av

 

‘ "p
. r11
Tr,

0.0!: '

 

..‘ van”.

in FLILSJ 2.... I
.flho.dtv.1.ron.tf..ulu
. . a n.......nflv.,.v.rxov1
v 5 -‘ . .

I:.- [P‘I .! ll?! -
2.3-1:!!!
13:...
.r I
I

.no.

.Lnya‘ -
.I. 53W. O

“3'
J“;-
3!:
:i"

7:; I

' i
2:: a
5.3%
J. '
.l

 

a”.

yr? .v. I .3! 1d»
5" v.vv$'t.b9.
' '0 Ithnffio f‘l.‘ .

lfi‘fit
"37*?

1'25
VAL!
2‘.
i
A
5‘

: s

‘vna‘bv
Viriv
Lt» 0...:

.

a.

VA:
-!:I‘llstlu
.2 "xix
. .9:

. A ;

(c x 0".
: '9...val
xvi.

910'.
A? (Pity...) .7‘!
‘:-\T n..¢D.tl-.I
7.!201522 .

if

5:}

I'l‘lr is
.1335 .3 f... 1.0.

. J»: 9155.391

L
in
.‘t

I.
1

§

4

5
7:9}

. n; .
'55"
c

9...}... .
Ever-.3203;
. . t.) .“Ertl’ttarrlmw. I; .
Q‘LL I...‘- ’0. I“: w“
iii E‘ . CI’YDI- I
I.‘l‘
‘ :1.
its,"

.. .»«J‘q.um_fi.rym£amuxr.

1.
I1. 11.! I... .0
53.1.7?! 3.1g

.t-lll

43%;!
flat!!! .I. IV. I

W" 6&5

MICHIGAN STATE

II II III III

 

 

 

 

 

 

 

 

 

 

 

IIIIIIIIIIIIIIIIIIIIIIIIIII

00895 7262

This is to certify that the

dissertation entitled

A KINEMATIC AND KINETIC ANALYSIS
OF THE OVERGRIP GIANT SWING ON
THE UNEVEN PARALLEL BARS

presented by

Winifred Witten

has been accepted towards fulfillment
of the requirements for

Ph- D degreein Exercise Science

 

 

i??&4 W/I-ZZU/

Major professor

 

MS U i: an Affirmative Action/Equal Opportunity Imtitution 0- 12771

 

LIBRARY
Michigan State
University

 

 

 

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

M.
DATE DUE DATE DUE DATE DUE
#994,849qu

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T" I

 

 

 

 

MSU is An Affirmative Action/Equal Opportunity institution
czmpma-pd

 

A KINEMATIC AND KINETIC ANALYSIS
OF THE OVERGRIP GIANT SWING ON
THE UNEVEN PARALLEL BARS

By

Winifred Hitten

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Physical Education and Exercise Science

1990

.J'

(5.x
4

",
. . Y—w v’
((0 /

'ABSTRACT
A KINEMATIC AND KINETIC ANALYSIS OF THE OVERGRIP GIANT SWING
ON THE UNEVEN PARALLEL BARS
8?

Winifred Hitten

This study investigated selected kinematic and kinetic
variables of the overgrip giant swing on the uneven parallel
bars. The subjects were 15 club gymnasts, who ranged in age
from 10 to 18 years. Four judges ranked the 15 overgrip
giant swings. Based on the judges' rating, three groups were
formed. Group 1, the highly skilled, and Group 3, the lower
skilled, were compared on kinematic and kinetic variables.

The top rail of the bars was instrumented with strain I
and torque gages to measure forces and torques directly on
the bar. Joint reaction forces and net muscle moments were
calculated for the elbow and shoulder Joints. Comparisons
between the two groups were made on absolute and temporal
occurrences of events within the skill, body segment
orientations, Joint angles, body segment velocities and
acceleration patterns and the path of the total body center
of gravity.

There were significant differences between highly
skilled and less skilled gymnasts for the horizontal

velocity of the total body center of mass (4.9 and 4.4

m/sec, respectively) and the relative time of occurrence of
the maximum vertical velocity of the center of mass.
Significant differences were shown between the groups for
peak velocity of the ankle and hip. Group 1 had a velocity
of 11.6 m/sec for the ankle and 6.2 m/sec for the hip. Group
3 showed velocities of 9.6 and 5.6 m/sec, respectively.
There were also significant differences for several
segmental velocities and accelerations (foot, shank and
arm), which were due to body position changes as the
gymnasts completed the descent phase of the skill.

There were no significant differences between the
groups on the kinetic variables. The total group mean
maximum resultant force on the bar was 3.1 times body.
weight, which occurred in the second quadrant of the skill.
The maximum mean torque on the bar was 41.3 Nm. Nean elbow
joint reaction forces were 1.6 times body weight for the
vertical force and 0.7 times body weight for the horizontal
force. The shoulder joint calculations were 1.6 and 1.3
times body weight for the vertical and horizontal forces

respectively.

ACKNOHLEDGEHENTS

The completion of this dissertation is the result of
the support, advice and help from a number of individuals. I
would like to formally thank each of them here.

First, I would like to express my gratitude to Dr.
Eugene H. Brown, Dr. V. Dianne Ulibarri, Dr. Crystal E.
Branta and Dr. Michael Kasavana for their time and help as
members of my doctoral committee. Additional thanks to Dr.
Brown and Dr. Ulibarri for their biomechanics expertise and
advice in designing and implementing the study.

Second, I would like to thank three important people,
who provided the technical help necessary to implement the
study. In addition to the moral support given over the
years, Dr. Charles X. Hitten, Eastern Nichigan University,
applied the strain gages to the bar. Robert Hells from the
3 Center for the Study of Human Performance provided the
electronic support to gather the force data and convert the
data to the computer for analysis. Dr. Nitten and Mr. Hells
spent many hours designing the hardware and calibrating the
instruments for use in this study. Dr. Hohammed Usmann, who
was in the Department of Engineering at MSU at the time of

the study, provided computer support for changes in the

iv

analysis program as well as providing many of the graphs
used in the study.

Third, I would like to give special thanks to the
American Athletic Institute and Mr. Larry Fie for the

donation of the graphite uneven parallel bar for use in this

study.

TABLE OF CONTENTS

LIST OF TABLES .

LIST OF FIGURES

CHAPTERS
I. Introduction
A. Statement of the Problem .
B. Purpose of the Study .
C. Need for the Study .
D. Hypotheses .
E. Limitations .
G. Definitions of Terms .

II.

III.

A.

MOOW

Review of Literature .

Gymnastics Studies that have Used
Instrumented Equipment .
1. General Calibration Procedures .

2. Calibration Procedures for Instrumented.

Bars .

. Biomechanical Studies of Skills on the

Uneven Parallel Bars.

. Studies of the Overgrip Giant on the

Horizontal Bar

. Studies of the Overgrip Giant Swing .

on the Uneven Parallel Ears .

. Summary of the Review of Literature :

Experimental Procedures .

A.

FUND O U

General Procedures .

1. Subjects. . . .
2. Filming of Performance Sequence .
3. Judges and Ratings of Performance .

. Instrumentation and Calibration of the

Bar

. Cinematographic. Techniques for Data

Collection and Analysis.

. Kinematic Analysis .
. Kinetic Analysis . . .
. Statistical Analysis .

vi

ix

.xii

H

bdbbem

ll

IV. Results and Discussion

NU OW>

. General Characteristics of the Subjects.
. Group Placement and Characteristics.
. Objectivity of Judges' Ratings

of Performances .
Reliability of Digitized Data.

. Kinematic Analysis .

1. Temporal Analysis . . .

2. Hip and Shoulder Angles .
a. Hip angle. .
b. Shoulder angle .

3. Center of Mass Analysis . . .
a. Path of the center of mass .
b. Peak resultant velocity of

the center of mass .

c. Acceleration of the center of mass .

d. Angular velocity and angular
acceleration .
4. Joint Analysis. . .
5. Body Segment Analysis .

. Kinetic Analysis .

1. Forces Applied to the Bar 2
2. Joint Forces and Moments.

V. Summary and Conclusions .

D.
E.
F.

. Summary of Procedures .
. Summary of Findings for the Kinematic

Variables. . .
1. Center of Mass. Analysis .
2. Joint Analysis.

3. Segment Analysis.

. Summary of Findings for the Kinetic

Variables . .

1. Forces Applied to the Ear .

2. Joint Forces. . . .
Conclusions.

Recommendations for Coaches. . .
Recommendations for Further Research .

VI. List of References.

APPENDIX A

Informed Written Consent Form .

vii

I119

120

.125

APPENDIX 9
Information Sheet for Data Collection

APPENDIX C
C.1 Judging Sheet for Giant Swings on the
Uneven Parallel Bars . . . . .

C.2 Summary of Judges' Ratings.

APPENDIX D
Statistical Summary Tables

viii

.127

.129
.130

.131

4.11

LIST OF TABLES

Summary of Instrumentation Studies .

' Summary of Uneven Parallel Bar Research

Means and Ranges of Selected Anthropometric
Measurements .

Selected Comparisons by Group

Analysis of Variance Summary Table and Objectivity
Calculation

Reliability Correlation Coefficients of Selected
Variables

Temporal Data for Giant Swings for the Total
Group, Group 1, and Group 3 . . . . . .

Mean Peak Resultant Velocity of the Center of Mass .

Mean Peak Resultant Velocity of the Selected Joints.

Mean Peak Resultant Acceleration of the Selected
Joints

Mean Peak Resultant Velocities of the Selected
Body Segments

Forces and Moments Applied to the Bar

Times of Occurrences of Peak Forces

Mean Peak Joint Forces and Moments at the Elbow and

Shoulder for the Total Group .

Mean Peak Velocities of Total Body Center of Hass

and the Times of Occurrences of Mean Peak Velocities
.131

Expressed as Percents of Total Time

Mean Peak Velocities for Total Body Center of
Mass Per Quadrant . . . . . . . . .

ix

18
21

57
59

86

. 61

. 63

72

. 81

. 83
. 98

98

.101

.132

.10

.11

.12

.13

.14

.15

.16
.17

.18

.19

Times of Occurrences of Mean Peak Velocities of
Total Body Center of Mass Per Quadrant Expressed
as Percents of Total Time. . . . . . . . . . . . . .133

Mean Peak Accelerations of Total Body Center of Mass
and Relative Times of Occurrences . . . . . . . . .134

Mean Peak Accelerations of Total Body Center of Mass

Per Quadrant . . . . . . . . . . . . . . . . . . .135
Times of Occurrences of Mean Peak Accelerations

of Total Body Center of Mass Per Quadrant . . . . .136
Mean Peak Resultant Peak Velocities of Joints. . . .137

Times of Occurrences of of Mean Peak Resultant
Velocities Expressed as Percents of Total Time . . .138

Mean Peak Resultant Velocities of the Selected

Joints Per Quadrant . . . . . . . . . . . . . . . .139
Times of Occurrences of Mean Peak Resultant

Velocities Expressed as Percents of Total Time . . .141
Mean Peak Resultant Accelerations of Selected

Joints . . . . . . . . . . . . . . . . . . . . .-. .143
Times of Occurrences of Mean Peak Resultant

Accelerations Expressed as Percents of Total Time. .144

Mean Peak Velocities of the Center of Mass of

Selected Segments . . . . . . . . . . . . . . . . .145
Times of Occurrence of Mean Peak Velocities of

Selected Body Segments Expressed as Percents

of Total Time . . . . . . . . . . . . . . . . . . .147
Mean Peak Accelerations of the Selected

Body Segments . . . . . . . . . . . . . . . . . . .149
Mean Peak Joint Reaction Forces . . . . . . . . . .150
Times of Occurrences of Mean Peak Joint Forces

and Moments Expressed as Percents of Total Time . .151
Joint Reaction Forces and Moments of the

Shoulder Per Quadrant. . . . . . . . . . . . . . . .152
Times of Occurrences of Joint Reaction Forces

and Moments of the Shoulder Per Quadrant Expressed
as Percents of Total Time . . . . . . . . . . . . .154

D.20 Joint Reaction Forces and Moments of the Elbow
Per Quadrant . . . . . . . . . . . . . . . . . . . .156

0.21 Times of Occurrences of Joint Reaction Forces

and Moments of the Elbow Per Quadrant Expressed
as Percents of Total Time . . . . . . . . . . . . .158

xi

LIST OF FIGURES

Figure

3.1 Skill Sequence on the Uneven Parallel Bars.

3.2 Skilled Performance of an Overgrip Giant Swing
on the Uneven Parallel Bars (Subject 16)

3.3 Unskilled Performance of an Overgrip Giant Swing
on the Uneven Parallel Bars (Subject 6)

3.4 Strain Gage Configuration

3.5 Calibration Device

3.6 Filming Area and Equipment Set-up .

3.7

In a. as .a o» co
(n a. co to e- «m

CDCDQG

.11
.12

abbhbbhifib
p
Q

.13

Performance of an Overgrip Giant Swing Showing
Quadrants (Subject 16). . .

Link Segment Model of Forearm and Arm .
Hip and Shoulder Angles (Subject 16).
Hip and Shoulder Angles (Subject 6)
Path of Center of Mass (Subject 16)
Path of Center of Mass (Subject 6)

Angular Velocity of Center of Mass of the
Whole Body (Subject 16) . . . . . .

Resultant Velocity of Ankle and Hip (Subject 16).

Resultant Velocity of Ankle and Hip (Subject 6)

Angular Velocity of Trunk and Shank (Subject 16).

Angular Velocity of Trunk and Shank (Subject 6)
Resultant Force Applied to the Bar (Subject 16)
Resultant Force Applied to the Bar (Subject 6).
Vertical Force Applied to the Bar (Subject 16).

Vertical Force Applied to the Bar (Subject 6)

xii

35

. 38

. 39

40
42
46

49
51
85
68
89
70

74
77
78
84
85
87
88
90
91

4.14 Horizontal Force Applied to the Bar (Subject 16)
4.15 Horizontal Force Applied to the Bar (Subject 6)
4.16 Moment Applied to the Bar (Subject 16).

4.17 Moment Applied to the Bar (Subject 6)

xiii

93

96
97

CHAPTER I

Introduction

The uneven parallel bar event is one of the most
dynamic and exciting events in women's gymnastics. The event
requires hanging, swinging, circling, kipping, and
supporting movements (Bowers, Fie, & Schmidt, 1981). In
recent years, large swinging movements, movements through
the handstand position, and releasing movements have been
added to the gymnast's requirements for success in this
event (Federation of International Gymnastics, 1985). In
addition, a routine is composed of movements that are
performed on both bars, change direction, and show flight.
The routine contains a minimum of ten skills, that come from
four difficulty groups ( "A“, "B", "C", and "D”). Age group
competitors are required to have a minimum of three “A“ ,
three ”B", and two "C" skills. A "D" level skill is from the
most difficult group. It is not a requirement of
competition until a gymnast reaches the elite level.

Prior to 1950, both men and women performed on the
parallel bars. Homen could narrow and lower the bars to
accommodate their smaller frames (Tonry, 1980). The two
wooden rails were oval-shaped and relatively stiff in
comparison with the modern apparatus. The uneven parallel
bars first appeared in Olympic competition in 1952 at
Helsinki (Bonniwell, 1983). In the early years of

competition on the uneven parallel bars, it was common to
see static pose handstands, leg circling, and climbing
movements interspersed with kipping and swinging movements.
More recently, the inclusion of releasing and large swinging
movements and the common practice of maintaining continual
movement during the routine has resulted in a more dynamic
character to the event.

The evolution of the skills on the uneven parallel bars
were accompanied by changes to the apparatus itself. In the
1950's and continuing into the late 1960's, the uneven
parallel bars were composed of one bar from the men's
parallel bars and an extension bar to achieve the
appropriate height for the second bar. Because the modified
parallel bars were not designed to be stable under large
lateral forces, many of the swinging movements performed by
the female gymnast would lift the base of the bars off the
floor. It was common to see gymnasts sitting on the base of
the modified parallel bars to stabilize the apparatus during
competition. The current practice of either cabling the bars
to the floor or structuring wide and weighted free standing
bases began in the 1970's. The rails themselves have become
more flexible and rounder. The material for the bars has
evolved from wood to a wood laminate over fiberglass and,
more recently, to a bar constructed from a graphite and
fiberglass composite with a wood laminate. With the

inclusion of the larger swinging and releasing movements,

the bars are being adjusted farther apart. This has caused
the impact skills, characterized by the hips hitting the
bars, to disappear from most modern uneven bar routines.

3 There has been very little research conducted on the
uneven parallel bars event. The majority of studies that
have been conducted have analyzed skills that are not
currently being used and/or utilize college women as
subjects (Landa, 1974; Hay, Putnam & Wilson, 1979; and
Evans, 1984). The current investigation examined the
overgrip giant swing. This skill has become a necessary
movement for the repertoire of the gymnast as it is the
basis for many large release movements so popular today.
This study used young female gymnasts, between the ages of
10 and 16, the age group that has the highest participation

rate in this country.

Statement_nf_tha_fiznhlam
The problem in this study was to investigate and
compare selected kinematic and kinetic characteristics
associated with the overgrip giant swing performed on the
uneven parallel bars by female, age group, Class I and Elite
gymnasts. A subproblem in this study was to develop a
technique for measuring forces generated on the bar and on

selected joints of the body by gymnasts circling the bar.

EHLRQS£_Q£_Lh§_SLnd1
The primary purpose of this study was to compare

selected kinematic and kinetic parameters obtained from two
groups of gymnasts who performed the overgrip giant swing on
the uneven parallel bars. A panel of Class I and Elite
judges identified one group as “highly skilled“ and the
other group as "less skilled". By comparing the highly
skilled and less skilled performers, it was the intent of
this study to identify critical factors that will aid the

teacher/coach in understanding this skill.

Need_fnr_tha_Stnd¥

The need for the study arose from the paucity of
quantitative research found in the literature on women's
gymnastics. Many of the studies that are available are
outdated because of the frequent changes in the sport,
especially in the uneven parallel bars event. Two studies
(Hay, et al., 1979; and Landa, 1974) analyzed skills that
are not frequently used today. Other studies (Sands, 1982;
1985a; 1985b; Bentham, 1987; and Hay, et al., 1979) used
either single subjects or very small populations. Still
other studies (Bovinet,1979; Evans, 1984; Hay,et al., 1979;
and Lands, 1974) used college aged women as subjects. While
this is a convenient population to draw from for university
research, it is not the age group with the majority of

participants. According to the United States Gymnastics

Federation (USGF) statistics, 86 percent of registered
gymnasts are between the ages of 10 and 16, with the average
being 12 years old (J. Clair, personal communication, June
26, 1990).

Equipment changes, such as the use of a graphite bar on
the uneven parallel bars, also necessitate further
investigation. Along with changes in equipment has come an
evolution of the type of skills that are being performed.
With additional stability to the apparatus, the movements on
the uneven parallel bars have become more similar to
horizontal bar skills performed by male gymnasts.

The overgrip giant swing was chosen for study because
it is a skill that has been widely performed in recent
years. It is being viewed as a basic prerequisite for more
advanced, large swinging, and releasing movements that
characterize today's uneven parallel bars routine.
Quantitative investigations are necessary to generate
knowledge about technique for teaching the overgrip giant
swing to young girls. The technique of the skill is
different from the same skill performed by men on the
horizontal bar. The proximity of the low bar of the uneven
parallel bars, places constraints on the execution of this
skill, with which the male gymnast does not have to contend.
Data concerning the optimal body position during the descent

phase of the circle need to be collected.

