly between these two groups (Wilks'
Lambda = .7950, p.< .0001). These items included the soccer ball
dribble, soccer ball throw, situp, sidestep, and flexed arm hang test
items. However, the discriminant function coefficients associated with
each of these items clearly indicated that the soccer ball dribble item
was the most powerful discriminator.

The discrepancy between the results of the univariate gland
discriminant function analyses may reflect the difference in the
mathematical techniques involved. Univariate {_values reflect separate
group comparisons for each dependent variable. Therefore, they do not
account for intercorrelations among the variables. Discriminant
function procedures (when using the stepwise method).attempt to

identify in a stepwise manner, the most parsimonious subset of

72

Table 2

Univariate §_Values and Standardized Discriminant Function
Coefficients for the Motor Abilities/Sport Skills Test Items

 

 

. Standardized
Discriminant
Dependent Variable Univariate F Function
Coefficient
Participation Main Effect (Wilks' Lambda = .7950 pK.0001)
1. Soccer dribble 22.69** 1.00
2. Basketball dribble 10.56** XX
3. Softball repeated throws 5.20* XX
4. Soccer ball throw 4.28* .30
S. Broad jump 0.46 XX
6. 60 yard shuttle run 2.85 XX
7. Situp 0.05 -.36
8. Sidestep 0.35 -.25
9. Flexed arm hang 2.54 .16
Gender Main Effect (Wilks' Lambda = .7352 23.0001)
1. Soccer dribble 47.47** .81
2. Basketball dribble 40.04** .52
3. Softball repeated throws 32.27** XX
4. Soccer ball throw 17.29** XX
S. Broad jump 11.96** .40
6. 60 yard shuttle run 5.93* .28
7. Situp 0.77 -.41
8. Sidestep 2.73 -.80
9. Flexed arm hang 6.63** .22
(Wilks' Lambda = .2483 pfi.00001)
Grade Main Effect (Wilks' Lambda = .7811 p§.0001)
1. Soccer dribble 33.93** -.18 -.70
2. Basketball dribble 79.77** .40 .30
3. Softball repeated throws 80.78** .41 .38
4. Soccer ball throw 52.17** .20 .59
5. Broad jump 36.08** .07 .08
6. 60 yard shuttle run 30.49** -.07 .70
7. Situp 28.24** .22 -.36
8. Sidestep 27.67** .28 -.27
9. Flexed arm hang 1.29 -.23 .39

 

*p_< .05. *fp < .01.
XX 2 item not included in the discriminant function equation

73

Table 3

Means and Standard Deviations for Main Effects Participation. Gender,
and Grade Level on the Motor Abilities/Sport Skills Battery Items

 

Dependent Variable

 

W

Participggts Nonparticipants

14. fl 1‘. .52
1. Soccer dribble 7.52 2.30 5.59 1.80
2. Basketball dribble 16.59 6.15 12.24 5.75
3. Softball throw 10.88 3.77 8.45 4.04
4. Soccer ball throw 196.98 74.61 156.06 64.11
S. Broad jump 47.38 9.68 43.75 8.92
6. 60 yard shuttle run 18.00 1.72 18.82 1.93
7. Situp 14.42 5.30 13.01 4.81
8. Sidestep 15.15 3.35 14.34 3.13
9. Flexed arm hang 10.03 7.78 7.37 6.38

Gender
Males Females

9. a M 5.9
1. Soccer dribble 7.41 2.18 5.67 2.02
2. Basketball dribble 16.32 6.38 12.41 5.64
3. Softball throw 10.78 3.91 8.51 3.97
4. Soccer ball throw 192.18 75.82 159.88 64.89
S. Broad jump 47.44 9.43 43 60 9.12
6. 60 yard shuttle run 18.14 1.76 18.70 1.95
7. Situp 14.10 5.25 13.29 4.91
8. Sidestep 14.58 3.44 14.90 3.07
9. Flexed arm hang 10.04 7.25 7.29 6.92

Grade
9. 9 .2.

9. 5.9 9. 5.9 M 59
1. Soccer dribble 7.94 2.22 7.86 2 01 6.75 1.54
2. Basketball dribble 20.56 3.62 18.18 4 76 14.43 4.30
3. Softball throw 13.75 2.76 12.32 3 12 9.66 2 44
4. Soccer ball throw 251.35 66.56 212.32 64.87 173.24 43.04
S. Broad jump 52.82 7.72 51.59 7.86 44.61 6.28
6. 60 yard shuttle run 17.47 1.13 17.06 1 30 18.14 1.50
7. Situp 16.80 3.98 16.80 3 79 14.66 3.32
8. Sidestep 16.95 2.58 16.27 2 53 15.41 2.31
9. Flexed arm hang 10.65 9.49 9.24 6 97 9.36 6.81

.1. 9

M 59 9. 5.9
1. Soccer dribble 5.81 1.94 4.31 1.29
2. Basketball dribble 11.47 4.84 7.19 3.21
3. Softball throw 7.12 2.45 5.32 2.40
4. Soccer ball throw 135.04 37.70 108.19 32.70
S. Broad jump 41.83 6.62 36.16 8.03
6. 60 yard shuttle run 19.26 1.76 20.20 1.65
7. Situp 10.82 5.12 9.40 4.03
8. Sidestep 13.62 2.64 11.44 2.88
9. Flexed arm hang 7.13 5.41 6.94 6.29

 

74

groups (Klecka, 1975). The first step involves choosing one variable
which has the highest value on the selection criterion. At each
subsequent step the variable added to the equation is the one which
will yield the best criterion score (maximize the statistical
difference between the groups) given the variable(s) already in the
equation.

Of the dependent variables included in these analyses, the soccer
ball dribble, basketball dribble, and softball repeated throw items
were highly correlated (Table 4). The fourth sport skill item, soccer
throw, also demonstrated a moderate correlation to two of those three
items. Therefore, when the strongest of these items was entered into
the discriminant equation little of the remaining variance could be
accounted for by these related variables.

Post hoc univariate E_tests for the main effect gender showed a
significant difference between males and females for seven of the nine
items (Table 2). ‘These included all four sport skill items (soccer
ball dribble, basketball dribble, softball repeated throws, and soccer
ball throw) and three of the five motor ability items (broad jump, 60
yard shuttle run, and flexed arm hang). Examination of mean values
indicated that males performed better than females on all of these
items.

The discriminant function analysis also identified seven of the
nine items as significantly related to discriminating between the
groups (Wilks' Lambda 2 .7352, p_< .0001). Specific items varied
somewhat from the univariate results. 'Two of the sport skill items

were not included in the discriminant equation while the two motor

75

 

 

om. amummwflm
mm. Om. asufim
mm.l Hm. om.| mauuszm coma 00
mm. mm. mo. qo.l gash cocoa
mm. «m. om. mm.1 00. zomnu Hams pmuuom
om. mm. on. oc.u 0“. mm. mzonnu
cmummamu Hamnumom
mm. cm. 00. 50.: 00. mm. ow. mannwuc Hamnumxmmm
mm. we. mm. No.u on. me. cm. on. massage umuuom
mam: aoummcfim asuwm manusnm neon zousu mzomnu massage
sum new» vmomm Hana commoqmu Hamnumxmmm
cmxmam oo umouom Hamnumom

 

>mouumm acmEmmmmm< mHwam uuoam\mmfiuwawn< nouoz ecu :o

a «Hams

WEQUH How Xflhumz GOfiumHmHHOUHmu—CH

76

ability items which were not significant in the univariate analysis
provided some discriminating power.

In the discriminant equation the soccer dribble and sidestep items
were weighted most heavily. Mean scores (Table 3) indicated that males
performed better on the first of these items while the second one
represents the only item on which females scored better than males.

The five additional items reflected motor ability and sport skill items.
Their coefficients were moderate to low for this equation and may be
inspected in Table 2.

Significant univariate Es were obtained for eight of the nine
motor items for the main effect grade level (Table 2). {dean values for
these items generally improved as grade level increased (Table 3).
Subjects"performance scores on one item, the flexed.arm hang, did not
differ significantly by grade level.

The discriminant function analysis for grade level resulted in
three statistically significant discriminant functions, at the .05
level of significance. However, the third function was very weak, with
an associated Wilks' Lambda of .90. Additionally, the subsets of
homogeneous grade levels produced by this function provided little
meaningful information. Therefore, only the first two functions will
be discussed further.

The first discriminant equation was associated with a Wilks'
Lambda of .2483 with p_<l.00001 (Table 2). It accounted for 89% of the
total discriminating power which could be generated by all possible
functions combined. This equation, in addition to providing

information which discriminated among the five grade levels, further

77

suggested a distinction between kindergarten, first grade level
subjects versus second, third, and fourth grade level subjects. The
basketball dribble and softball throw items were relatively stronger in
the equation than the others. ‘However, seven of the nine items
included in the analysis contributed a moderate amount of
discriminatory power to the equation, and no items were excluded from
the function.

The second discriminant function equation was associated with a
Wilks' Lambda of .7812, and a E< .0001. As in the first function
obtained, all items were included in the equation. Those items which
were weighted most heavily relative to the others were the soccer
dribble, soccer ball throw, and 60 yard shuttle items. This
discriminant function also discriminated among the grade levels, but
particularly separated the kindergarten subjects from the others and
the fourth grade subjects from the others.

The relationship between motor performance and participation
motivation as reflected in hours per week spent in gross motor activity
was investigated by an additional MANOVA. .All nine motor ability/sport
skill test items were used as the dependent variables. Independent
variables were hours of play (a continuous variable), gender, and grade
level. ”The general linear model for regression procedures was used so
that the independent variable, hours of play, could be included as a
continuous variable.

Results suggested that the main effect, hours of play, was
significant, Wilks' Lambda = .89, [(9, 222) = 3.18, g < .001.

Examination of the univariate analyses generated for each dependent

78

variable suggested that this variable was significantly related to
seven of the nine motor items (Table 5). Plots of the mean performance
scores for subjects at six values for hours of play suggested an
increase in performance on each of these items as number of hours of
play increased (Figure 7).

For several motor items a drop in performance was noted at 30
hours per week (see Figure 7). this may represent an artifact due to
the motor performance of the three kindergarten level subjects in this
subgroup. Across Grades K to 4, kindergarten subjects scored lowest on
all items. These three subjects demonstrated generally lower than
average motor abilities for this grade level.

The main effects, gender, Wilks' Lambda 2 .89, [(9, 222) = 3.00, E.
< .002 and grade level, Wilks' Lambda 2 .49, 5(36, 833) =- 4.88, p_ <
.0001 were significant. The relationships between these variables and
motor performance were examined previously with similar results (see
pages 74-77). 'Therefore followbup analyses were not conducted. In
addition, none of the interactions tested generated a significant E
value.

Relationship #3: Perceived Competence in Motor Ability to Demonstrated
Motor Ability

MANOVA - A 3 X 2 X 5 (Perceived Competence in Motor Ability by
Gender by Grade Level) MANOVA was performed to compare the relationship
between children's perceptions of their ability in motor activity to
their actual level of performance on selected gross motor tasks.

Within each grade level childrenhsrnean scores for perceived competence

were used to categorize them into the top, middle, or bottom one-third

Univariate Elvalues for the Motor Abilities/Sport Skills Test Items

79

Table 5

 

Hours of Play Main Effect

Univariate.E

 

1.

Soccer dribble

. Basketball dribble

. Softball repeated throws
. Soccer ball throw

. Broad jump

. 60 yard shuttle run

. Situp

. Sidestep

. Flexed arm hang

17.
16.

18.

