A 4.". anti...
3

 

.
.9 . $3.1...-

L... .\ a:

A.

23.1. . .5...
. J 1: I.
..

£3 .
7::
‘

. . l.|\r. v...
x‘..?.?.
1215.....12.

 

 

.
It .

.1. .. .} .33.
«.5.

r

.:-..l.. . r
u, r;

 

‘

12:)? Y,.:i\| I . (:31...

Al 5.
., , {.1317 t
'9)‘1.r.‘ , .1. 51:11}. -
.537!
u:.tn§3n§...v. 3|»
1

:
.v! 74;.»
.1 ' .v . 0.. x 1 .t 5333'. r
A\r. l...|.....! .
.tsg‘vne. t {1.315
a5!1..1...‘s‘.l.tft
. . 51;. is 6.: .I¢

 

 

;v.l...l .2. u‘
n f .

 

 

  

- :11: . V :1 , ‘
1.19.?

d. .J. :5

    

v . 7 f. ~:......ro 31;. .
. . . ..

  

 

A.‘Vu.. .7.
1...}. I!

 

.31. an
.5031. r

£125:
.- .3 )1 «it‘ll Hit.
.l- .n".."‘ 9-, I
In? in. ,.sr..5’vu;l
5.6.02: . 3 .031. it‘v
1.5.1.... . .1513!!!- ‘I\
luv, fl .75. u.)-
.)A r0... ‘- ‘. I’Lu
I!!!

 

{I3}: .2!- v...
. .v

0" lb.t)|||..
.r‘..‘3.-l4.

p.061 .
I): 3521'?!) tl'.f101|
.0... IIIs..{¢O [Iikll )

i Lt‘olvxi.

I'illllvl- ‘l .1 .

 

WERSWY LIBRARlES

lWW WWl WW

 

 

 

 

 

 

 

 

 

 

 

WWWIWWWWWWWWWWWWWWW WOOWWWWWWWWW

This is to certify that the

thesis entitled

THE RELATIONSHIP BETWEEN QUALITATIVE AND

QUANTITATIVE EVALUATION OF THROWING
presented by

RANDALL LEE BAKE R

has been accepted towards fulfillment
of the requirements for

 

MS degree inlhxsiaa). Education
and Exercise Science

 

Major professor

Date 27/773
/ I

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

 

 

W LIBRARY
MIchIgan State

UnhmnIty

 

 

 

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

DATE DUE DATE DUE DATE DUE ll

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

“WW W
W

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU I: An Affirmative Action/Equal Opportunity Institution
cmmd

 

THE REIATIQJSHIP BEIWEEN QJALITATIVE AND
WHATIVE EVALUATICN OF WING

by
Randall Lee Baker

A THESIS
Submitted to
Michigan State university
in partial fulfillment of the requirements
for the degree of

MAS'IEROFSCIENCE

Department of Physical Education and Exercise Science
1993

ABSTRACT

THE RELATIONSHIP BEIWEEN QWXLITA‘I‘IVE AND
QUANTITATIVE EVALUATION CF 'I‘HEKIHNG

By
Randall Lee Baker

This study examined the effect of gender, grade, and throwing
pattern on the distance children threw a ball. Children (n=303) in
grades K-5 were assessed on throwing pattern, throwing distance, grip
strength, push-ups, height, and weight. Results showed that boys threw
a ball farther than girls in each grade; older children (grades 3-5)
threw farther than younger children (grades K-Z); and, girls with a
mature pattern threw farther than girls with less mature patterns.
Throwing pattern in boys could not be analyzed because of the high
percentage (94%) who threw with a mature pattern. Regression analyses
revealed that grip strength, height, and push-ups also influenced the
distance throws of children. The results of this study, plus the
importance of factors not studied such as practice time and motivation,
led to the conclusion that the throwing performance of children is

dependent on more that gender, grade, and throwing pattern.

Copyright by
RANDALL LEE BAKER

1993

This thesis is dedicated to my wife, Dianna. ‘Without her help and

encouragement, this thesis never would have been written.

iv

ACKNCNLEIISEMENTS

I would like to thank Professor Crystal Branta for serving on my
committee.

Thank you to Professor Martha Ewing for serving on my committee and
providing valuable help in preparing the statistical portion of this
study.

A.special thanks to Professor John Haubenstricker for serving as
the chairperson of my committee and for the many hours he spent with me
planning and organizing the testing procedures, reviewing the throwers'

stages and his suggestions, support, and editing in writing this thesis.

TABLE OF CONTENTS

LIST OF TABLES
LIST OF FIGURES

CHAPTER 1: INTRODUCTION
Need for the Study
Purpose of the Study
Limitations
Hypotheses

CHAPTER 2: SURVEY OF RELATED LITERATURE
Methods of Evaluating Throwing
Accuracy
Distance
velocity
Fbrm
Factors Affecting Throwing Distance
Age
Gender
Strength
Body size
Form
Instruction
Justification of Testing Procedures
Strength
Grade versus Age
Modeling

CHAPTER 3: METHOD
Subjects
TEsting Procedure
Throwing Ability
Strength
Other Data
Treatment of Data

CHAPTER 4: RESULTS AND DISCUSSION
Descriptive Statistics
Predictability of Throwing for Distance
Differences in Throwing Distances for Gender and Grade
Differences in the Effect of Stage on Throwing Distance
Boys
Girls

vi

vii

ix

OUTU'IubH

10
12
12
15
23
23

24
32
38
41
45
45
45
45

48
48
50
51
54
55
57

59

64
71
75
76
78

Table of Contents (Continued)

CHAPTER 5: SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS 86
Summary 86
Summary of Results 88
Conclusions 90
Recommendations 92

APPENDIX A: Parental Information Sheet and Parental and Student
Consent Form 93

APPENDIX B: Means and Standard Deviations of All Independent
variables by Grade and Gender 95

LIST OF REFERENCES 100

vii

LIST OF TABLES

Table 1: Total Number of Participants and Dropouts in the
Study by Grade and Gender

Table 2: Testing Personnel and Their Duties

Table 3: Means, Standard Deviations, and t-values for the
Chronological Age of the Subjects (in months)

Table 4: Means and Standard Deviations for the Distance
Throw (in feet)

Table 5: Means and Standard Deviations for Throwing Stage
(range = l to 5)

Table 6: Test, Retest Correlation coefficients of
Strength Measures in Selected Grades

Table 7: Intercorrelation Matrix for Boys and Girls Combined
Table 8: Intercorrelation Matrix for the Boys
Table 9: Intercorrelation Matrix for the Girls

Table 10: Stepwise Multipde Regression of Predictor
variables on Throwing Distance for Boys and Girls Combined

Table 11: Stepwise Multipfle Regression of Predictor
variables on Throwing Distance for Boys

Table 12: Stepwise Multipfle Regression of Predictor
variables on Throwing Distance for Girls

Table 13: Signficant Differences of Throwing Distance
Between Grades K and 5

Table 14: The Number of Boys at Each Stage of Throwing
in Each Grade

Table 15: The Number of Girls at Each Stage of Throwing
in Each Grade.

viii

49

56

61

62

64
66
67
68

70

7O

72

77

79

List of Tables (Continued)

Table 16: Means of the Distance Thrown by Grade and Stage
for Girls

Table 17: Significant Differences of Throwing Distance for
Grades 1 to 4

Table 18: Significant Differences of Throwing Distance for
Stages 3 to 5

Table B - 1: Means and Standard Deviations for Height (in inches)
Table B — 2: Means and Standard Deviations for Weight (in pounds)

Table B - 3: Means and Standard Deviations for Grip Strength
(in K9)

Table 8 - 4: Means and Standard Deviations for Push-ups (number
completed in 30 sec)

Table B - 5: Means and Standard Deviations of the Retest of
Grip Strength (in Kg)

Table B - 6: Means and Standard Deviations of the Retest of
Push-ups (number completed in 30 sec)

ix

79

81

81
95

96

97

98

99

99

LIST OF FIGURES

Figure 1: Mean Distance throw of Girls and Boys in
Kindergarten through Fifth Grade 73

Figure 2: Gain in Mean Distance Throw of Girls and Boys
Between Successive Grades 74

CHAPTER 1

INTRODUCTION

The overhand throw is an important skill in many games and sports
in American culture. The ability to throw with force and with accuracy
is critical to success in sports such as baseball, football, and
basketball, as well as in many of the games and activities in which
children participate. Because of its importance in games and sports,
physical educators, coaches, and researchers have sought various means
by which to assess the throwing ability of individuals.

The research literature contains numerous investigations that focus
on the evaluation of throwing ability. These extensive investigations
have resulted in a variety of approaches to evaluate throwing ability.
The approaches may be grouped into two categories, those which focus on
the outcome of throwing and those that examine the process of throwing.

Achievement scores have long been used to determine throwing
ability. The most cannon score is the distance a ball can be thrown in
the air (AAHPER, 1965; Espenchade, 1960; Hanson, 1965; Hartman, 1943;
Jenkins, 1930; Keogh, 1965; Morris, William, Atwater, s. Wihrore,

1982). Measuring the accuracy with which a ball is thrown is another
method used to evaluate throwing (Fredrick, 1977; Hicks, 1930; Keogh,
1965; Miller, 1957). A third approach is to measure the velocrty of the
ball as it leaves the hand of the thrower (Glassow, Halverson, & Rarick,
1965; Glassow & Kruse, 1960; Halverson, Roberton, & Langendorfer, 1982;
Halverson, Roberton, Safrit, & Ibberts, 1977; Ebberton, Halverson,

Iangendorfer, & Williams, 1979).

2

The motion which produces the force to propel the ball also has
been examined. For example, anatomical, biomechanical, or
kinesiological variables have been used to examine parts of the throw
such as stride length (Schutzler, 1980) , increased range of motion
(Luedke, 1980) , rotation of various joints involved in throwing (Lyon,
1961: Ekern, 1970; Sanders, 1977; Vaughn, 1982), speed and
acceleratioi/deceleration of body parts involved in the throw (Deutsch,
1969), and the timing of the sequence of the body parts (Atwater, 1970).

Other investigators, especially those at Michigan State University
and the University of Wisconsin at Madison, have proposed developuental
sequences for fundamental motor skills including throwing. These
sequences organize the anatonical an} biotechanical information into
stages or steps through which performers progress as they develop a
better throw (Roberton, 1982; Seefeldt s. Haubenstricker, 1982).

Oonplicating the evaluation of throwing ability are several factors
which influence throwing performance. Factors which have been shown to
influence throwing performance are strength, body size, form and
instruction.

Strength has been shown to be a major factor in the performance of
many motor skills but rarely has been a good predictor for success in a
skill by itself (Johnson 5. Nelson, 1979). There is a wide range annng
the correlation coefficents counted between strength and throwing
distance. Correlation coefficents between strength and throwing
distance ranged between .31 and .42 for Sullivan (1970) and between .41
and .71 for Espenschade (1940). Significant relationships were found
between throwing velocity or distance and specific strength measures
such as shoulder strength (Brumfield, 1969) , grip strength (Richardson,

3
1976), and strength of wrist extension and elbow extension (Pegdagana,
Elsner, Roberts, Land, 5 Farewell, 1982). Several studies (Bagonzi,
1979; Rowlands, 1962; Sullivan, 1970) have shown that strength
development has increased a person's ball throwing velocity. On the
other hand, several studies (Barrow, 1960; Hardison, 1971; Staub, 1966;
William, 1985) report that strength building program have not
increased a person's throwing distance or velocity.

In examdning the relationship between body size variables and
throwing ability, Eoff (1985) found height, weight, arm length, and
subcutaneous fat to be predictive of throwing ability in young boys, and
arm length to be predictive of throwing ability in young girls. Mahnoud
(1979) reported significant relationships between the distance a ball
was thrown and height, hand, and forearm length. However, no
significant relationships between height or weight and throwing
performance were found by other researchers (Espenchade, 1940; Fredrick,
1977; Johnson, 1960; Seils, 1951). Other studies (Bowne, 1960;
Richardson, 1976; Sanders, 1977) alsoiconcluded that there was not a
significant relationship between arm length and throwing performance in
high school.or college students.

Biomechanical factors found to be associated with a better throw
are an increase in the range of trunk rotation (Bowne, 1960; Ekern,
1970; Singer, 1961); a decrease in the medial rotation of the arm
(Bcwne, 1960; Ekern, 1970); better ratings of formifor trunk and foot
action (Nelson, Tnonas, Nelson, 5 Abraham, 1986) ; rapid sequential
acceleration and deceleration of trunk and arm segments prior to release
(Atwater, 1970; Deutsch, 1969); an increase in stride length (Deutsch,

1969; Ekern, 1970; Schutzler, 1980); a greater forward flexion at the

4

hip joint at the point of release (Ekern, 1970; Lyon, 1961); and, a
greater range of movement in contributing joints (Singer, 1961).

Instruction has been found to improve throwing performance in sane
cases (Dusenberry, 1952; Hoffman, 1969; Luedke, 1980; Mah. , 1979;
Miller, 1978; Potter, 1963) while in other cases it did not
significantly increase throwing distance or velocity (Deatrick, 1977;
Dohrman, 1964; Glassow et a1., 1965; Halverson et a1., 1977; Nichols,
1971; Roberton et a1., 1979).

Need for the Study

Although the product and process of throwing have been studied
separately, there has not been an attempt to compare the quality of the
throwing motion as determined by a developmental sequence with the
resultant quantitative measure of the distance a ball is thrown. The
advancement from a lower developmental stage of throwing to a more
mature stage has not been verified to produce a concurrent improvement
in the distance a ball is thrown.

The primary reason for becoming aware of a developnental sequence
for a fundamental skill like throwing is to help the learner become more
proficient in the skill. A good evaluation describes the present skill
level of the learner, whereas the developmental sequence suggests a
systematic progression to help the student obtain an optimal skill
level. Evidence of an effective developnental sequence is that subjects
at a more mature stage generally achieve a better performance score than
subjects at a less mature stage. It is important that a.more mature
throwing pattern has an impact on performance, otherwise there would be
no need for teachers to instruct students to achieve a mature throwing

pattern. And there would be no need for correction of a thrower's

movement pattern. Emphasis then would be placed on throwing the ball as
far as possible regardless of the pattern, or upon other factors such as
increasing body strength. One way to evaluate a developnental sequence
is to determine if each successive stage reflects an increase in
quantitative performance when comparing children of similar age and/or
strength. Research is needed to determine if such a relationship exists
between the quality of the throwing motion and the outcome of the throw.
Purpose of the Study

The purpose of this study was to establish the validity of a
developnental sequence of throwing as a measure of throwing
performance. This study examined the relationship between the
quantitative aspects of throwing using the distance a ball was thrown
and the qualitative aspects of the novement content using the
developnental sequence for throwing proposed by reseachers at Michigan
State University. To acoonplish this purpose, the study required: (a)
information on the stage of throwing of males and females in
kindergarten through the fifth grade; (b) measurements of the children's
strength; (c) determinatim of the relationship between the
developnental stage of throwing and throwing distance; and, (d)
determination of the relationship between strength and throwing
distance.

Limitations

This study was limited to the use of an available sample of
students in kindergarten through fifth grade in a school with particular
characteristics. Thus the conclusions can pertain only to samples with
similar characteristics. This study also was limited to the oterhand

throw and the results should not be generalized to any other motor

6

skill. This study was limited by the fact that the nptivation of the
students could not be completely controlled even though an attempt was
made to encourage the subjects to do their best. Because of the use of
intact classes, the range of throwing stages, class size, and male to
female ratio were not controlled. The residual fallout effects caused
by the students who chose not to participate also might have influenced
the results of the study. A final limitation was that the developmental
sequence of throwing used to evaluate the quality of throwing has not
been validated with longitudinal data.
Hypotheses

This investigation tested three hypotheses. The first two
hypotheses have been shown to be true in previous studies and were
tested to see if they were true for this study also.

1) For each grade level, boys will throw a ball farther than will
girls.

2) Mothers of the same gender in a higher grade will throw a ball
farther than those in a lower grade.

The primary hypothesis of this investigation is that:

3) Children of the same grade and gender who have a more mature
pattern of throwing as determined by the Michigan state
University developnental sequence will throw a ball farther
than children who exhibit a less mature stage of throwing.

Factors other than form can affect the distance a ball is throm.

While the primary hypothesis proposed that stage of throwing is an
important variable in determining throwing distance, the strength,

weight, and height of a child also . y influence throwing performance.

7

Children who are taller, heavier, and/or who possess. greater strength
theoretically should be able to throw a ball farther than children who
are shorter, lighter, and/or weaker. The relationship of weight,
strength, and height to throwing form has not been established,

therefore their potential impact on throwing performance will need to be
controlled.

CHAPTER 2

SURVEY OF RELATED LITERATURE

This study examined the relationship between two nethods of
evaluating the throwing ability of individuals. The first means was a
distance throw. This is the oldest and nest comen nethod used to
evaluate throwing. A second means of evaluating throwing was to study
the netor pattern individuals use to throw a ball. Instead of being
concerned with an achievement score produced by the throw, this
evaluation is concerned with the nevenents involved in making the throw.

Research on throwing ability is complicated by the different
methods used to evaluate throwing ability. The assessnent of a netor
skill can be expressed in several ways, but nest often it is indicated
by an achievenent score (Wickstrom, 1983) . The nest comen measure of
throwing ability is the distance a person can throw a ball. However,
throwing ability also can be neasured by initial ball velocity, accuracy
or form. Glassow (cited in Halverson et al., 1977) pointed out that the
distance a ball is thrown is a function of initial velocity and the
angle of projection. She advocated recording both conponents of
distance to identify nere precisely the contribution of each. Film
analysis or a velocineter can be used to determine initial velocity
(Glassow et a1., 1965; Halversoi et a1., 1982; Halverson et a1., 1977;
Nichols, 1971; and Raberton et a1., 1979). Another achievement score
that is used to evaluate throwing is accuracy (Fredrick, 1977; Keogh,
1965; and Miller, 1957) . A fourth way to measure throwing ability is to
study the nevenent involved in producing the result. To some

8

researchers, the distance a ball is thrown is not as important as the
throwing motion used to propel the ball (Beach, 1950; Gutteridge, 1939;
Mahneud, '1979; lbberton & Halverson, 1984; Ryan, 1977; Seefeldt 8:
Haubenstricker, 19766; and Wild, 1938). Numerois biotechanical and
kinesiological factors also have been proposed to affect the throwing
motion.