There is a paucity of research using young girls as
subjects. Information concerning the forces that young girls
experience while performing skills on the uneven parallel
bars is nonexistent. In order to understand injury
potential, there is a need to generate data on joint forces
experienced by subjects performing on the uneven parallel

bars.

Blunthasaa
Three hypotheses were established to study the overgrip
giant swing as performed on the uneven parallel bars. These
hypotheses were:

1. There will be no difference, at the .10 level of
significance between the highly skilled and less
skilled groups in selected kinematic parameters.
These parameters include:

a. path and angular velocity of the gymnast's whole
body center of mass around the bar,

b. body position characteristics in each phase of
the overgrip giant swing, and

c. linear velocity and acceleration histories of
selected joints and angular velocity and
acceleration histories of selected body
segments.

2. There will be no differences at the .10 level of

significance in the force histories experienced at

the bar and at the elbow and shoulder joints between

and among the two skill level groups.

. There will be no differences at the .10 level of

significance in the net moments about the elbow

and shoulder joints between and among the two skill

level groups.

The study used:

1. Class I or Elite gymnasts in the state of

Michigan,

. volunteer subjects ranging in age from 10 to 16

years, and
four Class I or Elite judges, who are members of the
National Association of Women's Gymnastics Judges

(NAWGJ).

The results of the study were interpreted

with the following limitations:

1.

Data were collected on overgrip giant swings

performed in a noncompetitive environment.

. The use of Dempster's (1955) segmental data for

young girls, may have introduced a source of error.

. The model developed for joint reaction forces

assumed that the force of the bar acted at the

center of the wrist joint.

. Digitization of high speed film may have introduced

a source of error.

. The use of accelerations, generated from position

data, introduced a source of error.

. The analysis performed was two-dimensional. It was

assumed that the gymnasts circling the bar would

remain in the sagittal plane.

. The friction forces between the gymnasts' hands and

the bar were ignored.

. Overgrip giant swing - This movement is initiated

from a handstand position on top of the upper bar.
The gymnast rotates with the chest leading the
movement, one revolution around the bar, returning
to the handstand position. Sometimes, but not
consistently, this movement is called a backward

giant swing (See Figure 3.1, p. 35-36).

. Tap - This is a term used to denote a small, quick

body position change from an arch to a pike that
aids the gymnast in accelerating the upswing phase

of a movement.

. Force transducer - This is a device that measures

the force exerted by the gymnast on an external body
and gives an electrical signal proportional to the
applied force. The applied force causes a small
amount of strain in the transducer, resulting in an
unbalance of voltages proportional to the force
(Winter, 1979).

Strain gage - This is a small instrument that
measures the amount of deformation in a material and

translates this deformation into a measure of force.

. Age group gymnast — This is a gymnast who

participates in the USGF national program,
subscribing to the rules and regulations of that

governing body.

. Class I gymnast - At the time of this study, the

Class I gymnast was the highest level of age group

competition in the United States.

. Elite gymnast - This is a gymnast who has been

tested on compulsory elements and has achieved
scores that indicate her to be capable of

international level competition.

. Shoulder angle - The angle calculated between the

arm and trunk. Shoulder extension is a decrease in
the shoulder angle with full flexion being at 180
deg.

l0

9. Hip angle - The angle calculated between the trunk
and thigh. Hip flexion is a decrease in the hip

angle with full extension being at 180 deg.

CHAPTER II

Review of Literature

Kinematic descriptions of various gymnastic activities
have appeared in the gymnastics and biomechanics literature
since Cureton's (1939) early work on cinematographical
techniques for skill analysis. These studies have
contributed a great deal of knowledge about the angles of
joints during specific phases of movements, angles of take-
off from the floor and apparatus, and the velocities of body
segments. Vaulting appears to be the most frequently
analyzed gymnastic event. Skills on the horizontal bar
appear to be the second most frequently analyzed. There is a
lack of research for the pommel horse, balance beam, and
uneven parallel bars (Kreighbaum, 1983). The review of
literature that follows is organized in five parts:

1. gymnastics studies that have used instrumented

equipment,

2. biomechanical studies of skills on the uneven

parallel bars,

3. studies of the overgrip giant swing on the

horizontal bar,

4. studies of the overgrip giant swing on the uneven

parallel bars, and

5. summary of the review of literature.

11

12

G l' SI I' II | l H l I | l l E . |

In studying gymnastics movements, the direction and
magnitude of forces applied to the performer gives the
researcher important information concerning the skills to be
analyzed. Several methods of directly measuring the
horizontal and vertical forces exerted on gymnastics
equipment and indirectly on the performer have been reported
in the literature. Force transducers have been used in
several studies. The strain gage is the most common type of
transducer used to instrument gymnastics equipment. A
variety of skills on various pieces of apparatus have been
analyzed via strain gage technology.

In a study of women side horse vaulters, Kreighbaum
(1974) applied eight strain gages in a Wheatstone bridge
circuit to the bottom of a vaulting board. Deflections of
the board were recorded on photosensitive paper by a
galvanometric type of oscillograph. Results of the force
data, combined with kinematic data collected by filming the
vaulters, gave the researcher information concerning force-
time parameters of the vaulter-board interaction during
take-off and the mechanical responses of the board during
impact.

Wilkerson (1978) examined mean forces and impulses of
11 female gymnasts performing back handsprings on a balance

beam that had been instrumented with specially constructed

13

force platforms placed between the balance beam upright
supports and the beam. The force platform instrument
consisted of two mini-platforms that were bridged together
by steel for stabilization. Each mini-platform contained
strain gages for measuring pressure sensitivity in three
orthogonal planes.

Robertson, Paul, and Nicol (1985) studied "kip" actions
on the parallel bars, high bar, and rings. The authors
constructed instruments to measure the loads transmitted
through the hands to the respective apparatus so that the
overall pattern of resultant forces and resultant moments at
each major articulation could be assessed. For the rings,
the authors applied strain gages to a "proving ring" that
was mounted at ceiling level so it could rotate into the
line adopted by the cables at any instance. For the parallel
bars, the force measuring equipment was a custom made six-
quantity force transducer placed between the bar attachment
and the frame at each end of the bars. For the high bar,
strain gages were directly applied to the bar.

Several researchers have investigated a variety of
skills on rings that have been instrumented to collect force
data. Sale and Judd (1974) developed a method for directly
measuring the force developed on the still rings and for
synchronizing the force and cinematographic records. Since
force exerted on the rings is transmitted by the ring cables

to the suspension points, the authors connected a load cell

14

in series between the ring cable and the suspension point.
The load cell consisted of a 3/16" thick aluminum diaphragm
welded to an external ring. Eight strain gages were
attached to the diaphragm, four per side at 90 degree
intervals. Nissinen (1983) placed two Kistler force
transducers inside the metal framework of the rings just
above the attachment of the ring cables. The force signals
were recorded and stored on magnetic tape and then sent to a
computer for analysis. The author investigated reaction
forces, displacements of the total body center of gravity,
and segmental velocities and accelerations of subjects
performing giant swings on the rings. Cheetham, Mizoguchi,
and Sreden (1987) placed transducers in series with the
cables of the rings. They examined the forces on thirteen 10
to 12 year-old-male gymnasts performing dislocates on the
rings. The forces of the gymnasts were displayed on a
storage oscilloscope. The storage oscilloscope and gymnasts
were simultaneously video taped at 200 frames/sec via split
screen capability. This system offered immediate feedback to
the gymnast and coach.

The men's horizontal bar has been a frequently
instrumented piece of equipment. In 1976, Zinkovsky, Vain
and Term reported the use of strain gages in conjunction
with cinematography and electromyography. The authors also
attached an accelerometer to the back of their subject in

order to study the long underswing upstart (kip) on the

15

horizontal bar. The authors were interested in the patterns
of kinetic energy and muscular contractions. Kopp and Reid
(1980) compared the forces and torques applied to the
horizontal bar by gymnasts during the performances of two
styles of giant swings. Four gages were bonded to both ends
of the bar in such a manner that the resultant strain
measurement would be independent of the point of application
of force along the bar. Four torque gages were placed in a
similar manner. The vertical and horizontal forces and
torques were amplified on three channels of a Grass
Polygraph. A fourth channel was used to monitor the internal
timer from a LOCAM 16mm camera. This allowed the
synchronization of film analysis data with the strain and
torque gage output. A stroboscope was used to synchronize
the exact start and end of each trial. The force and torque
data were stored on a tape recorder for later analog to
digital conversion and computer analysis. Ishii and Komatsu
(1987) used strain gages applied to a horizontal bar and
cinematography to examine performances of the overgrip giant
swing. The four gages were placed 20 cm from each end of the
bar to collect horizontal and vertical forces on the
performances of overgrip giant swings by five gymnasts.

Two studies reported in the literature used strain
gages to examine forces generated on the uneven parallel
bars. Hay, et a1. (1979) applied strain gages to a Nissen

wood laminate bar to investigate maximum forces exerted by

16

women performing typical uneven parallel bar skills. These
strain gages were applied along the neutral axis at a
distance of 56 cm from one end of the bar to measure
horizontal and vertical bending . Evans (1984) studied
performances of the clear back hip circle on the uneven
parallel bars. In addition to film data, strain gages were
attached along the neutral axis of the bar 56 cm from each
end. The strain gages were connected to a galvanometric type
oscillograph, which measured bar deflections by means of a
light tracing on photosensitive paper. Evans (1984) reported
five prominent deflections in the horizontal direction. The
data for the vertical deflections was unreported due to

failure of the strain gages for this direction.
3 1 C 1.] I' E I

Several researchers have documented the methods that
they used to calibrate their instrumentation. Kreigbaum
(1974) calculated a calibration curve by vertical static
'loading of a vaulting board for nine positions, four inches
apart. The forces were determined by relating the position
of the vaulters' metatarsophalangeal joint on the board to
the calibration curve. Wilkerson (1978) calibrated a strain
gage type force platform, constructed for the balance beam,
by statically loading the beam in the vertical, horizontal,

and lateral directions. The static loading on the horizontal

17

and vertical axes was performed by manually pulling on the
beam in each respective direction. The force of pull was
determined by an electric cable tensiometer. The author
performed a series of tests to statically and dynamically
validate the instrument. A summary of the findings from
instrumentation studies reviewed in this section are found

in Table 2.1

C 1.] I' E l E I I E

In calibrating the horizontal bar, Kopp and Reid (1980)
applied known weights in the horizontal and vertical
directions and known clockwise and counterclockwise torques.
For each calibration, linear force-voltage and torque-
voltage relationships were found. The authors also found
that there was no interaction between the horizontal and
vertical force components and that the transducer output was
independent of the position of application of the load. In
order to determine the relationship between the strain and
electric signals, Ishii and Komatsu (1987) statically loaded
the bar in the middle and at 30 cm to the right and left of
center. The readings of these signals produced a calibration
curve. Hay, et a1. (1979) statically loaded an uneven
parallel bar at the geometric center of the bar and found
the system to respond linearly throughout the full range of

force from zero to failure. The authors derived a correction

18

Table 2.1 Summary of Instrumentation Studies

 

Author Event/ Instru- Calibration Number of
(Date) Skill mentation Method/ Subjects
Type Validation
Test

Sale, Judd Rings/ Load cell Static 4
(1974) Giant swing

Zinkovsky, *HBar/ Strain **NR NR
et al.(1976) Kip

Wilkerson Beam/ Force Static/ 11
(1978) Handspring platforms Dynamic

Hay, et a1. UBars/ Strain Static 3
(1979) Kips,beats gage

Kopp,Reid HBar/ Strain, Static ‘ 6
(1980) Giants torque

gages

Krieghbaum Vault/ Strain Static 8
(1983) Handspring gage

Nissinen Rings/ Transducer NR 60
(1983) Giant swing

Evans UBars/ Strain NR 4
(1984) Clear hip gage

circle

Robertson, PBars,Rings, Strain Static NR
et a1. HBar/ Kip gage,

(1985) transducers

Cheetham, Rings/ Transducer NR 13
et a1. Dislocate

(1987)

Ishii,. HBar/ Strain Static 5
Komatsu Giant gage

(1987)

 

*HBar - Horizontal bar.

**NR - Not reported.
UBars - Uneven parallel bars.
PBars - Parallel bars.

19

factor to enable the magnitudes of the recorded forces to be

adjusted whenever the bars were loaded eccentrically.

B' l . 15' l' ESl'JJ ll 1! 2 111B

There is a paucity of research on skills performed on
the uneven parallel bars. This is apparent from the number
of studies found in the literature (see Table 2.1). Also,
the studies that have been conducted investigated skills
that are seldom seen today (Landa, 1974; Hay, et al., 1979)
or are single subject studies of high level gymnasts (Sands,
1982, 1985a, 1985b; Bentham, 1987). Other studies (Landa,
1974; Ray, et al., 1979; Evans, 1984) have been conducted on
university women gymnasts, even though, most participants in
the sport of gymnastics today are below the age of 17 years.

The majority of the studies on skills performed on the
uneven parallel bars have used cinematography as the primary
method of data collection. In these studies frame rates
varied considerably. The lowest frame rate reported
(Bovinet, 1979) was 30 frames per second and the highest was
150 frames per second (Bentham, 1987). The most frequently
reported frame rates were 80 and 100 frames per second
(Sands, 1982, 1985a, 1985b 1987; Evans, 1984; Daniels,
1983).

In order to analyze movements on the uneven parallel

bar, several authors have divided the skills into

20

subcategories or phases. Sands (1985a) studied one gymnast
performing a low bar kip and release to a catch of the high
bar. The skill was divided into four phases:
1. from extension below the low bar to hip flexion,
2. from hip flexion to hip extension,
3. from release of the low bar to grasp of the high
bar, and
4. from the grasp of the high bar to the top of the
backswing.

Sands (1985b) used a similar procedure for his analysis
of a straddle back to handstand. Daniels (1983) and Evans
(1984) also subdivided the skills that they analyzed into
phases. Table 2.2 summarizes the research reviewed for
skills performed on the uneven parallel bars.

The identification of factors critical to the
performance of a skill is often the first step to the
biomechanical analysis of movement. Sands (1985b) identified
the hip extension and foot velocity as the critical factors
in good kip performance. Bovinet (1979) used three ability
groups with three subjects per group and identified the
critical factors of the glide kip as forceful hip extension
and the timing of the leg thrust. Daniels (1987) compared
two ability groups that performed a Stalder on the uneven
parallel bars. The performances of the groups were compared
through a one-way analysis of variance to identify the

critical factors. The author concluded that performance was

21

Table 2.2 Summary of Uneven Parallel Bar Research

 

Author Skills Number of
Subjects
Landa (1974) - Squat on LB' from long hang 4
Squat on LB from jump-cast
swing
Bovinet (1979) - Glide kip on LB 9
- Short kip to HB**
Hay, et a1. - Short kip, front hip circle, 3
(1979) cast wrap, hecht dismount

- Short kip, front hip circle,
cast handstand, pirouette,
short beat, cast over LB

Daniels (1983)
Daniels (1987)

Stalder to handstand 14
Stalder to handstand

Evans (1984) - Clear hip circle to handstand 4
Sands (1982) - Reverse hecht 1
Sands (1985a) - Kip to catch H9 1
Sands (1985b) - Straddle back to handstand 1
Sands (1987) - Giant, giant to dismounts 6
Bentham (1987) - Clear hip circle to handstand 1

 

* LB - Low bar.

** HB - High bar.

enhanced by a starting position that was at or near a
handstand. According to Daniels, this position maximized the
distance between the axis of rotation and the total body
center of mass. Furthermore, Daniels indicated that the
straddle-in action should be delayed as long as possible and
should be performed with a narrow straddle to maximize the

moment of inertia. To maximize further the radius of

22

rotation, Daniels suggested that shoulder extension should
be minimal and the hips should always be kept farther from
the rail than the shoulders. The straddle-out action should
be timed with the recoil of the rail on the upswing.
Extension of the legs should occur throughout the straddle-
out phase as it will have less effect on the path of the
total body center of mass with respect to the rail than a
faster, shorter opening. 3

In studying movements on the uneven parallel bars, a
number of authors have been concerned with the forces
exerted on a performer. Daniels (1987) calculated forces
indirectly from cinematographic data. The mean force
applied to the rail by a female gymnast performing the
Stalder to handstand was 2.2 times body weight, with a
maximum force of 3.3 times body weight. Smith (1981)
presented theoretically determined “centrifugal“ force
values of a gymnast performing two skills on the uneven
parallel bars. In the first skill, the gymnast swung from a
handstand to a short beat on the low bar. According to
Smith's calculations, the resultant bar reaction force
immediately before the gymnast contacts the low bar ranged
from 4.2 to 4.7 times body weight, depending on the height
of the gymnast. For the second skill analyzed, a long beat
from a handstand, Smith concluded that the maximum bar
reaction force occurred at the bottom of the swing and may

be as great as five times body weight.

23

Another topic investigated in uneven parallel bar
studies was that of bar deflection. Daniels (1987) compared
the pattern of bar deflections for each of her two groups.
The author concluded that the rail deflections were directly
related to the amount of angular momentum produced during
the performance of Stalder circles. Overall deflections of
the bar were greater in a more highly skilled group.

Sands (1987) reported a technique for recording bar
deflections and at the same time giving immediate feedback
to coaches and gymnasts concerning technique corrections. A
computer "joystick“ was attached to the upper rail of the
uneven parallel bars. A computer program “read" the bar
deflections in x-y coordinate form. The data were then
graphically represented on the computer screen for immediate
viewing. Sands attempted to develop models of bar
deflections for several skills, including the giant swing.
The author concluded that the deflection pattern of good
giant swings had two major bar movements and that his method
of analysis could identify giant swings in which the
performer did not “tap" well or arched excessively.

Two studies have directly measured force through the
application of strain gages on uneven parallel bars. Evans
(1984) reported a horizontal force of six times body weight
for the clear hip circle to handstand, while Hay, et
al.(1979) reported a maximum resultant force of 3500 N (5.7

times body weight) for an impact type skill on the low bar

24

and 2140 N (3.6 times body weight) for a swinging skill from
the high bar.

SI i' E Ii D . G' I S . II n . I J B

Giant swings are a basic skill in men's horizontal bar
progressions and have been described in various gymnastics
texts. Taylor, Bajin and Zivic (1972) describe the skill as

follows:

In the handstand position the gymnast
stretches from the shoulders with the center of
gravity held as far away from the center of the
bar as possible. From this position the gymnast
begins the swing forward and just before reaching
the vertical line under the bar, the leg swing is
slowed so the back is slightly arched as it passes
the vertical. Once past the vertical, a short,
quick swing of the legs, finished approximately 45
degrees after the vertical line, lifts the body in
a slightly piked position, decreasing the angle
between the arms and the body.

This position is maintained until the gymnast
is 45 degrees over the horizontal at which time
the body stretches, the head moves backward and
the chest is lifted upward over the bar. (Taylor,
et al., 1972, p. 218).