60-11")!-
3891-41-

65**

.OO**
.04*
.42**
.16*
.56

.20

 

{p < .05. *f2 < .01.

NUMBER OF CONES DRIBBLED PAST

NUMBER OF TIMES TARGET WAS HIT

SOCCER BALL DRIBBLE

 

N8 (12I)(62)(34) (11) (B) (14)

131
121
q
H-
1
10-9

d
9-
8‘

d

7-

r I I I I I
05 10 15 20 25 30

80

NUMBER OF CONES DRIBBLED PAST

19-

d

18-

17-

q

16"

15-I

.

14-I

d

13-l
124

d

11-I

BASKETBALL DRIBBLE

 

to?

fiIIIIr
51015202530

HOURS PER WEEK SPENT IN GROSS MOTOR ACTIVITY

SOFTBALL REPEATED THROWS

 

6%

I T r I I I
05 10 15 20 25 30

DISTANCE (In inches)

220 -

210-I
J

200!

d

190-

d

180‘
.1

170-1
.9

160--l

SOCCER BALL THROW

 

.1
ISO
:2

I r I I I i
05 10 15 20 25 30

HOURS PER WEEK SPENT IN GROSS MOTOR ACTIVITY

Figure 7. Performance on motor abilities/sport skills items by
hours per week spent in gross motor activity.

81

on mm on m— o— no
. . . . . ..

 

‘

nah—m

O—
N—
v—
o—
c—

on

SCINODSS 08 NI GBlETdWOD HSBWON

. Avoacchuv .h 059".

>:>:u< copes. 39.0 2. pzmmm 2mm; 5.. £50:
.1. .8 2: $3 88 cacuz

on ma ca .3 0— no
- - . . .

 

23¢ math—.5 Oz<> co

\

 

JIOK—

<5.
52
(SPUODGS) awu

Ind—

00 mm on n— 0— no
. . p p - _

 

m<<3q O<O¢m

Qua:

lav

 

(sewn!) SDNVlSICI

82

(Groups 1, 2, and 3, respectively) among their peers on perceptions of
ability in this domain.

Results indicated that all three main effects were significant.
The main effect, perceived competence, generated a Wilks' Lambda =
.8219, [(18, 424) = 2.42, p< .001. The gender effect resulted in a
Wilks' Lambda 2 .6986, _F_(9, 212) = 10.16, g< .001, and the main
effect, grade level, generated a Wilks' Lambda = .2266, £(36, 796) =
10.76, p < .0001. None of the interaction effects reached
significance.

In order to determine which items contributed most to distinguish-
ing among the children as grouped by perceived competence, univariate E
and discriminant function analysis procedures were used. A significant
difference was obtained for all sport skill items and one motor ability
item, the broad jump (Table 6). Idean values for performance scores on
all motor abilities/sport skills items suggested a decrease in per-
formance associated with a decrease in perceptions of ability (Table 7).

Discriminant function procedures resulted in one significant
function, associated with a Wilks' Lambda of .8712, p_< .002 (Table 6).
The softball repeated throws item emerged as the most powerful
discriminator among the dependent variables. .Additionally, the soccer
dribble, basketball dribble, soccer ball throw, and broad jump were
identified as strong contributors to the equation. Relative to group
differences, this function appeared to separate best those subjects who
were lowest on perceived competence (Group 3) from those of average to

high levels of perceived competence (Groups 2 and 1).

83

Table 6

Univariate §_Values and Standardized Discriminant Function Coefficients
for the Motor Abilities/Sport Skills Test Items

 

Standardized

Discriminant
Dependent Variable Univariate E Function

Coefficient

 

Physical Perceived Competence (groups=top, middle, and bottom thirds)

(Wilks' Lambda 3
.8713 p_< .002)

1. Soccer dribble 7.38** .62
2. Basketball dribble 8.50** .70
3. Softball repeated throws 3.63 -1.29
4. Soccer ball throw 10.08** .72
S. Broad jump 7.36** .57
6. 60 yard shuttle run 1.80 .31
7. Situp 2.86 XX
8. Sidestep 1.46 - .30
9. Flexed arm hang 0.26 XX

 

*p_ < .05. ”2 < .01.

XX — item not included in the discriminant function equation

84

Table 7

Means and Standard Deviations for Groups (top, middle, and bottom third
on perceived competence scores) on the Motor Abilities/Sport Skills

Test Items

 

Dependent Variable

 

Group 1 (top 1/3rd)

1. Soccer dribble
2. Basketball dribble

3. Softball repeated
throws

4. Soccer ball throw
5. Broad jump

6. 60 yard shuttle run
7. Situp

8. Sidestep

9. Flexed arm hang

M

7.

15.

10.

198.

47

18.
14.
15.

9.

20

99

28
02

.58

13

58

gp
2.60

6.73

83.84

10.52

174.
46.
18.
13.

14.

71

62

.59

Group 2

.17

.22

.78
.65
.88
.67
.99
.98

.05

9529.9

9 5.9
5.95 1.87
12.69 5.66
8.96 4.05
156.12 60.64
42.66 8.43
18.63 1.96
12.82 5.15
14.48 2.97
7.96 6.88

 

85

Other main effects in this model (gender and grade level) which
were found to be significant will not be discussed here. They were
examined previously with motor abilities/sport skill item scores as the
dependent variables, in the second relationship investigated.

Canonical correlation procedures were used to determine which
subset(s) of items from each set of item scores (perceived competence
and motor abilities/sport skills) best represented the relationship
between these two larger sets. Developed by Hotelling (1935),
canonical correlation finds a linear combination of weighted variables
from each set, called a canonical variate, which maximizes the
correlation between the two sets of variables. The correlation between
the first set of canonical variates is the first canonical
correlation. Subsequently, a second and successive set(s) of canonical
variates may be obtained, each being orthogonal to all previously
derived sets.

Grade levels with identical items on the perceived competence in
motor ability subscale were combined. 'Therefore, three canonical
correlations were applied to the data; one for kindergarten level
subjects, one for first and second grade level subjects, and one for
third and fourth grade level subjects. A significant canonical
correlation was obtained for third and fourth grade level subjects,
only. Subsequent discussion will refer to these subjects.

Results of the analysis for third and fourth grade level subjects
indicated one significant canonical correlation (set of canonical
variates), RC1 = .67, g < .03. The eigen value was .45 suggesting that

the set of canonical variates shared 45% of their variance.

86

Generally, the relative importance of each variable to the
canonical correlation is expressed by its canonical coefficient, or its
weighted contribution to the overall relationship between the
variates. However, several statisticians have argued that correlations
among the variables in a set may distort the interpretation of the
weights of each (Mintzes, 1979; Weiss, 1972). ‘This is similar to the
problem of interpreting beta weights in a multiple regression equation
when independent variables are highly correlated to each other and the
criterion. Therefore, a different statistic, the canonical loading is
recommended (Meredith, 1964; Weiss, 1972). Each variable's canonical
loading represents the correlation between the variable and the
artificially-constructed variate. Canonical loadings may be
conceptualized in the same way factor loadings represent a variablefis
relationship to a particular factor in a factor analysis study.
Canonical loadings for items on the two pairs of canonical variates
derived in this analysis are presented in Table 8.

Inspection of the loadings suggested that the canonical variate
for perceived competence was most reflective of items in the scale
which concerned being picked first and being good at sports in general.
The items with the highest loadings on the motor abilities canonical
variate reflected specific sport skills to a large extent. Overall,
the shared relationship appeared to be reflected by perceptions of
general ability in games and actual skill in sport tasks.

While one significant canonical correlation was produced in this
analysis, with a moderate percent of shared variance, this value for

shared.variance may be somewhat misleading. Several statisticians have

87

Table 8

Canonical Loadings for Perceived Competence Items and Motor
Ability/Sport Skill items on the Canonical Variate

 

 

Canonical
Items Variate
Loadings
Perceived Competence Items
1. do well at sports .36
2. good at catching .43
3. good enough at sports .41
4. good at throwing —.08
5. good at outdoor games .35
6. good at dribbling .41
7. better than others at sports .40
8. good at kicking .41
9. rather play than watch .47
10. good at running .08
11. good at new games .51
12. am strong .46
13. picked first in games .82
14. good at jumping .29
Motor Abilities/Sports Skills Items
1. Soccer dribble .61
2. Basketball dribble -.39
3. Softball repeated throws .35
4. Soccer ball throw .03
5. Broad jump .11
6. 60 yard shuttle run -.03
7. Situp .23
8. Sidestep .23
9. Flexed arm hang .43
3C1 = .67

 

88

recommended computation of the redundancy index to represent the actual
amount of variance in the original set of variables which is
predictable by knowing scores of the other set of variables (Cooley &
Lohnes, 1971; Karpman, 1981; Stewart & Love, 1968). 'The proportion of
the variance in the perceived competence items which may be predicted
by scores on the motor items was XML The redundancy of the motor
items or the proportion of their variance which may be predicted by
knowing perceived competence scores was XML Therefore, neither set of
variable appears to provide much power to predict scores on the other
set of variables.

Pearson product correlations were obtained between childrenfis
scores on seven perceived competence items and eight motor
abilities/sport skills test battery itemsn 'The perceived competence
items related to children's perceptions of their ability to

1. catch,

2. throw,

3. dribble a ball,

4. kick,

5. run,

6. jump, and

7. demonstrate strength.

Motor items used in the correlations were

1. softball repeated throws,

2. basketball dribble,

3. soccer ball dribble,

4. 60 yard shuttle,

89

5. flexed arm hang,
6. situp,
7. soccer ball throw, and

8. broad jump.

Average correlations within each grade level suggested that on an
item by item basis, children in each of the grade levels tested were
not very accurate in perceptions of their physical abilities and sport
skills. Correlations generally improved with grade level, but by
fourth grade a mean £_of only'.29 was obtained. Values for the
kindergarten through fourth grade level subjects were .14, .22, .24,

.16,and.29,respectively.

Summary of Results
The results of this study suggested that for these children in the

kindergarten through the fourth grade level:

1. Perceived competence in motor ability was not significantly
related to participation motivation, as measured by participation
in organized sports.

2. Males had a higher perception of their motor ability than did
females.

3. Perceptions of competence in motor ability decreased as grade
level increased for both males and females.

4t Motor ability, as measured by motor abilities/sport skills items,
was significantly related to participation motivation.

Participants in organized sports performed significantly better

90

than nonparticipants. 'The group difference was due primarily to
performance on the sport skills items.

Males performed the motor abilities/sport skills tasks
significantly better than did females. Males outperformed females
in all sport skill items, three of the five motor abilities items,
and were not significantly different from the females in two motor
ability items.

Performance on eight of the nine motor abilities/sport skills
items improved as grade level increased. Scores on the flexed arm
hang item did not change significantly as grade level changed.
Perceived competence in motor ability was significantly related to
actual motor ability. Several motor items contributed to this
relationship, however, the sport skill items contributed more
power to discriminate among levels of perceived competence than
did the motor ability items. IFurther, scores on the motor items
could be used to distinguish those subjects who scored in the
bottom third on perceived competence from those subjects in the
top two thirds better than they could be used to make any other
between-groups distinction.

Correlations between perceived competence in specific motor tasks
and actual performance in those tasks were low at each grade
level. However, a slight improvement was demonstrated with the
increased grade level. Additionally, only for the third and
fourth grade level subjects were significant amounts of shared
variance demonstrated between item scores for perceived competence

and motor ability.