This literature review will examine the different methods of
evaluating throwing with the exception of the use of evaluation by
biotechanical and kinesiological neans. Nest of the studies which
examine throwing by these means have used mature subjects who are
participating in a baseball program either at the high school, college
or professional level. They have examuned parts of the mature throw
such as stride length, the rotation of various joints, speed and
acceleration/deceleration of body parts, and the timing of the sequence
of body part movements. Review of all the investigations regarding
bionechanical and kinesiological implications of the throwing netion
does not fit into'the scope of this study. Readers interested in the
biotechanical aspects of throwing are directed to Hay (1985) , Gowitzke
and Milner (1988), and Kreghbaum.and Barthels (1985). A listing of
studies involving the bicmechanical analysis of throwing is provided by
Wickstron (1983) .

Research on throwing ability must also account for other factors
that may affect throwing. Such factors include age, gender, body size,
and strength. They stimulate questions such as: Is there a difference
in the throwing ability of boys and girls? Can a stronger, taller
subject generate nere power and therefore throw the ball farther than a

snaller, weaker subject even if the smaller subject has better form? Is

10

age a factor in throwing? Are there other neasures which give sone
subjects an advantage over others when throwing a ball for distance?

The review of literature will focus on methods of evaluating
throwing: (a) accuracy; (b) distance; (c) initial velocity; and, (d)
form. Following these four methods of evaluating throwing, factors
affecting throwing ability will be examined. These factors include:

(a) gender; (b) strength; and, (c) body size.
Methods of Evaluating Throwing

various approaches to evaluating throwing behavior have been
reported in the literature. The most common of these include the
assessment of accuracy, distance, initial velocity, and foam.

Accurfl

Accuracy is an important component in a mature throw, especially
when throwing is used in a gane or play situation. Accuracy is a major
way of evaluating throwing (Frederick, 1977; Hicks, 1930; Keogh, 1965;
Miller, 1957; van Slooten, 1973; and wester, 1939). While it is
difficult to summarize the results of the accuracy test studies because
of the different size balls (therefore different methods of throwing)
and the varietylof tests used, two general trends have been
identified. First, boys were found to be significantly more accurate in
throwing than girls of similar age. Second, both girls and boys showed
an improvenent in accuracy scores with increasing age.

Boys are more accurate than girls in throwing across the ages two
through nine years. Boys, ranging in age from two»to six years,
excelled over girls of the same age on a moving target accuracy test
(Hicks, 1930). Preschool boys from.three to five years of age scored

significantly higher than preschool girls on a test of throwing accuracy

ll

(Fredrick, 1977). In testing the throwing behavior of first grade
children, boys were found to be significantly nere accurate than girls
(Miller, 1957) . Keough (1965) noted a superiority of boys over girls
for children across the ages of 7, 8, and 9 years. In a throwing-for-
accuracy task involving boys and girls ages 6, 7, 8, and 9 years, the
boys were significantly better than the girls at each age level (Van
Slooten, 1973) .

Throwing accuracy increases for both boys and girls fron age two
years through grade 5. Improvement with age occurred regardless of the
testing procedures or equipnent used. In studying children aged two to
six years, Hicks (1930) noted the older children had the highest nean
scores. Children grouped in six nenth intervals from age three to five
showed significant improvenent from group to group (Frederick, 1977).
Keough (1965), testing children across the ages of 7, 8, and 9 years and
Van Slooten (1973) assessing children aged 6, 7, 8, and 9 years, both
found an annual improvement in accuracy performance. Mean total scores
increased fron grade three to grade five when boys were tested on their
throwing accuracy (Wester, 1939).

While accuracy is an important conponent in the mature throw, it
may be possible to obtain accuracy without using good form or producing
maximum force. In an accuracy test, the primary problem is for the
subject to hit a target and the form used to produce the results is
ignored. An investigation by Hicks (1930) provides a good example.
Hicks tested throwing accuracy in children from age three to six
years. He also made sole general observations and ornaments on their
style of throwing. He said, ”It is impossible to say that any one style

of throwing is always best, for good throwing is throwing that produces

12

good results"(p. 52). This is not the view that is taken here. There
is a correct style of throwing. There exists a style of throwing that
will result in a better throw. The reason throwing accuracy was not
selected as a measure for this study is because accuracy may be obtained
without necessarily requiring a mature throwing pattern.
Distance

The throw for distance has been an important tool for assessing
throwing ability (AAHPER, 1965; Espenchade, 1960; Hanson, 1965; Hartman,
1943; Jenkins, 1930; Keogh, 1965; Morris et a1., 1982) as well as a
measure of explosive strength (Fleishman, 1964). Many investigations
using the throw for distance to evaluate throwing ability have reported
that (a) boys exhibit better throwing ability than girls of the same
age, and (b) both boys and girls show a year-by-year improvement in
distance scores. The superior performance of boys over girls in the
distance throw and year-by-year improvenent in distance scores for both
girls and boys has been noted by Keogh, (1965); Hanson, (1965); Nichols,
(1971); and Hardin and Garcia, (1982). The trend has been shown to
begin with preschool children as young as three years of age (Frederick,
1977; and Johnson, 1960). Vincent (1968) reported that girls continued
to improve in a beanbag distance throw fron 7th grade through the 10th
grade after which their performance leveled off.
velocigy

SaIe studies have used ball velocity at the time of release as a
treasure of the force that subjects can produce during the throwing
netion. Mast of these studies originated at the University of Wisconsin
at Madison (Glassow, Halverson, & Rarick, 1965; Glassow & Kruse, 1960;

Halverson, lbberton, and Langendorfer, 1982; Halverson et a1., 1977;

13

Roberton et a1., 1979) . Researchers contend that the distance the ball
travels is a functim of the initial velocity imparted to the ball and
the angle'at which the ball is released. By examining ball velocity
they believe they are eliminating the errors the subjects produce in the
angle of release. When the release point is inaccurate, the actual
distance the subject throws a ball is less than the real distance the
subjects could throw the ball when the release point is correct. They
reasm that velocity is a nere accurate measure of the force generated
by the throwing netion than distance because the distance result is
complicated by the angle of release whereas initial velocity is ret.

While such reasoning is sound, the neasurenent of velocity requires
a large investment of neney for equipment and entails logistical
problems in performing the skill. Few physical educators have a
velocimeter to measure the velocity of a ball at release. Funds
generally are ret available for an expensive piece of equipment that
will receive limited use. Arether problem with velocity testing for
young children is the limited visual feedback they receive fron their
throw. Because velocity is measured in a horizontal plane, children are
instructed to throw a ball with as nuch force as they can against a
target on a wall and ret how far they can throw the ball. While
children can obtain sate auditory feedback from the soind of the ball
hitting the wall and sole kinesthetic feedback from how hard they feel
they are throwing the ball, children would nere likely use visual
feedback and it is difficult for than to know how far they are throwing
theball. Withadistance throw theycanseehowfar itgoesand
understand that a harder throw will produce a longer throw.

14

While there are some motivational and financial problems in using
velocity measurements to evaluate throwing ability, the investigations
assessing throwing velocity have verified the findings of studies using
the distance throw; namely, that boys throw faster than girls and that
throwing velocity increases from year to year. Glassow and Kruse (1960)
reported a yearly increase in velocity for girls in grades 1 through 8
at an annual rate of 3 to 4 feet per second per year. Halverson and
Roberton and their associates (Halverson et a1., 1977; Halverson et a1.,
1982, and Roberton et a1., 1979) completed an eight-year longitudinal
study of 22 boys and 17 girls that has been reported on three separate
occasions. The children were tested in kindergarten, first grade,
second grade, and seventh grade. The subjects were evaluated on both
ball velocity and developmental form. They concluded that (l) the
overarm throw is not fully developed by grade 7; (2) boys increased
their throwing velocities at a rate of 5 to 8'/sec/yr while girls
increased their velocities at the rate of 2 to 4.5'/sec/yr.; (3) girls
were more stable in maintaining their relative position than the boys;
(4) by grade 7 boys were 5 to 6 years ahead of the girls in developing
the motor components of the cverarm throw; and (5) boys reported
throwing nere often than the girls over the elenentary and middle school
years.

Since the general findings of distance and velocity throwing are
similar, both the distance a ball is thrown and the initial velocity of
a ball after being released can be used as valid measures of the force
imparted to a ball as a result of the throwing motion. However, when
the criterion of administrative feasibility is applied, the throw for

distance is the measure most often used in the school setting.

15

£953

A fourth way to evaluate throwing is through examination of the
netions used to propel the ball. Studies of qualitative skill
characteristics began in the late 19205 and 19305 with Burnside (1927) ,
Gesell (1928), H. M. Halverson (1931), Shirley (1931), McGraw (1935),
Ames (1937), Wild (1938), McCaskill and Wellman (1938), and Gutteridge
(1939) . Early researchers were concerned with defining the age at which
children exhibited certain behaviors or abilities rather than the
quality of their performance. Shirley (1931) examined 25 babies during
the first two years of their life and developed a sequence of intertask
behaviors that lead to walking. McCaskill and Wellman (1938) set up a
series of age-related tasks, each of increasing difficulty. The
children were evaluated on the number of tasks they could successfully
complete and canpared to the standards to see if they equalled, fell
below, or exceeded expectations for their chronological age.

Gutteridge (1939) established criteria for rating the performance
of children on individual netor skills. She examined children during
free play time and evaluated their performance on a 10-point scale. The
scale was divided into four main sections of reactions. At the first
level of the scale, re attanpt was made when an opportunity to perform a
skill was given. At the second level, a habit (skill) was in the
process of formation. In the third section of the scale, the basic
habit was achieved; and in the final sectim a skill could be perforned
with variations. With reference to throwing, Gutteridge concluded that
at two and three years of age re child could throw a ball well, and even
at for years only 20 percent of the children observed threw well,

although many were practicing. Fran five years to five years six nenths

1.6
of age, however, 74 percent of the children could throw well and by the
end of the sane year 85 percent were proficient. Gutteridge also noted
that the range of ratings in throwing was large at all ages. Even at

six years of age peformance in ball throwing ranged fron awkward to
excellent.

The now classic study of mnica Wild (1938) was the first published
report of developnental patterns in performing a skill, in her case the
overhand throw. After extensive analysis of the film records of 32
throwers ranging in age from 2 to 12 years, Wild suggested a four-stage
developrental sequence and tenatively assigned each stage to an age
level. The four stages she reported are described below:

Stage I is characterized by typical anteroposter ior
mevements, of which there is a preliminary incipient stage
with re body nevement. This stage can be assigned to ages
two to three or possibly up to four and is described as
follows: With the reverse arm movement, the trunk extends
with dorsal flexion of ankles and carries the shoulders
back. The trunk then straightens, carrying the shoulders
forward, and flexes forward with plantar flexion of ankles
as the arm swings forward over the shoulder and down in
front. Elbow extension starts early. Movements of body and
arm are alnest entirely in the anteroposterior plane over
feet which ranain in place; the body remains facing the
direction of throw all the tine; the arm is the initiating
factor. There is trunk left rotation toward the end with
the arm's forward reach.

Stage II is marked by the introduction of body and arm
nevements in the horizontal plane, as contrasted to the
anteroposter ior plane, and is assigned to ages three and
one-half to five years. The whole body rotates right, then
left above the feet; the feet remain together in place. The
arm neves either in a high oblique plane above the shoulder
or in a mere horizontal plane, but with a forward downward
follow-though. The elbow is much flexed; it may extend at
once or later. The body changes its orientation and then
reorientates to the throwing direction. The arm is the
initiating factor.

Stage III marks the introduction of stepping. It is
the right foot step-forward throw (right-handed thrower),
assigned to ages five and six. The weight is held back on
the left rear foot as the spine rotates right and extends;
the arm swings obliquely upward over the shoulder to a
retracted position with the elbow much flexed. The forward

l7

nevenents consist of stepping forward with right foot,
unilateral to the throwing arm, with spine left rotation,
early turning of the whole body to a partial left facing and
trunk forward flexion, while the arm swings forward either
in ‘an oblique-above-the-shoulder plane or in a sideways—
around-the-shoulder plane, followed by a forward downward
nevenent of follow-through. Elbow extension does not start
at once. This throw has both anteroposterior and horizontal
features.

Stage IV is the left-foot-step—forward (right-handed)
throw with trunk rotation and horizontal adduction of the
arm in the forward swing. This throw is the mature form and
all boys fran six and one-half years up exhibited it. The
girls had, in nest cases, attained the body and foot
nevements, but showed incompletely developed forms of the
arm nevement. Others showed decided regressions or
retardations. (p. 22)

In his study of throwing accuracy in young children, Hicks (1930)
also evaluated the "style of throwing." Components of the evaluation
included body nevenent, foot action, arm nevenent, path of the hand, use
of force, fixation of center, manner of holding the ball, and
recognition of errors. He did ret believe that any one style of
throwing was always best, maintaining that good throwing is throwing
that produces good results. Though one may speak in general of good or
poor style, highly successful results are often obtained by very
urertredox styles. In Hicks' study, the highest score on the neving
target test was made by a child who employed a peculiar underhanded
upward throw, radically different from the delivery of any other child,
and to the observer a very awkward manner of throwing. However, the
focus of his study was on accuracy and ret on form or distance. No
developrental progression was suggested.

There was little reported research on the developrent of netor
skills until Beach (1950) studied 43 girls and 40 boys, ages 2 through 6
years, to discover if there were, at each age level, discrete patterns

of performance in several fundanental :retor skills. She studied the

18
whole body involvenent. The initial stages of develogrent were
canprised of simple actions of the arm or leg with little ability to
implenent‘ adequately the denenstration trial into the desired netor
act. The advanced stages approached skillful adult performance
characterized by a series of canplex and integrated nevenents. Deach
concluded that the genetic development of netor skills proceeded
according to laws governing physiological maturation. Patterns of
performance increased in canplexity and were defined in terms of stages
of develogrent rather than by chronological age. Boys were
approximately one year in advance of girls and showed greater ability to
neve with an integrated total body pattern.

The work of Wild (1939) was the nest canplete work in the attempt
to identify the characteristics involved in the developnent of a netor
skill for decades until the study of :revenent patterns was revived by
researchers at Michigan State University in the late 19603 and early
705. Dead by Seefeldt, they examined extensive footage of film of
children performing motor skills. Evaluation of the films led to the
application of stage theory for analyzing netor behavior of children.
Stage sequences were reported for the skills of running (Seefeldt,
Reuschlein, & Vogel, 1972), jumping (Seefeldt, Reuschlein, & Vogel,
1972), throwing (Seefeldt, Reuschlein, & Vogel, 1972), catching
(Seefeldt, Reuschlein, 8. Vogel, 1972), kicking (Seefeldt &
Haubenstricker, 1974), repping (Seefeldt & Haubenstricker, 1976a) ,
punting (Seefeldt s. Haubenstricker, 1976b) , striking (Seefeldt &
Haubenstricker, 1976c), and galloping (Sapp, 1980).

The five developnental stages of throwing proposed by Seefeldt and
Haubenstricker (1982) are presented below:

19
Developnental Semence of Throwing

Stage 1. The throwing notion is essentially posterior-
anterior in direction. The feet usually remain
stationary during the throw. Infrequently, the
perforner may step or walk just prior to neving the
ball into position for throwing. There is little or re
trunk rotation in the nest rudimentary pattern at this
stage, but those at the point of transition between
stages one and two may evoke slight trunk rotation in
the follow-through phase. In the typical stage one,
the force for projecting the ball cares fron hip
flexion, shoulder protraction and elbow extension.

Stage 2. The distinctive feature of this stage is the
rotation of the body about an imaginary vertical axis,
with the hips, spine and shoulders rotating as one
unit. The performer may step forward with either an
ipsilateral or contralateral pattern, but the arm is
brought forward in a transverse plane. The netion may
resemble a sling rather than a throw due to the
extended arm position during the course of the throw.

Stage 3. The distinctive pattern in stage three is the
ipsilateral arm-leg action. The ball is placed into a
throwing position above the shoulder by a vertical and
posterior netion of the arm at the tine that the
ipsilateral leg is neving forward. This stage involves
little or re rotation of the spine and hips in
preparation for the throw. The follow-through phase
includes flexion at the hip joint and sore trunk
rotation toward the side opposite the throwing arm.

Stage 4. The nevement is contralateral, with the leg
opposite the throwing arm striding forward as the
throwing arm is moved in a vertical and posterior
direction during the "wind-up" phase. There is little
or no rotation of the hips and spire during the wind-up
phase; thus, the netioi of the trunk and arm closely
resemble those of stages me and three. The stride
forward with the contralateral leg provides for a wide
base of support and greater stability during the force
production phase of the throw.

Stage 5. The shift of weight is entirely to the rear
leg, as it pivots in response to the rotating joints
above it. The "wind-up" phase begins with the throwing
hand neving in a downward arc and then backward as the
opposite leg neves forward. Concurrently the hip and
spine rotate into position for forceful de-rotation.

As the contralateral foot strikes the surface the hips,
spire and shoulder begin de-rotating in sequence. The
contralateral leg begins to extend at the knee as the
shoulder protracts, the humerus rotates, and the elbow

20

extends, thus providing an equal and opposite reaction

to the throwing arm. The opposite arm also meves

forcefully toward the body to assist in the equal and

opposite reaction to the throwing arm.

During the 19703, researchers at the University of Wisconsin-
Madison also focused o1 developmental changes to describe the transition
of children's metor skills from immature to mature performance. The
Wisconsin approach to developmental sequencing was led by Lolas
Halverson and MaryAnn Roberton. They believed that if there were stages
(steps) in metor task developrent, perhaps these stages occurred only in
the conponents of the skill rather than in the total body
configuration. For example, Roberton (1982) divided the skill of
throwing into five different components and established a developmental
sequence for each of them. The components which she described as making
up the overarm throw are (l) trunk action; (2) preparatory arm
backswing: (3) humerus actioi; (4) forearm action; and (5) action of the
feet. Thus, in throwing, a child might meve ahead a stage in trunk
action while retaining the same stage of arm action. Another child
might stay at the same trunk action stage but progress in arm action.
The developnental stages for each of the components of throwing proposed

by Robertoi and Halverson (1984) are presented below.

DevelopuitalSegiereeforTrmMrActioiin
nirowingmdStrikirgfoerrce

Step 1. No trunk actioi or forward-backward
movements. Oily the arm is active in force
production. Sometimes, the forward thrust of the arm
pulls the trunk into a passive left rotation (assuming
a right-handed throw), but re twist-up precedes that

. action. If trunk action occurs, it accompanies the
forward thrust of the arm by flexing forward at the
hips. Preparatory extension sometimes precedes forward
hip flexion.

Step 2. Upper trunk rotation or total trunk "block"
rotation. The spine and pelvis both rotate away from

21

the intended line of flight and then simultaneously
begin forward rotation, acting as a unit or "block."
Occasionally, only the upper spine twists away, then
toward the direction of force. The pelvis, then,
remains fixed, facing the line of flight, or joins the
rotary movement after forward spinal rotatioi has

begun.