Hay (1985) added:

Passing through the horizontal once again and into
the fourth quadrant of the circle, the gymnast
arches his back and does further work against
gravity by pressing downward with the arms. This
downward pressing is usually preceded by a shift
in the grip to place the hands more nearly on top
of the bar than they were before - a shift that
makes it easier to press downward a moment later.
Ideally this hand shift takes place at a time when
the centripetal force required to maintain the
angular motion is equal to the component of the

25

gymnast's weight acting along a line joining the

center of gravity to the axis. Under such

circumstances, the pressure between the gymnast's

hands and the bar is at a minimum, and movements

of the hands about the bar are relatively easy to

make. (Hay, 1985, pp.320-321).

For many years and from different perspectives,
biomechanists have studied men performing the giant swing on
the horizontal bar. Cureton (1939) calculated "centrifugal"
forces and the velocities of the total body center of mass
about the bar. According to Cureton, peak forces occurred at
about 135 deg from the handstand position on top of the bar
and equaled 4.9 times body weight. Kopp and Reid (1980)
directly measured the forces applied by gymnasts during
giant swings on an instrumented horizontal bar and reported
a range of 3.5 to 3.7 times body weight in the subjects they
tested. The peak force consistently occurred in the third
quadrant, just past the bottom of the swing.

Herrmann (1968) produced velocity-time diagrams and
calculated hip angles for overgrip giant swing on the
horizontal bar. The author concluded that the changes in the.
body position during the upswing phase resulted in the
reduction of the moment of inertia and thereby increased the
speed of rotation of the body about the center of the bar.
In 1976, Borms, Moers, and Hebblelinck studied performances
of forward and backward giant swings. Angular changes in
body segments and angular velocities of the feet, center of

mass, and chest were examined. This study showed that the

maximum velocity of the center of mass of the body was 6.0

26

m/sec and of the shoulder was 3.5 m/sec. Both velocities
occurred during the third quadrant of the giant swing or .63
sec from the handstand position on top of the bar. The
maximum velocity of the feet, however, was 12.5 m/sec and
occurred at 1.03 sec from the beginning of the circle.

Boone (1977) filmed two male gymnasts of different
skill levels and subsequently examined angular velocity and
centripetal force at specific points in the giant swing on
the horizontal bar. In this study, the giant swings with a
higher angular velocity were associated with poorer
performances. This led to the conclusion that those gymnasts
who perform the giant swings with a higher angular velocity
must hold on tighter to the bar to counteract the
“centrifugal" forces. Cheetham (1984) also examined angular
velocity and the trajectory of the center of gravity of the
body and the beat characteristics of three types of giant
swings (overgrip giant, dismount giant, and wind-up giant).
He concluded that the path of the center of mass of the body
was different for each type of giant swing.

Several authors have attempted to model the overgrip
giant swing. Dainis (1975) used a three link model (arms,
trunk, and legs) to describe the differences between an
”ideal" and a "poor" overgrip giant swing. In his model, the
bar was considered as inflexible and grip friction was
neglected. The author concluded that a relaxed body on the

downswing caused a hyperextension of the hips in the first

27

quadrant of movement and this was indicative of poor
technique. A good overgrip giant swing should have
coordinated action between the shoulder and the hip
resultant forces. In a less skilled performance of the
overgrip giant swing, these forces do not act in unison and
do not produce as smooth a movement. Dainis (1974) also used
a four link system (arms, trunk, thighs, and shanks) in
developing a model from Lagranges's equations of motion.
Torques acting between the segments and the time increments
of torque applications were used as input into a computer
program. In developing the model, Dainis used two
performers, a skilled and an unskilled performer. The
skilled performer required 1.9 sec for completion of the
skill as compared with 1.8 sec for the less skilled gymnast.
At the completion of the skill, the gymnasts' angular
velocities were -101 deg/sec and -151 deg/sec, respectively.
Duck (1976) developed a three link model (arms, trunk, and
legs) based on Newtonian mechanics. Input variables
consisted of body segments, a friction constant of
proportionality, angular velocity and acceleration of the
shoulder and hip, and the initial velocities and
accelerations of the arms. Bauer (1983) used the equations
of a pendulum of variable length to construct a model for
various gymnastics movements including the giant swing.
From this model, the author could input anthropometric

variables for individual performers and predict performance.

28

The information in the literature concerning the
overgrip giant swing on the uneven parallel bars is sparse.
Gluck (1982) described the overgrip giant swing on the
uneven parallel bars as being mechanically more difficult
than the same skill on the horizontal bar. When comparing
the performance of an overgrip giant swing on the uneven
parallel bars by a female gymnast to the same skill
performed by a male gymnast on the horizontal bar, it is
evident that the female has to execute a skill of greater
difficulty. The female gymnast must contend with the
physical constraint of the low bar and rotate about a bar of
greater cross sectional area. This difference in bar size
may partially account for the skill being given a higher
difficulty rating, a “C", in women's gymnastics and only an
"A" rating in men's gymnastics. Piking and straddling are
the two methods for clearing the low bar on the downswing.
According to Gluck (1982), the pike method was preferred
because in this approach rotational inertia did not decrease
as much as in the straddle method and the bar deformed more
in the pike when compared to the straddle method.

George (1980) emphasized the importance of the starting

position and the fully stretched handstand, in order to

29

maximize the amplitude of the descent phase of the overgrip
giant swing. The second factor emphasized was the distance
of the center of mass of the body from the bar during the
descent phase. The greatest distance, according to George,
is realized by extending the shoulder-trunk angle and
flexing at the hips to miss the bar. Sands (1987) emphasized
that the completion of the overgrip giant swing with an
arched body position was related to the performer not
executing a “tap" at the appropriate time in the overgrip

giant swing.

Summazx_Df_tha_Rexian_nf_Litaratnna

A review of the literature concerning instrumentation
of gymnastics equipment indicated that this method of data
collection gives the researcher important information
concerning forces applied by the gymnast. Studies have been
conducted in which the vaulting board, balance beam,
parallel bars, high bar, rings, and horizontal bar had been
instrumented. (Kreigbaum, 1974; Wilkerson, 1978; Sale &
Judd, 1974; Zinovsky, et al., 1976; Kopp & Reid, 1960;
Nissinen, 1983; Robertson, et al., 1985; Ishii & Komatsu,
1987; Cheetham, et al., 1987). Force transducers have been
used to instrument the equipment, with strain gages being

the most frequently used transducer. Calibration of the

30

instrumented equipment is generally performed by statically
loading the equipment with known weights.

Two studies (Hay, et al., 1979; Evans, 1984) have
specifically instrumented the uneven parallel bars with
strain gages applied along the neutral axis for horizontal
and vertical bending. Each study synchronized the force data
with images from film and analyzed the following uneven
parallel bars skills: clear hip to handstand, kips, front
hip circle, cast wraps, long beats, and Hecht dismounts. The
number of subjects used in these studies were small. Hay, et
al. (1979) reported data for three subjects and Evans
(1984) reported analyzing four subjects.

The amount of force exerted by gymnasts on the uneven
parallel bars in the studies reviewed varied with the type
of skill being studied and the method of estimating the
force. The highest reported horizontal force was six times
body weight (Evans, 1984) and was directly collected from
the uneven parallel bars while the gymnast was performing a
clear hip circle. The lowest reported force was 2.2 times
body weight (Daniels, 1987) and was calculated indirectly
from film data of a gymnast performing a Stalder on the
uneven parallel bars.

Several studies of the overgrip giant swing on the
horizontal bar have been conducted. This skill is similar to
the overgrip giant swing on the uneven parallel bars, but

the female gymnast must change body position on the

31

downswing to compensate for the position of the low bar.
These studies reported variable force values. Cureton (1939)
calculated the peak force to be 4.9 times body weight, while
Kopp and Reid (1980) directly measured the forces to be 3.5
to 3.7 times body weight. There was also disagreement as to
the portion of the circle where the maximum force was
generated. Ishii and Komatsu (1987) reported the peak force
occurred during the third quadrant, just past the bottom of
the swing. Cureton (1939) and Kopp and Reid (1980) reported
the maximum force occurring during quadrant two, just prior
to the bottom of the swing.

Other variables reported in studies of the overgrip
giant swing on the horizontal bar included the hip and
shoulder angle and velocities and accelerations of body
segments, joints, and the total body center of mass as the
gymnast rotates about the bar (Borms et al., 1976; Boone,

1977; Herrmann, 1980; Cheetham, 1984).

CHAPTER III

Experimental Procedures

The purpose of this study was to compare selected
kinematic and kinetic parameters between two groups of
gymnasts performing the overgrip giant swing on the uneven
parallel bars. By comparing two skill groups, Group 1
(highly skilled) and Group 3 (less skilled), it was
postulated that critical factors would be identified that
could aid teachers and coaches in understanding the skill.
Subjects were assigned to the skill groups based on judges'
ratings of their overgrip giant swing performances.

This chapter is divided into the following sections:
general procedures, instrumentation and calibration of the
bar, cinematographic techniques for data collection and
analysis, kinematic analysis, kinetic analysis, and

statistical analysis.

Subjects.
The subjects for this study were 15 volunteer Class I

age group or Elite level gymnasts, who ranged from 10 to 16
years of age. Coaches from ten clubs throughout the state

of Michigan were asked to suggest subjects for this study.

32

33

Twenty gymnasts were invited to participate in the study and
17 actually arrived on the day of filming. Two more were
eliminated from the study because they were unable to
perform the required skill. The requirement for entry into
the study was that the gymnast must be able to perform an
overgrip giant swing on the uneven parallel bars without the
assistance of her coach. The subjects came from five
different gymnastics clubs in central and southeastern
Michigan. Prior to data collection, each subject, her
parents, and her coach were informed about the purposes of
the study and the requirements of participation. After
informing the subjects and their parents about the study,
the investigator asked them to sign an informed consent form
(see Appendix A).

For each subject, a brief history of their years in the
sport, their age, and place of participation was taken.
Percent body fat was determined by hydrostatic weighing and
height, weight, thigh length, calf length, shoulder-elbow
3length, elbow-wrist length, hand length, biacromial
diameter, and biiliac diameter were measured. The
anthropmetric measurements were taken at the Center for the
Study of Human Performance at Michigan State University and
followed the standards set by Lehman, Roche, and Martorell
(1988). This information was collected to compare the
groups, after they were formed on the basis of the judges'

ratings. It was the intent to show that the groups were

34

different because of the technique of skill performance and
not because of the above characteristics. The information

sheet used during data collection is shown in Appendix B.

Eilming_n£_Eaz£nrmsncs_Ssqusnca

Subjects were prepared for filming by placing joint
markers on the right side of the body on the following
landmarks: ankle, knee, hip, shoulder, elbow, and wrist. In
order to ensure as safe and familiar an environment for the
gymnasts as possible, the filming was performed in a
gymnastics facility. The bars were properly cabled into the
floor and there were numerous safety mats available. The
gymnasts' own coaches accompanied them on the day of filming
and were available to spot the gymnast if necessary.

Figure 3.1 shows the four skill sequence performed by
the gymnasts for this study. Because skills are linked
together in competition on the uneven parallel bars, each
gymnast was asked to perform a sequence of skills. The
skills occurred in the following order: kip, cast, clear hip
to handstand, overgrip giant swing, and flyaway dismount.
Since these skills are used commonly as transition skills
and are used frequently in compulsory routines, it was
expected that the gymnasts who volunteered for this study

could perform them in sequence.

35

Ass (Jae/KI

'L- Zlcuugz faunas: 1‘14;

a2, ’3

to. cu-asts, calm-sur- I111; 'tc: luauach-H:luaci

Figure 3.1 Skill Sequence on the
Uneven Parallel Bars

37

Judgss_and_Eatings_Df_Earfnrmanaa

In order to limit the judges' evaluations to the skill
of interest in this study, a video tape from each filmed
sequence was made of the overgrip giant swing isolated from
the other skills performed. The video tape was shown to
three Elite and one Class I NAWGJ judges, who were asked to
rate the performances of the gymnasts on the basis of
criteria established by the investigator. The criteria were
selected to correspond to deductions judges use to evaluate
performance of an overgrip giant swing on the uneven
parallel bars. Each criteria was assigned a score from 1.0
to 2.0, depending on the seriousness of the error. Each
judge was asked to score each criteria based on the number
assigned. The highest total score that a judge could award
was a 10. The judging criteria and score sheet appear in
Appendix C. Based on the judges' ratings, the top five
gymnasts were classified as the "highly skilled" group
(Group 1), the middle five were classified as intermediate
(Group 2), and the bottom five gymnasts were considered the
"less skilled” group (Group 3). Since it was the intent of
this study to differentiate skill groups that were diverse
in performance characteristics, group 2, the intermediate
group, was dropped from further study. Groups 1 and 3 were
compared on selected kinematic and kinetic variables.

Figures 3.2 and 3.3 show film tracings of two subjects, one

38

from each skill level group performing the overgrip giant
swing. Subject 16 was typical of Group 1, the highly skilled
group, while Subject 6 displayed a performance typical of
Group 3, the less skilled group. Both subjects used the pike
technique of clearing the low bar during the overgrip giant

swing.

 

‘3 fr
‘k‘g/L‘I
. Y

I .

 

Figure 3.2 Skilled Performance of an Overgrip Giant
Swing on the Uneven Parallel Bars (Subject 16)

39

Figure 3.3 Unskilled Performance of an Overgrip Giant
Swing on the Uneven Parallel Bars (Subject 6)

I I I I. I c Jo] l.e E II B

One 233.7 on American Athletic Institute Graphite X bar
with a wood laminated surface was instrumented for this
study. The 40 mm diameter bar was constructed from a
composite material consisting of graphite and fiberglass.

This is the latest bar produced and it has replaced the wood

40

laminated fiberglass bar that was oval in shape, measuring
45 mm by 40 mm. Four Measurements Group CEA-06—250UWw-350
electrical resistance strain gages (see Figure 3.4) were

bonded to the surface of the bar in a differential

 

 

a. strain gages to collect horizontal
and vertical forces

 

 

 

b. strain gages to collect torques

 

 

 

 

 

C D
c. placement of gages on the bar

Figure 3.4 Strain Gage Configuration

41

configuration. This configuration allowed the resultant
strain measurement in the vertical and horizontal direction
to be independent of the point of application of force along
the bar. The gages had a resistance of 350 ohms and a gage
factor of 2.075. One set of gages was placed on the top of
the bar to collect vertical strain values and another set
was placed orthogonally to the first set of gages to collect
horizontal strain values (see Figure 3.4 a). An additional
pair of gages were placed at a 45 deg angle to the long axis
of the bar for the purpose of collecting torque data (see
Figure 3.4 b). Figure 3.4 c shows the relative placement of
the pairs of gages along the top bar of the uneven parallel
bars.

A series of calibration tests were performed prior to
data collection. The bar was calibrated statically by
suspending known loads, ranging from 9.8 to 133.7 kg, from
the bar in the vertical and horizontal directions. It was
felt that this range of loads was sufficient since the
maximum weight of the subjects was 61.3 kg. A metal device
was constructed to hold the weights during calibration. The
device was constructed so that two clamps could be placed on
the uneven parallel bar and hold weights placed in a basket
suspended from the clamps (see Figure 3.5). The bar support
was modified to allow the bar to be rotated 90 deg for the
purpose of calibrating strain gages that measured the

horizontal forces. The calibrations included an analysis for

42

linearity and independence of vertical and horizontal

outputs. The strain gages applied at 45 deg, that measured

 

 

 

bar
L j _ I fl
7
clamps — weights

 

'\

basket

a. device suspended from uneven parallel bar rail to hold
weights

 

ueith’

 

 

II
. torque bar

D. weights applied to bar for torque calibrations
(viewed from above)

Figure 3.5 Calibration Device

43

torque, were calibrated by applying known torques to a
torque fixture mounted to the bar. Through these tests it
was found that the force varied depending on how far from
the gages the force was applied. As a result of this
finding, a second camera was used to determine the position
of each gymnast along the bar. Based on the gymnast's
position on the bar, calibration correction factors were
applied to the strain gage data.

Prior to filming, the bars were set at the farthest
distance apart and not adjusted for individual gymnasts. By
positioning the bars at the widest setting, the danger of a
gymnast striking the low bar was reduced. This approach also
provided a consistency to the data collection process. With
the bars in this position, the upright support was at an
eight degree tilt from vertical, which meant that the strain
gages were also rotated by eight degrees. A correction for
the eight degree rotation was included in the data analysis
software.

The strain gages were positioned to minimise channel
interactions (crosstalk). To verify this, experiments were
conducted by applying horizontal and vertical forces to the
bar. Results of these experiments showed nearly zero
interaction between the vertical and horizontal channels.

Based on the torque values obtained by Kopp and Reid
(1989), it was estimated that the maximum torques were

relatively small. This required the strain gages for

44

measuring torque to be about ten times the sensitivity as
the strain gages for measuring force. Because of this,
crosstalk on the order of 10% was experienced between the
horizontal force and torque. This interaction was corrected
to within 2-32 in the hardware by subtracting a portion of

the horizontal channel output from the torque channel

output. Final correction of the crosstalk was made in the
software. During the pre-filming calibration, loads were
applied first to the strain gages set to measure vertical
force, then to the strain gages set to measure the
horizontal force. As the forces were measured, variations
produced in the torque records were also measured. The
software then calculated a correction factor to be applied
to all subsequent torque outputs when forces were
simultaneously present.

The three pairs of strain gages that comprised half of
three Wheatstone bridge circuits were connected to three
pre-amplifiers. Included in the bridge circuits were three
pairs of fixed resistors to establish the reference side of
the bridge. In addition, three zero adjustments were used to
ensure that bridge outputs under zero conditions were near
the center of the pre-amplifier ranges. Final zero was
determined in software, prior to data collection for each
trial. The actual outputs of all three channels were sampled

and used as baselines.

45

During filming, the three pre-amplifier outputs were
sampled at a rate of 100 times/sec in sequence along with
synchronizing pulses from the timing lights to coincide with
the film images collected. The sampled data were converted
to digital form by an analog-to-digital converter (ADC) and
read by an IBM Model 9000 dedicated computer. A trigger
button, activated by this experimenter, started a 10 sec
software timer. During this period, incoming data were
stored in the computer's memory and then transferred to

floppy disk for storage.

0' l l' I l i f D l C J] I' I l J .

Film data were collected by two LOCAM 16 mm high speed
pin registration motion picture cameras. Kodak Ektachrome
Video News Film with an ASA rating of 400 was used in this
study. One camera, with a focal length of 30 mm and an
f/stop of 5.6, was placed perpendicular to the side of the
uneven parallel bars at a distance of 20.2 m from the center
of the bars. A second camera, with a focal length of 17 mm
and an f/stop of 4.0, was used to record the position of the
hands and was placed 10.6 m from the base of the bars. The
center of the lens of each camera was 2.3 m from the floor.
Both cameras had a shutter angle of 120 deg and were set at
100 frames/sec. The actual filming speed was calculated by
timing lights located in each field of view. The timing

46

lights had an effective frequency of 1000 cycles/sec and
were used as an aid in synchronizing the film and kinetic
data. A schematic of the filming area appears in Figure 3.6.
Prior to filming, a meter stick was held in the field
of view at a location where it was projected that the side
of the gymnast closest to the camera would form a sagittal
plane during the performance of the overgrip giant swing. A
second meter stick was placed in the field of view for
camera two for the frontal view of the gymnast. From these

reference measures a linear multiplier scale factor was'

 

 

 

 

 

 

 

 

 

 

 

. - ADC
W11 Com uter |
- ,__J P +r—'—
L______.
Camera 2
L-L-Uneven parallel bars
I 10.6 m J
20.2 m

 

Camera 1

 

 

Figure 3.6 Filming Area and Equipment Set-up

47

obtained. This scale factor was used in converting projected
film distances into actual distances. A plumb line was also
filmed in the field of view of both cameras as a vertical
reference for determining spatial orientation.