CHAPTER IV
DISCUSSION

Four hypotheses were stated for this investigation; two of which
were supported by the data collected, two were not supported. The
first hypothesis which predicted a significant relationship between
children's perceptions of competence and their motivation to
participate in sport, was not supported. The second and third
hypotheses were supported. They predicted significant relationships
between actual motor ability and motivation to participate and between
actual motor ability and perceptions of competence. 'The final
prediction, that the strength of the relationships predicted
(Hypotheses 1, 2, and 3) would increase as grade level increased, was
not supported.

Although Harter's developmental theory of motivation (Harter,
1981a, 1981b) suggests that the predicted relationship exists between
children's perceived competence in their motor ability and their
participation motivation, there are several tenable explanations why
support for this aspect of her model was not found. First, the basis
for her theory has been drawn primarily from work in the cognitive
domain. Within the physical domain several other related variables may
intervene. iFor example, results from this study and others (Alderman
& Wood, 1976; Sapp & Haubenstricker, 1978) indicate that two of the
most frequent reasons identified by children for pursuing sports are to

have fun and to be with friends. Perhaps sport participation for

91

92

children is mediated more by factors associated with social involvement
than by competence, at least in situations in which children are not
eliminated from teams and competition is not stressed as highly as in
many programs for older children. Some support for the application of
Harter’s theory to the motor domain has been suggested by other
researchers (Feltz & Petlichkoff, 1983; Roberts et al., 1981; Weiss,
Bredemeier, & Shewchuk, 1984).

A second possibility is that the strength of this relationship
increases with age, but is not strong enough during the early
elementary school years to be detected in a study such as this. In the
few previous studies utilizing participants and nonparticipants in
sport, older children served as subjects. Subjects in the Roberts et
al. (1981) study were the youngest, comprising fourth and fifth grade
levels. If this relationship begins to be significant at approximately
the fourth grade level, the pooling of subjects could have increased
the strength of this relationship. In the present investigation, fewer
fourth graders were assessed and their scores were pooled with those of
subjects in four lower grade levels. Additionally, an examination of
the relationship between perceived competence in motor ability and
participation motivation at individual grade levels suggested that as
grade level increased, the probability decreased that the mean
difference in perceived competence scores between.participants and
nonparticipants occurred by chance.

The assumption that participation or nonparticipation in sport is
a choice made freely by the child, may not be as accurate for young

children as it is for older children. This possibility may help to

93

explain a developmental relationship between perceived competence and
participation. Only 28% of the nonparticipants in this study rated not
being good enough at sports or not enjoying sports as very important or
somewhat important reasons for not participating. The reasons most
frequently given for not participating were: (a) programs were too
expensive» (b) available programs were too far from home, and (c)
subjects were not aware of programs or felt that programs available for
children their age did not provide sufficient choices. For
participants, the decision to become involved may have been their own.
However, many nonparticipants may also have had the desire to
participate and the perception that their motor skills were sufficient
for successful participation. Other factors (cost, transportation,
etc.) may have prevented them from pursuing this form of activity.

In contrast to the results above, some support for the possibility
that participation motivation may be related to perceptions of
competence in this young age group was obtained by using hours spent
per week in gross motor activity as a measure of participation
motivation. If hours spent in participation may be considered an
activity more directly controlled by the child, then its significant
relationship to perceptions of competence may support this aspect of
Harter's model. However, the amount of variance in perceived
competence scores accounted for by hours of play was small.

Future research involving the relationship between perceived
competence in motor ability and participation motivation in young
children should consider two factors. First, if the participation/

nonparticipation dichotomy is used, the researcher should establish

94

that the childfs status on this variable is the result of his or her
free choice. Second, alternative situations should be investigated
which may reflect childrenfls participation motivation. Examples would
include level or types of activity engaged in during recess at school,
level of participation in physical education class, and percent of free
time spent involved in gross motor activity.

The second hypothesis was supported by the results of this study,
i.e., a significant relationship was found between children's
demonstrated motor performance and their participation motivation.
Specifically, subjects who were participants in organized sports
performed the selected motor tasks better than did nonparticipants.
Follow-up tests suggested that the difference was due primarily to
scores on the sport skill items. Iflhile direct effects cannot be
inferred from the data collected in this study, these results may
suggest that involvement in sport may have a significant and positive
effect on skill level. ‘Ln this sample, soccer was the one sport or
one of the sports of involvement for 66% of all participants. Seventy-
five percent of the participants in third grade or younger participated
in soccer. The discriminant function analysis extracted the soccer
ball dribble item as the most powerful item in predicting group
membership, iJL, participant or nonparticipant. This skill would be
the one most likely to be affected by such involvement.

The results of this investigation agree in part with the only
other published study which investigated this relationship for children
of a similar age range. Magill and Ash (1979) compared the perceptual-

motor ability of first through fifth grade level participants and

95

nonparticipants. Their results suggested that participants did not
differ from nonparticipants on these tasks. While their tasks were
limited in number and scope of motor abilities, the present
investigation examined performance on a wider range of gross motor
abilities. At the univariate level of analysis, the present results
agree with those of Magill and Ash; none of the motor ability items
demonstrated a significant relationship to participation. However,
several items did contribute some discriminatory power in the
discriminant function equation.

The possibility that a relationship between fitness or motor
ability and participation in sport may become significant for older
children is suggested by the work of Guyot et a1. (1981) and Smoll and
Schutz (1984). Guyot and associates found a significant relationship
between motor abilities/fitness scores and participation for fourth
through sixth grade level males but not for females. Small and Schutz
found that this relationship was significant for 7th and 11th grade
level males and females but not for 3rd grade level subjects.

When hours per week.spent in gross motor activity was used to
reflect participation motivation the results also supported the second
hypothesis. (Hujdren who engaged in gross motor activity for greater
amounts of time per week performed better than children who were less
active in all but two motor items. 'That more of the motor ability
items were significant at the univariate level for this analysis than
in the sport participation analysis suggests two plausible
explanations. ZPerhaps children who are more competent in the more

"innate" motor abilities also are motivated to be active in their

96

free time» Alternately, activities of free play may be more effective
in increasing the performance of children on motor ability items than is
involvement in organized sport at these ages. Increased performance in
the sport skill items may have resulted from the probability that
participants in sport were included within that group of children
spending the greatest number of hours involved in gross motor activity
per week.

In a practical sense the support found for the second hypothesis
suggests the importance of emphasizing the develOpment of basic sport
skills for all primary and middle elementary grade level school
children. Particular emphasis should be on skills that are
prerequisites for the sports that are popular within the community.
Such an effort would help to raise the level of skill for those
children not yet involved in community sport programs.

The importance of emphasizing sport skills at an early age is
reinforced by the fact that a gender difference also existed for the
motor items. The difference was more evident in sport skills than
motor abilities, but in all items except two, males performed better
than females. .At these ages the difference may not be of practical
importance for game situations and during these years children may not
be eliminated from teams due to skill level differences. By the middle
school or junior high school years, as this difference increases,
females who try out for coed teams may be at a great skill disadvantage
if efforts are not made to improve individual skill levels.

Support for the third hypothesis resulted from a statistically

significant relationship which occurred between childrenfls scores on

97

motor performance tasks and their perceptions of motor ability.
Apparently children at these ages had a relatively accurate perception
of their ability to perform motor tasks. At least, they were aware of
their status as it related to being near the top, middle, or bottom
third of their grade level.

The results further suggested that those children at the bottom
one third of their grade level on motor performance and perceived
competence were more easily distinguished from all others than were the
top and average performers. ‘This would not surprise practitioners, who
routinely observe that the poorly skilled are made aware quickly of
their lack of skilln The fact that this may be demonstrated so early
and that sport skills were more strongly related to perceptions of
ability than were motor ability items suggests the practical need to
assess skill levels and then provide individualized, objectives~based
instruction. 1

These results are generally in agreement with other studies
involving older subjects within the physical (Boling & Kirk, 1982;
Guyot et al., 1981) and cognitive domains (Boersma & Chapman, 1978;
Harter, 1981b). 'However, some discrepancy is apparent between the
present results and those found by Harter (1982) and Harter and Pike
(in press) for children ages 4 to 12 within the physical domain. When
correlating children's perceptions of their motor competence to
teachers' perceptions of the children's competence she found low values
for the 4 to 7 year-old subjects and consistently moderate (.60s)
values for the 8 to 12 year-old subjects. This suggests an interaction

among age, perceptions of ability, and actual ability. This

98

interaction effect was not significant in the present study and may be
due in part to the fact that the measures of actual competence differed
in the two studies. Harter used teachers? perceptions of ability while
in the present study actual motor performance was measured. Harter's
subjects may have been more accurate at reflecting their teachers'
perceptions than in understanding their actual competence in specific
tasks. In addition, teachers as well as students may have been
thinking about childrenfls ability more in general than specific
competencies.

The discrepancy may also have occurred because the design of this
study and analysis was not powerful enough to detect such a change in
the relationships. In fact, follow-up correlational analyses did
suggest a stronger relationship with increasing age. 'The mean values
for correlations between specific perceived competence items were low,
but increased as grade level increased. Further, only at the third and
fourth grade level were canonical correlation techniques able to
produce a subset of perceived competence items that could account for a
significant amount of variance in a subset of motor performance items.

Further research should investigate the possibility that the
correlations between competence and perceptions of competence may be
increased by assessing tasks which are more relevant to subjects at
each specific grade level. For this study the desire to make
comparisons across five grade levels necessitated motor performance
tasks that 5 through 10 year-old children could perform. However,.some

tasks may be used more frequently than others at different grade and

99

age levels. 'This may affect how accurately children are able to judge
their competence.

The final hypothesis predicted that the strength of the
relationships examined in the first three hypotheses would increase
with age. Since no interaction effects were significant, this
hypothesis was not supported. Specific reasons why this may have
occurred were discussed relative to each hypothesis where apprOpriate,
in the preceeding pages.

In general, this prediction was based on the possibility that
accuracy and an understanding of the distinction between ability and
effort may occur earlier in the sport domain than in other domains.
This domain would appear to provide more opportunities for social
comparisons, more concrete, frequent, and contingent feedback.than the
cognitive domain (Klieber, 1981; Pascuzzi, 1981). Each of these
factors is important to developing perceptions of competence (Veroff,
1969; Harter, 1981a).

The work of several researchers (Harter, 19813; Horn, 1984;
Minton, 1979),:nay suggest one reason why greater accuracy did not
occur earlier. These researchers have found that in cognitive
situations students depend more on feedback from "authoritative
figures" than in sport situations. In sport, peer evaluations are more
salient; however, they may also be less accurate. In testing
situations for this study, the probability that cognitive competence
evaluations were frequently based on teacher feedback was evidenced by
verbal qualifications that children used when.answering math or reading

competence questions. 'For example, one child remarked "I know Ihn good

100

at reading because Ihn in the Redbird groupf' Another commented, "I
must be really good at arithmetic, Ihn in the 2nd grade workbook and
I'm only in lst gradeJ'

The fact that for young children, the types of gross motor
activities engaged in, the environment in which they occur, and
equipment available change frequently may actually hinder development
of accurate judgment of one's ability to perform specific motor tasks.
Nicholls (1978) identified the ability to relate temporally separated
outcomes and understanding the difference between ability and effort as
necessary to developing accuracy in self-perceptions. At the
elementary school level children receive reading and arithmetic
instruction and evaluation daily. However, the nature of childrenks
gross motor activities change as they move from the gym, to the
playground to home. They also vary depending on weather conditions,
availability of equipment, and play partners. Therefore, while oneks
ability to throw a stone farther than a friend on one occasion may
provide immediate and concrete performance information, the chance to
throw may not occur again for several weeks at which time the first
effort may have been forgotten. Increased time between performance
efforts may also make it easier to attribute one's poor performance to
lack of effort instead of lack of ability. Thus, developing accurate
perceptions of one's specific physical competencies may take more time
than for cognitive competencies. For children who become involved in
and devote regular practice time to one or more sports at an early age,
perceptions of motor competence may become accurate earlier than

perceptions of competence in other domains.