Step 3. Differentiated rotation. The pelvis precedes
the upper spine in initiating forward rotation. The
child twists away from the intended line of ball flight
and, then, begins forward rotation with the pelvis
while the upper spire is still twisting away.

Developental Sequereas for Backswing.
Exams, and Ebrearm Action
in the OverarnThrow for Fbrce

Preparatory arm backswing component

Step 1. No Backswing. The ball-in-the-hand neves
directly forward to release from the arm's original
position when the hand first grasped the ball.

Step 2. Elbow and humeral flexion. The ball neves
away from the intended line of flight to a position
behind or alongside the head by upward flexion of the
humerus and concomitant elbow flexion.

Step 3. Circular, upward backswing. The ball neves
away from the intended line of flight to a position
behind the head via a circular overhead mevement with
elbow extended, or an oblique swing back, or a vertical
lift from the hip.

Step 4. Circular, downward backswing. The ball neves
away from the intended line of flight to a position
behind the head via a circular, down and back metion,
which carries the hand below the waist.

Humerus (upper arm) action component during
forward swing

Step 1. Humerus oblique. The humerus meves forward to
ball release in a plane that intersects the trunk
obliquely above or below the l'erizontal line of the
shoulders. Occasionally, during the backswing, the
humerus is placed at a right angle to the trunk, with
the elbow pointing toward the target. It maintains
this fixed position during the throw.

Step 2. Hurerus aligned but independent. The humerus
neves forward to ball release in a plane horizontally
aligned with the shoulder, forming a right angle

22

between humerus and trunk. By the time the shoulders
(upper spine) reach front facing, the humerus (elbow)
has meved independently ahead of the outline of the

- body. (as seen from the side) via horizontal adduction
at the shoulder.

Step 3. Humerus lags. The humerus moves forward to
ball release horizontally aligned, but at the moment
the shoulders (upper spine) reach front facing, the
humerus remains within the outline of the body (as seen
from the side). No terizontal adduction of the humerus
occurs before front facing.

Forearm action component during forward swing

Step 1. No forearm lag. The forearm and ball meve
steadily forward to ball release throughout the
throwing action.

Step 2. Forearm lag. The forearm and ball appear to
'lag,' i.e., to remain stationary behind the child or
to meve downward or backward in relation to him/her.
The lagging forearm reaches its farthest point back,
deepest point down or last stationary point before the
shoulders (upper spine) reach front facing.

Step 3. Delayed forearm lag. The lagging forearm
delays reaching its final point of lag until the moment
of front facing.

DevelmtalseqaereeforActimcftteFeet in
Fbrceful Throwing aid Striking

Step 1. No step. The child throws from the initial
foot position.

Step 2. Halelateral step. The child steps with the
foot on the same side as the throwing hand.

Step 3. Oontralateral, short step. The child steps
with the foot or the opposite side from the throwing
hand.

Step 4. Oontralateral, long step. The child steps

with the opposite foot a distance of over half the
child's standing height.

Ibberton (1978) uses the component approach to stage theory because

her interpretation of stages, based on the writing of Piagetian

23
theorists Flavell and Wohlwill (1969) , Inhelder (1971), and Pinard and

Laurendeau (1969), requires that all of the different body parts
progress'at the same time. She also prefers to use the term "steps"
rather than ”stages" in her description of developmental levels.
Seefeldt and Haubenstricker (1982) agree with Roberton that all of the
subroutines described in their stages do not advance as an indivisible
unit. However, they have found sufficient cohesion between certain
combinations of the subroutines that listing them as a 'stage' appeals
to them as the least complicated way to describe a particular
developmental task. Haubenstricker and Seefeldt (1986) also write that
the medel does ret require simultaneous change in the movement patterns
of all body parts from one stage to the next. However, the total body
mevement configuration does change and is clearly distinguishable from
those in adjacent stages. Configurations that are not in full
compliance with one of the described stages are considered to be in
transition between stages.
Factors Affecting Throwing Distance

The ability to throw well is dependent on a variety of factors
including age, gender, strength, body size, and throwing form.
593

Of all the characteristics that subjects bring to studies of
throwing, age has the nest profound influence. For both boys and girls,
the older children become the farther they can throw (Hanson, 1965;
Hardin & Garcia, 1982; Keogh, 1965; Nichols, 1971). Improvement in
throwing distance was seen in six-menth intervals for boys and girls as
young as three to five years old (Frederick, 1977; Johnson, 1960) , and

for girls through the 10th grade (Vincent, 1968) .

24

m _

Another characteristic associated with performance on the distance
throw is the gender of the subjects. All of the studies which examined
the distance throw or initial ball velocity in children found a
significant difference in performance between boys and girls. Thomas
and French (1985) have done an extensive meta analysis on 64 studies
involving throwing and other motor tasks. They concluded that for most
'motor skills the performances of elementary age boys and girls are not
significantly different. However, for the skill of throwing, there was
a significant difference between boys and girls even at young ages.
Jehnson (1977), in testing 48 boys and girls between the ages of three
and six years, found that boys score significantly higher than girls on
:measures of throwing ability. Thus, while gender is not a significant
factor in most motor skills at young ages, it is in throwing performance
and must be considered in studies of throwing involving children.
Streggth

The selection of a strength test or battery of tests to
discriminate throwing ability is clouded by two factors; the specificity
of strength, and the interrelationshiptof different body parts during
the throwing motion. A.strength test is very specific in nature and
must be chosen according to the muscle groups to be tested and the type
of strength that is to be measured. It is also difficult to select
specific muscle groups for strength testing because most of the major
muscle groips of the body are involved in throwing. The arms apply nest
of the force used to propel the ball, but a substantial ameunt of force
is generated by the torso, the hips and the legs (Tbyoshima et a1.,

1974). If the limitations of selecting a strength test or battery of

25
tests are recognized, some discriminatory power may exist and be helpful
in analyzing performance. .

Strength is a major factor in the performance of many motor skills
but rarely has been a good predictor for success in a skill by itself
(Johnson & Nelson, 1979). An example of specificity is furnished by
Berger (1962) who in a study of strength found the relationship between
static or isoretric stength and dynamic or isotmic strength measures to
be low. He found that changes in muscle strength resulting from.dynamic
muscle training were more accurately measured by a dynamic strength test
than by a static strength test. Simri (1974) expanded on Berger's
specificity concept of strength and reports that strength is divided
into three independent factors of physical fitness: namely, dynamic
strength, explosive strength or muscular power, and static strength.
Jehnson and Nelson (1979) concluded that strength is specific, therefore
it is meaningless to use one type of strength to suggest abilities in
skills demanding another type of strength.

There are two ways investigators have tried to show a relationship
between strength and throwing ability. One way is to use some type of
(overload training program on the subjects and then test if the increased
strength also increases throwing velocity. The other way is to
determine the degree of relationship between specific strength tests and
throwing performance.

Programs for strength building in subjects of high school age and
older may facilitate an increase in throwing velocity. Bagonzi (1979)
studied 48 high school baseball candidates, ranging in age from 15 to 19
years. The subjects used a variety of overload training techniques

including weighted baseballs, free weight training, simulative isometric

26
exercises and cambinatims of the three. The training program lasted
for 18 weeks. At the end of the 18 weeks the results showed that
overload-training improved velocity and accuracy when compared with a
control group. Rowlands (1962) assigned six college baseball players to
an experimental group who were placed on a five-week weight training
program and six players to a control group. A cable tension test was
used to measure strength/loss of the shoulder medial rotator muscle.
Throwing power was computed from the velocity of baseballs thrown over a
100-foot distance in a horizontal trajectory at a target. The program
significantly improved both strength and throwing power. Sullivan
(1970) had 48 university students participate in a strength training
program four times a week for six weeks to determine its effect on
baseball throwing velocity. The training program included a weight
training group and a simulative training group which used a wall pulley
to simulate the baseball throwing metion. Both programs significantly
increased velocity, but the weight training program was significantly
better than the simulative training program.

However, not all studies reported gains in throwing velocity
resulting from a strength building program. Straub (1966) randomly
assigned 48 high school boys to participate in a six-week training
program using weighted balls to improve speed and accuracy. Overload
training had re differential effect on either high or low velocity
throwers. Control subjects who trained with regulation balls and
emphasized speed and accuracy threw as fast and as accurately as
subjects traired with progressive overload with an emphasis on speed or
accuracy. Using a weighted ball produced no long range improvement in

throwing speed or accuracy.

27

Barrow (1960) subjected the antagonistic muscles of 43 eighth grade
boys to a six-week resistive exercise program for five minutes a day.
The training program increased muscle strength and throwing accuracy,
but the mean gain for throwing distance was non-significant.

Hardison (197]) trairned seventh and eighth grade girls daily for
four weeks in buddy resistance exercises to increase arm-shoulder
strength. There was an increase in arm-shoulder strength, but the
increase in arm-shoulder strength did ret result in an improvement in
the distance girls could throw a softball.

Williams (1935) studied 14 college baseball players of which seven
participated in a weight training program three times a week for fonr
weeks. The other seven combined long distance throwing two times a week
with a weight training program. Williams found re significant
differences from the pre-test to the post test for either group. He
concluded that long distance throwing and weight training did ret
increase throwing velocity.

The for studies that did not produce significant changes in
throwing performance need to be examined as to the type of strerngth
overload used or the length of time the strength training took place.
Straub (1966) provided re evidence that his subjects actually increased
their strength using weighted balls. Barrow's (1960) subjects trained
for only five minutes a day for six weeks while Hardison's (1971) and
Williams' (1985) subjects only trained for four weeks. It is
questionable whether a strength training program of only five minutes a
day or one with a duration of only for weeks is of sufficient intensity
or duration to produce an increase in strerngth. The four studies of

Barrow (1960), Harrison (1971), Staub (1966), and Williams (1985)

28

provide evidence that strength training programs of short duration do
not produce a significant difference in throwing distance or velocity.
However, because of the questions about the appropriateness of the
length of their strength training programs it cannnot be concluded that
strength training does not affect throwing distance or velocity.

While comprehensive resistance training may produce greater
throwing velocity and distance, measures of strength are only moderately
correlated with throwing velocity or distance. Sullivan (1970) studied
48 university students enrolled in an experimental conditioning
course. He measured all subjects in throwing velocity, grip strength,
wrist flexion strength and medial arm rotation strength before and after
a six-week (unspecified) training program. He obtained a correlation of
.42 for the relationship between strength and velocity before strength
training, and .31 for the relationship after training. The hypothesis
that strength is an important component of throwing velocity was not
supported. Espenschade (1940) found slightly higher correlations
between grip strength and throwing distance when she studied 80 girls
and 85 boys over a two-and-a-half-year period beginning in the eighth
grade. The girls' correlation coefficient was computed over two years;
the first year it was .43 and the second year it was .60. The
correlation coefficients for boys, computed for three years were .53,
.71, and .41, respectively.

Other investigators, using regression analysis, have found strength
related to throwing ability. Brumrield (1969) tested 35 college
frestmen wonen on the overarm throw for accuracy and distance. The
variables studied were shoulder flexibility, shoulder strength, speed of

arm mevement, age, height, weight, physical education background,

29

athletic background, gender, and number and sex of children in the
family. The variables which showed significant relationships to
distarnce throwing were shoulder strength and athletic background.
Richardson (1976) tested 31 varsity high school baseball players to
determine if grip strength, range of wrist flexion, and length of
throwing arm were significantly related to throwing velocity. He found
that grip strength was significantly correlated to throwing velocity but
that wrist flexion and length of throwing arm were not. Richardson used
the stepwise multiple linear regession to conclude that throwing
velocity can be mederately predicted (shared variance of 36%) from grip
strength measurement.

At the elite level of performance, Pedegana, arnd associates (1982)
tested eight professional baseball players on the Cybex Dynamoneter and
determined that the wrist and elbow extensors have direct relationships
with throwing velocity. These investigators presented a positive view
of the relationship between strength and throwing velocity. They
reported:

"This study shows a positive correlation between
the strength of certain upper extremity muscle groups
arnd throwirng speed, a relationship which has rot been
clearly demonstrated in past research. The results
further show lack of perfect statistical correlation
between the strerngth of arm movenents and throwing
speed; however since it has been demonstrated by
Tbyoshima et a1. (1974) that only 53.1% of throwing
speed is the contribution of arm action and that
throwing is a conplex act involving all of the body
parts the lack of correlation is explicable on that
basis alone.

While this study has found, by isolatirng and
testing each upper extremity movement and statistically
correlating the strength of each movement with throwing
speed, that two movements, wrist extension and elbow
extension, appear to have more direct relationships
with throwing speed than do the others, it is also
apparent that these relationships are complex arnd
probably interactive. However, if the exact nature of

30

these relationships are not altogether clear, they

clearly exist and imply that modification of the

strength of certain muscle groups will likely modify

throwing speed" (p. 354).

How does strength affect throwing velocity or distance? Before the
answer to this question can clearly be obtained, the complexity of the
throwing motion must be addressed. Because one person is stronger than
arnother does not guarantee that the stronger individual will be able to
throw a ball farther. If the stronger person has an inferior throwing
pattern, the weaker person could compensate for a lack of strength by
using a more efficient pattern and the result could equal or exceed that
of the stronger person. However, if two people had bionechanically
equal throwing patterns, it would seem logical that the stronger person
would have the higher velocity. This logic explains Pedagana et al.’s
(1982) enthusiasm regarding their findirngs. They used a very small
sample of select elite players. The use of elite players would make it
easier to obtain a high correlation between a strength item and throwing
velocity because the variety of throwing patterns is reduced
dramatically. One could assume that the selection process for
professional baseball pitchers would eliminate the pitchers who do not
have a mature pitching form. Having all biotechanically proficient
throwers greatly limits the effect of throwing patterns since they all
are mature and efficient, thus allowirng other factors such as strength
to exert greater influence on throwing velocity. In contrast, Sullivan
(1970), who used university students, and Espenchade (1940) , who used
eighth grade boys and girls, did not have subjects who belonged to a
baseball team. They had subjects with a wider variety of throwing

skill, thus the corponent of form could carprise a greater portion of

the variability of the velocity scores. In other words, the correlation

31

between strength and velocity was smaller or not significant because the
factor of form was not controlled. -

The contribution and relationship of different body parts to
throwing performance were examined by Toyoshima and his associates
(1974) . They measured the throwing velocity of seven adult males with
five different weights of balls while limiting different parts of body
movement. They found that by immobilizing various body parts the
resulting velocity of the throw was subsequently reduced. The
percentage of reduction was fournd to hold constant throughout the
different weights of the balls. The mean velocity of the seven subjects
using an overhand throw with a step was considered 100 percent. When
the subjects were not allowed to take a step during their throw they
threw the ball 84 percent as fast. When their lower body was
immobilized, the velocity was 63.5 percent of the normal throw.
Immobilization of the upper body reduced the velocity to 53.1 percent.
When the upper arm was placed on the arm of a chair and imnoblized, the
velocity of the throw was reduced to 42.6 percent of the rermal throw.
Based on their findings, Toyoshima and his associates accorded 53.1
percent of the total throwing velocity to arm action and the renainirng
46.9 percent to the step, hip rotation, and trunk rotation. These
investigators reported that their results match those of Broer (1969)
who, in a study of two women throwing a tennis ball, concluded that
approximately 50 percent of the velocity of the overhand throw resulted
from the step and body rotation, while the remainder came from shoulder,
elbow, wrist, and finger action.

Toyoshima et a1. (1974) also measured the angular velocities of the

forearm during a normal throw and forearm throw. In normal throwing,

32

the results were 31.14 radians per second; in forearm throwing, they
were 15.57 radians per second. They concluded:

"It was verified, therefore, that the extension of
the elbow joint was performed at a higher speed in
rermal throwing than in the activity of maximal
voluntary effort using the elbow joint. This indicated
that the forearm was being swung like a whip by the
rotary actions of other parts of the body, such as hip,
trunk, and shoulder. Moreover, the greater the radius
of rotation, the greater will be the production of
speed in these ball-throwing situations.

"In summary, it appeared that the contribution of
the extention of the elbow joint to the speed of the
ball did ret result only from the power caused by
voluntary muscular contraction of the triceps brachii,
but also from the torque produced by rotation of the
body. The radius of rotation of the forearm in the
flipping motion was increased by body rotation, thus
resulting in a greater speed. It therefore would seem
that the contribution of the elbow joint in normal
throwirng performance may be less than the 42.6%
calculated in this study; as a matter of fact, a larger
percentage of the velocity of the thrown ball resulted
from the body rotation. Furthermore, it is interesting
to rote that the rapid arm action acts not on the basis
of conscious muscle contraction, but from the physical
phenorenon and reflection of the neuronuscular system"
(p. 174).

m size

In general skill testing, investigators have found low
relationships between the performance of fundamental skills and height
and weight. Keogh arnd Sudgen (1985) wrote:

"The development of maximal performance has been
studied primarily in terms of structural changes,
except for the development of muscle in relation to
functional strength. The principal firnding is that
structural differences areng children of the sate
chronological age cannot predict maximal mevenent
performarnce as defined here. Few of the measures of
height, body proportions, body conposition, and
physique for children at age 6 correlate substantially
with their maximal performance scores for runnirng,
jumping, throwing, and similar movenent tasks. Malina
and Rarick (1973) , in a critical review of related
literature, concluded that 'performance in motor skills
during elenentary school ages is largely unaffected by

33

body build and constitutional factors, except at the
extremes of the continuum' (p. 150, italics added).
They acknowledge that physique and related structural
characteristics can limit or enhance performance but
are not good predictors of maximal performance scores
(also see Malina, 1975)" (p. 254).

"Despite the generally neutral findings, structural
differences probably are important to maximal
performance for the movement skills we are reviewing
here. The problem seems to be in matching specific
structural characteristics to specific task
requirements. Fer example, it seems too much to expect
that a general measure of physique will predict a
specific movement achievement. But the general
physique category of mescmorphy does combine the body
proportions and composition that should contribute to
better performance in many play-game and athletic
skills. We should analyze task requirements more
carefully to find specific aspects of physique that
contribute to maximal performance. we also need to
include functional characteristics, which may be more
important. Asmussen (1973) stressed the functional use
of strength. This means that we must find ways to
assess the coordination of movement control. Some
individuals probably are better equipped than others
are to coordinate movement parts, in summating force,
although they can make great improvement with good
instruction. Our comments here do not change our
conclusion that structural differences among children
of the same chronological age do not predict maximal
performance" (pp. 257-258).