For analysis, the film was projected onto a drafting
table by a Vanguard Overhead Notion Analyzer (Brown, 1983).
With this set-up, the image size was controlled for clarity
with the intent of increasing the accuracy of data point
identification. X-Y Cartesian coordinates were obtained by
digitizing the film images with a Sonic Graf/Pen system. The
digitizing system was interfaced with an IBM-PC Computer and
an interactive data acquisition computer program. A FORTRAN
data analysis program, Bioanalysis, was used to analyze the
kinematic parameters selected. This program utilized a
fourth order Butterworth filter to smooth the raw data

points.

Kinematic_Anal¥sia

The overgrip giant swing on the uneven parallel bars
was divided into four phases following the convention
established in several studies of men's giant swings on the
horizontal bar (Kopp & Reid, 1980; Borms, et al., 1976;
Dainis, 1975). The quadrant location of the gymnast was
determined by the location of the subject's total body

center of mass calculated by the Bioanalysis computer

48

program. The four quadrants (see Figure 3.7) were defined as
follows:

Quadrant 1: the frame of film in which the center of mass
of the subject was closest to being vertical over
the bar to the frame of film where the center of
mass was located at 180°,

Quadrant 2: from the frame of film where the center of
mass of the subject was located closest to 160° to
the frame where the center of mass of the subject
was closest to being vertically (270°) below the
bar,

Quadrant 3: from the frame of film where the subject's
center of mass was located closest to 270° to the
frame of film where the center of mass of the
subject was located closest to horizontal (360°)
on the upswing phase of the skill, and

Quadrant 4: from the frame of film where the subject's
center of mass was located closest to 360° to the
frame of film where the subject's center of mass

again arrived over the bar to the 90° location.

49

 

 

Figure 3.7 Performance of an Overgrip Giant
Swing Showing Quadrants (Subject 16)

In order to compare the two skill level groups and to
compare the overgrip giant swing on the uneven parallel bars
with men's performances of this skill on the horizontal bar,

the following kinematic parameters were investigated:

50

1. Positions of selected body segments. (The segments
of interest in this study included the forearm, arm,
trunk and head, thigh, shank, and foot).

2. Angles between specific body segments. (The angles
of interest in this study were the shoulder angle,
and the hip angle).

3. Linear velocity of the center of gravity of
individual segments, individual joint centers, and
total body center of gravity.

4. Linear acceleration of the center of gravity of
segments, joint centers, and total body center of
gravity.

5. Time to perform the entire overgrip giant swing and

a temporal analysis of the individual quadrants.

K' I' E ] .

Data collected by the strain and torque gages were
transmitted to a computer program developed at the Center
for the Study of Human Performance at Michigan State
University. Force-time curves, normalized by body weight,
and torque-time curves were generated as well as vertical
and horizontal forces and moments. The resultant force and
the angle of the resultant force were calculated. By

synchronizing the kinetic data with the film data,

51

relationships could be drawn between kinematic and kinetic

variables in the performance of the overgrip giant swing.

Anthropometic, kinematic, and kinetic data were input

variables in a FORTRAN computer program developed by the

author to calculate joint forces and net muscle moments.

Figure 3.6 shows a link segment model of a gymnast's forearm

and arm that was used to calculate the joint forces and

muscle moments. The following assumptions were made with

respect to the model:

1.

The gymnast and the bar were bilaterally symmetrical
and a two-dimensional model in the sagittal plane

was valid.

. The arm and forearm could be treated as rigid

bodies.

. The joints, that link the segments together, could

be treated as frictionless and pinned.

. Each segment had a fixed mass located as a point

mass at its center of gravity.

. The location of the center of gravity of each

segment remained fixed within the segment during the

movement.

. The moment of inertia of each segment about its mass

center was constant during the movement.

. The forces and torques between the bar and hand were

considered to be transmitted directly to the center

of the wrist and forearm, respectively.

52

 

Figure 3.6 Link Segment Model of Forearm and Arm

)
)

NH

QWNg'U FWIBRUAA

53

Model Nomenclature

Forearm

Arm

Bar and hand

acceleration due to gravity (~9.8 m/sz)
mass (kg)

Torque (Nm)

Force (N)

Segment length (m)

Percent of segment length to center of gravity
Center of gravity

Horizontal acceleration (m/sz)

Vertical acceleration (m/sz)

Angular acceleration (rad/s2)

Figure 3.6 (Continued)

6. The use of Dempster's (1955) data to represent the
segment mass proportions and center of gravity
locations on female gymnasts did not substantially
alter the magnitude of kinematic and kinetic

parameters from their true values.

Linear equations of each segment's mass center may be

expressed as:

(1)

(2)

SF = ma, where a acceleration (m/secz)

ZFx = mX (cg) ZFr = m? (cg)

in - Fxs = m1Xi(cg) Fri - Frs - mig = sYi(cg)
in = Fxs + m1Xi(c¢) (3) Pr: = Fvs + mig + miYi(cg)
sz - in = m2Xz(cg) F72 - Fri - mag = m2?rz(cg)
sz = F31 + m2X2(cg) (4) F22 = Fri + mag + m2Y2(cg)

54

Angular equations of motion of the forearm and arm may

be expressed as:

EN = IS, where I = moment of inertia (kg m2)

(5) H1 = I161 + Fvs (1-p1) L1 cos 61 - Fxs (l-pi) L1 sin 61
+ Fv1 (p1L1 cos 61) + Fx1 (p1L1 sin 61) + Na
(3) H2 = 1262 + FY1 (1-p2) L2 cos 62 - Fx1 (1-p2) L2 sin 62

+ Fv2(p2 L2 cos 62) + Fx2 (p2 L2 sin 62) + H1

SI l' l' ] E 1 .

The judges‘ ratings of the subjects' giant swings were
used to divide the subjects into two groups, highly skilled
and less skilled. The objectivity of the judges' ratings was
obtained by calculating an intraclass correlation
coefficient as described by Safrit (1976) and Wilkerson
(1976). The selected kinematic and kinetic variables for
each group were compared by using a one way analysis of
variance (ANOVA). One way ANOVA was used to determine if the
groups were significantly different (p < .05) in height,
weight, segment lengths, and percent body fat; parameters
which could account for differences in kinematic or kinetic

variables. A p < .05 level of significance was used because

55

a more stringent test was desired to show group

comparability than was used to show skill performance

differences.

The analysis of variance technique was used to compare

the two groups on the following variables:

1.
2.

temporal comparisons of the two skill groups;
peak resultant, horizontal and vertical velocities

and accelerations of segments;

. peak resultant, horizontal and vertical velocities

and accelerations of the center of mass;

. peak resultant velocities of joint centers (ankle,

knee, hip, shoulder, and elbow):

. shoulder, elbow, and hip joint angles;

. peak resultant, horizontal, and vertical forces

generated on the bar by each group: and

. horizontal and vertical forces and moments exerted

on the elbow and shoulder joints.

This statistical treatment, calculated by the

Statistical Package for the Social Sciences (SPSSX) on a VAX

computer was used to determine if performance differences

between the two groups were significant at the .10 level to

identify critical features associated with skillful

execution of the overgrip giant swing on the uneven parallel

bars.

CHAPTER IV

Results and Discussion

The data for this study of the overgrip giant swing on
the uneven parallel bars were collected from two sources:
cinematographic records of performance and strain gage
records of force and torque applied to the top bar of the

uneven parallel bars. The results are presented in this

chapter under the headings of general characteristics of the
subjects, group placement and characteristics, objectivity
of the judges' ratings, reliability of digitized data,

kinematic analysis, and kinetic analysis.

Gansral_Chazactaristics_nf_tha_Suhiacts

The performance of the overgrip giant swing on the
uneven parallel bars by 15 female, Class I and Elite
gymnasts were filmed for possible inclusion in this study.
The gymnasts were from five clubs located in Michigan.
Gymnasts ranged in age from 10 years and 6 months to 16
years and 5 months, in weight from 29.5 kgs to 61.3 kgs, and
in height from 134.3 cm to 167.3 cm. The means for age,
weight, and height were 13 years and 3 months, 41.1 kgs, and
149.3 cm, respectively. The subjects had participated from
four to 10 years in gymnastics with a total group mean of

6.7 years.

56

57

The subjects were weighed hydrostatically and their
percent body fat calculated. The total group body fat ranged
from 13 to 22 percent with a mean of 17.5 percent. The
following anthropometric measurements were taken: biiliac
diameter, biacromial diameter, hand length, elbow-wrist

length, shoulder-elbow length, calf length, and thigh length
(see Table 4.1).

Table 4.1 Means and Ranges of Selected Anthropometric

 

 

 

 

Measurements
4AEntira;§rouy

Measurement Mean Minimum Maximum

(cm) (cm) (cm)
Biiliac Diameter 23.7 21.0 26.1
Biacromial Diameter 33.5 28.3 39.0
Hand Length 16.2 14.5 17.5
Elbow-wrist Length 23.2 19.5 26.0
Shoulder-elbow Length 31.3 27.4 34.9
Calf Length 34.3 30.0 36.4
Thigh Length 36.5 31.0 45.0

 

Grnua_2lacament_and_Charactaristics

A panel of four Michigan judges, three elite and one
Class 1, independently rated the 15 gymnasts based on the

following characteristics that had been predetermined by

56

this researcher: bent arms, arched body position, bent legs,
separated legs, forced return to handstand, and general
impression. A sample judging sheet and a summary of the
judges scores appear in Appendix C. A weighted scale was
devised that associated with each characteristic a set
number of points, resulting in a total of 10 points. Each
judge awarded a score out of 10.0 for each gymnast after
having viewed the subjects' individual video tapes of their
overgrip giant swing isolated from the other skills in the
sequence. The judge was given three times to view each trial
before awarding a score. The ranking of the gymnasts was
obtained by averaging the two middle scores of all judges.
Subjects with the five highest scores were assigned to Group
1 - the highly skilled group. Subjects with the lowest five
scores were assigned to Group 3 - the less skilled group.
Group 2 was the intermediate skilled group and subjects in
this group were not analyzed further.

A one way analysis of variance (ANOVA) was used to
determine if there was any significant difference between
the groups on the following variables: age, height, weight,
years in gymnastics, percent body fat, thigh length, calf
length, shoulder-elbow length, elbow-wrist length, hand
length, biacromial diameter, and biiliac diameter. While
there are individual differences between the subjects in
this study, the ANOVA results revealed that there were no

significant differences at the p < .05 level between Groups

59

1 and 3 on the above variables. Table 4.2 gives a summary of

selected group characteristic.

Table 4.2 Selected Comparisons by Group

 

 

Variable Group Mean Minimum Maximum
Aze (Yrs) 1 13.35 12.17 15.33
2 12.20 10.50 14.58
3 14.19 10.50 16.42
Years in 1 7.20 4.00 9.00
gymnastics 2 5.70 4.00 7.50
3 6.87 4.00 10.00
Height (cm) 1 151.00 145.50 155.70
2 143.00 134.30 155.00
3 153.84 136.00 167.30
Weight (kg) 1 39.90 34.60 44.06
2 36.24 29.46 45.09
3 47.15 31.70 61.34
Body Fat (1) 1 17.53 13.00 22.00
2 18.14 16.00 20.30
3 16.83 13.50 21.33

 

Obiactixit2_nf_JudgasL_Batings_nf_2er£nrmancaa

The procedure for evaluating the objectivity of the
judges ratings was taken from Safrit (1978) and Wilkerson
(1978). A two-way analysis of variance was obtained through
the latest version of the Statistical Package for the Social

Sciences (SPSSX) program on a VAX mainframe computer and the

60

intraclass correlation coefficient was manually calculated.
The analysis of variance summary table and objectivity

calculation appear in Table 4.3. The objectivity score was
.935. This supported the technique of using judges ratings

to assign gymnasts to skill level groups.

Table 4.3 Analysis of Variance Summary Table and
Objectivity Calculation

 

 

 

 

 

 

 

 

Source of Variation dF SS MS
Between Subjects (n-l) 13 106.09 8.16
Between Judges (l-l) 3 53.20 17.73
Interaction (n-l)(l-1) 39 20.55 .53
Total 55 179.84
0: a3; = MSs_:_HSsxa. = 6_lfi_:_.53.= .935

032 + oZaxa.+ a£1 H38 3-13

l l
B li lil'l E DI 'li I D l

Two of the subjects' trials were digitized twice and
each trial was analyzed by the Bioanalysis computer program.
A Pearson product-moment correlation was calculated by the

SPSSX program for the linear velocity of the toe, ankle,

61

knee, and hip for Subjects 1 and 4. In addition,
correlations were performed on acceleration values
calculated from displacement data for both subjects for the
arm, and forearm. The results, which appear in Table 4.4,
show that the reliability for the velocities were very high,
ranging from .969 for the hip of Subject 1 to .997 for the
knee of Subject 4. The reliability coefficients for the
accelerations, however, are much more varied. The lowest
correlation coefficient was .315 for the vertical
acceleration of Subject 1's arm, while the highest
coefficient was .883 for the horizontal acceleration for
Subject 4's arm. The acceleration values are subject to
variability, since there is error introduced due to the
digitizing process as well as from calculating accelerations
from position data. Since the arm and forearm accelerations

and masses are relatively small compared to the magnitude of

Table 4.4 Reliability Correlation Coefficients of Selected

 

Variables
Velocity: Toe Knee Ankle Hip
Subject 1 .994 . .983 .989 .969
Subject 4 .987 .997 .996 .995
Accelerations: Arm Arm Forearm Forearm
X Y X Y
Subject 1 .785 .315 .367 .553

Subject 4 .883 .662 .787 .686

 

62

the forces exerted and the mass of the whole body, it is
assumed that the variable accelerations would not
substantially alter the joint reaction force and muscle

moment calculations.

8' I' E ] .

The digitized cinematographic data files were
transferred to a mainframe VAX computer and analyzed by a
FORTRAN program (Bioanalysis), that was developed at
Michigan State University. The Bioanalysis program used the
digitized, position data to calculate velocities and
accelerations of the total body center of mass, the body
segments, and the selected joint centers (ankle, knee, hip,
shoulder, and elbow) for each subject. For each of the
previously mentioned variables, peak values, peak values per
quadrant, and the time of occurrence of peak values
expressed as a percent of total time for the cycle of the
overgrip giant swing were calculated. Statistical analysis
consisted of a one way analysis of variance on each variable
to determine if a significant difference existed between
Group 1 (highly skilled) and Group 3 (less skilled). In this
section, the results will be presented under the following

sub-headings: temporal analysis, position data for shoulder

63

and hip angles, center of mass analysis, joint analysis, and

body segment analysis.

W

The giant swing from handstand to handstand took a mean
time of 2.26 seconds for the total group. The range for the
total group was 1.80 to 2.87 seconds. Table 4.5 gives the
mean times for each group and the relative proportions for
each quadrant. The mean total times and times for individual
quadrants for Groups 1 and 3 showed no statistical
difference (p< .10).

Although not statistically different at p < .10, there
was an indication of a trend between the groups in the first
and fourth quadrants. The gymnasts in Group 3 frequently did
not end the previous skill in a vertical handstand position
and therefore began the overgrip giant swing from a lower

starting position. This in turn, created a shorter time

Table 4.5 Temporal Data for Giant Swings for the Total
Group, Group 1, and Group 3

 

 

 

Total_11ms_isen9
Mean S.D. Min Max
Total Group 2.280 .324 1.80 2.87
Group 1 2.304 .355 1.97 2.87
Group 3 2.256 .332 1.80 2.65

Ralatixa_Iime1Quadrant_Lx_nf_tntal_cxcle_t1mal
Quad 1 Quad 2 Quad 3 Quad 4
Total Group 42.88 15.48 16.18 25.68
Group 1 44.44 15.71 18.06 23.78
Group 3 40.87 15.25 16.31 27.57

 

64

period for Quadrant 1. Consequently, in Quadrant 4, the less
skilled gymnasts averaged a longer time to bring their
center of mass over the high bar to complete the overgrip
giant swing.

Men's overgrip giant studies reveal similar temporal

data. Boone (1977) recorded times of 2.40 and 2.04 sec for

his two male subjects to complete the overgrip giant swing,
while Ishii and Komatsu (1987) reported a time of 2.2 sec
for their subject. The slightly longer period of time for
the overgrip giant swing on the uneven parallel bars
reported in the current study as compared to those reported
for men on the horizontal bar, may be accounted for by the
difference in the size of the bars and the obstacle

presented by the low bar.

His—and_5hnuldez_Anzlas
Him_angla. Figure 4.1 shows a graph of the hip and

shoulder angles of Subject 16, who was typical of the higher
skill level (Group 1) and Figure 4.2 shows Subject 6, who
was typical of the less skilled gymnasts (Group 3). These
graphs show shoulder and hip angles throughout the overgrip
giant swing. The vertical lines on the graph separate the
quadrants, each quadrant being labeled Q1, Q2, Q3, or Q4. A
horizontal line was drawn at 180 deg to indicate when either
the hip or shoulder was fully extended. The graphs of the

hip angles for the two gymnasts show a similar shape,

65

although Subject 6 began the skill in a hyperextended hip

position.

downswing phase of the circle.

Both gymnasts showed maximum hip flexion on the

Subject 16 reached maximum

hip flexion (88.7 deg) just prior to the end of Quadrant 1,

while Subject 6 reached her maximum pike (110.2 deg) just

after the beginning of Quadrant 2. Studies of men's giant

Mr fl

Angle (Degrees)

300

 

 

 

 

 

 

 

 

 

 

 

ll Shoulder
I'.
I
I
'I
I
I
I
~."
01 02 03 Q4
0 iff T'Vf *iW‘fi ‘*fV *fiv'rf*WV1‘Vfii
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Time (Sec)

Figure 4.1 Hip and Shoulder Angles (Subject 16)

3.5

. Angle (Degrees)

66

 

 

 

 

 

 

 

 

 

 

 

 

 

o V i V r Y w j V V j 1 I r ' Y I V U ' f ‘ V V 1 i I T ' V

0.0 0.5 1.0 1 .5 2.0 2.5 3.0 3.5

‘fime (Sec )

Figure 4.2 Hip and Shoulder Angles (Subject 6)

swings have shown a different pattern during this quadrant.
Men do not have the physical constraint of the low bar on
the descent phase. Borms, Moers and Hebblelinck (1976)
reported that their male subjects were slightly

hyperextended at the end of Quadrant 1.

67

During Quadrant 2, the hips again extend to reach a
nearly straight position as the body passes under the bar.
Subject 16 reached an angle of 191.8 deg at the end of
Quadrant 2, while Subject 6 achieved a maximum angle of
176.1 deg during this quadrant. George (1980) emphasized the
importance of achieving a straight body position before the
end of Quadrant 2. Subject 16 achieved this goal, but
Subject 6 did not. Subject 6 began an extending movement,
but did not achieve a fully extended position during
Quadrant 2.