101

Summary

Overall, the findings of this study suggest that for young
children, perceptions of competence in motor ability is not a
significant criterion for participation in organized sport. However,
those who are participants demonstrate a higher level of skill than
nonparticipants, and, in a general sense, these children demonstrate an
understanding of their skill level within their own grade level.

If, as has been suggested by results from other studies (Feltz &
Petlichkoff, 1983; Roberts et al., 1981; Weiss et al., 1984),
perceptions of ability do become significantly related to involvement
within a few years, then some direction for practitioners may be
suggested by these results. During these early years, efforts to
improve actual skill levels and the children's understanding that their
level of skill has improved and is sufficient for effective
participation in sport, would be most beneficial to future development.

From a theoretical standpoint, results of this endeavor support
the relationship proposed by Harter (1981a) between perceived
competence and actual competence. Contrary to this, however, a
significant relationship between perceived competence and participation
motivation was not supported for this age group. .Possibly during these
years a portion of the developmental model may need to be modified
(Figure 8). It may be that within sport situations, young children
participate for a variety of reasons (of which perception of competence
may be one but not yet a major contributor). Participation
subsequently affects skill level, which feeds into onehs perceptions of

skill and ability. In time, the connection between perceived

102

£96.20 959» com oco .EOEOt _oU_m>;a or:
:23? 602023.: 8:2an3 *o _oooE roots:

*0 «$556.6 out: 9350 3:30:29. 233.com .m 0.5mm

 

oucoaanou
.3285;

 

’

 

 

l

 

oucoaanOU
.oeau<

 

 

 

Aoogmznoamo am.»
*0: cozooccofl

’

I

I

1.,

 

 

r

cozofiozcom
:

820262

 

103

competence and motivation becomes established, although at different
rates for different children.

The possibility that the direction of the relationship which may
become established between perceived competence and participation
motivation may be bidirectional should also be investigated. In older
children perceived competence may affect choice to become involved in
organized sport. But participation may also affect perceptions of
competence. iFor example, being relegated to sitting on the bench or
receiving primarily negative feedback from the coach may lower
perceived competence. Perceptions of competence may increase for those
who are stars on the team or who receive primarily positive feedback
from coaches and peers. Certainly, further research is warranted to
establish such relationships, the nature of their cause and effect
status, and the timing of such events. The need for a longitudinal
approach is evident.

In conclusion, this research effort may be considered exploratory
in nature. Until recently, little effort had been made to theorize
regarding the developmental relationships among the variables
investigated here. Research involving young children within the
physical domain is limited and usually was indirectly related to the
major research questions being tested.

The psychological characteristics and motor abilities children
bring with them to situations which involve motor skill instruction are
important. Each affects how the teacher or coach should structure the
session. (Ruldren who want to learn and expect to enjoy the situation

may require the instructor to attend minimally to motivational

104

techniques, allowing greater emphasis on development of skill.
Conversely, children who expect to have little success and to find
little pleasure in the situation may require the teacher to place
greater emphasis on changing perceptions. Knowledge of childrenks
levels of motor skill provides information the instructor can use to
prescribe successful and challenging experiences. Learning experiences
need to be appropriately matched to childrenfis levels of skill to
maximize improvement.

Understanding how children's psychological characteristics and
motor abilities change, which may be changed by means of specific
treatments, and how each may affect the subsequent use of skills
learned in a movement setting have theoretical as well as practical
importance. Such information would further our understanding of
development in general and the interrelationships among various
domains. In applied situations this information could influence the
selection of instructional goals for children at various developmental
levels and the expectations of what may be reasonable improvement.
Specific learning tasks may also be prescribed more appropriately. At
some levels of development change may be facilitated more easily in one
domain than another. .At other levels we may provide for the
development of skills in more than one domain via tasks that are
mutually beneficial. Continued research investigating these
interrelationships seems warranted. Many questions are still to be
answered, and more to be raised, particularly from a developmental

approach.

APPENDICES

APPENDIX A

PILOT STUDY

105

PILOT STUDY

The items in the Assessment Battery for Motor Abilities and Sport
Skills were chosen to represent general motor ability in the
investigation of the relationships among actual motor competence,
perceived motor competence, and participation motivation. ‘The
literature clearly suggests that a general attribute of motor ability
does not exist, per se (Singer, 1975; Sage, 1977). Therefore, it was
deemed necessary to measure a number of tasks representative of several
more specific motor abilities. Specific motor abilities which have
been identified by factor analysis studies involving children include:
speed; power; agility; eye-hand, eye-foot, and general body
coordination; flexibility; and balance (Rarick & Dobbins, 1975;
Peterson, Reuschlein, & Seefeldt, 1974). Items such as the 30 to 50
yard dash, broad jump, shuttle run, etc., have commonly been used to
represent these factors (Smoll, Schutz, & Keeney, 1976; Glassow,
Halverson, & Rarick, 1965). Safrit (1973) noted that a problem communi
to many motor ability tests used to judge motor performance is that
they do not relate well to actual sport and game situations.
Kirkendall, Gruber, and Johnson (1980) noted that although individual
items (measuring speed, agility, etc) are not very predictive of
demonstrated skill in sports, when tested in ways more similar to game
situations, correlations increase substantially. Many single tasks may
be judged as measuring more than one component (e49, basketball
dribble test measures a skill specific to a sport as well as eye-hand
coordination; soccer ball dribble test measures a skill specific to a

sport, agility, and eye—foot coordination). A battery composed of

106

items representing motor ability factors and sport skills was
therefore developed to be representative of many, though certainly not
all, components of general motor ability.

The items included in the preliminary assessment battery were
chosen with six additional criteria in mind. The purpose of the pilot
study was to determine how well each item could meet four of the six
criteria. There were as follows. Each item had to:

1. be appropriate for the age range of 5 to 10 years
(appropriate to children's cognitive as well as motor abilities),

2. be reliable,

3. be valid, and

4. in conjunction with other items in the assessment battery,
not require a total administration time in excess of 30 minutes.

The nine items included in the preliminary test battery were:

1. standing broad jump

2. flexed arm hang

3. side-step test

4. sixty yard shuttle run

5. sit-up test

6. basketball dribble test

7. soccer ball dribble test

8. soccer ball throw test

9. softball repeated throw test

Methods:
Subjects: A total of 88 females and 106 males were tested on all

nine items of the test battery. All were volunteers from an elementary

107

school in Carterville, Illinois, enrolled in grades kindergarten
through four. Ages ranged from a mean from kindergarten subjects of
67.5 months to a mean of 119.5 months for fourth grade level children.

Procedures: During the months of December and January (1983-1984
school year) children were tested on the battery items as part of their
physical education classes. Each day two stations were set up in the
gymnasium. The physical education specialist assisted in the testing
by administering one item while this investigator administered the other
(NHL Upon entering the gym children were asked to remove their shoes
and socks and to report to the station at which they were to begin
testing that day. Assignment to stations was random and changed daily.
After being tested in one station the student reported to the other
station for a second test. .All physical education classes met two to
three times per week for 30 minutes each class period. Testing for the
pilot study was accomplished for each grade level over a period of
approximately 2 l/2 weeks.

Ten first grade level children who were members of a class not
included in the total pilot study were chosen randomly as subjects in
an examination of the test—retest reliability of the items. ‘These
children were taken in pairs to the gymnasium during free time and
tested on all items by this investigator. The order of presentation
of the items was counterbalanced and randomized for each pair of
subjects. Subsequent to initial testing a retest was administered one
week later.

In addition to the actual testing, all subjects were given a rank

order score based on their general level of motor performance as

108

demonstrated in their school physical education class. Rankings were
generated within each grade level by the childrenhs physical education

specialist.

Results:

The results of the pilot study will be addressed as they related
to compliance of the test items with the criteria noted previously.
All data referred to are included in Appendix C.

1. Appropriateness of items across age range of 5 to 10 years

a. Cognitive requirements: For the sit-up test, directions
appeared to be somewhat difficult for kindergarten level subjects to
comprehend. Some children did not seem to understand the concept of
doing as many sit-ups as possible in 60 seconds with a rest within that
period of time, as needed. Protocol for this item was adjusted, as
noted below in section 1c. None of the other items appeared to be too
difficult for the youngest children to understand. In addition to
verbal directions given, all tasks were demonstrated as well for the
kindergarten and lst grade level children. Demonstrations were
supplemental for children in grades 2 through 4 as deemed necessary.

1% Physical requirements: Inspection of the means and standard
deviations for each item across the 5 grade levels suggested an
appropriate age related increase for all items. Idinimum and.maximum
values obtained suggested the lack of a floor or ceiling effect for
most items. For one item, the flexed arm hang, several subjects at
each grade level were unable to maintain the starting position, and
mean values were low. The possiblility that this occurrence was sample

specific was suggested when these scores were compared to scores of

109

children the same age who were part of a larger group to whom the same
test was administered. Since mean values and standard deviations were
acceptable, this item was maintained. Examination of scores for this
item within the major study suggested that the low values obtained in
the pilot study may have been specific to that sample.

c. Protocol: Throughout the pilot study several aspects of
administration and scoring were examined. For the ball dribble (with
hands) three different balls were tried; a junior basketball, fully
inflated 9 inch playground ball, and a semi-deflated 9 inch playground
ball. The 9 inch playground ball (inflated to a pressure of 2 1/2 lbs)
provided most control and did not hurt the hands of young children
while still offering a challenge for older children.

The sit-up test allowed children to perform as many sit-ups as
possible within 60 seconds. However, two scores were recorded, (a)
number done before a rest was taken, if taken and (b) total number done
in 60 seconds. 'This was done to examine a possible recovery factor
which may interfere with the strength and endurance factors desired. A
significant difference was not found between the two, although some
children did stop, rest, and then continue. Additionally, some younger
children did not seem to understand the concept of doing as many sit-
ups as possible within 60 seconds with a rest within the test as
needed. Therefore to alleviate the possibility of confounding the
recovery effect with strength and endurance and to eliminate confusion,
a 30 second sit-up test was used to replace the 60 second test.

When test-retest reliabilities were obtained, that for the 60

yard shuttle was relatively low, .57. Since the distance covered was

110

judged as being short enough to be considered more of a speed than an
endurance measure, a second trial was added (as is typically done in 30
and 50 yard dashes) to increase this value.

The protocol for the side-step test indicated penalties for cross-
over steps. During the pilot study some younger children, although
demonstrating appropriate use of sliding steps during practice,
sometimes reverted to galloping, particularly to one of the two
directions. Therefore, verbal reminders were given when such movements
occurred and lines crossed with such movements were not be counted.

One change was made in the scoring of the soccer ball dribble test
to increase objectivity. Pilot procedures called for awarding a point
for each cone dribbled past, as in the basketball dribble. However,
younger children sometimes skipped cones due to a lack of control,
which made judgments regarding awarding points difficult. Scoring
procedures were therefore adjusted to award a point each time the ball
crossed an imaginary line formed by any two adjacent cones in the row
of cones.

Procedures for all items not specifically addressed above were
maintained as originally formulated. Protocol for all items may be

found in Appendix B.