 

 

The relationship of biological variables to throwing performance
has been investigated. Johnson (1960) administered a number of
performance tests on fundamental skills to 2,459 boys and 2,195 girls in
Grades 1 to 6. In addition to the skill testing, he collected data on
the ages, heights, and weights of the subjects. He concluded that the
age, height and weight of boys and girls in grades 1 to 6 appear to have
low relationships with performance on tests of fundamental skills.
Frederick (1977) evaluated the throwing ability of 3, 4, and 5 year-old
black and white girls and boys, but found no Signficant relationships
between weight, height and throwing performance level. Espenschade

(1940) studied 80 girls and 85 boys for a two-and-a—half-year period

34

beginning in the eighth grade. She evaluated the relationship between
weight, height, stem length/height, and stem breath/length with the
distance throw, but all, the relationships were non-significant.

Nelson, Thomas, Nelson, & Abraham, (1986) examined gender
differences in children's throwing performance, including biological and
environmental variables. They studied 100 kindergarten children, 48
girls and 52 boys. Throwing performance was evaluated by throwing a
beanbag for distance. The form of the throw was evaluated using the
scales developed by Roberton (1984) for the trunk and feet. Biological
factors measured were height, weight, body mass index, ponderal index,
skinfolds (triceps, subscapilar, suprailiac, and calf), body diameters
(biacromial, bi-iliac, biepicorndylar width of the humerous, and
bicondylar width of the femur), girths (biceps and calf), arm length,
forearm length, and somatotype. Environmental factors consisted of
whether the child had older siblings, an adult ma1e(s) in the haze, and
whether the child played with other children.

In a forward stepwise regression only two variables significantly
predicted throwing performance for boys, estimated leg muscle and
shoulder/hip dianeter ratio. However, they accounted for only 18
percent of the variance in throwing. For the girls, two biological
(estimated arm muscle and shoulder/hip dianeter ratio) and two
envirormental variables (older brother and playirng with other children)
were significant predictors of throwing performance. These variables
accounted for 48 percent of the throwing variarnce. Four variables that
predicted throwirng distance regardless of gender were joint diameters,
shoulder/hip ratio, sum of skinfolds, and playing with other children.

The lirear composite accounted for 41 percent of the variance (R2) .

35

Nelson and associates (1986) then employed an ANCOVA using gender
as an independent variable, throwing for distance as a dependent
variable and the three biological variables (joint diameters,
shoulder/hip'ratio, and sum of skinfolds) previously identified as
covariates. The results of this analysis indicated the overall test of
the model was insignificant. When biological factors were not taken
into account, girls' performance was 57 percent of that of the boys',
but when they were considered, girls' performance was 69 percent of that
of the boys'. Boys differed from.girls in certain specific growth
characteristics-boys had significantly greater diameters for the elbow
and knee, and more estimated arm muscle while the girls were
significantly more endomorphic and had a greater sum of skinfolds,
indicating more fat.

The results of this study support the prediction of Thomas and
French (1985) that prior to puberty, throwing is a task in which some of
the small differences in specific growth characteristics between boys
and girls influence performance. Having a more robust skeleton (larger
joint diameters), a greater shoulder/hip ratio, a smaller sum of
skinfolds, and more estimated arm muscle positively influernces
performance in throwirng for distance. This does rot mean that
biological variables are the major contributors to performance
differences in throwing, they are not. However, they do appear to make
some contributions.

Results of other investigations regarding the influence of
biological factors on throwing performance are mixed. Sanders (1977),
when studying college baseball players, found that age, height, weight,

and length of arm span were not significant as predictors of baseball

36

throwirng velocity. Kang (1982) studied 8th grade boys and girls and
determined that band size had re significant effect upon throwing
performance. Bowne (1960) selected 42 high school girls, one-third of
whom were average throwing skill (v=42-46'/sec) , one-third of medium
skill (v=50-54'/sec), and one-third highly skilled (v=S8-62'/sec). She
took numerois structural measures including lengths for the hand,
shoulder, pelvis, foot, stature, sitting height, trunk height, leg,
upper arm, forearm, and thigh. She concluded that it was not possible
to distinguish between velocity groups on the basis of the scores for
any structure length variable. Richardson (1976) tested 31 varsity high
school baseball players an arm length, wrist flexion, and throwing
velocity. Arm lerngth and wrist flexion were rot significantly
correlated with throwing velocity.

Seils (1951) tested first, second, and third grade boys and girls
on a variety of motor skills and physical growth variables. Very low
insignificant relationships between age, height, weight and throwing
ability were found. However, a correlation of .42 for the boys and .38
for the girls was obtained between throwing performance and a measure of
skeletal maturity. Hoff (1985) studied tl'e throwing ability of first
and forth grade boys and girls as represented by velocity, distance,
accuracy, and quality of throw and their relationship to the structural-
maturational variables of height, weight, arm length, and subcutaneous
fat. Arm lerngth was predictive of the throwing ability of first grade
girls. Arm length and weight were predictive of the throwing ability of
fourth grade girls. All of the structural and maturational variables
contributed to the prediction of throwing distance for both first and

fourth grade boys.

37

Mahmoud (1979) evaluated girls and boys ages four to six years to
determine the relationship between throwing distance scores and some
selected temporal and kinematic factors as well as certain
anthroporetric measures, and to determine the best combination of
factors for predicting throwing distance and improvement based upon
distance gain. The following variables were measured: height, weight,
upper arm, forearm, and hand and forearm length. Film analysis yielded
force time, angle of ball release, velocity of the ball at release, as
well as segment orientation and stride length. Mahmoud reported
significant relationships between the distance a ball was thrown and
sex, age, height, hand and forearm length, angle of release, velocity of
the ball, and wrist angle on the pre-test. On the post-test a
significance relationshipnwas reported between the distance a ball was
thrown and sex, angle of release, stride length, velocity of the ball
and improvement; but the anthropometric measures of height, and hand and
forearm length were not significant.

The inconsistencies associated with the significance of structural
variables with throwing distance leads to»the conclusion that structural
variables alone are not sufficient to differentiate between good and
poor throwers. This is especially true of subjects in grade seven and
older. Some structual variables of young children (age 4 to grade 4)
have been found to have predictive or significant structural
relationships with throwing ability. A possible explanation is that
structual variables are indicative of the total maturation of the young

subjects resulting in greater strength and a more mature nervous system.

38

Form

 

Biamechanical factors found to be associated with a better throw
are an increase in the range of trunk rotation (Bowne, 1960; Ekern,
1970; and Singer, 1961) ; a decrease in the medial rotation of the arm
(Bowne, 1960; and Ekern, 1970); better ratings of form for trunk and
foot action (Nelson et a1., 1986); rapid sequential acceleration and
deceleration of trunk and arm segments prior to release (Atwater, 1970;
arnd Deutsch, 1969) an increase in stride length (Deutsch, 1969; Ekern,
1970; arnd Schutzler, 1980); a greater forward flexion at the hip joint
at the point of release (Ekern, 1970; and Lyon, 1961) and a greater
range of movement in contributing joints (Singer, 1961).

An increased range of trunk rotation was found to be associated
with a better throw when Bowne (1960) selected 42 high school girls,
one-third of whom were average throwing skill, one-third of medium
skill, and one-third highly skilled. In evaluating the subjects' form
she reticed they increased velocity when there was an increase in the
range of trunk rotation. Singer (1961) in her four-year longitudinal
study of four girls also reported that subjects with greater throwing
velocity showed a greater rotation of their torso. Ekern (1970)
compared the throwing performances of boys and girls (2 each) in grades
2, 4, and 6 on the basis of selected measures taken from film records of
the overarm throw with a particular emphasis on the preparatory phase of
the throw. Selected body levers were studied noting the time of
entrance, duration of action, morent arm lengths and ranges of
meverent. The position and direction of movement among selected body
parts, the path of the ball, the center of gravity changes of the body

and the bases of support were investigated. The results of this study

 

39

irndicate that the boys in the preparatory phase used greater reverse
spinal rotation than the girls. They also used pelvic and spinal
reverse and forward rotations more effectively.

Bowne (1960), in her study of 42 high school girls grouped
according to their throwing velocity, noticed they irncreased velocity
when there was a decrease in the medial rotation of the arm. Ekern
(1970) in her comparison of the throwing performances of boys and girls
(2 each) in grades 2, 4, arnd 6 also reported a decrease in time used for
medial rotation of the humerus in boys compared to girls and when older
students were compared to younger ones.

Roberton's (1984) components of trunk and foot action
(corresponding with trunk rotation and stride length) were also used to
compare good throwers with poor ones. Nelson et a1. (1986) evaluated
the throwing differences in boys and girls at 5 years of age. One of
the means by which they evaluated the throwing performance was
determined by components in trunk and foot action as proposed by
Roberton. The correlations were moderately high between distance thrown
and ratings of form for trunk rotation (.67) and foot action (.64).

Another means for evaluating good form is the determination of the
sequential acceleration and declaration of trunk and arm segments prior
to release. Atwater (1970) examined three groups of subjects (5 in each
groip): skilled men (college baseball players v= 110-125'/sec) ,
skilled women (v=70-80'/sec), average wolen (v=40—50'/sec) . She fonnd
that the subjects with the fastest ball velocity at release were those
with the most rapid sequential acceleration and deceleration of trunk
and arm segments prior to release. She concluded that while the range

and sequence of joint actions was similar for all skilled men and

40

skilled wonen, the rate at which these actions took place was faster in
men than women. Deutsch (1969) took three groups of five women
representing expert, average, and poor levels of throwing to determine
the mechanical and muscle action differences underlying skill in
throwing a baseball for speed both overhand and underhand. Stride
length, horizontal arnd vertical displacement of the right hip and
shoulder, body lean arnd right arm angle at the left foot down and last
ball contact positions, arnd ball speed were derived graphically from
successive movie frames. Overhand ball speed correlated significantly
with stride length and the differernce in right arm angle during the
throw. Throwing fast apparently depended on successive and concentrated
hip, shoulder, and hand (ball) acceleration of sufficient intensity for
the reaction to produce regative acceleration in the preceeding segment.

The advantage of an increase in stride length was noted by Deutsch
(1969) in a study of college women when overhand ball speed correlated
significantly with stride length. In a study of boys and girls in
grades 2, 4, and 6, Ekern (1970) noted that both boys and older children
used larger working bases with more advantageous foot placerent.
Schutzler (1980) compared stride length to velocity between groups of
major league, triple A arnd college pitchers. He found velocity was
significantly different between groups but stride length was rot. But
when the 15 fastest pitchers were compared with the 15 slowest pitchers
irregardless of their group, the ratio of stride length/height was
significantly greater for the faster pitchers.

A greater forward flexion at the hip joint at the point of release
was noted to be a characteristic of both highly skilled throwers and

developing throwers. Lyon (1961) compared one major league pitcher with

41

seven University of Wisconsin pitchers and found that the, subjects with
the greatest forward flexion at the hip joint had the highest
velocities. Ekern (1970) in her study of second, fourth, and sixth
grade boys and girls also noted that the boys and older students
exhibited greater forward trunk inclination.

Proper form is not guaranteed as a function of age or maturation.
Halverson, Ibberton and Langendorfer (1982) reported that the overarm
throw was not fully developed in male or female seventh graders. Leme
(1973) tested mature females and discovered that similar movement
patterns are used in the development of throwirng regardless of age.
Less than complete developnent of throwing exists in sore adults. She
also noted that one subject regressed in throwing velocity while
learnirng to use a rew, more developed, throwing pattern. Dusenberry
(1952) also reticed sore decreases in distance due to changes in the
manner of throwing.

Instruct ion

 

Providirng instruction in throwing has produced mixed results,
soretimes even in the same study. Hoffman (1969) taught first through
third grade boys and girls 30 minutes a day for a six-week period with
many opportunities for individual help. Significant improvenent was
reported for first grade girls and second grade boys. Greater, but ren-
significant, improvenent was reported for second grade girls and third
grade boys and girls. No gain was was reported for the first grade
boys. Luedke (1980) instructed secornd and fourth grade students in two
different programs. They included basic instruction and instruction

emphasizing increasing range of motion (IRM) . The 1R4 instruction

42

signficantly improved the velocity scores of second graders, but not
fourth graders. -

Teaching has produced increases in throwing ability in preschool
children, in children in grades three through seven, and in college
women. Dusenberry (1952) paired fifty-six 3, 4, 5, 6, and 7—year-ol-ds
into control annd treatment groups on the basis of age, gender, race, arnd
the average of five throws. The treatment groups met 2 times a week for
3 weeks. He concluded that at a seven per cent level of signficance,
the specific training irncreased learninng over maturation and gym
practice. He also noted that boys improved more than the girls and that
the older children (5-6) profitted more from the instruction than did
the younger children (3-4). Mahmoud (1979) reported that preschool
children exhibited significant changes in throwing distance, form and
stride length from pre- to post-test after instruction for 15 minutes a
day, 3 days a week, for 4 weeks. Potter (1963) trained college women by
two methods. Ore group practiced throwing arnd the other was trained
isonetrically. The women trained 3 times a week for 5 weeks. The
throwirng group showed a significant mean gain in throwing distance over
the isonetric group.

On the other hand, several investigations failed to demonstrate
impovenent in throwing due to teaching for subjects ranging from
kindergarten through eighth grade. Halverson et a1. (1977) gave
kindergarten students 120 minutes of guided practice over an eight-week
period annd reported that throwirng velocity did not significantly
change. Roberton et a1. (1979) also evaluated these sane children as
second grade boys and girls and reported that there was no evidernce of

long term effects due to the instructional program they received in

43

kindergarten. Nichols (1971) instructed a group of four to seven year
olds, five minutes a day, two days a week, for six weeks and reported no
significant difference between the distance throw of this group and a
control group which had regular free time. Deatrick (1977) divided
seventh grade girls into three groups for a fivedweek training period.
Group I trained on the Apollo Exercisor, while group II trained by
practicing softball throwing and group III maintained its regular
physical education program. She reported a non-significant relationship
between the training programs. Dohrman (1964) conducted a study with
eight-year-old boys and girls. He divided 100 students into two
treatment groups. Each group was to receive special training in
throwirng and kicking. One group received the training during the fall
while the second group participated in the training during the spring.
He concluded that throwing and kickirng programs, in addition to regular
physical education, do not result in greater improvements in throwing or
kicking ability.

In an effort to determine the effectiveness of various programs of
motor skill instruction for three- and four-year-old children, Miller
(1978) compared four different programs. She compared a traditional
style of teaching, a teaching program which also included parental help,
a free play program in which the children used the sane equipnent but
were not exposed to any formal program of instruction, and a control
group which had no access to the equipment but was used to control for
the effects of maturation. Miller found that the traditional and parent
instructional groups performed significantly better than the free play

group, indicating that programs of directed practice and instruction

44 ‘
were more effective than programs of free play in increasing fundamental
skill level of young children.

Glassow and associates' (1965) two—year, longitudinal study
included overhand-throw ball velocities for all children and filmed
content measures for a randomly selected group from three grade
levels. First-, third-, and fifth-grade records were obtained following
a year in a special instructional program. Second-, fourth-, and sixth-
grade records on the sane children were collected after a second year in
the program. The authors reported no consistent differences between
gains in overhand ball velocities for the experimental groups and those
made by previous children in school used for control comparison. When
throwing form was considered, however, the children did not show steady
improvement across ages in all groups. For instannce, after two years in
the special instructional program, a higher percentage of second- and
fonrth-grade girls used a ninety degree preparatory turn than the older
third- and fifth-grade girls did after one year in the program. Also, a
greater percentage of second-grade boys and girls took a forward step
and had a longer stride than the older third-grade children. Apparently
the special program did have an influence on the throwing ability by
improving the throwing form of the experimental groups, a finding not
evident when only the velocity achievement scores were examined.

In summary, instruction can play an important role in improving
throwing ability. However, because of the complexity of throwing, the
readiness and the motivation of the students, and the length and quality
of the instruction, it is important to realize that not all instruction
will have a positive influence on every aspect of throwing ability every

time .

45

Justification of Testing Procedures

Strength
The hand grip test was chosen because it is not directly related to

body size and is the best single indicator of body strength. In
addition, grip stength (has been found to be related to throwing
performance (Espenschade, 1940; Richardson, 1976). On the other hand,
grip strength is a measure of static strength rather than the explosive
strength or muscular power used in the throw for distance. Thus a
second test, one measuring dynamic strength was also used. The best
test measuring extensor strength of the arm and shoulder muscles was the
push-up. Shoulder strength was found to be significantly related to
throwing deistance in college freshmen women (Brunnfield, 1969) .
Grade verses Age

Hanson (1965) evaluated boys and girls from grades 1 through 6 on
age, height, weight, intelligernce and numerous motor performance
tests. She examined the interrelationships between the different
variables and concluded that the within grade performannce of children
was consistent enough to be evaluated as a group. She found low single
variable interrelationships between age, height, weight, intelligence
and motor performance tests. Relationships derived from multiple
correlations (age, weight, height, and intelligence) (age, height, and
weight) were significantly low to negate the need for classification
indices to evaluate the motor performance of children within a single
grade level.
Modeligg

The formation of a proper modeling procedure for the subjects was
primarily motivated by Bandura's (1977) theory of self-efficacy.

46

Bandura's theory of self-efficacy is defined as the strength of a
person's conviction that he or she can successfully execute a behavior
required to produce a certain outcone. In this case, self-efficacy
would entail that the children would have enough confidence in their
ability to throw effectively that they would give their best effort in
their distarnce throw. In the developnent of self-efficacy, Feltz,
Landers, and Raeder (1979) examined the effectiveness of participant,
live, and videotape modeling on enhancing self-efficacy in the learning
of a high-avoidance springboard-diving task. They found that the
participant-modeling treatment produced more successful dives and
stronger expectations of personal efficacy than either the live-modeling
or videotape-modeling treatments. However, no differences were obtained
between the live-modeling and videotape-modeling treatments. Gould and
Weiss (1981) tested the effects of model similarity and model talk on
self-efficacy and muscular endurance. Specifically, subjects who
observed a model of similar sex perceived to be similar in athletic
ability demenstrated greater muscular leg endurance than subjects who
observed a dissimilar model of the opposite sex perceived to be superior
in athletic ability.

It was determined a videotaped model would be an effective modeling
procedure for the study, because Feltz et a1. (1979) did not find a
difference between live-modelirg and videotape-modeling treatments. The
videotaped model was a consistent medel, providing the sane information
to each of the participants.

Both a male and female student model were videotaped because Gould
and Weiss (1981) showed a significant differennce in performance between

subjects who viewed similar and dissimilar models. A sixth grade boy

47

and girl, both of whom could throw using a mature pattern, were chosen
to demonstrate all of the testing procedures on videotape. Thus each of
the subjects could watch a model of the same gender, nearly the same
age, perform all of the testing procedures. The videotape would also
make certain that the same information would be provided to each of

them.

 

CHAPTER3

METHOD

The purpose of this investigation was to investigate the
relationship of throwing form as described by a developmental sequence
and throwing achievement as defined by the distance a ball can be
thrown. To accomplish this, the throwing motion of the throwers was
tested as well as the distance they threw the ball.