During Quadrant 3, Subject 16 and Subject 6 again piked
to 130.6 deg and 133.3 deg, respectively. Cheetham (1984)
reported that there was very little pike necessary during
this phase for the overgrip giant swing on the horizontal
bar and Borms (1976) reported that his subjects only showed
a 10 deg change from a straight body position during the
same phase of the giant swing on the horizontal bar. The
female gymnasts pike much more during this phase to
compensate for the inability to maintain a long body
position during the descent phase of the skill due to the
proximity of the low bar. Both Subjects 16 and 6 extended
their hips to arrive on top of the bar in a handstand
position. Both gymnasts hyperextended their hips in Quadrant
4, with Subject 8 showing slightly more hip hyperextension

during this phase of the skill.

66

Sh9n1d§z_angle. The graphs of the shoulder angle also

show similarities in the two representative subjects'
patterns (see Figures 4.1 and 4.2). Since the gymnasts
performed a sequence of skills, the beginning position of
the overgrip giant swing was dependant upon the finishing
position of the previous skill. Both gymnasts began the
overgrip giant swing with the shoulders in an extended
position. Subject 16 showed a beginning shoulder angle of
132.9 deg, while Subject 6 had a shoulder angle of 55.2 deg.
Both gymnasts rapidly began to flex at the shoulder joint so
that maximum flexion (181 deg and 183 deg, respectively) did
not fluctuate very much until the end of Quadrant 2. George
(1980) emphasized the importance of maintaining an open
shoulder angle in order to achieve a longer radius of
rotation of the center of mass of the body during the
descent phase. Subject 16 came closer to achieving this goal
than Subject 6.

During Quadrants 3 and 4, the shoulders of Subjects 16
and 6 extended again until the gymnasts began to return to
the handstand position on top of the bar. Kopp and Reid
(1980) indicated that this shoulder extension during
Quadrants 3 and 4 served to decrease the moment of inertia
of the gymnast to generate an increase in angular velocity.
These factors contribute to the gymnast being able to

complete the overgrip giant swing.

69

CantaLQLMasLAnalxsis
Bath_nf_the_canter_n£_mass. Figures 4.3 and 4.4 show

the path of the center of mass of the two representative

gymnasts, as they performed the overgrip giant swing. The

horizontal and vertical lines were drawn to depict the

occurrences of the quadrants. The smaller closed figure

 

 

 

 

 

 

2.0 J
1.5 "
s A I Q! 01.
o a . .
c
'fi 3 1.0 {3
4:
é o
| E
>- ., , :12 as
.5
0" 'fi - ITj—TO o r.1fitUjfi Tfi—‘H
0.0 0.5 1.0 1.5 2.0

X—Coordinote

(motor. )

Figure 4.3 Path of Center of Mass (Subject 16)

70

 

 

 

 

 

 

2.0
1.5 -
‘ m 01.
.8 A - rm
3 u - \J
‘fi 3 1.0 III '
g 4; 1 02 03
o .
| E
>- v *
a.s -
a.“ I o a Trfi' ‘T'jj T—T ‘I—T fj
0.0 0-5 1.0 1.5 2.0
X-Coordinote
(motes-I)

Figure 4.4 Path of Center of Mass (Subject 6)

shows the path of the wrists around the bar during the
performance of the skill. The larger closed figure shows the
path of the total body center of mass. While a difference in
height of the two gymnasts, (154.3 cm for Subject 18 and

137.5 cm for Subject 6) may have influenced the relative

71

size of the paths of the center of mass as the gymnast
performed the overgrip giant swing, the pattern of the path
was quite different and was of interest in this
investigation. It can be seen by these figures that Subject
16, the more highly skilled gymnast, had a higher relative
starting position of the center of mass, as recommended by
George (1980), and the ending position of the circle was the
same as the beginning position. In Figure 4.4, the center of
mass of Subject 6 was closer to the bar at the beginning of
the cycle. Then, on the downswing, the center of mass moved
farther from the bar which enabled the gymnast to complete
the skill. The ending position, however was lower than the
highest position of the center of mass of Subject 6 during
Quadrant 1. For both subjects, note that the position of the
center of mass is farther from the bar during Quadrants 1
and 2, the descent phase, than Quadrants 3 and 4, the ascent
phase.

Eaak_rasultant_Ialncitx_nf_the_centar_nf_mass. The mean
peak resultant velocity (m/sec) of the center of mass was
calculated for each subject and the groups were compared for
the mean peak resultant velocity, mean peak resultant
velocity per quadrant, and the time of occurrence of peak
values expressed as a percent of total time. The mean peak
resultant velocity for the total group was 5.60 m/sec, the

means were 6.15 and 5.05 m/sec for Group 1 and Group 3,

72

respectively. Borms, et a1. (1976) reported a maximum
resultant velocity of the center of mass to be 6.0 m/sec for
the male subjects in their study and the maximum resultant
velocity occurred in Quadrant 2. The mean peak values for
the velocities in the horizontal and vertical directions can
be seen in Table 4.6. There was a significant difference (p
< .10) between the groups for mean peak horizontal velocity
and the relative time of occurrence for the mean peak
vertical velocities. (See Appendix D, Table D.1)

The mean peak horizontal velocity was 4.63, 4.89, and
4.38 m/sec for the Total Group, Group 1, and Group 3,
respectively. The mean peak vertical velocity for Group 1
was .3 m/sec. and occurred 57.9 percent through the cycle.
For Group 3, the mean peak vertical velocity of -2.9 m/sec

occurred 40 percent through the skill.

Table 4.6. Mean Peak Resultant Velocity of the Center of

 

 

Mass
Vx’ Vv Vs
- (m/sec) (m/sec) (m/sec)
Total Group +4.63 (.46) -1.30 (5.40) 5.60 (1.41)
Group 1 4.89 (.37) .30 (6.64) 6.15 (1.91)
Group 3 4.38 (.42) -2.90 (3.89) 5.05 (0.23)
+Mean (S.D).

* p < .10.

73

Examining the mean peak velocity of the center of mass
per quadrant (see Appendix D. Table D.2) revealed a
significant difference (p < .10) between the groups for
horizontal velocity in Quadrant 3. The mean peak horizontal
velocity for Group 1 was 4.3 m/sec and 3.6 m/sec for Group
3. A significant difference (p < .10) was found for the mean
peak resultant velocity in Quadrant 4. These results
indicated that the more highly skilled gymnasts, as compared
to the less skilled, had a greater increase in velocity of
the center of mass during this critical, upswing phase of
the skill.

AnnalazaLinn_n£_tha_nantez_n£_massn The accelerations
(m/secz) were also calculated (see Appendix D, Table D.4).
Mean peak resultant accelerations of the total movement of
the overgrip giant swing were 768.8, 1068.4, and 485.3
m/sec2 for the total group, Group 1, and Group 3,
respectively. There were no significant differences in the
mean peak resultant accelerations of the center of mass
about the bar. In Quadrant 4, a significant difference (p <
.10) was found between the groups for the relative time of
occurrence of mean peak resultant accelerations for the
center of mass. Appendix D (Tables D1-D.6) contains summary
tables of the statistical comparisons for peak velocity and
acceleration.

WW Angular

velocity (deg/sec) and angular acceleration (deg/sec?) of a

Angular Velocity (Deg /Sec.)

74

line segment connecting the center of the bar and the center
of mass of the gymnast were calculated. The mean peak
angular velocity for the total group was 262.3 deg/sec and
occurred at 73.6 percent of the time of the cycle. The

representative graph of the angular velocity for Subject 16

appears in Figure 4.5. The angular velocities reported in

 

 

 

 

 

 

 

 

fi‘jj’fijfijjj“ ‘1'1VTV'TTV‘W Tif‘

00.? .5 1.0 115 2.0 2.5 3.0 3.5
Time (See)

Figure 4.5 Angular Velocity of Center of Mass
of the Whole Body (Subject 16)

75

this investigation were somewhat lower than those reported
in other research on men's overgrip giant swings. Cheetham
(1984) reported a mean peak angular velocity of 318 deg/sec,
which occurred at the end of Quadrant 3. Boone calculated
angular velocities of 291.0 deg/sec and 304.9 deg/sec for
the two male subjects in his study. The female gymnasts
lower maximum angular velocity may be related to their use
of a larger size bar and to the technique of piking on the
downswing.

The mean peak angular acceleration of the total group
was 1159.60 deg/sec2 and occurred in Quadrant 1, 33.0
percent into the cycle. There were no significant

differences between the groups on these variables.

Joint_Anal¥Sia

Mean peak resultant velocity and acceleration values
were calculated for selected joints of the body (ankle,
knee, hip, and shoulder) and compared using a one way ANOVA.'
A significant difference (p < .10) was found between Group 1
and Group 3 on mean peak resultant velocity of the ankle,
knee, and hip. Table 4.7 summarizes the mean peak
velocities of selected joints. The total group mean peak
resultant velocity for the ankle was 10.47 m/sec, while
Group 1 was 11.03 m/sec and Group 3 was 9.64 m/sec. For the

hip, the mean peak resultant velocities were 5.98, 8.22 and

5.57 m/sec.

respectively.

for the Total Group, Group 1, and Group 3,

Table 4.7 Mean Peak Resultant Velocity of the Selected

 

 

Joints
Ankle*** Knee* Hip*** Shoulder Elbow
(m/sec) (m/sec) (m/sec) (m/sec) (m/sec)
Total Group +10.47 7.16 5.76 3.64 2.66
(.793) (.323) (.427) (.657) (.359)
Group 1 11.03 7.30 6.22 3.91 2.64
(.372) (.191) (.326) (.751) (.235)
Group 3 9.84 6.95 5.57 3.24 2.68
(.338) (.369) (.186) (.071) (.543)
*Mean
(S.D.).
* p < .10 *** p < .01.

There also was a significant difference (p < .10) between
the groups for the time of occurrence of peak velocities of
the ankle and shoulder expressed as a percent of total time.
Tables D.7 and D.8 in Appendix D give the complete
statistical information for the peak velocities and their
relative times of occurrence.

When examining the mean peak resultant velocities per
quadrant, a significant difference (p < .10) was found
between the groups for the hip and elbow in Quadrant 2; for
the ankle and shoulder in Quadrant 3; and for the knee, hip,
and shoulder in Quadrant 4. Tables D.9 and D.l0 in Appendix

D show the statistical data for mean peak resultant

‘IIIOC,'

Resultant velocity

77

velocities per quadrant and the times of occurrence of peak
values expressed as a percent of total time.

Figures 4.6 and 4.7 show the resultant linear velocity
of the hip and ankle joints for Subject 16 and Subject 6,

typical of Group 1 and Group 3, respectively. The graphs are

r \g/g

E233 ..

 

 

 

 

 

 

 

 

 

fij'fij‘ VVVjI—V‘f‘

2.0 2.5 3.0 5.5

Time (Sec )

Figure 4.6 Resultant Velocity of Ankle and Hip
(Subject 18)

76

 

10

 

 

 

 

 

 

 

 

 

Viv‘jfij‘v ‘jjfitjwvv

c’00 ' '01: 1.0 ' 1 11.5 2.0 235 5.0 3.5

lesul tant vein: 1 ty (elsec)

Time (Sec )

Figure 4.7 Resultant Velocity of Ankle and Hip
(Subject 8)

similar, but Subject 16 had a larger peak resultant velocity
for the ankle (10.77 m/sec) and hip (5.73 m/sec) than
Subject 6 for the ankle (9.0 m/sec) and the hip (5.40
m/sec). For all gymnasts, the peak resultant velocity of the
ankle occurred in Quadrant 3, as the gymnast began the

upswing phase of the skill. The timing of the maximum

79

velocity of the ankles agrees with the results of Borms, et
a1. (1976), although his gymnasts showed a larger resultant
velocity (12.5 m/sec). In the graph of Subject 16, there
was a slowing of the ankles at the end of Quadrant 1.
Following an increase in velocity, the ankles slowed to 3.67
m/sec in Quadrant 2 before the rapid rise to the peak
velocity in Quadrant 3. During a similar time of
-performance, Subject 6 did not show the same pattern of
slowing. The velocity of the ankle of Subject 6 remained
relatively constant. This slowing of the ankle of Subject 16
in Quadrant 2, was the beginning of the “tap" that is
emphasized by coaches when they teach the overgrip giant
swing. The gymnast is asked to reach her feet back under the
low bar after she clears that bar. This extends the gymnast
at the hips and delays the feet in their cycle around the
bar. When the total body center of mass travels under the
high bar, the gymnast is encouraged to be in an extended or
slightly hyperextended hip position. During Quadrant 3, the
gymnast then begins the piking action and the velocity of
the feet is greatly enhanced. Subject 16 has managed to
"tap" during her performance of the overgrip giant swing,
whereas Subject 6 did not demonstrate this technique.
Subject 6 maintained a slight pike at the hips throughout
Quadrants 1 and 2. Therefore, she did not have sufficient

range of motion in the lower extremities to gain the

80

velocity necessary to carry her to a handstand above the bar
with a minimum of body position errors.

Both gymnasts showed two peaks in resultant velocity of
the hip; one occurred midway through Quadrant 2 and a lesser
peak occurred at the end of Quadrant 3. This pattern
indicated that the maximum resultant velocity of the hip
occurred before the body passed under the bar, slowed
somewhat while passing under the bar, then increased again
as the body began the upswing phase of the skill. These
occurrences indicated that the increase in resultant
velocity of the feet, during Quadrant 3, successfully
transferred some momentum to the hips to increase the
resultant velocity of the hips for the upswing phase of the
skill.

The mean peak resultant velocity of the shoulder of the
total group in this investigation agreed with the results of
Borms, et al. (1976). The mean peak resultant velocity of
the shoulder in this study was 3.6 m/sec, while Borms
reported a resultant velocity of 3.5 m/sec. The peak
resultant velocity of the shoulder in both studies occurred
in Quadrant 2.

Peak resultant acceleration values for the ankle, knee,
hip, shoulder, and elbow for the Total Group, Group 1, and
Group 3 appear in Table 4.8, with more detailed statistical
information appearing in Appendix D, Table D.11. There were

no significant differences in the peak resultant

61

acceleration values for these joints, but there was a
statistical difference (p < .10) between the groups in the
the time of occurrence of peak values expressed as a percent

of total time. Appendix D, Table D.12 contains maximum and

Table 4.8 Mean Peak Resultant Accelerations of the Selected

 

 

Joints
Ankle Knee Hip Shoulder Elbow
(m/secz) (m/secz) (m/secz) (m/secz) (m/secz)
Total 91.02+ 67.67 48.71 74.89 51.58
(13.4) (18.9) ( 8.1) (75.1) (20.3)
Group 1 94.78 65.72 49.56 99.54 44.12
(16.6) (18.4) ( 9.02) (91.9) ( 4.8)
Group 3 85.39 70.58 42.45 37.90) 82.77
( 3.4) (19.6) ( 4.63) ( 3.4) (30.2)
+Mean.
(S.D).

minimum values for mean peak resultant acceleration, as well
as mean and standard deviation values for the relative times

of occurrence of peak accelerations.

Mammals

Mean peak resultant, horizontal, and vertical
velocities and accelerations were calculated for the trunk,
thigh, shank, foot, arm, and forearm. There were significant
differences (p < .10) between the groups for the resultant
velocity of the trunk, foot, arm, and shank: the vertical

velocity of the shank and foot; and horizontal velocity of

62

the foot. Table 4.9 gives a summary of mean peak resultant
velocities for the body segments. Table 0.13 in Appendix D
shows the complete data for horizontal and vertical
velocities for each of the body segments. Also in Appendix
D, Table D.14 gives the time of occurrence of peak
velocities of each body segment expressed as a percent of
total time.

The differences between the groups in linear velocity
of the body segments seemed to indicate that the velocities
of the foot, shank, trunk, and arm are critical features of
the overgrip giant swing. By quickly rotating these segments
around the bar, fewer body position errors occur and fewer
deductions are seen by the judges. This is especially true
in the case of the trunk and arms. Both gymnasts show hip
hyperextension and arm flexion during Quadrant 4. However,
the more highly skilled gymnast (Subject 18) was able to
minimize the time interval in which these errors occurred.
This shorter time interval perhaps caused less severe
deductions from the judges since it is more difficult to see
errors in performance when done for a shorter period of
time.

The mean, maximum, and minimum horizontal, vertical,
and resultant accelerations for each body segment were
calculated. While there was a significant difference (p <
.10) for the mean peak vertical accelerations of the foot

and trunk and resultant acceleration of the foot, it was

83

Table 4.9 Mean Peak Resultant Velocities of the Selected
Body Segments

 

 

Total Group Group 1 Group 3
(m/sec) (m/sec) (m/sec)
Trunk“I 4.64+ 4.86 4.45
( .33) ( .32) ( .19)
Thigh 5.80 5.88 5.73
( .34) ( .37) ( .33)
Shank*** 8.56 8.94 8.18
( .50) ( .20) ( .41)
Foot** 9.72 10.33 9.12
( .92) ( .48) ( .86)
Arm* 3.08 3.42 2.74
( .64) ( .77) ( .17)
Forearm 1.94 2.07 1.82
( .27) ( .22) ( .28)
*Mean.
(S.D.).

*p < .10 ** p < .05 *** p < .01.

felt that the means were misleading due to acceleration sign
changes. Therefore, ranges of the peak accelerations are
reported. The peak resultant accelerations of the foot for
Group 1 ranged from 89.2 deg/sec2 to 122.2 deg/secz, while
for Group 3, the resultant accelerations ranged from 72.7
deg/sec2 to 95.3 deg/secz. The highly skilled gymnasts were
able to accelerate their feet to a greater magnitude than
the less skilled group. This greater acceleration of the
feet at the bottom of the swing again showed the existence
of a “tap" in the highly skilled group. Complete statistical
information is located on Table 0.15, Appendix D.

Graphs of the angular velocity of the trunk and shank

are shown in Figures 4.8 and 4.9. The plots of the trunk and

64

shank velocities were similar for both subjects. The angular
velocity of the trunk increased throughout Quadrants 1 and
2, reached a maximum at the end of Quadrant 3, and was of
similar magnitude for Subjects 18 and 6 (458.92 deg/sec2 and
448.00 deg/secz, respectively). Subject 16 showed a higher

Ms

 

 

 

 

 

000
:‘J‘. — runk
'8‘ ‘ " " M"
g .
400- '
Q 'l.
Ch ‘ ”
8 ’ r." '
V 200. o" I.
O; ‘ "It‘g' 0 0.
‘3 l. '. a t‘ .0
0 o. " ‘ ‘ .. 0
=> I :
.s ‘ ‘. a“
c:
E -200. '0'
i 01 02

 

 

 

 

0.0 0.: 1T0 ifs 2.0 2.:
Time (See)

Figure 4.8 Angular Velocity of Trunk and Shank
(Subject 18)

Anguolr Velocity (Deg /Sec)

85

 

   
   

 

 

I
I
I
I
I

 

 

 

 

 

V V V T I V V U j V j V U ' V U ‘U

0.0 ' 015' 1.0 1.5 2.0 2.5 3.0 3.5

Time (Sec ) '

Figure 4.9 Angular Velocity of Trunk and Shank
(Subject 6)

positive angular velocity (577.31 deg/sec) than Subject 8
(490.13 deg/sec) for the shank in Quadrant 3 and a greater
negative value (-208.87 deg/sec and -98.46 deg/sec,

respectively) for the first negative peak in Quadrant 2.