2. Test-retest reliability
Scores for five male and five female first grade level children
resulted in generally moderate to high Pearson product correlation

values. The range was from .57 to .95 with only two scores below .79.

111

The average across all nine items was XXL From this information it

was judged that the items chosen demonstrated sufficient reliability.

3. Validity

The validity of this battery of items to provide objective values
for motor abilities and sport skills was examined by comparing the
scores that subjects attained on the test items to their rank order
score within their grade level. Rank order scores were converted to
percentiles and then to Z scores and used as the dependent measure in a
series of multiple regression analyses. These analyses suggested that
at each grade level some combination of items was able to account for a
significant amount of the variance in rank order scores. The average
R2 a .664, while individual grade level values were .56, .66, .77, .59,
and .75 for Grades K, 1, 2, 3, and 4, respectively.

Discriminant function analyses were also performed on these data.
Subjects} 2 scores were used to classify children into one of three
categories, top, middle, or bottom one-third of their class relative to
overall motor performance. Classification (Group 1, 2, or 3) was the
criterion variable, while the nine test items were the predictors in
the model. One significant (p_<l.05) discriminant function equation
was obtained for each grade level, with the exception of kindergarten.
At this grade level a Wilks' Lambda of .3854, chi-squared :- 23.84, p <
.16 was obtained prior to extracting any discriminant functions. This
suggested that the variation to be accounted for among the three groups
was small. However, the two disciminant function equations which were
produced correctly classified 78% of the subjects. The average percent

of correct classification for Grades 1 through 4 was 79.5. Therefore

112

subjects'scores on these motor items were judged to relate adequately
to their overall motor performance as evaluated by their physical

education specialist.

4. Administration time

During the test—retest portion of the study it was found that 2
first grade level subjects could perform the full battery of tasks in
approximately 30 minutes. Kindergarten level children were expected to
take slightly more time to complete the test items while older children
were expected to finish in less time. The inclusion of all items
therefore was not expected to require too much time within the demands

of the total dissertation study.

APPENDIX B

PILOT STUDY - SUMMARY STATISTICS

113

PILOT STUDY - SUMMARY STATISTICS

Item Grade M S_D Minimum Maximum
Sidestep 4 18.63 2.16 14.0 24.0
3 16.44 1.68 12.5 20.0
2 14.43 1.92 11.0 19.5
1 13.85 2.07 ~6.0 17.5
K 11.48 1.92 7.5 16.5
Softball Throw 4 13.61 2.66 5.5 18.0
and Catch 3 11.90 2.57 7.0 18.5
2 10.03 2.93 4.0 17.0
1 7.24 2.65 2.0 14.5
K 4.91 1.39 2.5 9.5
Situps to 60 4 29.87 8.30 10.0 50.0
3 32.44 9.26 8.0 47.0
2 27.08 8.72 7.0 45.0
1 23.38 6.84 7.0 37.0
K 11.03 8.19 O 28.0
Broad Jump 4 51.43 6.60 34.7 64.7
3 49.83 9.11 27.3 71.2
2 45.61 7.06 34.5 63.2
1 43.29 8.04 29.0 59.7
K 35.34 8.71 17.5 55.0
Soccer Dribble 4 8.71 1.94 4.7 13.3
3 7.79 1.74 4.3 11.7
2 7.77 1.66 5.0 12.0
1 6.42 1.85 3.0 12.0
K 4.92 1.72 2.0 9.0
60 yd Shuttle 4 17.52 1.40 14.5 20.9
3 17.88 1.26 15.3 20.7
2 18.05 1.14 15.8 20.5
1 19.15 1.57 16.1 23.0
K 21.45 2.15 16.9 26.1
Soccer Throw 4 262.35 51.87 164.7 374.0
3 221.27 48.32 128.3 322.7
2 191.54 50.23 100.0 283.7
1 141.52 37.22 76.0 217.0
K 112.07 35.11 39.7 182.0

114

Item Grade M_ §D_ Minimum Maximum
Flexed Arm 4 8.32 10.23 0.0 52.4
Hang 3 13.62 12.02 0.0 56.1
2 8.04 6.54 0.0 27.3
1 6.26 6.21 0.0 26.3
K 4.44 4.48 0.0 17.8
Basketball 4 19.46 3.73 12.0 29.5
Dribble 3 16.80 4.65 8.0 28.0
2 16.07 4.87 6.5 27.0
1 9.73 3.21 4.0 16.5
K 6.27 2.59 3.5 14.0
Age 4 119.50 6.67 110.0 133.0
3 105.10 5.53 97.0 126.0
2 93.70 4.59 84.0 107.0
1 81.05 6.14 73.0 96.0
K 67.47 5.18 61.0 79.0

Test-Retest Results: Pearson Product Moment Correlations
N210 first grade subjects (five females, five males)

1. Soccer Dribble .79
2. 6O yd Shuttle .57
3. Soccer Ball Throw .59
4. Flexed Arm Hang .91
5. Sidestep Test .86
6. Softball Throw .89
and Catch
7. Situps to Stop .82
Situps to 60 sec. .82
8. Broad Jump .85

9. Basketball Dribble .95.

115

Multiple Regression Analyses

I.V.s were ordered: Broad Jump, flexed arm hang, situps (to 60 sec),

basketball dribble, softball throw, 60 yard shuttle, soccer

throw, soccer dribble, sidestep test.

D.V. was 2 score: determined by a transformation of rank order score,

as assigned by child's physical education specialist.

R2

 

Grade N _1: p
4 38 (9,28)=9.12 .75 .0001
3 45 (9,35)=5.52 .59 .0001
2 37 (9,27)=10.03 .77 .0001
1 42 (9,32)=6.95 .66 .0001
K 32 (9,22)=3.13 .56 .01

Discriminant Function Analyses

Method used for the discriminant analyses was the direct method, i.e”

all variables were entered into the analysis at the same time.

Group = 1, 2, or 3 (child's position in the top, middle, or bottom one-
third of their class based on rank order scores assigned by
their physical education specialist)

Predictors = Broadjump, flexed arm hang, situps (to 60 sec),
basketball dribble, softball throw, 60 yard shuttle, soccer
throw, soccer dribble, sidestep test.

 

Grade Wilks' Lambda Xf p_ % of subjects
classified correctly
4 .194 50.85 .0001 84.2
3 .397 35.06 .009 68.9
2 .213 46.34 .0003 78.4
1 .225 52.20 .0001 88.1
K .385 23.84 .16 78.1

 

Amoscfiucou umwav

 

mm. 08. mm. mo.l x
mm. Ha. nm. em. umH
sq. Om. um. we. cam
am. we. mm. mm. cum amummesm
ON. cm. oH. ma. :uq
mm. ms. 08. me. ma. x
mm. om. am. am. 0H. umfi
as. ms. as. as. NN. cam meme eta
ma. am. mm. mH. 0H. ohm taxman
Hm. mm. mm. mm. ofi. nus
mm. mm. om. 0H. mo. as. x
08. ms. mm. 00. ad. mm. umH
on. so. #8. mm. «N. on. bum 3ocnh
as. om. cm. mo. wN. mm. com coooom
mm. mo. Hm. on. mo. mm. nus
mm.n «5.: o~.| HN.I Hq.l mm.l mq.| x
Nq.l mo.l am.| Hm.l No.1 am.l Hq.| umH
mm.u mm.l Ho.| mc.n mm.: 08.1 m<.| mam mauoacm
mm.t os.- 55.. w~.- 50.- mm.- as.- can as as
«0.: q©.I mq.n Nq.| Hq.| om.l Hq.1 sue
me. om. mm. mm. 0H. He. mm. mm.: a
He. mm. as. am. On. «q. we. nm.| umH
On. 08. me. no. Om. mm. ca. om.l cam mannfiua
mo. Om. «N. am. mm. mH. No. Hm.l cum ooouom
80. ca. mm. 08. co. om. mm. «8.: sum
mannwmo mesh .oom co zomne noumowflm mam: zomne oauuszw
Hamnuoxmmm umomm om masuwm Hamnuwom ac< moouom a» co

UOXOHM

WGOHUQHOHHOUHOHGH EOHH

 

me. x
cm. omm
Om. vcm mash
ow. ohm umomm
mm. nus
Hm. as. x
00. No. umH
we. mo. cam
«N. am. cum masufim
mo. Hm. cos
we. Hm. NH. x
mo. 00. we. umH
mm. mm. as. vcm zomnh
so. we. 08. ohm Hamnuwom
no. mm. mm. :uq
manage: mesh .omm cc 3omse aoummvwm woo: 3oune oauumzm
Hamnumxmmm cmomm ou masuflm Hampumom Em< moouom a» Do

emxmaa

APPENDIX C
INFORMATION LETTER FOR PARENTS
AND

CONSENT FORM

118

Information Letter for Parents

Dear Parent(s) or Guardian,

In the past few years educators have demonstrated a greater
commitment to documenting the progress of their students in all subject
areas. In physical education and sport it has been assumed that
involvement enhances development in several areas besides actual skill
in sports, such as leadership, self-confidence, and cooperation.

Rarely has such change actually been measured. One reason is that
little is known about how such qualities develop in children, within
the normal environment.

I am currently a doctoral student specializing in the development
of motor ability in children. As a former elementary physical
education teacher and director of pre-school programs, I have a strong
interest in research that has practical application. In my
dissertation study I will be investigating the development of
childrenfls self-concept of ability in motor skills. Self-concept of
ability will also be compared to children's actual motor abilities and
choice to participate in sports and games.

I will be working with children in kindergarten through fourth
grade at Winkler and Parrish elementary schools, and would like to ask
you to allow your child to participate in this study. Children will be
asked to 1) complete the attached questionnaire (at home, with parental
assistance), time involved: approx. 10 minutes; 2) take a test of
perceptions of ability (at school; 3rd and 4th grade children will
receive the test as a group, younger children will be tested
individually). time involved: approx. 20 minutes; and 3) take a test
of sport skill/motor ability (at school, to be administered in pairs of
subjects), time involved: approx. 30 minutes. In the perceptions of
ability test, children will be asked to respond to statements that
indicate what they are like» Sample questions might ask them how good
they feel they are at playing outdoor games, making new friends, or
kicking balls. Nine short test items comprise the motor skill battery,
examples of which include: soccer ball dribble, softball throw, broad
jump, and shuttle run.

All of the information collected during this study will be kept
strictly confidential. The names of all subjects will be replaced with
identification numbers as soon as all testing is completed. ley
group information will be referred to in any publication of the results
of this study. Individual parents or children will be free to
discontinue participation at any time. Upon completion of the study,
information concerning the findings will be made available to all
interested parties involved, within the confidentiality limitations.

119
-2-

The principals of the schools to be involved in this project (Mr.
Drake and Mr. Nelson) as well as the district superintendent, (Dr.
Thomas) have reviewed this study and procedures and have agreed to
cooperate with men In addition, this project is being conducted
through the joint cooperation of the physical education departments of
Michigan State and Southern Illinois Universities. If you will allow
your child to participate, please read and sign the attached consent
form and assist your child in completing the participation
questionnaire. Please have your child return these forms to his or her
classroom teacher within two weeks. If you have any questions
regarding this study, please feel free to contact me at the telephone
number or address listed below.

Your cooperation in this project will be greatly appreciated. The
results of studies such as this help to provide information through
which educators may continue to strive to provide a better educational

environment for all children.

Beverly D. Ulrich
202 North Oakland
Carbondale, IL 62901 457—8906

120

Parental Consent Form

I have read the information contained in the accompanying letter
concerning the motor development study which will be conducted
with children in kindergarten through fourth grades in the
Carbondale elementary schools. I will give my permission for my
child to participate as a volunteer
in this study being conducted by Beverly D. Ulrich.