Subjects

Students in kindergarten through fifth grade at a private Christian
elementary school provided an available sample for this study. All the
students participated in the study unless they were physically disabled
or elected not to participate. The subjects came from predominantly
white, middle class, Protestant families residing in a small Midwest
community.

The procedure for participating in the study included the signing
of a consent form by the parent and a consent form by the student prior
to testing. Parents of the students were contacted by letter explaining
the rationale for the testing and the procedures that were to be used.
(A copy of the parent consent form is located in Appendix A.) The
students were given the opportunity to chose whether or not to
participate in this study after previewing a demonstration of the
testing procedures on videotape. Those wto decided not to participate
were returned to their classroom and excluded from testing. Children

wno decided to participate signed a consent form. (See Appendix A.)

48

49

The sample size and the number of drop outs for_ each grade and sex
are listed in Table 1. Of the 307 subjects agreeing tanparticipate,
four were dropped from.the study because they had incomplete data. The
physical education teacher completed a subjective evaluation of the
students who elected not to participate in the study. Most of them were
rated below average in athletic ability and, more importantly, below
average in self confidence regarding their physical abilities. In
examining the data, more than half of the drop cuts were girls in the
kindergarten, first, or second grade. It appears some of the younger
girls were reluctant to put themselves in a situation they perceived as
competitive and male-orientated. The three boys in the fourth grade who
dropped out had joined together and made a groupndecision not to

participate in the study.

 

Table 1

Total NUmber of Participants and Dropouts in the Study by Grade and
Gender

 

 

 

Males Females
Grade
Non 'Non Tkxal

___ Partkjgxumzs Partnjuxumzs Partkjgemmzs PartkaEumzs IPartkfipants
K 27 3 l9 5 46

l 29 l 31 S 59

2 l9 0 28 9 47

3 22 0 28 2 48

4 22 3 26 l 48

S 32 0 24 l 55

Tbtal 151 7 156 23 307

 

50

Testing Procedures _

The testing procedures included calculating the ages of the
subjects based on birthdates obtained from school records, measurirg
their height and weight, and dividing them.into squads based on grade
and gender which rotated through five testing stations. Information
concerning height and weight was collected within two weeks of the
administration of the test for throwing ability.

In order to facilitate testing and to increase efficency, the
subjects were placed in squads. The squads contained four to six
subjects of the same grade and gender. The squads rotated as a unit
through the various stations. When the subjects were placed into
squads, they were informed that they were not competing with anyone else
in their squad nor were the squads competing against each other. They
were just told to do their very best on the tests. They were asked to
encourage the members of their squad to perform.their best at each
station. They also were reminded to do the best they possibly could and
not to compare their performance to anyone else's performance.

The five stations were as follows. The subjects received
instruction regarding the procedures used to test their throwing ability
and strength at station 1. The subjects participated in an organized
warm—up at station 2. The actual testing of the subjects' throwing
ability occurred at station 3. Grip strength testing occurred at
station 4. Finally, the subjects were tested on their ability to do
push-ups at station 5. After the students completed the teSting at

station 5, they were excused to return to their classroom.

51

Throwing Ability

Throwing ability was tested in two ways, the distance the ball
traveled .in the air and the form that was used in throwing the ball.
Both of the variables were simultaneously tested, throwing distance was
measured immediately after the throws were made, while the form was
determined via examination of a video tape of each subject's throws.

‘ Testing the throwing ability of the children required three steps
which were accomplished at three stations. The subjects were provided
an explanation and demonstration about the testing procedures so they
understood what was required of them. This was accomplished by viewing
a videotape at station 1. The students were allowed time to warm—up
their throwing arms and practice throwing at station 2. Finally, the
subjects participated in the actual testing at station 3 where they
threw the ball as far as they could and were videotaped for future
evaluation of the stage of throw they used.

The method cnosen to explain the testing procedure was a videotape
of a sixth grade child denonstrating the different activities involved
in the study while an explanation of each of the testing procedures was
given. Two videotapes were made; one with a male model for the boys to
view and one with a female model for the girls. The modeling was
motivated by the theory of self-efficacy (Bandura, 1977) and the finding
that the more similar the model the greater the performance (Gould &
Weiss, 1981). A videotape was used instead of a live model because
videotape-modeling is as effective as live-modeling (Felt: et a1.,
l979) , the demonstrations and instructions would be uniform across
gender squads, and the different stations coild be viewed without

physically moving the subjects about.

52

In preparation for the distance throw, the subjects watched a
videotape demonstration of the testing procedure. The videotape showed
a sixth grader demonstrating a mature (stage 5) throwing pattern who was
the same sex as the subjects who were going to perform the distance
throw test. The demonstrator on the videotape told the subjects how to
perform.the test. Emphasis was placed on taking their time and throwing
the ball as far as possible before it bounced. After viewing the
videotape, the subjects were given the opportunity not to participate in
the study.

After they watched the videotape, the students who elected to
participate in the study proceeded to the warm-up station. The warmrup
took place outside the gym next to the testing station. To warm.up,
they threw a ball against the school wall. The warm-up consisted of a
minimum of 10 and a maximum.of 15 throws. The subjects were told to
throw the ball as hard as they could on their last three warm-up
trials. They used a 9-inch rag ball during warm-up as well as during
the testing. After the subjects had warmed up, they proceeded to the
throwing station and waited until the preceding group'was finished
throwing.

At the throwing station, the subjects' throwing ability was
assessed by the horizontal distance they could throw the ball in the air
and the stage they used in making the throw. The instructions to the
subjects were, "From this line on the ground, (pointing to the
restraining line) throw the ball as far as you can that way (pointing to
the direction they were to throw the ball) in the air.” The children
threw'from.behind a restraining line. Their throws were measured from

the restraining line to the spot on the ground where the ball landed.

53
The longest of the three throws the subjects made was used as their

score. The longest of the three throws was used in evaluating an
individual's maximum performance because maximum performance is commonly
used in sports performance.

Three individuals were needed at the throwing station: a camera
operator (and recorder), the test administrator, and a person to
identify and measure the point where the ball landed. The camera
operator was responsible for videotaping the thrower. The test
administrator was at the restraining line making sure the subjects were
throwing in the correct sequenoe and that they understood the directions
given to them on the video. He had the subjects verbalize what they
were to do to verify that they understood the directions. He also cued
the camera operator when the subject was ready to throw. The third
menber of the administration teanm was positioned in the landing area.
Markers, identified with a number and a letter to designate the thrower
and the number of the throw, were used to mark the spot where the balls
landed. After each squad completed its throws, this person and the test
adnministrator measured the distance the ball traveled in the air to the
nearest foot. The camera operator recorded the distance scores on a
score sheet according to subject number and the number of the throw.

The subjects' throwing performance was recorded on videotape for
each trial. The developmental stage the subjects used in achieving
their best distance throw was determined. The stage was not determined
during the throw itself but on review of the videotaped throws. The
investigator's assignment of stages was verified by faculty experts from

Michigan State University.

54

Strength
The rationale for including strength tests to help evaluate the

distance throw is that given two persons of similar height, weight, age,
and throwing form the one who is stronger should throw the ball
farther. Tne relative strength of the throwers might affect their
performanoe. The strength tests used in the investigation included a
static test—hand grip, and a dynamic test—push-ups. These two
measures were selected because of their potential contribution to
throwing performance as well as the feasibility of their use with
elementary school children.

The hand dynamoneter test was used to assess dominant hand grip
strength. Tne hand used for throwing was noted during the throwing
procedure and that hand was tested for grip strength. When the hand
used for throwing was not consistent from trial-to-trial, the hand used
for the longest of the three throws was considered the dominant hand.

In the administration of the hand dynamoneter test, the tester set
the pointer to zero and placed the dynamoneter in the subject's hand.
The tester then adjusted the handle of the dynamoneter so the first
joint of the index finger formed a 90 degree angle as it rested on top
of tie handle and it felt comfortable in the subject's hand. The
subjects were encouraged to squeeze as hard as they possibly could.
They squeezed as sharply and steadily as possible, making certain that
no part of the arm tomched the body. Three trials were taken with a
minimum rest period of a one minute between squeezes. The three trials
were recorded to the nearest one-half kilogram. The score for each
subject was the mean of the three trials. The sanne dynamoneter was used

for all subjects .

55

The subjects were tested on the number of push-ups they could do
during a 30-secord period. A piece of foam rubber 4 inches thick, 8
inches wide and 5 feet long was used to help standardize the push-up
procedure. The foam rubber was placed underneath the subjects with
their arms straddling it. The subjects started by lying on the foam. A
push-up was counted every time the subjects pushed their bodies up so
that their arms were fully extended and their knees were clear of the
foam rubber. On the downward phase of the push-up, part of the head
(for example the forehead, nose or chin) and part of the chest or
abdomen had to touch the foam before another push-up could be counted.

A subject's score was the number of full extensions completed during the
30-second time period. A video recording was taken of each group so
that performances could be scored at a later tinne. A description of the
stations used in the study, the number of individuals needed to operate
each of the stations, and the duties of the individual personnel are
identified in Table 2.

Other Data

The subjects' grade and gender were noted and recorded on the
sheets used to record their throwing distance. The subjects' height and
weight were obtained and recorded within two weeks of the testing.
Standing height was measured with the students standing fully erect in
stocking feet and stretched to the fullest height while keeping the
heels flat on the floor. The students stood against a wall while the
tester placed a wooden triangle with a right angle against the wall and
on top of the subjects' head. Standing height was read to the nearest

quarter inch from a tape measure fixed to the wall. The subjects were

 

Table 2

Testing Personnel and Their Duties

Station Workers

#1 Video Supervisor

#2 Warm-up Supervisor

#3 Throwing Test Administrator

Camera Operator

Spotter

#4 Dynamometer Tester

#5 Push-ups Tester

56

Duties

Supervise showing the video tape on
the instructions of the throwing
procedures. Make sure the correct
one is played and the students are
listening and understand the
instructions.

Make sure students are throwing so
they have a proper warm-up and keep
other students finished with the
warm-up in line, not watching the
thrower.

Make sure subjects understand
instructions and exhort them to make
their best throw. Hold the end of
the measuring tape in the correct
place.

Film the thrower, making sure the
entire body is filmed. Record the
distance of the throws when the
squad was finished.

Mark tie place where the ball hits
the ground with the correct marker
and help measure the distance.

Demonstrate the procedure of the
hand dynamoneter and record the
scores.

Give instructions for the proper
push-up, time the alloted 30
seconds, and count and record the
number of push-ups.

57

weighed to the nearest one-half pound on a balance scale without shoes
while in their daily school clothes.
. Treatment of Data

Means and standard deviations for the independent and dependent
variables were calculated. The variables included throwing hand grip
strength, push-ups, developnental throwing stage, height, weight, age,
and the distance the ball was thrown. In addition, the means and
standard deviations on these variables were computed for each grade, and
for gender within each grade.

Pearson correlation coefficients were calculated to determine
relationships between developmental throwing stage, hand grip strength,
push-ups, height, weight, age, and the distance the ball was thrown.
Three intercorrelation matrices were generated, one matrix for all the
subjects and one for each gender. In addition, correlation coefficents
were obtained to determine test/retest reliability of the push up and
hand dynamoneter strength tests.

Three stepwise multiple regression analyses were run to predict
throwing distance based on the variables of age, grade, stage, height,
weight, push-ups, and hand grip strength. Regression equations were
generated for all the subjects combined, for the girls, and for the
boys. The regression equations also provided appropriate covariates for
the ANOOVAs that were run. Any variables which entered and remained in
the final equations were considered important variables which had to be
accounted for either as main effects or as covar iates for the gender of
the subjects involved in the ANCOVA.

To test the first hypothesis, that for each grade level boys will
be able to throw the ball farther than girls, a 2 X 6 ANCOVA was run

58

using gender and grade as the two factors. The variables that entered
the regression equation for all subjects combined, except for gender and
grade, were used as covariates in the ANCOVA.

The second hypothesis was that members of the same gender in an
older grade will be able to throw the ball farther than those in a
younger grade. This hypothesis was tested by running a one-way ANCOVA
using grade as the factor with the appropriate covariates as determined
by the appropriate regression equation.

Testing the third hypothesis, that children of the same grade who
have a more mature form of throwing as measured by developnental
sequence will be able to throw the ball farther than children at a less
mature stage of throwing, required a number of steps to test. First, a
stage by grade by gender cross tabs program was run to determine the
frequency of subjects in each cell of the matrix. Blank cells were
noted and cells were collapsed based on the findings. Next, separate
ANGDVAS were run for the boys and girls using develognental stage as the
independent variable with appropriate variables as covar iates.

For all of the ANCIJVAs run, the main effects and two-way
interactions were examined. men the main effects were significant, the

post hoc multiple comparison method of Scheffe was applied.

CHAPI'ER4

RESULTS AND DISCUSS ION

This study investigated the relationship between developmental
stage of throwing and the distance a ball was thrown by children. In
addition, this research effort examined the effect that gender, height,
weight, age, grade, push-ups and grip strength had on a distance throw.

In this chapter, the results ad discussion will be presented
together in five sections. First, descriptive statistics of the
variables will be reviewed. Second, findings from regression analyses
to determine the predictability of throwing distance and to identify
covariates will be discussed. Third, the results from the analyses of
variance will be presented and examined in relationship to the effect of
gender and grade on throwing distance. erth, the results from the
analyses of covariance will be reviewed with regard to the effect of
stage on throwing distance. Finally, the findings will be discussed and
compared to the results of previous investigations.

Descriptive Statistics

The means and standard deviations of all the variables were
calculated to provide an overview of the relationships between the
variables within and across the different grade and gender groups. The
tables containing the means and standard deviations for throwing hand
grip strength, push-ups, height, and weight are located in Appendix B
(Tables 8-1 through 3-4) .

59

60
An examination of the difference between the mean age of the boys
and girls for each grade showed that the chronological age of the boys
was sightly greater than that of the girls in every grade, but only in
the fourth grade was the difference more than 2 months (Table 3). A 3-
test comparing the ages of boys and girls showed that only in the fourth

grade was there a significant difference between the ages of boys and

 

Table 3

Means, Standard Deviations, and t-values for the Chronological Age of
the Subjects (in months)

 

Total/ Standard
Grade Gender Means Deviation Cases t
KINDERGARTEN TENTH. 76.93 3.81 46 1.156
BOYS 77.26 3.51 27
GIRLS 76.47 4.26 19
191‘ GRADE TOTAL 88.68 4.28 59 .994
BOYS 89.24 3.76 29
GIRLS 88.13 4.73 30
2ND GRADE TOTAL 99.70 4.71 47 .987
BOYS 100.53 4.85 19
GIRLS 99.14 4.62 28
3RD GRADE TOTAL 111.79 4.05 48 1.417
BOYS 112.68 4.16 22
GIRLS 111 . 04 3 . 87 26
4TH GRADE TOTAL. 124.44 4.30 48 2.060*
BOYS 125.82 3.40 22
GIRLS 123.27 4.68 26
5TH GRADE TOTAL 134.64 5.48 55 .828
BOYS 135.16 6.26 32
GIRLS 133 . 91 4 . 17 23
ALL GRADES TOIM. 106.27 20.54 303
sons 106.99 21.67 151
GIRLS 105.56 19.45 152

 

* t is significant p<.05;

61
girls. Since only one of the grades was significantly different in age
between the boys and girls it was decided that grade would be an
appropriate means for dividing the subjects.
The mean performance of the subjects in the distance throw
increased for each of the grades tested (Table 4). However, when
examining the gender groups separately, the sane trend existed but an

increase did not occur between each grade level. The boys' performance

 

Table 4

Means and Standard Deviations for the Distance Throw (in feet)

 

 

Total/ Standard
Grade Gender Means Deviation Cases
KW mom. 36.17 11.86 46
BOYS 43.15 9.90 27
GIRLS 26.26 5.76 19
lST GRAm TOTAL 49.56 21.12 59
BOYS 67.69 14.18 29
GIRLS 32.03 7.21 30
2ND GRADE TOTAL 56.49 24.72 47
BOYS 82.68 14.43 19
GIRLS 38.71 9.60 28
3RD GRADE TOTAL 73.73 24.98 48
BOYS 97.14 16.77 22
GIRLS 53.92 11.92 26
4TH GRADE TOTAL. 75.48 25.49 48
BOYS 97.00 16.57 22
GIRLS 57.57 12.03 26
5TH GRADE TOTAL 83.35 27.16 55
m 102.22 16.46 32
GIRLS 57.09 13.78 23
ALL. GRADES TOTAL 62.67 28.28 303
BOYS 81.07 26.13 151
GIRLS 44 . 39 15 . 86 152

 

 

62

increased from kindergarten through the third grade, -plateaued at grade
4, and increased again in grade 5. The mean performance of the girls
increased. across grades until they experienced a plateau between grades
4 and 5. At all grade levels, the boys threw the ball substantially
farther than the girls.

The average stage exhibited by the children also increased across
grade levels (Table 5) . However, because of the high percentage of boys

 

Table 5

Means and Standard Deviations for Throwing Stage (range = 1 to 5)

 

Total/ Standard
Grade Gender Means Deviation Cases
KINDERGARTEN TOTAL 3.48 1.44 46
BOYS 4.11 1.28 27
GIRLS 2.58 1.17 19
lST GRADE TOTAL 4.07 1.28 59
BOYS 4.90 .31 29
GIRLS 3.27 1.36 30
2ND GRADE TOTAL. 4.13 1.17 47
BOYS 4.89 .32 19
GIRLS 3.61 1.26 28
3RD GRADE TOTAL 4.54 .74 48
mYS 4.82 .66 22
GIRLS 4.31 .74 26
4TH GRADE TOTAL 4.71 .68 48
BOYS 5.00 .00 22
GIRLS 4.46 .86 26
5TH GRADE TOTAL 4.84 .37 55
3023 5.00 .00 32
GIRLS 4.61 .50 23
ALL. GRADES TOTAL. 4.30 1.11 303
BOYS 4.78 .69 151

GIRLS 3 . 83 l . 23 152

 

63

using the mnature stage level (stage 5) , the increasewas due primarily
to the improvenent in throwing form by the girls. The mean stage for
girls increased each year from 2.58 (5 maximum) in kindergarten to 4.61
in grade 5. In contrast, the mean stage for kindergarten boys already
was 4.11. The first, second, and third grade boys leveled off with
means of 4.90, 4.39, and 4.32, respectively. All the boys tested in
grades 4 and 5 used the mature pattern.

Tie means for height and weight systematically increased across the
grade levels. Boys were taller than the girls at all grade levels
except grade 4 where the mean height was approximately even. The boys
also were heavier than the girls except at grades 3 and 4 where the
girls were approximately 4 and 6 pounds heavier, respectively, than the
boys. (See Tables B—1 and B-2) .