66

The plots of the angular velocities of the trunk and
lower limb of a male gymnast analyzed by Ishii and Komatsu
(1987) showed a different pattern than shown in the current
study. In their investigation, the angular velocity of both
limbs increased constantly until midway through Quadrant 2.
The authors concluded that the segments were accelerating
due to gravity.-They further showed that the angular
velocity of lower limb continued to increase constantly
until midway through Quadrant 3 where it began to decrease
gradually until the completion of the skill. The angular
velocity plots for the female gymnasts in the current study
were very different from those reported by Ishii and Komatsu
for male gymnasts. The female gymnasts showed a bimodal
path, which corresponds to the body position change

necessary to clear the low bar on the downswing.

Kinetic_Anal¥ais

This section on kinetic analysis is divided into two
main headings: forces applied to the bar and joint forces
and moments.

Earnes—Annlisd_tn_tha_Bar

The force data were collected by applying strain gages
to the upper bar of the uneven parallel bars. The sampled
data were changed to digital form by an analog-to—digital

converter and then transmitted to a IBM 9000 dedicated

87

computer. A computer program was written to generate force-
time and torque-time histories. Figures 4.10 and 4.11 show
the resultant force (normalized by body weight) applied to

the bar by Subjects 16 and 6, respectively. Subject 16

 

I501

 

-gp
suit
'15.!
.an.
.35.. Peak Ieese I 1459.5 I
.m'.
.au.
~400‘ i

Resultant Force (100*N/Body Weight)

 

 

 

 

 

 

 

 

 

Time (See)

Figure 4.10 Resultant Force Applied to the Bar
(Subject 16)

88

fig

3501?
'0‘
2501

 

150‘
100'!
5.1
.-

 

'10”
.1504
-2...
'250‘ Peat terse 8 1203.! I
.3...
.3...
.q...

Resultant Force (100*N/Body Weight)

 

 

 

 

 

 

 

 

Time (See)

Figure 4.11 Resultant Force Applied to the Bar
(Subject 6)

produced a peak resultant force of 1489.6 N or 3.40 times
body weight. Subject 6 produced a peak resultant force of

1203.9 N or 3.94 times body weight. The peak resultant force

89

occurred midway through Quadrant 2 for all gymnasts. The
force on the bar then began to decrease and level off as the
gymnast's center of mass passed under the high bar at the
beginning of Quadrant 3. The force continued to decrease for
Subject 6. However, another small peak occurred for Subject
16 prior to the end of Quadrant 3. It was hypothesized that
this peak represented a grip change or sliding of the hands
around the bar by the highly skilled gymnasts timed to
correspond with a recoil action of the bar. Sands (1987)
noted a deflection of the bar at this point and he concluded
that this was related to the "tap" the gymnast performed
earlier in the overgrip giant swing. The "tap" may allow the
gymnasts to accelerate their feet and eventually result in
an unweighting of the hands so that the gymnast may slide
the hands around the bar in order to secure a better grip
for the upcoming handstand position.

Resultant forces reached a minimum near the beginning
of Quadrant 4. For Subject 18, this minimum resultant force
occurred just after the center of mass passed the height of
the bar (the beginning of Quadrant 4) while for Subject 6
this minimum occurred just before the end of Quadrant 3.
During Quadrant 4, another small force was applied to the
bar. At this point in the overgrip giant swing, the gymnast
pulled with the arms to help bring her body over the bar
into a handstand position. Ishii and Komatsu (1987) also

noted an application of a small force near the end of the

overgrip giant swing and concluded that it was due to body
position changes or an application of force to the bar in
order to increase the rotational velocity.

Figures 4.12 and 4.13 show the vertical forces on the

 

4... Peak Ieeee I .1451.3]
3501 91 02 Q3 9‘
130.
ml
lflh
5‘.‘ l

 

 

 

 

 

 

Vertical Force (100tN/Body Weight)
I
F :-

 

 

 

 

Time (See)

Figure 4.12 Vertical Force Applied to the Bar
(Subject 16)

91

is

«n. Imnlfluue I-HDJAII
3" 01 02 03 04

 

1.01

 

-1501

-801
'm'
.301

Vertical Force (100*N/Body Weight)
2

 

— q—
l l fi
2

“s ‘ ‘e‘

 

 

 

 

 

 

 

Time (See)

Figure 4.13 Vertical Force Applied to the Bar
(Subject 6)

bar produced by Subject 18 and Subject 6, respectively. A
negative force on the graph represents force being applied

to the bar in a downward direction. Subject 16 produced a

92

maximum vertical force of 1451.3 N directed downward or 3.4
times body weight, while Subject 6 produced a maximum
vertical force of 1204.4 N or 3.9 times body weight. Ishii
and Komatsu (1987) found that the peak vertical velocities
of his five male subjects ranged from 3.5 to 4.9 times body
weight, with the unskilled gymnast producing the lowest
force. This peak force occurred midway through Quadrant 2
for the female gymnasts in this study, while the male
gymnasts in Ishii and Komatsu's (1987) study produced their
maximum vertical force just after passing under the high bar
(at the beginning of Quadrant 3). Other studies (Cureton,
1939; Boone, 1977) reported that the maximum force occurred
at the bottom of the swing. As the force on the bar
decreased, Subject 16's plot showed a short period of time
where the force was constant just as the center of mass
passed under the bar at the end of Quadrant 2. This short
period of constant force, being applied as the gymnast's
center of mass passed under the bar, may be related to the
"tap“ observed by Sands (1987). The force for Subject 6,
however, continued to decrease.

Both graphs show a small peak of vertical force being
produced midway through Quadrant 3, which perhaps was a
recoil of the bar, then a minimum was reached shortly after
entering Quadrant 4. The application of a small force during
Quadrant 4, as previously discussed, was especially evident

on the vertical force graphs.

93

Figures 4.14 and 4.15 are plots of the horizontal force
applied to the bar by Subjects 16 and 6, respectively. As
seen in Figures 4.14 and 4.15, a negative horizontal value

indicated a force that tends to displace the bar to the left

riw/

an. Ieah Ieeee e-uquJ
3'1 01 02 l 03 04

 

 

 

'1.1
.m.
ea”.
.3...
o'd
0‘”.

 

Horizontal Force (100tN/00dy Weight)
1

 

 

 

h- —

1 I 1
‘ .5 I 0.5 3 3.5 0

 

 

 

 

 

 

Time (See)

4.14 Horizontal Force Applied to the Bar
Figure (Subject 16)

94

, l...

egg. Peek lease 0 442.0 I

 

 

 

 

 

 

 

 

 

g m. 01 02 03 ‘ G4

3 m.

g 3501

5» 200‘

fig lflh

S 10%

a "‘

59'; "

v '5.-

o -1001

3 use.

:9. -m-

0.. -1501

3 40m

3””

- «moi

L——_ ._._

" ' I T T j I
.5 I 1.5 2 2.5 3
Time (See)

Figure 4.15 Horizontal Force Applied to the Bar
(Subject 8)

and a positive value to the right. Subject 16 produced a
maximum of -1104.4 N or 2.6 times body weight. This occurred

at the beginning of Quadrant 2, on the subject's downswing

95

phase of the skill. A second, positive peak of horizontal
force occurred at the beginning of Quadrant 3, just after
the gymnast's center of mass passed under the bar. This peak
was 664.7 N or 1.5 times body weight. Figure 4.15 shows a
similar pattern for Subject 6, who had a maximum horizontal
force of -743.0 N or 2.4 times body weight. The occurrence
of this peak force was at the end of Quadrant 1 and was
slightly sooner than observed for Subject 16. The second
peak, which was 503.1 N or 1.7 times body weight, occurred
at the beginning of Quadrant 3. During the upswing phase of
the skill, the horizontal force gradually decreased until
the giant swing was completed.

Figures 4.16 and 4.17 are graphs of the moment applied
to the bar by the gymnasts performing the overgrip giant
swing. Subject 16 (see Figure 4.16) had a peak moment of
39.43 Nm. Subject 6 (see Figure 4.17) had a peak moment of
31.17 Nm. The peak moments for both gymnasts occurred midway
through Quadrant 2 and decreased until the beginning of
Quadrant 4. The peak moment of force corresponded to the
peak vertical and resultant forces produced by the gymnasts
in this study.

No statistical differences were found between the
groups on peak moments or peak resultant, vertical, or
horizontal forces applied to the bar or on the relative
times of occurrences of the peak forces or moments. Table

4.10 contains a summary of the Total Group, Group 1, and

96

Group 3 peak moments and peak resultant, vertical, and
horizontal forces. In addition, the relative time of the

occurrences of each peak force and torque is reported in

 

egg. Peak eemost I 39.43 h

m. . 93 04

300s
150‘
I 5"

.zs—v MN,»

'I.‘
-1501
noel
.2501
.3...

AA AA

 

Moment (1000*Nm/Body Weight)

.0... 1

 

 

 

 

 

 

 

 

 

Time (See)

Figure 4.16 Moment Applied to the Bar
(Subject 16)

97

is

«In hall ease-t - I ’J? in
150. 01 02* 03. Q4

 

353.1/\_\

'1504
-2001

Moment (1000*Nm/Body Weight)

.300 .
'35. .
«I. .

 

 

 

H I—
1

 

 

h‘l

 

 

 

Time (Sec)

Figure 4.17 Moment Applied to the Bar
(Subject 8)

Table 4.11. For each of the above variables, the mean,

standard deviation, and range were shown in Tables 4.10 and

4.11.

98

Table 4.10 Forces and Moments Applied to the Bar

 

Tota1___Grnun_1_Groun_3 Total___Grnua_1_Grnun_3

Vertical_Eorcer Horizontal_2ozcaf
Mean 3.082 2.286 3.872 1.979 1.532 2.426
SD 2.598 3.687 0.107 1.838 2.352 0.065
Min -4.280 —4.280 3.710 -2.660 -2.660 2.380
Max 4.170 4.170 3.990 2.790 2.790 2.540

Moment_Lle. Resultant_£nrcef
Mean -41.3 -44.0 -38.6 3.090 2.294 3.888
SD 9.3 6.2 11.7 2.596 3.683 0.118
Min -55.5 -49.1 -55.5 ~4.270 -4.270 3.710
Max —29.5 -35.5 -29.5 4.160 4.160 3.980

 

+Normalized by body weight.

Table 4.11 Times of Occurrences of Peak Forces*

 

Total Group 1 Group 3 Total Group 1 Group 3
Iartical_Enrca Horizontal_£nnca

Mean 56.26 58.66 53.88 48.27 51.42 45.12
SD 4.96 4.40 4.64 6.35 6.40 50.19
Min 48.00 54.30 48.00 41.00 45.30 41.00
Max 65.50 65.00 60.40 61.70 61.70 53.60

Moment Resultant_£nrca
Mean 56.71 58.94 54.48 56.33 58.68 53.98
SD 5.06 4.30 5.18 5.26 4.54 5.30
Min 48.00 54.30 48.00 47.50 54.30 47.50
Max 65.50 65.50 81.90 65.90 65.90 61.50

 

* expressed as a percent of total time

The force data obtained in this study showed lower

resultant, vertical, and horizontal forces compared to

99

forces reported in studies of the overgrip giant swing
performed by men on the horizontal bar. Kopp and Reid (1980)
reported a resultant force of 3.6 times body weight while in
the current study the same variable was found to be 3.1
times body weight. Ishii and Komatsu (1987) found a mean
horizontal force of 2.5 times body weight and a mean
vertical force of 4.3 times body weight. The current
investigation found these variables to be 2.0 and 3.1 times
body weight, respectively. The peak torques found by Kopp
and Reid (1980) were 23.3 Nm while in the current study the
peak torque on the bar was reported to be 41.29 Nm. It was
hypothesized that this difference in torques can be
explained by the larger bar size for the women. Since the
female gymnasts' hands are smaller in relation to the bar
size, they must grip the bar more tightly, as compared to
the men gripping the horizontal bar. In the study by Ishii
and Komatsu (1987), the peak resultant force occurred just
after the gymnast passed under the high bar, or at the
beginning of Quadrant 3. Kopp and Reid (1980) reported that
the maximum force occurred just before the gymnast reached
the bottom of the swing, or at the end of Quadrant 2. For
the current study, the peak resultant force occurred midway
through Quadrant 2. The body position change, during the
descent phase, necessary to clear the low bar also changed

the timing of the maximum force.

100

mm

The forces and moments measured at the bar and
accelerations calculated from film data were read into a
FORTRAN computer program written by this researcher. The
program used equations (1) - (6), expressed in Chapter 3, to
calculate the muscle moments and joint forces at the elbow
and shoulder. Each of the forces and moments were halved to
represent the forces on one side of the body.

No significant differences between the groups on the
mean peak joint forces, moments, or relative time of
occurrence of the peak forces were found. Table 4.12 reports
the mean peak forces and moments of the total group for each
joint. Table D.16 and D.17 in Appendix D reports the forces
and the relative times of occurrence for the individual
groups.

When examining the mean peak joint reaction forces per
quadrant, (see Appendix D, Table D.18), there was a
significant difference (p < .10) between the groups for the
horizontal force for the shoulder in Quadrant 1. The forces
were -0.5, -0.4 and -0.6 times body weight for the Total
Group, Group 1, and Group 3, respectively. The less skilled
gymnasts in this investigation experienced significantly
more horizontal force at the shoulder joint than the more
highly skilled gymnasts. This may be due to the less skilled
gymnasts piking earlier during the descent phase of the

overgrip giant swing. There was also a significant

101

difference (p < .10) between the groups for the time of
occurrence of peak moments expressed as a percent of total
time.

For the elbow joint there were significant differences
(p 4 .10) between the groups for the vertical force in
Quadrant 4. This force occurred as the gymnast attempted to
attain the handstand position above the bar. The less
skilled gymnasts experienced a higher vertical force at the
elbow during this phase. This could be due to the slower

vertical movement of the hips and feet.

Table 4.12 Mean Peak Joint Forces and Moments at the Elbow
and Shoulder for the Total Group

 

Mean SD Min Max
FY Elbow* -1.56 1.33 -2.09 2.22
FX Elbow+ -0.74 1.11 -1.44 1.36
Moment Elbow '-54.11 55.28 -97.80 96.80
Flughoulder* 1.62 1.28 -2.01 2.17
FX Shoulder+ -1.28 .09 -1.41 -1.14
Mgsegt Shoulder 97.04 27.24 50.80 134.30
m

 

+Normalized by body weight.

There was a significant difference (p < .10) between

the groups for the relative time of occurrence of the

102

vertical force in Quadrant 1, for the relative time of
occurrence of the moments, and the relative time of the mean
peak horizontal force in Quadrant 2. Complete analysis for
the individual groups appear in Tables D.18 - D.21 for the
forces per quadrant and relative time of occurrences for the

forces for the shoulder and elbow joints.

CHAPTER V

Summary and Conclusions

Summarx_nf_£rnnaduras

The purpose of this study was to identify critical
elements of the performance of the overgrip giant swing on
the uneven parallel bars by comparing highly skilled and
less skilled gymnasts on selected kinematic and kinetic
parameters. The kinematic variables included temporal
analysis; position data for shoulder, hip, and center of
mass; velocity and acceleration of the center of mass;
angular velocity and acceleration of a line segment
connecting the center of the bar and the center of mass:
linear velocity and acceleration of the ankle, knee, hip,
shoulder, and elbow joints; and velocity and acceleration of
the center of mass of the trunk, thigh, shank, foot, arm,
and forearm. The kinetic data, which were collected by
instrumenting the high bar with strain and torque gages,
included the following analyses: resultant, vertical, and
horizontal forces: moments applied to the bar: and
calculated joint reaction forces for the elbow and shoulder
joints. A comparison was made between the results obtained
in this study with the results obtained from investigations
of male gymnasts performing an overgrip giant on the

horizontal bar.

103

104

The subjects for this study were 15 female gymnasts
ranging in age from 10 to 16 years. They were all Class I or
Elite gymnasts from five gymnastics clubs in Michigan.
Anthropometric data and hydrostatic weight were collected at
the Center for the Study of Human Performance at Michigan

State University on the day that kinematic and kinetic data

were collected. The subjects were then transported to a
local gymnastics club where they were prepared for filming.
The preparation included placing joint markers on the right
side of their bodies (center of the wrist, elbow, shoulder,
hip, knee, and ankle joints) as reference points for
digitizing. Following a practice trial, the gymnasts were
asked to perform a sequence of four skills (long hang kip,
cast clear hip to handstand, overgrip giant swing, and
flyaway dismount) on the uneven parallel bars. The sequence
of skills were filmed with two LOCAM 16mm pin registration
motion picture cameras, one perpendicular to the plane of
movement of the gymnast and the other placed to record a
frontal view. The cameras were set at 100 frames/sec and the
filming speed was calibrated from timing lights placed in
the field of view. Kodak Video News high speed film with an
ASA of 400 was used in this investigation.

The film was transferred to video tape, which was
edited so that the overgrip giant swings were the only skill
left on the tape. Four Elite and Class I judges then

independently evaluated the 15 overgrip giant swings and

105

awarded a score to each gymnast based on criteria set by the
experimenter. These criteria were similar to deductions that
judges use in evaluating uneven parallel bar routines. Skill
groups were then determined on the basis of the judges'
ratings. The top five scoring gymnasts were placed in Group

1 (highly skilled) and the five lowest scoring gymnasts were

placed in Group 3 (less skilled). In order to compare
diverse performance, Group 2, the intermediate group was
omitted from further study.

After processing, the high speed film was projected
. from overhead onto a drafting table by a Vanguard Motion
Analyzer and digitized using a Sonic Graf/Pen system. The
digitizing system was interfaced with an IBM-PC, which
collected the X and Y coordinates in a data acquisition
computer program. A FORTRAN Bioanalysis program, developed
at Michigan State University, was used to analyze the

kinematic data.

5 E 2' li E ll 3' I' n l I]

Centar_n£_Mass_Anal¥Sis

This study found a significant difference (p < .10)
between Group 1 and Group 3 for the mean peak horizontal
velocity of the total body center of mass, which was 4.89
and 4.38 m/sec, respectively. There was also a significant

difference (p < .10) for the timing of the peak vertical

106

velocity of the total body center of mass. The peak vertical
velocity occurred at 57.9 and 40.3 percent of the way
through the giant swing for Group 1 and Group 3,
respectively. The highly skilled gymnasts were able to
generate a significantly (p < .10) higher horizontal
velocity of the total body center of mass (4.9 and 4.4 m/sec
for Groups 1 and 3, respectively) and that mean peak
velocity occurred later in the cycle. This enabled the
gymnast to complete the overgrip giant swing without major
form breaks.

When the peak velocity data of the total body center of
mass for each quadrant were examined, there was a
significant difference (p < .10) between the groups for the
horizontal velocity in Quadrant 3. The velocities were 4.33
and 3.64 m/sec for Group 1 and Group 3, respectively. There
was also a significant difference (p < .10) between the
groups for the mean peak resultant velocity during Quadrant
4. These velocities were 3.87 m/sec for Group 1 and 3.21

m/sec for Group 3.