 

The study has been explained to me and I understand what my
child's participation will involve.

I understand that participation by my child is subject to his or
her verbal consent. Further, I understand that I (or my child) am
free to discontinue participation at any time.

I understand that the identity of the subjects involved will be
treated with strict confidence, and that subjects will remain
anonymous. ‘Within this restriction, results of the study will be
available at my request.

Signed

 

Date

 

APPENDIX D

MOTOR ABILITIES/SPORT SKILLS ASSESSMENT BATTERY: PROTOCOL AND SCORE SHEET

121

MOTOR ABILITIES/SPORT SKILLS ASSESSMENT BATTERY - PROTOCOL

— All subjects will perform the test items in bare feet.

— Encouragement will be offered by the test administrator
throughout the testing session.

- Items will be administered to each subject in one of 4
counterbalanced orders, randomized by subject.

Standing Long Jump

Equipment: mat, masking tape, tape measure, pointing stick or yard
stick

Description: Subject stands with toes behind the take-off line,
‘with feet several inches apart. Prepartory arm and leg
movements are encouraged and demonstrated for subjects.
Take-off and landing are on two feet.

Scoring:'Two practice and three trial jumps are given. Distance is
measured from take~off line to point where body touches the
mat nearest to the line, calculated to the nearest 1/2 inch.
Three trial jumps are recorded, final score is the mean of
the three trials.

Flexed Arm Hang

Equipment: doorway gym bar approximately 1 1/2 inches in diameter,
stopwatch

Description: Height of the bar is adjusted so it is approximately
6 inches above the subjectds height. Subject grasps the bar
with a pronated (palms away) grip, and is helped (boosted or
may use a bench) into a position in which chin is above the
bar, with chest close to it and arms fully flexed. Position
is maintained as long as possible. Trial ends when performer
rests chin on bar or elbows extend beyond 90°.

Scoring: One trial is given and recorded to nearest tenth of a
second. (Spotter/timer should stop legs from swinging when
subject begins to hang freely. Raising the knees or kicking
legs to remain hanging is not permitted)

Side-stepiTest
Equipment: masking tape, stopwatch
center line side line

"tick mark"

08

 

 

 

 

122

Description: From a standing position astride the center line,
subject slides (sidestep) to the right touching foot to floor
outside the right side line; without turning moves in
opposite direction to touch left foot outside left side line.
Sliding to right and left continues, touching as often as
possible in 10 seconds. Cross-over steps may not be used,
and one point is deducted each time a cross—over step is
utilized. 'Use of a gallop to move from side to side is also
not allowed, and lines crossed in this manner are not
counted.

Scoring: Each trip from the center line across a marker counts one
point. For example, moving to the right the prformer crosses
a tick mark for one point, the outside marker for two, back
across the tick for three, across center line for four,
across left tick mark for five, and so on until 10 seconds
have passed. (To simplify scoring, scorer may begin counting
each of the three longer lines by "2" starting with the first
sideline crossed, then add an additional "1" point at the end
if the last line crossed was a tick markJ Subjects receive
a 5 second practice, followed by two 10 second trials.
Subject's score is the mean of the two 10 second trials.

4. 60 Yard Shuttle
Equipment: masking tape, stopwatch, two cones
cone cone
4 foot

< 15 yards > starting
line

 

 

Description: Two cones are placed so their outer edges are 15
yards apart. A starting line is placed on the floor, even
with the outer edge of one cone. Subject runs two laps
around the cones. Runner is asked to keep turns close to but
not touching markers.

Scoring: Time is recorded to the nearest 1/10th of a second. Two
trials are given; subjectds score is the mean of both trials.

5. Basketball Dribble Test

Equipment: 9 inch red playground ball inflated to 2 1/2 pounds of
pressure, six cones, masking tape, stOpwatch

A 8' A 8' A 8' A 8' é 8' A 5' <--suba‘ect
A? 3/ III
\/ fl. _——-—

begins
here

123

Description: Six cones are placed in a straight line, first one is
5 feet from the starting line, rest are 8 feet apart. On
"go" subject dribbles with either hand in and out of cones,
continuing for 30 seconds. (Cones are placed in a line
parallel to and 6 feet from a wall. Tester moves in a line
parallel to and 6 feet from the subject as subject dribbles
through the course, on the side of the cones which is
opposite to the wall. Tester insures that a loose ball does
not get farther away from the subject than 6 feet, by
stopping it.)

Scoring: One point is awarded for each cone passed in 30 seconds.
A cone is considered passed when the subjectds body is even
with it. Two trials are given, with score being the mean of
both trials.

Soccer-ball Dribble Test

Equipment: One semi-deflated indoor soccer ball (size five,
inflated to 4 pounds of pressure), five cones, masking tape,
stopwatch

Description: Five cones are placed in a straight line, 10 feet
apart with the first one 5 feet from starting line. on "go"
subject dribbles soccer ball with feet in and out of the
cones for 30 seconds. As noted above for the basketball
dribble, cones are placed 6 feet from a wall, and tester runs
parallel and along the side of the subject to keep "loose"
balls from getting too far away.

Scoring: Subject receives one point each time the ball passes
between two cones in 30 seconds. 'Three trials are given, and
subject's _s___core is the mean for all three trials.

A 10' Q 10' & 10'/&\10' A 5' .. <—-subject

begins
here

124

7. Softball Repeated Throws Test

Equipment: Two indoor softballs, masking tape for wall target and
throwing area, table turned on its side for a backstOp

target F\\\\\‘\\\\\\\

backstOp I

throwing area ~‘\\\\‘\J

15 feet ‘\\\\ 9 feet

Description: a. Target area is outlined with masking tape on the
wall as shown (10 feet X 5 1/2 feet).
IL Throwing area is marked on floor, 9 feet from the target
(5 1/2 feet X 5 1/2 feet).
c.‘Table is placed on its side, 15 feet from the throwing
area to serve as a backstop.
d. Child stands in throwing area with softball in hand. On
"go" child throws ball overhand at target, attempts to catch
rebound in air or after bouncing, repeating as often as
possible in 30 seconds. Child may leave throwing area to
catch but must return to execute throws. .All throws must hit
within target to score a point. A practice of 10 seconds
preceeds two 30 second trials.

Scoring: One point is scored for each throw in which the ball hits
within the target boundaries. Subjectds score is the mean of
two trials.

 

 

 

 

 

8. Soccer Throw—in for Distance

Equipment: one size five soccer ball, tape measure, masking tape

Description: From behind the restraining line, subject utilizes a
two-handed overhead throw for distance» One step is taken
prior to throwing, but restraining line may not be stepped on
or over until after the ball is released.

Scoring: Two practice trials, followed by three test trials are
given. Distance is measured from a line perpendicular to the
restraining line, to the nearest inch. Subjectds score is
the mean of the three test trials.

125

9. Sit-up Test

Equipment: mat, stopwatch

Description: To assume the starting position, the subjects lie on
their backs with knees flexed, feet on the mat approximately
12 to 15 inches from the buttocks. Hands are clasped behind
the head. Feet are held by a partner to keep them in
constant contact with the mat. The subject, by tightening
his or her abdominal muscles, curls to a sitting position,
touching both elbows to knees. Chin should remain tucked
into chest. One sit-up is completed when the subject returns
to the original starting position (midback must make contact
with the mat). Testing begins when tester says "go".

Scoring: One 30 second trial is given. Subjectfs score is the
total number of sit—ups completed within that time period.

126

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.H
m. NH as m. me NH HE M. me me He m. «a He
amok Bough sash songs
soummowm umooom cmoum Hamnumom
_ a z .N
_ m m, «d
w. my we as as as ms ma H8 m. up He osmz
manafiun maaziau< was «Hosanm «Hanson to
umuoom emxmam -usm as as magnumxmmm
mum:
< Enos

 

 

 

 

 

 

Hmmmmmmoom I wmmeh<m Hzmxmmmmm< mAAme Hmomm\WMHHHAHm< mOBOE

APPENDIX E

QUESTIONNAIRE REGARDING PARTICIPATION IN PHYSICAL ACTIVITY AND SPORT

127

QUESTIONNAIRE REGARDING PARTICIPATION IN
PHYS_ICAL ACTIVITY AND SPORT

Dear Parent(s) or Guardian:

The questions on this survey were written to obtain responses
from your child's point of view. However, for many children within
the range of ages being surveyed, some questions may be too difficult
for their reading level while other questions require recall that may
be beyond their current memory. Please assist your child in responding
to the enclosed questionnaire. In some cases only a few of the questions
may apply to your child. but all levels of response are important to
this study. Your assistance in helping your child complete and return
this questionnaire to his or her teacher within two weeks is greatly
appreciated.

Sincerely,

Beverly D. Zirich

(For additional information regarding this survey and its purpose see
enclosed materials.)

128

Child's Name
(firs: name and first initial
of last name only)

 

1. Have you been involved in one or more community or school sponsored sport

programs in the past year, other than phySIcal education class? (from Spring
1983 through Spring 1984)

yes no

(If your answer was ye_s, please continue with question #2. If your answer
mggomtoquestionltl-

 

 

2. If you answered 1e_s_ above, please complete one set of the following questions
for each sport in which you were a participant.
( A. Name the sport .
l 8. Did you complete the season? yes no
C. How many hours a week did you spend at practice?
D. Would you like to play this sport again next year? yes no
E. Would you like to try a different sport instead? yes no

 

F. Below are some reasons people give for participating in sports. Read
each item and decide if that item describes a reason why ygg participated
in this sport. Mark an "X" to indicate if that reason is very important,
somewhat important, or not at all important for you.

Very Somewhat Not at all
Important Important Important
; a. To improve my skills, or a. D U D
4_ learn new skills
5
mg“ b. To have fun b- D D D
c. 18::23: my parents thought c. D D D
a d. To be physically fit d. D D D
e. To compete e. D E] D
f. Likb: :éin’hISaniends or f. D E] D
g. For excitement g- D D D
I like the challenge h. D D D
i. I had nothing else to do i. D D D
j. Other (please specify) 1. D E] D

 

 

From the reasons listed above, go back and circle the reason that was
t the most important to you.

Spor‘l’ *3

B.

I"???

 

 

129

 

-2-

Name the sport .

Did you complete the season? yes no

How many hours a week did you spend at practice?

Would you like to play this sport again next year? yes no

Would you like to try a different sport instead? yes no

Below are some reasons people give for participating in sports. Read
each item and deCIde if that item describes a reason why Lo_u participated
in this sport. Mark an "X" to indicate if that reason is very important.
somewhat important. or not at all important for you.

 

 

Very Somewhat Not at all
Important Important Important

a. To improve my skills, or a.
learn new skills

b. To have fun b- D D E]
c. 18:33: my parents thought c. D U D
d. To be physically fit d. U D D
e. To compete e. D D D
‘- Lit.:t“.9:.§:‘°"d‘°' ‘ U '3 D
g. For excitement g. D C] D
h.‘ I like the challenge In E] D D
i. I had nothing else to do i. D E] D
j. Other (please specify) i- D D D

 

From the reasons listed above, go back and circle the reason that was
the most important to you.

Name the sport .
Did you complete the season? yes no
How many hours a week did you spend at practice?

 

 

Would you like to play this sport again next year? yes no
Would you like to try a different sport instead? yes no

 

Below are some reasons people give for participating in Sports. Read
each item and decide if that item describes a reason why y_o_u participated
in this sport. Mark an "X" to indicate if that reason is very important,
somewhat important, or not at all important for you.