Both boys and girls increased their grip strength from kindergarten
throgh grade 5. The boys were stronger than girls at all grade
levels. The range of the difference between boys and girls increased
from 1.62 kg in kindergarten to 5.63 kg in grade 5. The pattern for
push-ups was different. Mean performance improved from kindergarten
until grade 4 and then decreased. Peak perfornmance for the boys
occurred in grade 3 and for the girls in grade 4. Boys did more push-
ups than girls at all grade levels. (see Tables B—3 and B-4)

A test, retest correlation analysis was perfornmed on the strength
nmeasures. The retests were conpleted four weeks after the original
strength measures were obtained. Throwing hand grip strength was
retested on tte students in kindergarten, secod grade, and fourth
grade. The first graders, third graders, and fifth graders were
retested on their push-ups. The results (Table 6) showed a significant

64
Table 6

Test, Retest Correlation Coefficients of Strength Measures in Selected

 

 

Grades
Grip Strength Push-ups
Grade R Grade R
K . 880* 1 . 664*
2 . 939* 3 . 902*
4 . 907* 5 . 852*

 

* p < .01 (2 - tailed)

correlation (p < .01) for every group tested. The first graders
retested on push-ups had the lowest correlation coefficient of any group
(r = .66). This lower correlation emphasizes the need for careful
control of the push-ups test for younger children. The rennaining
coefficients ranged from .852 to .939. Thus the performance of the
children on the strength tests were judged to be consistent and
acceptable.
Predictability of Throwing for Distance

A selected number of growth, strength, and perfornmance variables
were examined for their ability to account for the variablility in a
throw for distance. A determination of the relationship among these
variables was important for predicting a distance throw and to provide
covariates in tre analysis of the variance of throwing for distace.
The predictability of throwing for distance was studied in three
groups: girls, boys, and all subjects combined so that the correct

65

variables could be controlled according to the subject group being
analyzed.

Three intercorrelation mnatrices were obtained, one for all the
subjects and one for each of the gender groups. The correlation
coefficients of the variables are listed in Tables 7, 8, and 9. The
intercorrelation matrix for all subjects (Table 7) shows a correlation

coefficient of .588 between throwing style and throwing distance. r-

fir

Althogh modest in magnitude, it is comparable to the relationship of
age (r = .586), grade (r = .570), and height (r = .592) to throwing

distance. Its relationship to throwing distance is greater than that of

 

weight (r = .440), grip strength (r = .428), or push-ups (r = .214) to r."
throwing distance. A connparison of the boys' matrix with the girls'
matrix (Tables 8 and 9) reveals similar coefficients for most of the
variables. The major exception to the general trend is stage, whose
correlations with the other variables are noticeably higher for the
girls than for the boys. For example, the correlation between stage and
the distance throw was .42 for the boys and .61 for the girls. The
higher correlations are a result of the greater variability of stage for
the girls when compared to the boys. The percentage of boys using a
stage 5 throwing pattern was so large (88%) that there was very little
variability of stage for tie boys. The snmall variability of stage and
the greater variance of throwing distance resulted in a snmaller
correlation value for the boys.

Three stepwise regression analyses were run to deternmire which of
the variables of height, weight, push-ups, and grip strength studied
would be important for predicting throwing distance and for use as

covariates in further analyses. To qualify as a covariate the variable

 

 

66

«3. gm. ZN. So... NS. 5:. e: . 86:63

2:. m3. ems. 5.. mg. 8.. see. 5886 8.6
gm. me... 8m. 03. me... oh. 8m. 38E. 8563
Sm. Mae. 8.... ea. am. we... mme. e86
$6.- Gee. oee. 8N. NS. 5.. 5.. 23m:
«8. mow. Nam. 8m. New. Se. 2m. 32%
m2. 3.. 2m. mam. ewe. Be. as. 886
3:. 62.. mam. ma... 5.. e8. e8. 62

 

mm: 5883 86E. mmflm 2663 cameo: 6636 6%
noose 3.6 8:865

 

phoebe—co macho ocm moon new imam: Squmaouuocumucn
n ~33.

67

 

 

 

meg. new. eee. Hee.n eee. eee. gee. eeeneeee
meg. Nee. «em. eee. gee. eee. eee. eueceeee eeee
pee. wee. eee. gee. age. ape. eee. gouge eoeeeeee
eee. New. eee. eee. eee. Hem. eee. emcee
gee.) eee. nee. eee. Nee. eee. eee. name»:
age. gee. eee. eee. Nee. eee. eee. eeeeee
eee. eee. ape. gee. eee. eee. eee. mambo
Hen. eve. eee. eee. eee. eee. eee. «me
me: decade 38E. Made 23...: Emma 886 ewe
named me no 8:663

 

moon 05 ecu 559‘ 833030335

o manna.

68

 

 

Ge. e3. eee. Nee... e8... eee. NS. 8.768
Be. eee. Ge. Re. eee. eee. eee. decade and
2:. eee. Be. on. eee. ewe. 5.. 385. 8.53m
eee . Be. Be. 8e. eee. eee . 2e. ease
Nee... 2a. Nee. 8e. eee. 3.. 2e. £262
8.... e2. eee. eee. nee. eee. eee. ”Beam:
:8. eee. e2. e3. 3.. eee. fie. weed
NS. eee. 5.. 2e. 3». eee. see. «me
mo: fimcmhm 665. «meow £533 2&6: mambo 6%

-68 966 «Seems .

 

mHqu 05 HOW Nahum: SmuMHOHHAHVHUHCH

o canon.

69

must enter the regression equation and contribute an .increase of at
least .05 to the adjusted R2. The value of a .05 increase in the
adjusted, variance was used to eliminate variables that did not
contribute a substantial gain in throwing distance. One regression was
run for all the subjects, one for the boys, and one for the girls.

The variables that entered the regression equation for boys and
girls conbined were grip strength, push-ups, height, and weight (Table
10). However, of these four variables, only grip strength added a
minimum of .05 to the adjusted R2. Grip strength had the highest
individual correlation with distance throw with an R = .675 and an
adjusted R2 of .454. In the analyses involving the performance of boys
and girls combined, grip stength was the only variable used as a
covariate.

In the regression equation for the boys the variables that entered
were height and push-ups (Table 11) . Both variables added at least .05

 

Table 10
Stepwise Multiple Regression of Predictor Variables on Throwing Distance
for Boys and Girls Conbined

 

Regression
Variable Mnltiple R R2 Adjusted R2 SEE F Ratio Ocefficient
Grip Strength .675 .456 .454 7.696 3.26
Push-ups .701 .492 .488 3.786 .65
Height .709 . 503 .498 4.465 2.28
Weight . 730 . S33 . 526 19 . 46 -4 . 341 -. 50

Constant -4.129 —85.07

 

70

to the adjusted R2, so height and push-ups were used as covariates in
analyses of the boys. Height alone accounted for an R = .719 and for an
adjusted m2 of .514. Height and push-ups connbined not a multiple R of
.763 and an adjusted 32 of .576.

The regression equation for girls entered height first, push-ups
second, and grip strength third. However, neither push-ups nor grip
strength raised the adjusted R? the required .05. Therefore height was

 

Table 11

Stepwise Multiple Regression of Predictor variables on Throwing Distance
for Boys

 

 

 

 

Regression
variable Multiple R R2 Adjusted R2 SEE F Ratio Coefficient
Height .719 .518 .514 13.443 3.90
Push-ups .763 .582 .576 16.95 4.767 .84
constant -8.751 -l37.89
Table 12

Stepwise mltiple [Egressicn of Predictor Variables on Throwing Distance
for Girls

 

 

Regression
variable Mu1tiple R R2 Adjusted R2 SEE F Ratio. coefficient
Height .688 .474 .471 5.318 1.90
Push-ups .720 .519 .512 2.927 .45
Gkip Strength .732 .536 .527 10.92 2.355 .83

 

71
the only variable used as a covariate for the girls accounting for an R
of .688 and an adjusted R2 of .471.
Differences in Throwing Distace for Gender and Grade

A grade by gender (6 X 2) ANCOVA was used to evaluate the first two
hypotheses: l) for each grade level, boys will throw a ball farther
than girls and 2) boys or girls in a higher grade will be able to throw
a ball farther than their gender counterparts in a lower grade. The
ANCOVA showed significant gender _F_(l,302) = 641.36, p < .001 and grade
£6,298) = 103.53, p < .001 main effects in the distance students can
throw the ball. The two-way interaction between gender and grade also
was fond to be significant £6,298) 8 9.55, p< .001. The findingthat
the boys threw significantly fartter than the girls was expected and
agrees with the numerous studies examined by Nelson and French (1985) .

Because the two-way interaction between gender and grade was
significant, the gender groups had to be examined simultaneously. A
Tukey multiple conparison test was run to determine which grades were
significantly different from each other. The means and standard
deviations for each grade are displayed in Table 4. The significant
differences (p < .05) between grades is shown in Table 13.

The mean distance subjects threw the ball increased each grade from
kindergarten through fifth grade. The mean distance throws of
kindergarten, first, and secod graders were significantly different
from each other and from grades 3, 4, and 5. However, there was no
significant difference in mean throwing distance between grades 3, 4,
and 5. Even trongh no significant differences existed between the
third, fourth, and fifth grades, the trend of an increase in throwing

72
Table 13

Significant Differences of Throwing Distance Between Grades K and 5

 

Grade
Grade K 1 2 3 4 5
K .—
l * _.
2 it * .....
3 i * * ......
4 t i * as .._..
5 w w w: ns ns ..

 

* 9 Significance at the p4 .05 level

perfornmace in succeeding grades reported in previois studies (Hanson,
1965: Keogh, 1965) was confirmed in the present study.

To examine the significant effect of the interaction between gender
and grade, two graphs (Figures 1 and 2) were plotted. Figure 1 shows
tte mean throwing distances of boys and girls for each grade . Figure 2
examines the gain each gender group made between each of the successive
grades studied. In looking at Figures 1 and 2, there are two distinct
sections. The first section is between kindergarten and third grade
where both genders are increasing their throwing distances every year.
The secod section is between grades three and five wrere a leveling off
of throwing distance occurs. The increase between the last two years is
smaller than the smallest increase of any of the previous years. In
fact, the gain from third grade to fifth grade for the boys is

approximately oe third of the smallest gain of any year between

Mean Distance Throw (in feet)

73

 

100 ~—

90

80

70

60

50

 

 

 

 

40 -- 0/
, ' ’ -—- Boys
’ o
3o _ ’2 o—---o Girls
(V I
l l T T
K 1 2 3 4 5
Grade
Figure 1. Mean distance throw of girls and boys in Kindergarten

through 5th Grade.

Gain in Mean Distance Throw (in feet)

74

 

 

 

 

 

Successive Grades

Figure 2. Gain in mean distance throw of girls and boys between
successive grades.

75

kindergarten and third grade. For the girls, the gain from third grade
to fifth grade is appnoximately one half of any of the previous years.

Even though both gender groups exhibited improvement from grades
kindergarten through third grade and a leveling off between third and
fifth grades, there was a difference within each section between the
boys and girls. In the section which includes kindergarten through the
third grade, the boys increased the most between kindergarten and first
grade (24.5 feet) and had steady improvement from first grade to second
grade (15.0 feet) and from second grade to third grade (14.5 feet). The
girls showed the steady improvement in the first two years with an
imemovements of 5.8 feet from kindergarten to first grade and 6.7 feet
from.first grade to second grade. Then the girls experienced their
biggest increase in throwing distance between second grade and third
grade with an increase of 15.2 feet-more than twice the increase in any
other year. So the gender trends for throwing distance in kindergarten
through the third grade was for the boys to increase the most between
kindergarten and first grade and the girls to increase the most between
second and third grades.

The boys and girls also differed in their increases in the fourth
and fifth grades. The boys did not increase their throwing distance
from the third to the fourth grade while they did increase 5.0 feet
between the fourth and fifth grades. The girls, on the other hand,
increased the distance they threw the ball 3.3 feet between the third
and fourth grades but did not increase their distance between the fourth
and fifth grades. The differences between the boys and girls resulted

in a significant interaction between gender and grade.

76

Differences in the Effect of Stage on Throwing Distance

The third hypothesis, that subjects of the sane gender in the sane
grade wto have a more mnature form of throwing (as determined by the 160
developmental sequence) will throw a ball farther than boys or girls do
exhibit a less mature stage of throwing, was examined separately for
boys and girls.

_B__oyg

Because of the large percentage of boys using the stage 5 throwing
pattern, tne anticipated 6 X 5 grade by stage ANCOVA could not be run.
In checking the crosstabs table (Table 14) , the only grade that had two
or more throwers in stages 1, 2, or 3 was kindergarten, so it was the
only grade that could be examined for the effect of stage on throwing
distance. It was also noted that stages 1, 2, and 3 in kindergarten
only had two subjects in each cell, which were not enough to run
analyses. Because the first three stages represent an immature throwing
pattern, these cells were collapsed into one (n = 6). An ANCOVA was
then run on the kindergarten boys comparing throwing distance by stage
using push-ups and height as covariates. The main effect of stage was
not significant, 52,25) 8 1.71, p = .205, in determining throwing
distance for kindergarten boys. The covariates of push-ups, F_(l,26) =
.25, p a .635, and height, §(1,26) = .01, p = .933, also were shown to
be non-Signficant.

Even tl'otgh stage was not shown to be a significant indicator of
throwing ability, the general trend of the boys showed an improvement
fromstagesl, 2, and3conbined (n=6:M=36.8), stage4 (n86,

M =- 42.7), and stage 5 (n =- 15: M = 45.9). Leme (1973) noticed that

sone women wro moved from a lower stage of throwing to a higher one lost

77

Table 14

The Number of Boys at Each Stage of Throwing in Each Grade.

 

Grade
Stage
K 1 2 3 4 5

5 15 26 17 20 22 32
4 6 3 2 1

3 2

2 2 l

l 2

 

distance in their performance throw. It has been postulated that
performance may decrease during the transitional tinne when a person is
changing from one stage to another because time is needed to coordinate
the new movements in the pattern ad make it more efficient.

Throwing performance, may be greater when using a relatively stable
less mature stage to its full potential than when an individual is in
transition to a more mature stage. The kindergarten boys are all in the
process of forming their motor patterns and constantly refining them.
Stage 1 provides the thrower with a stable base to generate force to the
ball. With each succeeding stage, the base is less stable and either
more body parts are moving or the body parts are moving through a
greater range of motion. In young throwers, such as kindergarten boys,
the trannsiticn from one stage to another calls for ccntinnous learning
to control the new movements and at any given tinme there will exist a
wide range of efficiency in throwing perfornmance at a particular stage
level. Stage 1 provides the kindergarten thrower with the stable base

78

to involve the basic arm movenents to perform a fairly good thrcw.
Kindergarten boys using the fourth and fifth stage of throwing have been
able to increase their throwing distance but they have not yet been able
to fine tune the acceleration and deceleration of body parts to take
full advantage of the improved stage of throwing. Thus the kindergarden
boys have shown an increase in throwing distance for the different
stages but the distances have not been found to be significantly
different.
one

In investigating the effect of throwing stage on throwing distance
for the girls, the anticipated 6 X 5 grade by stage ANCOVA could not be
run because several open cells were noted (Table 15) . Kindergarten was
dropped from the initial analysis because there were no stage 5 girls.
Fifth grade was also dropped because all the girls were throwing with a
stage 4 or 5 pattern. Stages 1 and 2 were dropped because of empty
cells and innsufficient rnumbers in others. The largest section of grade
by stage without any open cells was grades 1 throgh 4 and stages 3
throngh 5. An ANCOVA was run using grade (1,4) and stage (3,5) with
height as a covariate. In a second analysis a one way ANCOVA was run on
the fifth grade girls to determine if there was a significant difference
between stage 4 and 5 throwers. Height was also used as covariate in
the secod ANCOVA. Finally, a one way ANCINA of throwing distance by
stage with height as the covariate was run for all the girls.

Three general trends about stage and throwing distance can be
observed when examining the means of throwing distance for girls by
stage and grade listed in Table 16. First, the higher grades (3 ad 4)

have the largest percentage of throwers using the most skilled stage.

79

 

Table 15

The Nomber of Girls at Each Stage of Throwing in Each Grade.

 

 

 

 

Grade
Stage
K 1 2 3 4 5
5 4 6 12 l7 l4
4 5 l4 13 10 5 9
3 6 4 5 4 3
2 3 2 l
1 5 6 4
Table 16
Means of the Distance Throw by Grade and Stage for Girls
Grade
Stage Row
1 2 3 4 Totals
5 34.8 44.3 59.2 61.7 55.5
(n=4) (n=6) (n=12) (n=l7) (n=39)
4 35.2 39.8 48.2 51.2 41.6
(n=14) (n=13) (n=10) (n35) (n=42)
3 29.5 31.8 52.5 42.3 38.4
(n=4) (n85) (n=4) (n=3) (n=l6)
Column 34.1 39.3 53.9 57.3 46.7
Totals (n=22) (n=24) (n326) (neZS) (n=97)

 

80

(Notice the larger number of stage 5 throwers when the higher grades are
compared with the lower grades.) The secod trend is that or each grade
the students using a more mature stage throw the ball farther. The two
exceptions to this trend are the first grade stage 5 throwers who did
not throw as far as the first grade stage 4 throwers, and the third
grade stage 3 throwers who threw farther than the stage 4 throwers. The
third trend is that for each stage, girls in the higher grades threw the
ball farther than girls in the lower grades. Once again the stage 3
throwers in the third grade were the exception. They threw the ball
farther than the fourth grade stage 3 and stage 4 throwers.

The trends in the throwing distance of the girls in grades 1
through 4 using stages 3, 4, and 5 were examined in the first ANCOVA.
Height was used as a covariate. The main effects for stage, F(2,94) =
9.04, p < .001, and grade, 30,93) = 6.26, p = .001, were significant.
The covariate, height, §(l,95) = 69.10, pg< .001, also was signficant.
The two—way interaction between stage and grade, §(6,90) = 1.13, p_=
.353, was not significant. Thus both stage and grade were shown to have
significant effects on the distance a ball could be thrown by first,
second, third, and fourth grade girls.

The Scheffe multiple comparison test was used to compare the
different levels of grade and stage. The results of the significant
differences between the grades and stages in throwing distance are given
in Tables 17 and 18, respectively. In reviewing the differences between
the grades, there was a signficant increase in the throwing distance
between second and third grade. Thus, the mean throwing performance for
grades 1 and 2 was significantly different from that of grades 3 and 4.