Joint_Anal¥Sis

There were significant differences (p < .10) between
the groups for the mean peak resultant linear velocities of
the ankle, knee, and hip joints. Group 1 had a mean peak
resultant velocity of 11.03 m/sec for the ankle while Group

3 showed a mean peak resultant velocity of 9.64 m/sec. The

107

mean peak resultant velocities for the knee were 7.30 m/sec
for Group 1 and 6.95 m/sec for Group 3. The mean peak
resultant velocities for the hip were 6.22 and 5.57 m/sec
for Group 1 and Group 3, respectively. There was a
significant difference (p < .10) between the groups for the
time of occurrence of peak values of the ankle expressed as
a percent of total cycle time. The mean peak resultant
velocity of the ankle occurred at 65.1 and 58.4 percent
through the cycle of the overgrip giant swing for Groups 1
and 3, respectively. There were also significant differences
(p <.10) between the groups for the time of occurrence of
peak values of the hip and shoulder expressed as a percent
of total time. The mean peak resultant velocity occurred at
55.3 and 48.6 percent of the cycle for the hip and at 61.0
and 48.5 percent for the shoulder for Groups 1 and 3,
respectively.

When examining the mean peak resultant velocity of the
joints by quadrant, significant differences (p < .10) were
found between the groups in Quadrant 2 for the hip and
elbow; in Quadrant 3 for the ankle, shoulder and hip; and in
Quadrant 4 for the knee, hip, and shoulder.

The mean peak resultant velocities of the ankle, knee
and hip joints were of greater magnitude and occurred later
in the performance of the overgrip giant swing of the more
highly skilled gymnasts when compared to the less skilled

group. It was concluded that the gymnasts who “tapped" at

108

the appropriate time were able to achieve greater mean peak
resultant velocities and achieve them later in the overgrip
giant swing, thereby arriving in the handstand position on

the top bar without severe form breaks.

SegmentJnalysis

An examination of the mean peak velocities of the
segments of the body revealed a significant difference
(p < .10) between the groups for the mean peak vertical
velocity of the foot and mean resultant velocity of the
shank. Group 1 had a mean peak vertical velocity of the foot
of 10.4 m/sec, while Group 3 had a velocity of 9.0 m/sec.
For the mean peak resultant velocity of the shank, Group 1
had 8.9 m/sec and Group 3 had 8.2 m/sec. There was also a
significant difference (p < .10) between Groups 1 and 3 for
the mean peak resultant velocity of the trunk (4.9 and 4.5
m/sec, respectively), mean peak vertical velocity of the
shank (7.7 and 7.1 m/sec, respectively), and mean peak
horizontal velocity (10.3 and 9.1 m/sec, respectively) and
mean peak resultant velocity (11.7 and 10.4 m/sec,
respectively) of the foot. There was also a significant
difference (p < .10) in mean peak resultant velocity of the
arm for Groups 1 and 3 (3.4 and 2.7 m/sec, respectively).

There were significant differences (p < .10) between
the groups for the relative time of occurrence of the mean

peak resultant velocity for the shank; the mean peak

109

vertical velocity of the shank; and the mean peak resultant
velocity of the trunk. The examination of the mean peak
horizontal, vertical, and resultant acceleration data showed
the following significant differences (p < .10): in the mean
peak resultant and vertical acceleration of the foot; and

the mean peak vertical acceleration of the trunk.

S E E' l' E I] K' l' M . l]

Eorcas_Annliad_tn_the_Baz

While there were no significant differences between the
groups for the horizontal, vertical and resultant forces
applied to the bar, it was interesting to compare the forces
obtained in this study with forces reported in the
literature available for the overgrip giant swing and other
similar skills. The current study found that the mean peak
resultant force on the bar was 3.1 times body weight. Kopp &
Reid (1980) recorded a mean peak resultant force of 3.6
times body weight. Hay et al. (1978) found a resultant force
of 3.6 times body weight for a long swing type movement on
the uneven parallel bars. Each of these studies, in which
the researchers directly measured forces by instrumenting
the bar resulted in force values considerably lower than
those reported in studies that indirectly calculated the
force. Smith (1981) theoretically determined a force of 5.0

times body weight applied to the bar for a skill similar to

al

110

the overgrip giant swing. In his early work, Cureton (1939)
estimated that the peak force on the bar to be 4.9 times
body weight for a male gymnast performing an overgrip giant
swing.

The present investigation found the mean vertical and
horizontal forces to be 3.1 times body weight for Group 1

and 2.0 times body weight for Group 2. Ishii and Komatsu
(1987) found the mean peak vertical and horizontal forces
for men to be 4.3 and 2.5 times body weight, respectively.
Ishii and Komatsu (1987) reported that the occurrence of the
‘ peak resultant force was at the beginning of Quadrant 3,
just after the gymnast passed under a vertical line
projected by the bar. Kopp and Reid (1980) reported the peak
force occurred at the end of Quadrant 2, just before the
body passed under the bar. In the present study, the mean
peak resultant force occurred earlier, just after the
gymnast had cleared the low bar on the descent phase of the
skill. It was hypothesized that since a body position change
was necessary in order for the gymnast to clear the low bar
during the descent phase of the skill, the expected pattern
of the peak force occurring at the bottom of the swing was
altered. The gymnast must pike during Quadrant 1 in order to
clear the low bar. Having cleared the low bar, the gymnasts
again extended at the hips during Quadrant 2. This extension

of the hips may account for the earlier production of the

111

mean peak resultant force as compared with the men's
overgrip giant swing on the horizontal bar.

The mean peak torque found in the current study was
41.3 Nm with a range of 30.3 to 55.5 Nm. The mean peak
torque reported by Kopp and Reid (1980) was 23.3 Nm. It was
hypothesized that the larger bar of the uneven parallel bars
was the factor that created this difference. Since the
female gymnasts' hands are smaller in relation to the bar
size, they must grip the bar more tightly, as compared to

the men gripping the horizontal bar.

Joint_Enrsss.

Using the equations (1)-(6) reported in Chapter 3,
joint reaction forces were calculated for the elbow and
shoulder joints. There were no significant differences
between the groups for mean peak vertical or horizontal
force or for moments at the elbow and shoulder joints. The
total group mean peak vertical force on the elbow joint was
1.8 times body weight. The range of vertical forces was 1.2
to 2.0 times body weight. The total group mean peak
horizontal force on the elbow was 0.7 times body weight,
with a range of 0.2 to 1.2 times body weight. The vertical
force for the shoulder joint for the total group was 1.6
times body weight with a range of 1.3 to 2.0 times body

weight. The horizontal force of the shoulder joint was 1.3

112

times body weight for the total group with a range of 1.1 to
1.4.

When examining the joint reaction forces per quadrant,
a significant difference (p <.10) occurred in the horizontal
force of the shoulder and in the moment of the elbow in

Quadrant 1.

Gymnastics is a sport that has aesthetic qualities,
which are reinforced by the rules. The table of deductions
in the Federation of International Gymnastics Code of Points
(1985) emphasizes that the aesthetics of the sport require
straight body lines, straight arms, and straight legs even
though the ease of execution may be enhanced by the athlete
performing a skill with one or more deviations from the
ideal. Often, when a new skill is performed, a less than
perfect body line is acceptable, until gradually the
execution improves with each succeeding generation of
gymnast. This is the case with the overgrip giant swing
being performed on the uneven parallel bars. When first
performed, it was executed with a very large arch in the
back during the ascent phase. This arch helped to compensate
for the relatively small radius of rotation due to the
gymnasts' necessity to clear the low bar. The giant swing

was looked at as being a very difficult, high level skill on

113

its own. As time went on, some gymnasts began to execute the
skill with noticeably less deviations in body line. They had
an excellent model to follow as the skill has been performed
by men on the horizontal bar for a very long time. Like the
men, the female gymnasts also started to use the overgrip
giant swing as a means of generating speed for the execution
of large release movements. The goal today is not only to
execute the skill, but to execute it with speed and force to
be able to use the skill to lead into more difficult
movements.

The female gymnast has one very large constraint placed
upon her that her male counterpart does not have. That
constraint is the proximity of the low bar of the uneven
parallel bars. The male gymnast is fully extended, even
hyperextended slightly, during the descent phase of the
skill in order to maximize the velocity of the downswing.
Due to gravitational and relatively constant frictional
forces and air resistance, the velocity of the ascent phase
of the circle is always less than the descent phase. The
gymnast usually compensates for this decrease in velocity by
decreasing the length of the radius of rotation during the
ascent phase. The most acceptable method for accomplishing
this decrease in length of the radius for men on the
horizontal bar is to decrease the shoulder and hip angles
after passing under the bar (Gluck, 1982). Cheetham (1984)

has stated that for regular overgrip giant swings, these

114

angular decreases are very slight. When the gymnasts on the
horizontal bar want to increase the angular and linear
velocity of the giant swing for large release moves and
dismounts, they decrease the hip and shoulder angles even
more than for the regular overgrip giant swing.

The female gymnasts performing the overgrip giant swing
on the uneven parallel bars produced slightly less linear
and angular velocities during the descent phase than their
male counterparts. In the current study, the total group
mean peak resultant velocity was 5.6 m/sec and occurred
during Quadrant 2. Borms (1976) reported a peak resultant
velocity of 6.0 m/sec., which also occurred in the second
quadrant. While the differences were not statistically
significant, it was interesting to note the trends between
the groups on this variable. Group 1 achieved mean peak
resultant velocity of 6.2 m/sec at the end of Quadrant 2,
while Group 3 only had a mean peak resultant velocity of 5.1
m/sec, which occurred earlier in Quadrant 2. There was a
significant difference (p < .10), however, between the
groups in the horizontal velocities, which were 4.9 and 4.4
m/sec for Group 1 and Group 3, respectively. Even more
interesting was the difference in the timing of the peak
vertical velocity, which showed a significant difference (p
< .10) between the groups. The mean peak vertical velocity
occurred slightly before the end of Quadrant 2 for Group 1,

but at the end of Quadrant 1 for Group 3. The highly skilled

115

gymnasts were better able to increase the peak vertical
velocity by starting in a higher position on the bar, as
recommended by George (1980), and by keeping the shoulder
angle more flexed during the descent phase. In addition, the
highly skilled gymnasts resumed a more extended hip position

earlier in the cycle, after clearing the low bar. Achieving

this extended hip position well before reaching the bottom
of the swing has an important function. Sands (1987)
discussed the importance of a "tap" when the body is
positioned below the bar. He suggested that the "tap" was
used to achieve greater velocity of the body on the ascent
phase of the swing.

In the present study, there was a significant
difference (p < .10) between Groups 1 and 3 on the mean peak
resultant velocity of the ankle joint. The velocities were
11.0 and 9.6 m/sec, respectively. There was also a
significant difference between the groups on the timing of
this peak velocity. Group 1 was able to produce the peak
velocity later in Quadrant 3 than was Group 3. It was
observed that the highly skilled gymnasts were able to
achieve a straighter body position before the bottom of the
swing, while the less skilled gymnasts maintained a slight
pike throughout Quadrant 2.

Sands (1987) studied rail movement of several skills,
including the overgrip giant swing. He determined patterns

of bar movements of gymnasts performing the overgrip giant

116

swing and observed that gymnasts who did not "tap"
effectively at the bottom of the bar showed a smoother curve
on the upswing phase of the skill. This was also seen to be
true for this investigation. The highly skilled gymnasts
showed a second peak in force on the graphs of the resultant
force. This second peak indicated that the bar was in the

process of recoiling and that another force had been added.
This was likely the result of the gymnast pulling on the bar
to aid in the upswing phase of the skill. Borms, et al.
(1976) observed this phenomenon in men's giant swings,
stating that the forceful arm pull resulted in an upward and
forward movement of the shoulders. In their study this
forceful arm pull occurred in Quadrant 4. In the present
study, this second peak occurred in Quadrant 3 indicating
that the female gymnast must exert this force earlier, since
she has less velocity coming through the bottom of the
swing.

The phase of the giant swing that receives the most
deductions from the judges is the ascent phase. Large
deductions (up to .3 for each error) are incurred for
bending the arms, bending the legs, and arching the body.
None of the gymnasts in this study performed deduction free
giant swings. Those who were rated the highest, however,
were gymnasts who maintained straighter body positions for a
longer period. Subject 16, who performed the best overgrip

giant swing on the uneven parallel bars, did show some arch

117

at the end of the overgrip giant swing, but for a very brief
period of time. It should be emphasized that the ascent
phase is dependent upon the descent phase.

The gymnasts in this study applied a mean peak
resultant force to the bar of slightly more than three times
their body weight. This figure was slightly lower than that
reported in the literature for men's giant swings. The
maximum force for the subjects in this study, also occurred
sooner, in Quadrant 2, than was reported in the studies on
the horizontal bar. It was hypothesized that this difference
was due to the body position change in order to clear the
low bar during the descent phase.

The stress placed on the elbow and shoulder joints of
the total group was of interest in this investigation. The
gymnasts in this study experienced a mean peak vertical
force of 1.6 times body weight on each elbow and shoulder.
The mean peak horizontal forces were 0.7 and 1.3
respectively. These figures agree with Robertson et al.
(1985) who found that a performer experienced 1.5 times body
weight on one arm while performing swinging skills on the

rings.

The purpose of this study was to identify critical

elements of the overgrip giant swing on the uneven parallel

116

bars. Based on the findings of this study, the author would

recommend the following suggestions to coaches for teaching

the overgrip giant swing:

Instruct gymnasts to:

1.

Start in as high a body position on the bar as

possible.

. Extend the handstand, so the body is completely

straight with the shoulders elevated.

. Keep the extended body position as long as possible

on the downswing.

. Pike the hips and keep the shoulder joints as

straight as possible, when clearing the low bar.

. Extend the hips as quickly as possible, as soon as

the low bar is cleared. This will put the

gymnast into position to perform a “tap".

. Shorten the radius of rotation during the

ascent phase by decreasing the angles of the
shoulder and hip joints, rather than by
hyperextending (arching) the hips. Once gymnasts
learn the recommended technique for shortening the
radius, they will be able to increase their velocity
during the ascent phase to set up more advanced

skills.

119

There has been a paucity of research in women's
gymnastics especially on recently evolving skills on the

uneven parallel bars. This author encourages others to

conduct additional research on high level skills performed
by highly skilled gymnasts as well as skills performed by
developing gymnasts. The largest percentage of participants
in the sport of gymnastics is children between the ages of
10 and 16. More research needs to be done to determine the
effects of gymnastic skills on the development of those
participants.

There needs to be additional data analyzed on joint
reaction forces and muscle moments. Also the long term
effects of forces and torques applied to the joints of young
gymnasts needs to be studied. In addition, because of the
inordinate amount of back problems in this population, there
need to be methods developed for studying the forces and

torques applied to the lower back of gymnasts.

List Of References

Bauer, W. L. (1983). Swinging as a way of increasing the
mechanical energy in gymnastics maneuvers. In H. Matsui
5 K. Kobayash (Eds), Biomechanins_!lll_fi.(Pp 801- 806).
Champaign, IL: Human Kinetics.

Bentham, R. J. (1987). A unified simulation technique
applied to single bar gymnastics skills. In T. B.
Hoshizaki, J. H. Salmela, 5 B. Petiot (Eds.),

D1agnnst1cs1_tteatmant_and_anal¥s1s_nf_gxmnastics
talent (PP. 67-75). Montreal: Sport Psyche Editions.

Bonniwell, R. (1983, November). Women s gymnastics: report
of the committee chairman and team manager.

Internationa1_Gymnast, pp. 859- $60.

Boone, T. (1977, February). Understanding the biomechanics
of the over - and reverse grip giant swing.

International_fixmnast. pp. 58- 59

Borms, J., Moers, R., 5 Hebblelinck, M. (1976).
Biomechanical study of forward and backward swings. In
P. V. Komi (Ed. ), Biomechanics_1=6.(99 309- -313).

Baltimore: University Park Press.

Bovinet, S. L. (1979). The dynamics of the kip on the uneven
parallel bars. In J. Terauds 5 D. B. Daniels (Eds.),
° ' (pp. 37-44). Del Mar, CA:
Academic Publishers.

Bowers, C,, Fie, J.,5 Schmid, A. (1981). Judging_and
cnaching_nnmen;s_gymnastics.(2nd ed.). Palo Alto, CA:
Hayfield Publishing Co.

Brown, E. (1983). Design of a film projection system for
analysis of cinematographic images. Jouzna1_n£_flaalth.
Ehxs1ca1_Educatinn1_Eec:eatinn_5_Danca., 56110). 150-
153. '

Brown, E., 5 Abani, K. (1985L Kinematics and kinetics of
the dead lift in adolescent power lifters. Medicina_and.

Sc1anca_1n_Spnnt_and_Exatciaa., 11 (5), 554- 586.

Cheetham, P. (1984). Horizontal bar giant swing center of
gravity motion comparisons. In J. Terauds, K.
Barthels, E. Kreighbaum, R. Mann, 5 J. Crakes (Eds.),

Santts_h1nmachan1cs (pp. 99-108). Del Mar, CA: Academic
Publishers.

120

121

Cheetham, P.J., Mizoguchi, H. , 5 Sreden, H. I. (1987,
December). Preliminary investigation of forces produced

by junior male gymnasts on rings. USGE_Iechnique, 1
(4). 4- 6.

Cureton, T. K. (1939). Elementary principles and techniques
of cinematographical analysis as aids in athletic

research. Reseatch_0uaztatlx. 10. 4-22.

Dainis, A. (1974L Analysis and synthesis of body movements
utilizing the simple n- -link system. In R. C. Nelson 5
C. A. Morehouse (Eds. ). Biomechanics_1!.(pp. 513- -518).
Baltimore: University Park Press.

Dainis, A. (1975, February). Dynamical analysis of ordinary
grip giant swings. menast. pp. 38-39.

Daniels, D. B. (1983). Biomechanical analysis of the Stalder
on the uneven parallel bars. In J. Terauds (Ed. ).
(pp. 209-230). Del Mar, CA:
Academic Publishers.

Daniels, D. B. (1987). Identification of critical features
in gymnastics skills through biomechanical and
statistical analysis. In T. B. Hoshizaki, J.H. Salmela,
5 B. Petiot (Eds.), D1agnnst1cs1_tzeatment_and_analxsis.
o£_g¥mnast1n_talsnte (pp. 76-95). Montreal: Sport
Psyche Editions.

Dempster, W. T. (1955). mm
opazatoz" (WADC Technical Report, pp. 55-159). Ohio:
Wright-Patterson Air Force Base.

Duck, T. A. (1976).
hnt1znnta1_hat.. Unpublished doctoral dissertation,
Pennsylvania State University, University Park, Pa.

Evans, G. G. (1984). Hip circle to handstand on the uneven
parallel bars. In J. Terauds, K. Barthels, E.
Kreighbaum, R. Mann, 5 J. Crakes (Eds.). Shanta
biomechanics (pp. 109-120). Del Mar, CA: Academic
Publishers.

Federation of International Gymnastics, (1985). Coda_nt
Enints_:_flomen£EIG. Switzerland: Author.

George, G. (1980, November). "Uneven , parallel
bars...overgrip giant swing...the descent phase...'see

how they fall'". USGE_menastics. pp. 16-17.

122

Gluck, M. (1982). _Mechan1cs_fot_gxmnast1cs_cnach1ng1_tnols
fint_skill_anal¥sis. Springfield, MA: Charles C. Thomas.

Hay, J. G. (1985).Iha_h1nmechan1cs_n£_spntts_tachn1quas,
(3rd Ed. ). Englewood Cliffs, NJ: Prentice-Hall.

Hay, J. G. Putnam, C. A. 5 Wilson, B. D. (1979). Forces
exerted during exercises on the uneven bars. Medicine

and_Science_1n_Snntts.,11 (2), 123- 130.