Very Somewhat Not at all
Important Important Important

a. To improve my skills, or a. D D D
learn new skills

 

130

El [:1 C]

b. To have fun b. D D D
c. [82:23: my parents thought c. D [:1 D
d. To be physically fit d. D D D
e. To compete e. D D D
- f. Li's: ntthflmzngiends or f. [3 D D
g. For excitement g. D E] D
h. I like the challenge - h. D D D
i. I had nothing else to do i. C] D D
j. Other (please specify) i. D D D

 

Franthereasonslistedabove,gobackandcirclethereason thatwas
themostimportanttoyou.

If you chose to become involved in an organized sport program but did n0t
complete the season, please answer the following questions.

A. Name the sport .

8. Below are some reasons that peeple give when they stop participating
in a sport. Read each item and decide if that item describes a reason
why m stopped participating. Mark an "X" to indicate if the reason
was very important, somewhat important, or not at all important for you.

 

Very Somewhat Not at all
Important Important Important
a. I did net have enough fun a.
b. Practice was boring b.
c. - There was no teamwork c.
d.lwasnotasgoodasl d.

wanted to be

e. I wanted to play a different e.

DUE] [:lClElD
DUE] DDDD
DUE] DUDE

sport
f. I did not like the coach f.
g. It was not exciting enough 3.

h.
i.

i.

131

-4-

The training was too hard
I did not like the pressure

Other (please specify)

 

h.

i.

Cl
Cl

Cl
G

El
E]

El Cl C]

From the reasons listed above, go back and circle the reason that was
the most important for you.

If you answered 92 to question number one, please read the items below and
decide if that item describes a reason why you did not participate in an organ-

ized sport in the past year.

important, somewhat important, or not at all important for you.

b.

C.

e.

f.

I do not enjoy playing sports
I did not make the team

No sports are available for
children my age

The available programs were
too far from my home

The available programs cost
too much money

I am not very good at sports

Playing on a team takes too
much time

My parents thought I should
not play on a team

Other (please specify)

 

b.

d.

Very
Important

DC] DDDDDDD

Somewhat
Important

DC! DDDDDDD

Mark and "X" to indicate if that reason was very

Not at all
Important

CID CIDCIDCIEIEI

From the reasons listed above, go back and circle the reason that was
the most important for you.

6.

8.

10.

132

-5-

Place a check beside each age IISted below, if when you were that age. you
partimpated in at least one community or school sponsored sport program.

Age Name the sport or Sports

5

 

 

 

 

9

 

IO

 

How important is being good at sports to you?
very important somewhat important not at all important

If you could choose to spend free time at school doing one of the following,

which would you choose lst, 2nd, 3rd, 4th, and 5th (please number the activities
below).

_taking music or art lessons

_reading new books in the library

_playing games which involve arithmetic or spelling
_playing games in the gym

_working on a science project

How many hours a week do you spend playing physical games. such as basket-
ball, tag, kickball, jumping rope, frisbee, playing catch, baseball, etc.?

5 or less [0 15 20 25 30 or more

 

 

 

Name the sport programs available to children your age, in your area.

 

 

 

What sport or game do you like to play best of all? Why?

 

APPENDIX F
SAMPLE PICTURE PLATES FROM THE
PICTORAL SCALE OF PERCEIVED COMPETENCE AND
SOCIAL ACCEPTANCE FOR YOUNG CHILDREN

AND FROM PLATES ADDED BY THE INVESTIGATOR

134

 

 

 

 

 

 

 

 

 

 

LIST OF REFERENCES

138

LIST OF REFERENCES

Alderman, R., & Wood, N. (1976). An analysis of incentive motivation
in young Canadian athletes. Canadian Journal ongpplied Sport
Sciences, 1, 169-176.

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of
personality change. Psychological Review, 84) 191—215.

 

Boersma, F. J., & Chapman, J. W. (1978). Comparison of the Student's
Perception of Ability Scale with the Piers-Harris Children's Self-
Concept Scale. Perceptual and Motor Skills, 42, 827-832.

 

Boling, R., & Kirk, P. (1983). Differences between actual motor
abilityyand physical self-concept (perceived motor performance/
body image) of fifthegrade boys. Mississippi State, MS:
Mississippi State U. (ERIC Document Reproduction Service No. ED.
229-372)

 

 

Breer, P. E., & Locke, E. A. (1965). Task experiences as a source of
attitudes. Homewood, IL: Dorsey Press.

 

Bressan, E. S., & Weiss, M. R. (1982). A theory of instruction for
developing competence, self-confidence and persistence in physical
education. Journal of Teaching in Physical Education, 2, 38-47.

Brown, E. Y., Morrow, J. R. Jr., & Livingston, S. M. (1982). Self
concept changes in women as a result of training. Journal of
Sport Psychology, 4, 354-363.

 

 

Byrne, B. M. (1983). Investigating measures of self-concept.
Measurement and Evaluation in Guidance, 16, 115-126.

 

Carlson, R. K. (1970). Raising self-concepts of disadvantaged
children through puppetry. Elementary English, 41, 349-355.

 

Chapman, J. W., & Boersma, F. J. (in press). Academic self-concept
in elementary learning disabled children: A study with the
Student's Perception of Ability Scale. Psychology in the Schools.

 

Connell, J. P. (1982). A model of the relationships among children's
self-related cognitions, affects, and academic achievement.
Dissertation Abstracts International, 42, 49498. (University
Microfilms No. 82-09945)

 

Cooley, W., & Lohnes, P. (1971). Multivariate data analysis. New
York: John Wiley.

 

139

Coopersmith, S. (1967). Antecedents of self-esteem. San Francisco,
CA: W. B. Freeman.

 

Corbin, C., Landers, D., Feltz, D., & Senior, K. (1983). Sex
differences in performance estimates: Female lack of confidence
vs. male boastfulness. Research anrterly for Exercise and
Sport, 54, 407-410.

 

Corbin, C., & Nix, C. (1979). Sex-typing of physical activities and
success predictions of children before and after cross-sex
competition. Journal of Sport Psychology, 1, 43-52.

 

Corbin, D., Stewart, M., & Blair, W. (1981). Self—confidence and
motor performance of preadolescent males and females studied in
different feedback conditions. Journal of Spgrt Psychology, 3,
30-34.

 

Covington, M. V., & Omelich, C. L. (1979). It's best to be able and
virtuous too: Student and teacher evaluative response to successful
effort. Journal of Educational ngchology, 11, 688-700.

 

Dauer, V. (1971). Dynamic physical education for elementary school
children (4th ed.). Minneapolis, MN: Burgess.

Davis, J. W. (1981). The effects of a unified as compared to a
conventional arts program measured by the "How I see myself
scale," and the "Tennessee self-concept scale." Dissertation
Abstracts International, 49, 1545A. (University Microfilms No.
81-208116)

 

 

Duquin, M. E. (1978). Attributes made by children in coeducational
sport settings. In D. M. Landers, & R. W. Christina (Eds.),
Psychology of motor behavior and sport, 1977, (pp. 462-469).
Champaign, IL: Human Kinetics.

 

Dweck, C. S., & Reppuci, N. D. (1973). Learned helplessness and
reinforcement responsibility in children. Journal of Personality
and Social Psychology, 25, 109-116.

 

Felker, D. W. (1974). Building positive self-concepts. Minneapolis,
MN: Burgess.

 

Feltz, D., & Petlichkoff, L. (1983). Perceived competence among inter-
scholastic sport participants and dropouts. Canadian Journal of
Appliengport Sciences, 8, 231-235.

 

 

Festinger, L. (1954). A theory of social comparison processes.
Human Relations, 1, 117-140.

 

140

Glassow, R., Halverson, L., & Rarick, G. L. (1965). Improvement of
motor development and physical fitness in elementary school
children. (Cooperative research project #696). Madison, WI:
University of Wisconsin. (ERIC Document Reproduction Service
No. ED 002 927)

 

 

Griffin, N. S., & Keogh, J. F. (1981). Movement confidence and
effective movement behavior in adapted physical education.
Motor Skills: Theopy into Practice,_5, 23-35.

 

Griffin, N. S., & Keogh, J. F. (1982). A model for movement
confidence. In J. A. S. Kelso & J. E. Clark (Eds.), The
development of movement control and co—ordination, (pp. 213-
238). New York: John Wiley.

 

Guyot, G. W., Fairchild, L., & Hill, M. (1981). Physical fitness,
sport participation, body build, and self-concept of elementary
school children. International Journal of Sport Psychology, 12,
105-116.

 

Haas, H., & Maehr, M. (1965). Two experiments on the concept of
self and the reaction of others. Journal of Personality and
Social Psychology, 1, 100-105.

 

 

Harter, S. (1978). Effectance motivation reconsidered: Toward a
developmental model. Human Develppment,.2l, 34-64.

 

Harter, S. (1979a). Perceived cognitive competence as a determinant
of children'sypreference for challenge, Unpublished manuscript,
University of Denver, Denver, CO.

 

 

Harter, S. (1979b). Perceived Competence Scale for Children, Manual.
Denver, CO: University of Denver.

Harter, S. (1981a). The development of competence motivation in the
mastery of cognitive and physical skills: Is there still a place
for joy? In G. C. Roberts & D. M. Lenders (Eds.), Psychology of
motor behavior and sport, 1980, (pp. 3-29). Champaign, IL: Human
Kinetics.

 

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

 

Hatter, S. (1982). The perceived competence scale for children. Child
Development, 53, 87—97.

 

Harter, S. (1983, August). Personal communication.

Harter, S., 8 Pike, R. (in press). The pictoral scale of perceived
competence and social acceptance for young children.

141

Harter, 8., Pike, R., Efron, C., Chao, C., & Bierer, B. (1983).
Pictoral Scale of Perceived Competence and Social Acceptance
for Young Children, Manual. Denver, CO: University of Denver.

 

 

Herkowitz, J. (1978). Sex-role expectations and motor behavior of the
young child. In M. V. Ridenour (Ed.), Motor development: Issues
and applications, (pp. 83-98). Princeton, NJ: Princeton Book Co.

 

 

Horn, T. (1984, July). Children's perceived sport competence:
Informational sources and developmentalyprogression. Paper
presented at the Olympic Scientific Congress, Eugene, OR.

 

 

Hotelling, H. (1935). The most predictable criterion. Journal of
Educational Ppychology,‘24, 139-142.

 

 

Hughes, 0. A. (1973). A comparison of the effects of four teaching
techniques of body conditioning upon physical fitness and self-
concept. Dissertation Abstracts International, 34, 3957A.
(University Microfilms No. 73-31255)

 

Karpman, M. (1981). Redundancy in canonical analysis. Research
Quarterly for Exercise and Sport, 52, 291-292.

 

Keogh, J. F., Griffin, N. S., & Spector, R. (1981). Observer
perceptions of movement confidence. Research Quarterly,
465-473.

2.

 

King, B. R. (1982). The influence of motor ability, self-concept,
peer relationships and reported interests on children's choices
in physical education. Dissertation Abstracts International, 4;,
1079A. (University Microfilms No. 82-20939)

Kirkendall, D., Gruber, J., 8 Johnson, R. (1980). Measurement and
evaluation for physical educators. Dubuque, IA: wm. C. Brown.

 

 

Klecka, W. R. (1975). Discriminant analysis. In N. Nie, 0. Hull, J.
Jenkins, K. Steinbrenner, & D. Bent (Eds.), Statistical package
for the social sciences (2nd ed.), (pp. 434-467). New York:
McGraw-Hill.