However, the increases between first and second grade and between third

81
Table 17

Significant Differences of Throwing Distance for Grades 1 to 4

 

Grade
Mean Grade 1 2 3 4
34 . l 1 --
39 . 3 2 ns —-
53 . 9 3 * * --
S7 . 3 4 * * ns --

 

* - Significance at the p < .05 level

 

Table 18

Significant Differeces of Throwing Distance for Stages 3 to 5

 

Stage
Mean Stage 3 4 5
38.4 3 --
41.6 4 ns --
55.5 5 * * —-

 

* - Significace at the p < .05 level

82

,a

ad fourth grade were not significant. In contrasting the different
stages, mean throwing distance for stage 3 was not significantly
different from that of stage 4 but it was significantly different from
that of stage 5. Also, the mean throwing distance performance for
stages 4 and 5 were significantly different.

Tte second ANOOVA for the girls examined those in fifth grade,
which had only stage 4 and 5 throwers. In comparing the two stages a
difference approaching significance was fonnd for the mnain effect of
stage §(1,22) = 4.21, p = .054. A nonsignificant difference was
obtained for the covariate of height _F_‘_(1,22) = 1.17, p = .293. These
results, in general, support the hypothesis that a more mature stage
will result in a loger distance throw.

The results of this ANCOVA provide more evidence of the difference
in throwing distance between girls using the stage 5 throwing motion and
time using stage 4. These results are in agreennent with the findings
of the ANCOVA for the first through fourth grade girls which also found
a significant difference in throwing for distance between stage 5 and
stage 4 throwers. Tl'e general trend shows that as the girls get older
more of then throw at the mnature stage 5, and therefore can throw a ball
farther. Of the fifth grade girls in this study, 60 per cent throw
using the stage 5 pattern. The trend would seem to indicate that a
number of girls would have been using the stage 5 pattern for a period
of time and thus shonld have been able to improve their efficiency in
the use of this pattern. If there is a difference between a stage 5
thrower ad a stage 4 thrower, they snould have been able to demonstrate
it. The ANCOVA results verify that this is the case.

 

83

The final comparison for the girls examined their throwing distance
by stage. The cell for stage two girls had an n of 6 which was
considered small especially conpared to the 56 girls in stage 4 and 53
girls in stage 5. A comparison was mnade of the stage 2 throwers with
matched pairs of stage 1 and stage 3 throwers. The stage 2 subjects
were matched by grade, height, weight and strength in that order. The
mnatched pairs of stage 1 and 3 throwers were combined with the stage 2
throwers and were placed in a one way ANOVA of throwing distance by
stage. The ANOVA showed no significant difference between the stages
3332,15) = .27, p = .767. Since the ANOVA did not produce a significant
difference between stages, stages 1, 2, and 3 were collapsed into one
stage for comparison in the ANCOVA.

The one way ANCIJVA with stage as the main effect and height as the

covariate found the main effect of stage £32,149) 22.33, p< .001) to

be significant. The covariate of height 3(1,150)

46.35, p< .001) was
also found to be significant. The Scheffe multiple conparison test
showed that all three stages—l, 2, and 3, combined: stage 4: and stage
5 were significantly differently from each other.

The third ANCIJVA, used to evaluate the effect of the stage pattern
on the girls‘ throwing distance, also supports tl'e hypothesis that a
more mature stage will increase throwing distance. It is not surprising
that the covariate, height, was found to be signficant because of the
large difference of height between kindergarten and fifth grade girls,
and the trends previoisly stated about the increase in distance throws
with increased age. It is important to note that for the entire age

group and taking the covariate of height into account, the stage pattern

84

of the thrower still had a signficant influence on the distance the
girls could throw the ball.

Although it cannot be stated that every girl who uses a more mature
throwing pattern, as determined by her throwing stage, will throw
farther than one who uses a less mature stage, certainly evidence is
provided in this study to support the hypothesis that the stage of
throwing individuals use has an important influence on the distance they
can throw a ball.

This study provides support for the use of the Michigan State
University developmental sequence of throwing in assessing throwing
performance. The study has shown differences in throwing distance
between girls using a more mature throwing pattern and girls using a
more rudimentary throwing stage. However, the stage sequence is not a
panacea, explaining all there is to the throwing behavior. It mnust be
remembered that throwing is a complex motion requiring precise
coordination of many body parts. The stage sequence explains the major
movements that the body must make to produce a better throw, but it
cannot explain every minute detail of the throw. As it was previously
noted, there are a number of other factors that affect throwing
performance such as strength and body size. Because of its complexity,
throwing performance cannot be easily explained and compartmentalized.
The 360 stages of throwing do provide a way to explain sone of the
differences in throwers and thereby provide a sequential progression to
follow in helping students develop the proper throwing patterns of a
mature throw. Within these limits it can be a valuable tool for the

teacher /coach .

85

This study does point to sone limitations of the stage theory. In
this study, the boys had already reached the mature stage by the first
grade. If the students have already reached stage 5, the stage theory
cannot provide any additional information regarding the differences
between the throwing distance of the boys in the first through fifth

grades. It is noted that there still exists a great difference in the

_ _Il'-I‘

throwing distance of the boys even though they are in the sane grade and
use the sane throwing form. There must be a refining of the stage 5

pattern in the timing and sequencing of the motion, along with physical
and mnaturation factors such as height and strength, that goes beyond the

 

easy categorizing of the motion into stages, and this refining adds to a
length of tie throw.
This study provides information, within the linmits of the stage
sequence that students (particularly girls) wto use a more mature fornm
of throwing do indeed experience an increase in throwing distance. This
study provides evidence that the stage sequence for throwing does
provide important information to tne teacher/coach by providing

important clues which can lead the performer to a more efficient throw.

CHAPTER 5
SUD‘MARY, CQCLUSICNS, AND WTICNS

Summary

T'ne purpose of this study was to evaluate the relationship between
the qualitative throwing motion used by children and the distance they
could throw a ball. The study also examined the influence of grade,
gender, grip strength, push-ups, height, and weight on throwing
distance.

The subjects were 303 students in a private Christian elementary
school in kindergarten through fifth grade. Tl'ere were 151 boys and 152
girls participating in the study. The students cane from predominantly
white, middle class, Protestant families residing in a small Midwest
community. All the students in kindergarten through the fifth grade
were encouraged to participate, but a signed consent form from a parent
and from do students thennselves were required to participate.

The students watched a vidmtape showing a 6th grade model of their
gender performing all the tasks they were required to do during the
testing. The students then signed a consent form if they wished to
participate. They proceded to a warm-up area where they made 10-15
throws with a rag baseball similar to tie oe used for the distance
throw. Next, they went to the station where they threw the rag baseball
for distance. They were reminded again that the throw was going to
nmeasured by the distance the ball traveled in the air. All the subjects
made three throws, with the logest throw being used in the study as the
subject's score. The subjects were videotaped at the time of their

86

87

distance throws and the videotape was reviewed at a later date to
determine the stage they were using to throw the ball. The subjects
then went inside the school to have their throwing hand grip strength
and push-ups performance assessed. Push-up performance was videotaped
and the videotape was reviewed at a later time tolverify the correct
number of push-ups for each subject. The subjects' height and weight
were obtained within two weeks of the other tests. Grade, gender, and
age in months also were noted for each subject.

Correlation coefficients were obtained between each of the
variables. In the analysis inwolving all the subjects, grip strength
had the highest correlation coefficient with throwing distance at
.675. Stage also had a moderate correlation with throwing distance at
.588.

The major hypothesis was that the higher the stage of throwing
used, the longer the distance throw. The range of stages exhibited by
the boys was very limited with only the kindergarten boys having stage
1, 2, and 3 throwers. Thus, the only comparison that could be made
between throwing stage and throwing distance was with the kindergarten
boys. The resulting ANCOVA showed that there were no significant
differences in throwing distance among the throwing stages for the
kindergarten boys.

The range of stages was greater for the girls, although very few
girls threw using stage 1 or 2 patterns. In comparing the girls in
grades 1 through 4 who used the stages 3 through 5, stage 5 was
significantly different from stages 3 and 4, but stage 3 ad 4 were not
significantly different from each other. The fifth grade girls used

only stage 4 and stage 5 throwing patterns. When their performance was

88

analyzed, a significant difference in throwing distance was found
between stage 4 and stage 5 throwers. Finally, the throwing distance of
all the girls was compared with stages 1, 2, and 3 combined into an
immature pattern group. The results showed that there was a significant
difference between stages 5, 4, and the immature stages in throwing
distance.
Summary of Results

T'ne purpose of this study was to investigate l) the difference
between girls and boys in the performance of a distance throw: 2) the
relationship of grade level to the distance throw; and 3) the
relationship of developnental throwing stage to the distance a ball can
be thrown. The following results were obtained:

1. ANOVA procedures, yielded significant gender and grade
effects. A significant grade by gender interaction also was
fond. Boys threw the ball significantly farther than the
girls.

2. Children in kindergarten, first, and secod grade were
signficantly different from each otter and from children in
grades 3, 4, and 5 in how far they could throw a ball.

3. There was no significant difference in mean throwing distance
between third, foirth, and fifth graders.

4. The range of stages for the boys was so limited that the effect
of throwing stage on the distance throw could only be analyzed
at tne kindergarten level.

5. campariscn of kindergarten boys' distance throws with their
throwing stage showed an improvement in mean scores from stage

to stage (collapsing stages 1, 2, and 3). Men adjusted for the

10.

89

covariates of push-ups and height, however, the mean scores for
throwing distance by stage were not significantly different.

For girls in grades 1 through 4, throwing distance for first and
second graders was significantly different from that of the
third and fourth graders. However, the mean throwing distance
of first graders was not significantly different from that of
the second graders, and the mean throwing performance of the
third graders was not significantly different from.that of the
fourth graders.

For girls in grades 1 through 4, girls with a stage 5 throwing
pattern could throw a ball significantly farther than girls with
a stage 3 or stage 4 pattern, but the means for throwing
distance of the stage 3 and stage 4 throwers were not
significantly different.

In comparing fifth grade girls, stage 5 throwers conld throw a
ball significantly farther than stage 4 throwers. The covariate
height, however, was not significant.

When examining all the girls by stage, and collapsing stages 1,
2, and 3 into»a new stage '3", a significant difference was
found between the new stage '3', stage 4, and stage 5

throwers. The covariate of height also was significant.

The covariates were significant factors in the throwing
performance of children when the ANOOVA.was used to compare
students of more than one grade. When ANCOVA involved students
within one grade (kindergarten boys-push-ups and height, and

fifth grade girls—-height) the covariates were not significant.

90

11. Of the variables stndied, push—ups had the lowest correlation

coefficient with throwing distance (r= .307, boys and girls
conbined: .262, boys: .198 girls).

Conclusions

Tl'e results suggest the following conclusions in reference to the

hypotheses proposed :

1.

The hypothesis that boys would throw a ball farther than girls
in their own grade was supported. For grades kindergarten g
throngh fifth grade, the boys threw a ball significantly farther I
than the girls. Even trough the analyses used the covariate of

grip strength to factor in sonne of the physical differences

 

n1..-“

between boys and girls, the boys still threw signficantly
farther than the girls.

The secod hypothesis proposed that menbers of the sane gender
in a higher grade will throw a ball farther than those in a
lower grade. This hypothesis was partially supported. The mean
distance throw increased for each of the grades with two
exceptions—the third and forth grade boys were approximately
equal as were the fourth ad fifth grade girls. Kindergarten,
first, and secod grades were significantly different from each
other and from each of the other grades, but the third, fourth,
and fifth grades were not significantly different from each
other. For the girls, the only successive grades that were
significantly different were secod and third grade. For the
boys, kindergarten and first grades were significantly different ‘
from all the other grades, but not from each other. The only

other grades that were significantly different from each other

 

91

were the second and fifth grades. Again there was no
significant difference between the third, fourth, and fifth
grades. Under the conditions of this study, the lower grades
(kindergarten through second grade) supported the hypothesis of
the higher grades throwing farther but the third, fourth, and
fifth grades did not.

The third hypothesis stated that children of the same grade and
gender who have a more mature stage of throwing will throw a
ball farther than children who use a less mature stage of
throwing. The m using height as a covariate showed that
stage 5 girls threw the ball significantly farther than stage 4
girls in grades 1 through 5. There was an improvement in the
mean throwing distances of stage 4 throwers in conparison with
stage 3 throwers, but the differences were not significant in
the grades one throigh four. There were not enough stage 1 or
stage 2 throwers to make any comparisons. men comparing all of
the girls, a category of imnature throwers was formed which
included stage 1, 2, and 3 throwers. In the ANCOVA conparing
the inmature stage category with stage 4 and stage 5, a
significant difference was found between all three groups of
throwers.

The kindergarten boys showed an increase in mean throwing
distance between the immature groip, stage 4, and stage 5
throwers but none of the three groups were significantly
different frou each other. Ninety-fox percent of the boys in
grades me through five used the mature throwing pattern (stage
5) . The large percentage of boys exhibiting the stage 5

 

92

throwing pattern precluded analyses using developmental throwing
stage as a main effect.

The narrow range of throwing stages used by the boys prevented
testing the hypothesis that a more mature stage of throwing
produces a farther distance throw. However, the differences in
throwing distance demonstrated by the girls who used the
different stages of throwing pattern lend strong support to this
hypothesis and suggest that the method of throwing a ball is
inportant to the distance a person achieves with one's throw.

Recormendations

 

a result of this study, the following recounendations are made:
This study should be replicated using younger boys to provide
the full array of stages for analysis of throwing behavior.

A longitudinal study with girls should be conducted, working on
their form to determine if an increase in their stage of
throwing increases their throwing distance.

A larger sample of throwers needs to be tested to see if the
percentages reported by Way et a1. (1979) for each stage of
throwing need to be revised.

Highly motivated girls need to be challenged in their throwing
ability at a young age to see if they achieve the same

performance results as boys.

 

APPENDICES

APPENDIX A

Parental Information Sheet and

Parental and Student Consent Forms

93

Parental Information Sheet:

Dear Parent,

I am planning to m the throwing ability of students in air elementary schoolin
fulfillment of my master's thesis at Michigan State University. I am determining
the relatiorship between the quality of the throwing motion and the distance the
ball is thrown. I am also examining the effect strength, height and weight play in
the process. Ihopethisstudywillprovideabetterunderstandingofhowtoplan,
implement and evaluate effective shysical. education instruction as it relates to
throwing.

Ifyouallow yourchildtopardcipateinthisstudy,he/she willbeevaluated
according to his/her height, weight, strength, the distance he/she can throw a ball,
and the form he/she uses in throwing the ball.

If yoi allow your child toparticipateinthissmdy, he/she willbe videotaped as
he/she performs a maximal throw for distance. Videotaping will permit me to
evaluate your child's throw according to the developmental stages of throwing
formulated by researchers at Michigan State University. Strength will be
measuredbythe numberofprsh-urs thathe/shecandoin30secondsandbya
hand dynamometer test. In the hand dynamometer test your child willsqueeze a
hand dynamometer as hard as he/she can. The dynamometer will register a
reading corresponding to how hard he/she squeezed it.

For this study, your child will be given an identification number to protect his/her
anonymity. We will use this identification number rather than the child's name
when reporting the results of this study.

These tests willbe administered in the school'sgymnasium orcnthe athletic field
by the phym'cal education irstructcr and study author, Randy Baker. Your child
willbetoldtheplrposeofthestudyand willbegiventestinstructions. Students
willbe informed thatthey maychocsenottoparticipateinthesetests.

I believe that this study willhave substantialbenefitsinlearning more aboutthe
fundamental motor skillof throwing. By learning more about the different factors
that effect motor skills teachers can be helped to im grove physical education and
sports grograms. Please help to achieve this goal by permitting your child to
participate in this study. Thank you.

Thetestingdatefcryoirchild'sclass, , is
If you have any questiors, please feel free to call me either at schoolor at home.

 

Sincerely,

Randall L. Baker

94

Parental Consent Form:

Motor Development Thesis Study

Consent Form

The purpose and extent of involvement in this project have been explained to
my satisfaction. I agree to my child(ren)'s participation in this goject. I
understand the risks involved and am free to discontinue participation at any time
without recrimination.

I understand thatif my child isinjured asaresultof mychild's participation
in this research project, emergency medical care will be provided if necessary, but
these and any other medical expenses must be paid from my own health irsurance
program.

I understand the results of my child(ren)'s participation will be treated with
strict confidence. I also understand that I must give my written permission before
such films may be used for educationalor other purposes.

Child(ren) '3 Name:

 

Parent's Signature:

 

Date:

 

 

Student Consent Form:

I have seen the videotape explaining what I am going to do in the
throwing study and I am willing to participate in this study.

 

APPENDIX B

Means and Standard Deviations of
All Variables

By Grade and Gender

97

Table 3 - 3

Means and Standard Deviations for Grip Strength (in Kg)

 

Total/ Standard
Grade Gender Means Deviation Cases
KINDERGARTEN TOTAL 11.12 2.15 46
BOYS 11.79 2.43 27
GIRIS 10.17 1.19 19
lST GRADE TOTAL 13.30 2.52 59
BOYS 14.51 2.09 29
GIRLS 12.12 2.36 30
2ND GRADE TOTAL 15.36 3.92 47
BOYS 18.06 3.87 19
GIRLS 13.53 2.75 28
3RD GRADE TOTAL 17.72 2.60 48
BOYS 19.02 2.41 22
GIRLS 16.61 2.25 26
4TH GRADE TOTAL. 20.69 3.24 48
BOYS 22.27 3.08 22
GIRLS 19.35 2.76 26
5TH GRADE TOTAL 23.03 5.05 55
BOYS 25.39 4.42 32
GIRLS 19.76 3.97 23
ALL GRADES TOTAL 16.93 5.36 303
BOYS 18.56 5.77 151
GIRLS 15.30 4.37 152

 

96

Table B - 2

Means andStandard Deviations for Weight (in pounds)

 

Total/ Standard
Grade Gender Means Deviation Cases
KINDERGARTEN TOTAL 52.12 7.78 46
BOYS 53.14 8.62 27
GIRLS 50.67 6.34 19
lST GRADE TOTAL 57.38 7.89 59
BOYS 59.20 6.40 29
GIRLS 55.62 8.85 30
2ND GRADE TOTAL. 64.27 12.10 47
BOYS 68.99 13.67 19
GIRLS 61.07 9.93 28
3RD GRADE TOTAL 74.01 14.75 48
BOYS 71.91 10.02 22
GIRLS 75.79 17.82 26
4TH GRADE TOTAL 84.12 16.03 48
BOYS 80.92 15.34 22
GIRLS 86.83 16.39 26
5TH GRADE TOTAL 91.83 18.38 55
BOYS 95 . 24 18 . 83 32
GIRLS 87.08 16.99 23
ALL GRADES TOTAL 70.77 19.57 303
BOYS 72.00 19.68 151
GIRLS 69 . 55 19 . 45 152

 

99

Table B - 5

Means and Standard Deviations of the Retest of Grip Strength mi Kg)

 

 

Total/ Standard
Grade Gender Means Deviation Cases
KINDERGARTEN TOTAL 10.87 2.24 45
BOYS 11.60 2.50 26
GIRLS 9.88 1.32 ]9
2ND GRADE TOTAL 14.83 3.92 45
BOYS 17.38 3.90 18
GIRLS 13.13 2.92 27
4TH GRADE TOTAL 20.94 3.64 47
BOYS 22.53 3.76 22
GIRLS 19.53 2.93 25
Table B - 6

Means and Standard Deviations of the Retest of Push-ups (nmnber

completed in 30 sec)

 

Total/ Standard
Grade Gender Means Deviation Cases
lST GRADE TOTAL 8.39 5.22 51
BOYS 8.90 5.20 21
GIRLS 8.03 5.28 30
3RD GRADE TOTAL 11.50 9.23 46
BOYS 15.68 10.38 22
GIRLS 7.67 6.01 24
5TH GRADE TOTAL 10.80 7.90 55
BOYS 11.62 8.25 32
GIRLS 9.65 7.40 23

 

LISTOFREFEm

LISTOFREFERENCEB

AAHPER youth fitness test manual, (1965). rev. ed., Washington: AAHPER.