 

 

Kleiber, D. A. (1979, October). Sex differences in leisure behavior:
Leisure as a context for androgeny. Paper presented at the SPRE
Research Symposium of the National Recreation and Park Association
Convention, New Orleans.

 

Kleiber, D. A. (1981). Of joy, competence, and significant others in
children's sports. In G. C. Roberts, & D. M. Landers (Eds.),
Ppychologyrof motor behavior and sport, 1980, (pp. 30-36).
Champaign, IL: Human Kinetics.

 

142

Kleiber, D. A., & Hemmer, S. (1981). Sex differences in the
relationship of locus of control and recreational sport participa-
tion. Sex Roles, 2, 801-810.

Kleiber, D. A., & Roberts, G. C. (1983). The relationship between
game and sport involvement in later childhood: A preliminary
investigation. Research Qparterly for Exercise and Sport, 54,
200-203.

Kukla, A. (1978). An attribution theory of choice. In L. Berkowitz,
(Ed.), Advances in experimental social_psychology (Vol. 11), pp.
113-144). New York: Academic Press.

Kun, A. (1977). Development of the magnitude-covariation and compensa-
tion schemata in ability and effort attributions of performance.
Child Development, 42, 862-873.

 

Lamb, M. E. (1976). The role of the father: An overview. In M. E.
Lamb (Ed.), The role of the father in child development, (pp. 1-61).
New York: Wiley.

Lever, J. (1976). Sex differences in the games children play. Social
Problems, 22, 478-487.

Lewko, J., & Ewing, M. (1980). Sex differences and parental influence
in sport involvement of children. Journal of Sport Psychology, 2,
62-68.

Maehr, M., Mensing, J., & Nafzger, S. (1962). Concept of self and the
reaction of others. Sociometry, 22, 353-357.

 

Magill, R., & Ash, M. (1979). Academic, psycho-social, and motor
characteristics of participants and non-participants in children's
sport. Research Quarterly, 29, 230-240.

Martens, R., & Seefeldt, V. (1979). Guidelines for children's sports.
washington, DC: American Alliance for Health, Physical Education,
Recreation, and Dance.

Martinek, T. J., & Zaichkowsky, L. D. (1977). Manual for the
Martinek-Zaichkowsky Self-concept Scale for Children.
Jacksonville, IL: Psychologists and Educators.

Masche, K. (1970). Effects of two different programs of instruction
on motor performance of 2nd grade students. Research Qmarterly,
2, 406-411.

Masters, J. C. (1971). Social comparison by young children. Young
Children, 22, 37-60.

Maul, T., & Thomas, J. R. (1975). Self-concept and participation in
children's gymnastics. Perceptual and Motor Skills, 42, 701-702.

143

McClintock, C. C., & Moskowitz, J. M. (1976). Children's preferences
for individualistic, cooperative, and competitive outcomes.
Journal of Personality and Social Psychology, 34, 543-555.

McClintock, C. C., Moskowitz, J. M., & McClintock, E. (1977).
Variations in preference for individualistic, competitive, and
cooperative outcomes as a function of age, game class, and task
in nursery school children. Child Development, 4g) 1080-1085.

Meredith, W. (1964). Canonical correlations with fallible data.
Psychometrika, g2, 55-65.

 

Miller, S. E. (1978). Facilitation of fundamental motor skill
learning in young children. Dissertation Abstracts International,
49, 740A. (University Microfilms No. 79-17753)

 

Minton, B. (1979). Dimension of information‘underlying children's
Judgments of their competence. Unpublished master's thesis,
University of Denver, Denver, CO.

 

Mintzes, J. (1979). Overt teaching behaviors and students' ratings
of instruction. Journal of Experimental Education, 4g, 145-153.

Nicholls, J. G. (1978). The development of the concepts of effort
and ability perception of own attainment, and the understanding
that difficult tasks require more ability. Child Development, 423
800-814.

Nicholls, J. G. (in press). Achievement motivation: Conceptions of
ability, subjective experience, task choice, and preference.
Psychological Review.

Orlick, T. D., & Botterill, C. B. (1975). Every kid can win. Chicago,
IL: Nelson Hall.

Pangrazi, R. (1982). Physical education, self-concept, and achievement.
Journal of Physical Education, Recreation, and Dance, 53, l6-18.

Pate, T. L. (1978). An investigation of the effects of creative drama
upon reading ability, verbal growth, vocabulary development, and
self-concept of secondary school students. Dissertation Abstracts
International, 3g) 6506A. (University Microfilms No. 78—05471)

 

Peterson, K., Reuschlein, P., & Seefeldt, V. (1974). Factor analysis
of motor performance for kindergarten, first, and second grade
children: A tentative solution. Paper presented at the annual
meeting of the American Alliance for Health, Physical Education,
Recreation, and Dance, Anaheim, CA.

 

Piers, E. V. (1969). Manual for the Piers-Harris Children's Self-
Concept Scale. Nashville, TN: Counselor Recordings and Tests.

 

144

Rarick, G. L., & Dobbins, D. A. (1975). Basic components in the motor
performance of children six to nine years of age. Medicine and
Science in Sports, 1, 105-110.

 

 

Roberts, G. C., Kleiber, D. A., & Duda, J. L. (1981). An analyis of
motivation in children's sport: The role of perceived competence
in participation. Journal of Sport Psychology,_3, 206-216.

 

Roberts, G. C., & Pascuzzi, D. L. (1979). Causal attributions in
sport: Some theoretical implications. Journal of Sport Psychology,
1, 203-211.

 

Rogers, M., Miller, N., & Henningan, K. (1981). Cooperative games as an
intervention to promote cross-racial acceptance. American
Educational Research Journal, lg, 513-516.

 

Rohrbacher, R. (1973). Influence of a special camp program for obese
boys on weight loss, self-concept, and body image. Research

Quarterly, 44) 150-157.

Rosenberg, M. (1979). Conceiving the self. New York: Basic.

 

Ruble, D. N., Boggiano, A. K., Feldman, N. S., & Loebl, J. H. (1980).
Developmental analysis of the role of social comparison in self-
concept evaluation. Developmental Psychology, 19, 105-116.

 

Ruble, D. N., Feldman, N. S., & Boggianao, A. K. (1976). Social
comparison between young children in achievanent situations.
Developmental Psychology, 12) 192-197.

 

Safrit, M. (1973). Evaluation in physical education: Assessing
motor behavior. Englewood Cliffs, NJ: Prentice Hall.

 

Sage, G. (1977). Introduction to motor behavior; A neurophysiological
approach (2nd ed.). Reading, MA: Addison-Wesley.

 

Sapp, M., & Haubenstricker, J. (1978, April). Motivation for joining
and reasons for not continuing in youth sports_programs in
Michigan. Paper presented at the American Alliance for Health,
Physical Education, and Recreation National Convention, Kansas
City, MO.

 

Scanlan, T., & Passer, M. (1981). Determinants of competitive per-
formance expectancies of young male athletes. Journal of
Personalipy, 42, 60-74.

 

 

Schempp, P., Cheffers, J. T. P., & Zaichkowsky, L. D. (1983). Influence‘
of decision-making on attitudes, creativity, motor skills, and
self—concept in elementary children. Research Quarterly for
Exercise and Sport, 24) 183-189.

 

 

 

 

145

Seefeldt, V. (1984, April). NASPE Justification Project. Paper
presented at the annual meeting of the American Alliance for
Health, Physical Education, Recreation, and Dance, Anaheim, CA.

 

Seefeldt, V., & Gould, D. (1980). Physical and_psychologica1 effects
of athletic competition on children and youth. Washington, DC;
ERIC Clearinghouse on Teacher Education.

 

Selby, R., & Lewko, J. (1976). Children's attitudes toward females in
sports: Their relationship with sex, grade, and sports participa-
tion. Research Qparterly,‘4l, 453-463.

 

Shephard, R. J., Lavellée, H., Jequier, J-C., Rajic, M., & LaBarre, R.
(1980). Additional physical education in the primary school - A
preliminary analysis of the Trois-Rivierres Regional Experiment.
In M. Ostyn, G. Beunen, & J. Simona (Eds.), Kinanthropometry II,
International series on sport science, Vol. 9. Baltimore, MD:
University Park Press.

 

 

Singer, R. (1975). Motor learning and human performance (2nd ed.).
New York: Macmillan.

Smith, T. (1982). Self-concepts and movement skills of third grade

children after physical education programs. Perceptual and Motor
Skills, 24, 1145-1146.

Smoll, F. L. (1974). Motor impairment and social development. American
Corrective Therapy_gourna1, 3g, 4-7.

 

Smoll, F. L., & Lefebvre, L. M. (1979). Psychology of children in sport.
International Journal of Sport Psychology, 49, 173-177.

 

Smoll, F. L., & Schutz, R. W. (1984, July). .Age trends and sex
differences in_physical fitness: Athletes and nonathletes.
Paper presented at the Olympic Scientific Congress, Eugene, OR.

Smoll, F., Schutz, R., & Keeney, J. (1976). Relationships among
children's attitudes, involvement, and proficiency in physical
activities. Research Qparterly, 41, 797-803.

Sonstroem, R. J. (1982). Exercise and self-esteem: Recommendations
for expository research. goest, 33, 124-139.

Sonstroem, R., & Kampper, K. (1980). Predictions of athletic participa-
tion in middle school males. Research Qparterly,.§1, 685-694.

Spink, K. S., & Roberts, G. C. (1980). Ambiguity of outcome and causal
attributions. Journal of Sport Psychology, 3, 237-244.

 

State of Michigan (1976). Joint legislative study on youth sports
programs: Phase I. East Lansing, MI: State of Michigan.

 

146

Stewart, D., & Love, W. A. (1968). A general canonical correlation
index. Psyoholical Bulletin, 19, 160-163.

Stipek, D. J. (1981). Children's perceptions of their own and their
classmates' ability. Journal of Educational Psychology, 13,
404-410.

Suls, J., & Sanders, G. S. (1979). Social comparison processes in the
young child. Journal of Research and Development in Education,
l3, 79-89.

Veroff, J. (1969). Social comparison and the development of achievement
motivation. In C. P. Smith (Ed.), Achievement-related motives in
children, (pp. 46-101). New York: Russell Sage Foundation.

Weiner, B. (1972). Theories of Motivation: From mechanism to cogni-
tion. Chicago, IL: Markham.

Weiner, B. (1974). Achievement motivation and attribution theory.
Morristown, NJ: General Learning Corp.

Weiss, D. J. (1972). Canonical correlation analysis in counseling
psychology research. Journal of Counseling Psychology, lg,
241-252.

weiss, M., Bredemeier, B., & Shewchuck, R. (1984, July). The dynamics
oflperceived competence, perceived control, and motivational
orientation in youth sports: A causal analysis. Paper presented
at the Olympic Scientific Congress, Eugene, OR.

Wiggins, R. R. (1973). Differences in self-perceptions of ninth grade
boys and girls. Adolescence, g, 491-496.

White, R. W. (1959). Motivation reconsidered: The concept of competence.
Psychological Review, 44, 297-333.

wylie, R. (1974). The self concept: A critical survey of pertinent
research literature. Lincoln, NE: University of Nebraska Press.

Yeatts, R. R., & Gordon, I. J. (1968). Effects of physical education
taught by a specialist on physical fitness and self-image.
Research Quarterly, 32, 766-770.

Zaichkowsky, L. D., Zaichkowsky, L. B., & Martinek, T. J. (1980).
Growth and development. St. Louis, MO: C. V. Mosby.