Ames, L. B. (1937). Individuality of motor skill development. Joirnal
of the American Physical Therapy Association, 4_6_ (2), 122-127.

Asmussen, E. (1973). Growth in muscular strength and power. In G. L.
Rarick (E‘d.) , Physical activity: Human growth and development. New
York: Academic Press.

Atwater, A. E. (1970). Movement characteristics of the overarm throw: a
kinematic analysis of men and women performers. Dissertation
Abstracts International, 11, 5819A.

 

Bagmzi, J. A. (1979). The effects of graded weighted baseballs, free
weight training, and simulative isometric exercise on the velocity
and accuracy of a thrown baseball. Dissertation Abstracts
International, 40, 1343A.

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

Barrow, R. A. (1960). Effect of strength development of antagonistic
muscles on throwing performance. Completed Research in Health;
Physical Education, Recreation, and Dance, 3, 28.

Berger, R. A. (1962). Comparison of static and dynamic strength
increases. Research Mterly, £1 329-330.

Bowne, M. E. (1960) . Relationship of selected measures of acting body
levers to ball throwing velocities. Research Quarterly, fl: 392-402.

Broer, M. R. (1969). Efficiency of human movement. Philadelphia: W.
B. Saunders.

Brumfield, M. D. (1969). A comparison of the relationship of shoulder
flexibility and other selected factors to throwing performance by
college women. Dissertation Abstracts International, 32, 4807A.

Burnside, L. H. (1927). Coordination in the locomotion of infants.
Genetic Psychological Mungraphs, g (5) , 238-372.

Beach, D. F. (1950). Genetic develoglent of motor skills in children 2
through 6 years of age. Unpiblished doctoral dissertation,
University of Michigan, Ann Arbor, MI.

100

101

Deatrick, D. J. (1977). The use of the Apollo Exerciser in improving
softball throwing ability in seventh grade girls. ' (From Completed
Research in Health, Physical Education, Recreation, and Dance, _2_0_,
102)

Deutsch, H. M. (1969). A comparison of women's throwing patterns.
Dissertation Abstracts International, 30 632A.

 

Dohrmann, P. (1964). Throwing and kicking ability of 8-year-old boys
and girls. Research Quartegy, _3_5, 464-471.

 

Dusenberry, L. (1952). A study of the effects of training in ball
throwing by children ages three to seven. Research giarterly, 23, 9-

 

Ekern, S. R. (1970). An analysis of selected measures of overarm
throwing pattern of elementary boys and girls. Dissertation
Abstracts International, 31, 633A.

 

Eoff, N. L. (1985). The relationship of structural-maturational
variables to throwing performance Dissertation Abstracts
International, g, 2960A.

 

Espenschade, A., (1940). thor performance in adolescence. Morographs
of the Society for Research in Child Development, 1, 1-126.

 

Espenschade, A., (1960). Motor development. In W. R. Johnson (Ed.),
Science and Medicine of Exercise and Sports (pp. 419-439). New
York: Harper 5. Brothers Publishers.

Feltz, D. L., Landers, D. M., & Raeder, U. (1979). Enhancing self-
efficacy in high-avoidance motor tasks: A comparison of modeling
techniques. Jwrnal of Sport Psychology, 1, 112-122.

Flavell, J. & Wohlwill, J. (1969). Formal and functional aspects of
cognitive development. In D. Elkind & J. Flavell (Eds.). Studies in
cognitive development. New York: Oxford University Press

Fleishman, E. A. (1964). The structure and movement of physical
fitness. Englewcod Cliffs, NJ: Prentice-Hall.

Frederick, S. D. (1977) . Performance of selected motor tasks by three-,
four-, and five-year-old children. Dissertation Abstracts
International, 394 748A.

Gesell, A. (1928). Infancy and human growth. New York: Macmillian Co.

Glassow, R. B., Halverson, L. E., & Rarick, G. L. (1965). Improvement
of motor development and physical fitness in elementary school
children. Cooperative Research Project #696. Madfscn: University
of Wisconsin.

 

 

102

Glassow, R. B. 8 Kruse, P. (1960). rotor performance of girls age 6 to
14 years. Research Quarterly, 2, 426-433. '

60116, D., & Weiss, M. (1981) . The effects of model similarity and
model 'talk on self-efficacy and muscular endurance. Journal of Sport
Psychology, _3_, 17-29.

Gowitzke, B. A., & Milner, M. (1988). Scientific bases of human
movement. Baltimore: Williams & Williams.

Gutteridge, M. V. (1939) . A study of motor achievements of young
children. Archives of Psychology, 244, 1-178.

Halverson, H. M. (1931). An experimental study of prehension in infants
by means of systematic cinema records. Genetic Psychology
Nbrographs, _l_(_l_, 107-286.

Halverson, L. E., Lbberton, M. A., & Langendorfer, S (1982).
Development of the overarm throw: movement and ball velocity changes
by the seventh grade. Research _glarterly, _53, 198-205.

Halverson, L. E., Roberton, M. A., Safrit, M. J., & Roberts, T.
(1977) . Effect of guided practice on overhand-throw ball velocities
of kindergarten children. Research Quarterly, 48, 311-318.

Hansom, M. R. (1965). Motor performance testing of elementay school age
children. Dissertation Abstracts International, 27, 102A.

Hardin, D. H. 8. Garcia, M. J. (1982). Diagrostic performance tests for
elementary children (Grades 1-4) . Journal of Physical Education,
Recreation and Dance, 53, 48-49.

Hardison, D. J. (1971). The effect of arm-shoulder strength training of
selected muscle groups on softball distance throw for seventh and
eighth grade girls. Corpleted Research in Health, Physical
Education, moreaticn, and Dance, _14, 126.

Hartman, D. M. (1943) . The hurdle jump as a measure of the motor
proficiency of young children. Child Develoment, 14, 201-211.

Haubenstricker, J. 5. Seefeldt, V (1986) . Acquisition of motor skills
during childhood. In V. Seefeldt (Ed.) , Physical activity 8 well-
beim (pp 42-102). Washington: AAHPERD.

Haubenstricker, J., Seefeldt, V., Fomtain, C. s. Sapp, M. (1981). A
developmental sequence for kicking. Paper presented at the annual
convention of the Midwest District of the American Alliaroe of
Health, Physical Education, Recreation, and Dance. Chicago,
Illirois.

Hay, J. G. (1985). Thejbiomechanics of sports techniques (3rd ed.) .
Englewcod Cliffs, N. J.: Prentice-Hall, Inc.

103

Hicks, J. A. (1930). The acquistion of motor skill in young children.
In G. D. Stoddard (Ed.), University of Iowa Studies in Child
Welfare. University of Iowa, Iowa City, Iowa.

Hoffman, H. A. (1969) . The effect of instructioi and practice of
throwing ability of primary grade children. Dissertation Abstracts
International, _Zfl, 3761A.

 

 

Inhelder, B. (1971). Criteria of the stages of mental developrent. In
J. Tanrer s. B. Inhelder (we), Discussions on child develogtent.
New York: International Universities Press.

Jenkins, (1930) . A comparison study of motor achievements of children,
five, six, and seven years of age. New York: Bureau of
Publications, Teachers College, Columbia University.

Johnson, B. L. & Nelson, J. K. (1979). Practical measurements for
evaluation in physical education (3rd ed.) . Minneapolis: Burgess

Publishing Company.

Johnson, R. D. (1960). Measurement of achievement in fundamental
skills: elementary-school children. Dissertation Abstracts

Johnson, W. D. (1977) . A comparison of motor creativity and motor
performance of young children. Dissertation Abstracts International,
_3_8_, 4024A.

Kang, S. J. (1982). Effect of different ball sizes and weights and
throwing methods on distaroe and speed in ball throwing.
Dissertation Abstracts International, 1;, 1469A.

Keogh, J. (1965) . Motor performance of elementary school children. Los
Angeles: Department of Physical Education University of California.

Keogh, J. 5. Sugden, D. (1985). mtcr skill develcplent. New York:
Macmillan.

Kreghbaum, E. s. Barthels, K. M. (1985). Biotechanics: A qualitative
approach for stifling human movement. Minneapolis: Burgess
Publishing Company.

Leme, S. A. (1973). Developtental throwing patterns in adult female
performers within a selected velocity range. Completed Research in
Health,Physica1 Education, Recreation! and Dance, L6, 168.

Luedke, G. C. (1980). Range of motion as the focus of teaching the
overhand throwing pattern to children. Dissertation Abstracts
International, _4_2_, 1534A.

Lyon, W. R. (1961) . A cinematographical analysis of the overhand
baseball throw. Completed Research in Health, Physical Education,
Recreation, and Dance, _4_, 81.

104

Mahmoud, N. (1979). The effects of instruction and practice on the
overarm throwing pattern of preschool children. Dissertation
Abstracts International, fl, 4958A.

 

 

Malina, R. M. (1975). Anthroporetric correlates of strength and motor
performance. Exercise and Sport Sciences Reviews, 3, 249-274.

Malina, R. M. & Rarick, G. L. (1973). Growth, physique, and motor
performance. In Physical activity: Human growth and development,
Ed. G. L. Rarick. New York: Academic Press.

McCaskill, C. L. & Wellman, B. L. (1938). A study of comon motor
achievements at the preschool ages. Child Development 2, 141-150.

McGraw, M. B. (1935). Growth, A study of Johnny and Jimmy. New York:
Appleton-Century.

Miller, J. L. (1957). Effect of instruction on development of throwing
for accuracy of first grade children. Research Quarterly, E, 132-
137.

Miller, S. E. (1978). The facilitatioi of fundamental motor skill
learning in gang children. Unpublished Doctoral Dissertation,
Michigan State University, East Lansing, MI.

Morris, A. M, Williams, J. M., Atwater, A. E., & Wilmore, J. H.
(1982) . Age and sex differences in motor performance of three
through six year-old children. Research Quarterly, 2' 214-221.

Nelsom, J. K., Thomas, J. R., Nelson, K. R., & Abraham, P. C. (1986).
Gender differences in children's throwing performance: Biology and
enviroment. Research Quarterly, 5_7_, 280-287.

Nicl'ols, B. A. (1971) . A comparison of 2 methods of developing the
overhand throw for distance in four, five, six and seven year old
children. Dissertation Abstracts International, 3_2_, 2469A.

Pedegana, L. R., Elsner, R. C., Roberts, D., Lang, J., 5 Farewell, V.
(1982) . Relationship of upper extremity strength to throwing
speed. American Joirnal of Sports Medicine, 12, 352-354.

Pinard A. & Laureendeau, M. (1969). "Stage“ in Piaget's cognitive-
developmental theory: Exegesis of a cooept. In D. Elkind & J. H.
Flavell, (£218.), Studies in Cognitive Develogent, (pp. 121-170).
New York: Oxford University Press.

Potter, D. L. (1963). A ompariscn of the effect of skill practice
sessions and upper extremity isometric exercise sessions om a sofball
throw for distaroe. Completed Research in Health, Physical
Education, Recreation, and Dance, 6, 94.

105

Richardson, W. R. (1976) . A study of the relationship between grip
strength, wrist flexion, arm length and the velocity of a thrown
baseball in male high school varsity baseball players. _Clmpleted
Research in Health, Physical Education, Recreationy and Dance, 1_9,
161-162.

Roberton, M. A. (1978). Stability of stage categorizations in motor
development. In D. M. Landers & R. W. Christina (Eds.), Psychology
of motor behavior and sport-1977, (pp. 494-506). Champaign,
Illirois: Human Kinetics Publishers.

 

Raberton, M. A. (1982). Describing "stages” within and across motor
tasks. In J. A. Scott Kelso, & J. E. Clark (Eds.), The develogtent
of motor control and co-ordination. New York: John Wiley & Sons,
LTD.

 

Roberton, M. A. (1984). Changing motor patterns during childhood. In
J. R. Thomas (Ed.), Motor develogent during childhood and
adolescence. Minneapolis, m: Burgess.

Roberton M. A. & Halverson, L. E. (1984). Develging children—their
changing movetent, a guide for teachers. Philadelphia: Lea &
Eebiger.

Roberton, M. A., Halverson, L. E., Langendorfer, 8., 5. Williams, K
(1979) . Longitudinal changes in children's overarm throw ball

velocities. Research Erterly, g9, 256-264.

Rowlands, D. J. (1962) . The effect of weight training exercise upon the
throwing power and strength of college baseball players. C_2_on_pleted
Research in Health, Physical Educatian Recreation, and Dance, 3, 96.

Ryan, T. M. (1977). A compariso'i of selected basic gross motor skills
of moderately retarded and rormal children of middle childhood age
utilizing the Ohio State University Scale of Intra Gross Motor
Assessment. Dissertation Abstracts International, 38_, 2650A.

Sanders, J. A. (1977) . A practical application of the segmental method
of analysis to determine throwing ability. Dissertation Abstracts
International, 33, 4661A.

Sapp, M. M. (1980) . The developient of galloping in yourg children: A
preliminary study. Unpublished masters profect, Michigan State
University, East Lansing, MI.

Schutzler, L. L. (1980). A cinematographic analysis of stride length in
highly skilled baseball pitchers. Completed Research in Health,
Physical Education, Recreation, and Dance, 33, 165.

Seefeldt, V. 8 Haubenstricker, J. (1974). A develogtental sequence for
kickigg. Unpublished manuscript, Michigan State University, East
Lansing, MI.

106

Seefeldt, V. & Haubenstricker, J. (1976a) . A developmental seqpence for
_hgpgigg. Unpublished manuscript, Michigan State University, East
Lansing, MI.

Seefeldt; V. & Haubenstricker, J. (1976b) . A develogental seglgnce for
m. Unpublished manuscript, Michigan State University, East
Lansing, MI.

Seefeldt, V. & Haubenstricker, J. (1976c). A developlental sequence for
strikigg. Unpublished manuscript, Michigan State University, East
Lansing, MI.

Seefeldt, V. & Haubenstricker, J. (1976d) . A developmental segience for
throwigg (rev. ed.). Unpublished manuscript, Michigan State
University, East Lansing, MI.

Seefeldt, V. & Haubenstricker, J. (1982). Patterns phases or stages:
An analytical model for the study of developmental movement. In J.
A. Scott Kelso, & J. E. Clark (Eds.), ghie development of movelent
control and co-ordinaticn. New York: John Wiley & Sons LTD.

 

Seefeldt, V., Reuschlein, S. & Vogel, P. (1972). Sequencing motor
skills within the physical education curriculum. Paper presented at
the annual convention of the American Alliance of Health, Physical
Education, Recreation, and Dance. Houston, Texas.

Seils, L. G. (1951). The relationship between measures of physical
growth and gross motor performance of primary-grade school
children. Research giarterly, _2_2_, 244-260.

 

Shirley, M. (1931). The first two years: A study of twenty-five
babies, in postural and locototor development. Minneapolis:
University of Minnesota Press.

Simri, U. (1974). Assessment procedures for human performance. In L.
A. Larson (Ed.) Fitness, health, and work capacity: International
standards for assessment. New York: Macmillian Publishing.

Singer, F. (1961). Comparism of the development of the overarm

throwing patterns of good and poor performers (girls). go_mpleted
Research in Health, Physical Education, Recreation, and Dance, 3, 82.

Straub, W. F. (1966). The effect of overload training procedures upon
the velocity and accuracy of the overarm throw. Dissertation
Abstracts International, Er 955A.

 

Sullivan, W. J. (1970). The effects of three experimental training
factors upon baseball throwing velocity and selected strength
measures. Dissertation Abstracts Internatiomal, 3;, 2162A.

Thomas, J. R. & French, K. E. (1985). Gender differences across age in
motor performance: A meta-analysis. Psychological Bulletin, _93,
260-282.

107

Toyoshima, S., Hoshikawa, T., Miyashita, M., 8 Oguri, T. (1974).
Contribution of the body parts to the throwing performance. In R. C.
Nelson 8 C. Morehouse (Eds.), Biomechanics IV. Baltimore:
University Park Press.

Van Slooten, P. (1973) . Performance of selected motor-coordination
tasks ty youngboys and girls in sex socio—economicproups.
Unpublished doctoral dissertation, Indiana University, Bloomington,
IN.

 

Vaughn, R. E. (1932). The relationship of selected kinematic and
anthropmetric measures to throwing velocity of baseball pitchers.
Dissertation Abstracts International, fl, 2931A.

Vincent, M. F. (1968). Motor performance of girls from 12 throdgh 18
years of age. Research Quarterly, 33, 1094-1100.

 

Way, D., Haubenstricker, J ., 8 Seefeldt, V. (1979). Performance
standard of selected motor skills for primary grade children.
Unpublished manuscript, Michigan State University, East Lansing, MI.

 

Wester. B. (1939). A comparison of the accuracy of throwing of third,
fourth, and fifth grade boys. Unpublished master's thesis.
University of Iowa, Iowa City, IA.

 

 

Wickstrom, R. L. (1983). Fundamental motor patterns (3rd ed.).
Philadelphia: Lea 8 Febiger.

Wild, M. R. (1938). The behavior pattern of throwing and sote
observations concerning its course of developrent in children.
Research 3034arterly, 3(3) , 20-24.

Williams, T. A. (1985). The effect of long distaroe throwing and weight
training on throwing velocity. Completed Regejrch in Health,
Physical Education,Recreation, and Dance, _21, 126.

           
   

   

 

HICHIGQN STRTE UN

Ill)" will)! mumllw

2 00908

   

 

 

 

 

 

LIBRARIES
WIHIHHWI
7747