Herrmann, S. (1968). Motion recording of gymnastic exercises
by means of high speed camera shooting and their
analyses. In J. Wartenwiler, E. Jokel 5 M. Hebblelinck
(Eds.), B1nmechan1ns_l (pp. 320-323). New York: S.

Karger.

Ishii, K., 5 Komatsu, T. (1987). Changes of kinematic
parameters and forces on the horizontal bar backward
giant swing. In T. B. Hoshizaki, J. H. Salmela, 5 B.
Petiot (Eds. ), Diagnnst1cs1_tteatment_and_analxs1s_nt

gymnastics_talant (pp. 107-117). Montreal: Sport Psyche
Editions.

Kopp, P. M. 5 Reid, J. G. (1980, June). A force and torque
analysis of giant swings on the horizontal bar.
Canad1an_Jnuzna1_nf_App11ed_Spozt_Sciences 5 (2), 98-
102.

Kreighbaum, E. (1974). The mechanics of the use of the
reuther board during side horse vaulting. In R. C.
Nelson 5 C. A. Morehouse (Eds. ). Biomechan1cs.l¥.(pp.
137- -143). Baltimore: University Park Press.

Kreighbaum, E. (1983). Biomechanics research in gymnastics:
past, present and future. In J. Terauds (Ed.),
(pp. 193-207). Del Mar, CA:

Academic Publishers.

Landa, J. (1974). Shoulder muscle activity during selected

skills on the uneven parallel bars. Resaazch_Qnattezlx.
45. 120-127.

Lohman, T. G., Roche, A. F., 5 Martorell, R. (Eds.). (1988).
Anthtnmnmatt1c_standatd1zatinn_zefatanca_manual.
Champaign, IL: Human Kinetics.

Nissinen, M. A. (1983). Kinematic and kinetic analysis of
the giant swing on rings. In H. Matsu. 5 K. Kobayahi,
(Eds. ), Biomechanics_!111_A1_ptnceed1ngs_n£_the_8th
internat1nnal_congtass_n£_h1nmachanics (pp 781- -785).

Champaign, IL: Human Kinetics.

123

Robertson, E. A., Paul, J. P., 5 Nicol, A. C. (1985).
Investigation into the biomechanics of certain men's
Olympic gymnastics. In D. A. Winter, R. H. Norman, R.
P. Hells, K. C. Hayes, 5 A. E. Patla, (Eds. ),
W (pp. 377- -.382) Champaign, IL: Human

Kinetics.

Safrit, M. J. (1978). Beliehilii¥11henzxu Washington, DC:
AAHPERD Publications.

Sale, D. G. 5 Judd, R. L. (1974). Dynamometric
instrumentation of the rings for analysis of gymnastic
movements. Medieine_end_Se1enee_in_Sports, 8(3), 209-
216.

Sands, B. (1987, February). Analysis of uneven bar rail
during giant swing and flyaway. USECA_NeweieLtena pp.
1-11.

Sands, B. (1985). Aspects of the kip catch. Ieehninne111he
Qffin1al_Enbiinetinn_n£_the_flSGE1.5 (1). 4 3

Sands, B. (1985, July). Aspects of the straddle back to
handstand. USGE_IeehniQne1 5 (2). 4 9

Sands, B. (1982, December). Biomechanical graphic analysis
of the reverse hecht. .a first stage. USGE_Ienhninal
Jnn:nei.pp. 5-7.

Smith, T. (1981). Centrifugal forces during swinging from
handstand on the asymmetric bars. Inierneiinnal

Gxmnes11.1eehnieai_Snpniement. 8, 38-41.

Taylor, B., Bajin, B., 5 Zivic, T. (1972). lemnie
gymnasties_£nr_men_and_wnmen. Englewood Cliffs, NJ:

Prentice-Hall.

Tonry, D. (1980). SpnrLs_illnstra1ed1_wnmen;s_gymnasties_2.
New York: Lippincott 5 Crowell.

Vercruyssen, M. (1984). Anthropometric profile of female
gymnasts. In J. Terauds, K. Barthels, E. Krieghbaum, R.

Mann 5 J. Crakes (Eds.). Spnzts_hinmeehanina (99.121-
137). Del Mar, CA: Academic Publishers.

Wilkerson, J. D. (1978). K1nema11e_and_kine11e_analxsis1n£

Unpublished doctoral dissertation, Indiana University.
Bloominton, IN.

Winter, D. A. (1979). B1nmeehenies_n£_human_mnxement. New
York: John Wiley 5 Sons. .

124

Zinkovsky, A. V. Vain, A. A. 5 Torm, R. J. (1976).
Biomechanical analysis of the formation of a gymnastic
skill. In P. V. Komi (Ed. ), Binmenhnnins_!;3_.(PP 322-
325). Baltimore: University Press.

APPENDIX A

125

APPENDIX A
Informed Iritten Consent Form

For Parents/Participants in a Study of
Giant Swings on the Uneven Parallel Bars

The purpose of this study is to examine the biomechanics of
a giant swing on the uneven parallel bars. Volunteer
subjects between the ages of 10 and 16 will be filmed
performing a previously learned sequence of skills,
including the giant swing, on the uneven parallel bars. The
top bar of the uneven parallel bars will be instrumented
with strain gages to directly measure the forces and torques
on the bar while the gymnast performs. Data obtained from
the film and instrumented bar will be used to identify
critical factors in performing good giant swings and to
calculate typical forces and torques on the elbow and
shoulder joints while performing the skill. The study will
be conducted at the Human Performance Laboratory at Michigan
State University and at Great Lakes Gymnastics Club, Lansing
Michigan.

1. I have freely consented to allow my child to take part
in a scientific study being conducted by Winifred
Mitten, a doctoral candidate in the School of Health
Education, Counseling Psychology and Human Performance
at Michigan State University.

2. The study has been explained to me and to my child. He
understand the explanation that has been given and what
participation in the study will involve.

3. My child and I understand that she is free to
discontinue her participation in the study at any time
without penalty.

4. I understand that my child will be performing skills
that she has previously learned on the uneven parallel
bars and that proper safety procedures and matting will
be provided.

5. I understand if my child is injured as a result of
participation in this study, Michigan State University
will provide emergency medical care if necessary, but
these and any other medical expenses must be paid from
my own health insurance program.

126

 

 

 

 

 

 

6. I understand that my child's name will not be associated
with any publication and/or presentation of the data
collected in this study.

7. I understand that film and other measurement data will
be collected and may be used for demonstrations,
instruction and/or study.

8. I understand that my child's participation in this study
does not guarantee any beneficial results for her.

9. I understand that, at our request, my child and I can
receive additional explanation of the study after
participation is completed.

Signature of Parent Signature of Witness

Date Date

Signature of Gymnast Signature of Hitness

 

 

Date

Date

APPENDIX B

127

APPENDIX 8

INFORMATION SHEET FOR DATA COLLECTION
Biomechanical Study of Giant Swings
A. General Information:

 

Address:

 

 

Phone Number:

 

Date of Birth:

 

Height:

 

 

Height:

Age:

 

8. Gymnastics History:

Years of participation in gymnastics?

 

Class I or Elite?

 

Coach

 

C. Anthropometric Data:
Thigh length
Calf length
Shoulder-elbow length

Elbow-wrist length

 

Hand length

Biacromial diameter

 

Biiliac diameter

 

128

D. Underwater weighing data:

Body density (g/ml)

 

Relative fat (1)

 

Lean body mass (kg)

 

D. Filming Record

 

 

 

 

 

 

Location

Camera Film rate
Shutter angle f/stop

Lens Focal length

 

 

Type of film

Camera height

 

Camera-subject distance

 

Shooting details (trials, etc) Comments

APPENDIX C

C.1 Judging Sheet for Giant Swings on the Uneven

Subject 3

129

APPENDIX C

Parallel Bars

 

Arms Bent

Body Position:
Arch

Leg Position:
Legs Bent

Legs Apart

Arriving to Handstand:

Using Force

General Impression:

Judge

2.0

2.0

2.0
1.5

1.5
1.0

 

Comments:

 

 

 

 

 

 

 

 

fins-".1.- m

139

Ratings

C.2 Summary of Judges'

J2 J3 J4 Ave.

JI

 

Subject
Number

888888666658554

4.150924769984480

999888778777676

399840851793832

989888668673674

275580999743615

988755544444332

APPENDIX D

131

APPENDIX D
Statistical Summary Tables
Table 0.1 Mean Peak Velocities of Total Body Center of

Mass and the Times of Occurrences of Mean Peak
Velocities Expressed as Percents of Total Time

 

. Mean SD Min Max
Vx * (m/sec)
Total Group 4.633 .460 3.880 5.450
Group 1 4.866 .371 4.550 5.450
Group 3 4.380 .423 3.660 4.960
V? (m/sec)
Total Group -1.301 5.401 -5.110 9.070
Group 1 .296 6.641 -4.850 9.070
Group 3 -2.900 3.891 -5.110 4.020
Va (m/sec)
Total Group 5.601 1.410 4.580 9.470
Group 1 6.152 1.912 4.580 9.470
Group 3 5.050 .239 4.760 5.320
2 Time Vx
Total Group 57.65 6.500 49.20 71.50
Group 1 60.92 6.956 53.90 71.50
Group 3 54.36 4.464 49.20 61.10
1 Time Vv **
Total Group 49.07 12.67 27.30 73.10
Group 1 57.66 10.352 44.30 73.10
Group 3 40.26 6.493 27.30 46.30
X'Time Vs
Total Group 55.95 10.26 36.40 73.10
Group 1 60.42 6.61 49.60 73.10
Group 3 51.48 10.50 36.40 61.80

 

I"p < .10, ** p < .05, ***p < .01.

132

Table D.2 Mean Peak Velocities+ of Total Body Center of
Mass Per Quadrant

 

Total Group 1 Group 3 Total Group 1 Group 3

 

Quadrant 1 Quadrant 2

Vx Vx

Mean -1.98 -2.03 -1.92 4.57 4.77 4.38
SD .39 .47 .36 .47 .48 .42
Min -2.84 -2.84 -2.53 3.88 4.22 3.88
Max -1.63 -1.63 -1.70 5.45 5.45 4.96
VY V?

Mean -3.84 -3.97 -3.72 -4.74 -4.96 -4.51
SD .74 .63 .89 .87 1.16 .46
Min -5.11 -5.02 —5.11 -6.95 -8.95 -5.02
Max -2.91 -3.44 -2.91 -3.95 -3.95 -3.97
Va Va

Mean 3.20 2.41 3.99 5.39 5.90 4.91
SD 2.63 3.63 .90 1.26 1.69 .34
Min -4.00 -4.00 3.16 4.39 4.58 4.39
Max 5.13 5.02 5.13 8.83 8.83 5.32

Quadrant 3 Quadrant 4

Vx xx Vx

Mean 3.99 4.33 3.64 -2.04 -2.04 -2.03
SD .46 .43 .18 .14 .16 .14
Min 3.43 3.74 3.43 -2.20 -2.18 -2.20
Max 4.82 4.82 3.81 -1.81 -1.81 -1.85
V? V?

Mean 4.45 5.20 3.71 3 18 3.37 2 98
SD 1.83 2.36 .75 52 .67 25
Min 3.04 3 52 3.04 2 56 2.59 2 56
Max 9.07 9 07 4.90 4 39 4.39 3 22
Va Vs *

Mean 5.00 5.88 5.33 3.54 3.87 3 21
SD 1.70 2.25 .54 .56 .51 42
Min 3.84 3.84 3.88 2.75 3.47 2 75
Max 9.47 9.47 5.27 4.74 4.74 3 90

 

 

*p < .10, ** p < .05, ***p < .01.

133

Table 0.3 Times of Occurrences of Mean Peak Velocities of
Total Body Center of Mass Per Quadrant Expressed
as Percents of Total Time

 

Total Group 1 Group 3 Total Group 1 Group 3

 

Quadrant 1 Quadrant 2

2 Time Vx 1 Time Vx

Mean 36.03 36.88 35.68 57.08 59.22 54.94
SD 7.39 8.07 7.50 5.15 4.18 5.56
Min 26.90 26.90 28.30 49.20 53.90 49.20
Max 47.40 45.20 47.40 64.60 64.60 63.90
2 Time Vv X Time Vv

Mean 41.41 43.58 39.24 49.23 52.98 45.48
SD 6.79 6.34 7.21 7.74 6.89 7.41
Min 34.00 38.60 34.00 37.30 44.30 37.30
Max 52.00 52.00 51.90 60.00 80.00 57.50
2 Time Va 2 Time sz '
Mean 41.62 43.58 39.66 52.66 55.50 50.22
SD 6.70 6.34 7.16 7.55 7.01 7.85
Min 34.00 36.60 34.00 43.80 47.70 43.80
Max 52.00 52.00 51.90 64.60 64.60 61.40

Quadrant 3 Quadrant 4

2 Time Vx 1 Time Vx

Mean 60.37 62.88 57.86 83.58 86.56 80.56
SD 6.05 5.84 5.70 6.52 3.09 7.98
Min 51.60 57.00 51.60 69.30 83.80 69.30
Max 71.50 71.50 86.70 91.70 91.70 90.60
1 Time V? 1 Time Vr

Mean 68.29 70.02 66.56 76.59 78.02 75.16
SD 6.08 5.19 6.98 4.01 2.79 4.63
Min 60.70 61.00 60.70 69.30 74.70 69.30
Max ' 75.60 73.80 75.60 82.20 82.20 80.60
2 Time Va 2 Time Va

Mean 63.97 66.12 61.82 76.07 77.50 74.64
SD 6.47 6.90 5.91 4.04 2.15 5.19
Min 52.90 57.00 52.90 69.30 74.60 89.30
Max 73.10 73.10 69.50 80.80 80.20 80.60

 

*p < .10, *’ p < .05, ***p < .01.

134

Table D.4 Mean Peak Accelerations+ of Total Body Center of
Mass and Relative Times of Occurrences

 

 

Mean SD Min Max

Ax

Total -20.39 475.73 -475.73 1047.31
Group 1 46.13 634.79 -475.73 1047.31
Group 3 -86.90 308.54 -340.61 293.46
Av

Total 237.45 964.89 -496.81 2578.68
Group 1 600.39 1238.71 -496.81 2578.86
Group 3 -125.49 480.68 -370.94 732.33
As V
Total 766.84 716.11 286.37 2522.39
Group 1 1088.38 947.56 286.37 2622.39
Group 3 465.30 169.17 296.17 734.35
2 Time Ax

Total 54.71 21.92 16.80 90.10
Group 1 62.00 27.53 16.80 90.10
Group 3 47.42 13.78 35.90 88.90
2 Time Av

Total 59.32 14.57 35.90 78.10
Group 1 62.60 18.07 36.70 78.10
Group 3 56.04 13.88 35.90 69.40
X Time As

Total 61.48 12.91 35.90 76.50
Group 1 66.26 10.36 51.60 76.50
Group 3 58.70 14.51 35.90 69.40
*(m/sec2)- If

*p < .10, ** p < .05, ***p < .01.

135

Table D.5 Mean Peak Accelerations+ of Total Body Center of
Mass Per Quadrant

 

 

 

Total Group 1 Group 3 Total Group 1 Group 3
Quadrant 1 Quadrant 2
Ax Ax
Mean - 92.39 -19.02 -165.75 75.13 10.36 139.89
SD 296.16 374.94 208.12 253.66 303.78 204.96
Min -359.33 -359.33 -340.61 -262.24 -262.24 -220.50
Max 456.67 456.67 183.45 356.26 356.26 293.46
Av Av
Mean —30.97 222.35 -284.29 -110.03 -309.77 89.71
SD 405.15 448.99 85.51 531.57 665.32 305.57
Min -370.94 -290.63 -370.94 -1423.59-1423.59 -236.86
Max 773.52 773.52 -189.70 432.97 203.67 432.97
As An
Mean 407.12 474.74 339.50 439.09 550.94 327.24
SD 182.66 229.50 104.81 367.22 509.67 111.23
Min 242.52 285.67 242.52 264.46 266.37 264.46
Max 852.90 852.90 503.60 1460.28 1460.28 523.05
Quadrant 3 Quadrant 4
Ax Ax
Mean -11.46 43.67 -66.59 -17.10 43.49 -77.70
SD 439.42 613.14 255.87 233.71 255.47 220.12
Min -475.73 -475.73 -277.42 -284.17 -253.18 -284.17
Max 1047.31 1047.31 208.94 300.75 300.75 165.46
Av Av
Mean 264.63 612.81 -83.54 85.11 79.66 90.55
SD 948.52 1227.81 462.31 349.39 265.26 241.09
Min -496.81 -496.81 -353.21 -282.50 -282.50 -214.95
Max 2578.88 2578.88 732.33 792.30 792.30 324.42
As An
Mean 721.25 1048.18 394.32 333.90 399.45 268.36
SD 743.45 968.01 198.38 198.20 246.51 95.07
Min 222.75 232.49 222.75 200.33 236.72 200.33
Max 2622.39 2622.39 734.35 832.77 832.77 431.28
+(m/secz).
*p < 10, ** p < .05, ***p < .01.

136

Table D.6 Times of Occurrences of Mean Peak Accelerations
of Total Body Center of Mass Per Quadrant
Expressed as Percents of Total Time

 

Total Group 1 Group 3 Total Group 1 Group 3

 

Quadrant 1 Quadrant 2

2 Time Ax 2 Time Ax

Mean 32.39 30.54 34.24 51.35 54.08 46.62
SD 6.58 8.45 4.17 8.64 10.18 6.75
Min 16.80 16.80 29.90 37.30 42.50 37.30
Max 40.40 38.90 40.40 64.20 60.00 64.20
2 Time Av X Time Av

Hean 31.44 33.02 29.86 51.63 54.32 48.94
SD 13.37 11.85 15.99 8.38 7.63 9.05
Min 8.60 16.80 8.60 40.60 42.60 40.60
Max 51.60 46.90 51.60 60.00 64.20 62.00
X Time AXa 1 Time As

Mean 29.99 27.32 32.66 51.63 54.32 48.94
SD 11.02 10.25 12.28 8.38 7.60 9.06
Min 16.60 16.80 19.30 40.80 42.60 40.60
Max 51.60 39.10 51.60 64.20 62.00 64.20

Quadrant 3 Quadrant 4

2 Time Ax 2 Time Ax *

Mean 67.49 69.38 65.60 86.48 92.28 80.68
SD 7.21 6.18 8.37 10.01 8.34 8.47
Min 56.30 60.50 58.30 68.90 79.30 68.90
Max 79.00 76.10 79.00 99.50 99.50 91.10
2 Time Av 2 Time Av

Mean 68.50 68.50 68.50 85.85 88.92 82.38
SD 6.13 7.12 5.82 9.69 8.74 10.39
Min 60.50 80.50 61.40 88.90 76.90 68.90
Max 78.60 78.60 77.20 99.50 99.50 97.20
2 Time As 2 Time As

Mean 87.34 -66.38 66.30 86.34 87.14 85.54
SD 6.16 5.53 7.22 9.90 9.97 10.94
Min 58.30 60.50 58.30 68.90 76.50 68.90
Max 76.80 73.10 76.80 99.50 99.50 97.20

 

I"p < .10, '* p < .05, ***p < .01.

137

Table D.7 Mean Peak Resultant Velocities* of Selected

 

Joints
Mean SD Min Max

Ankle ***

Total 10.47 .793 9.29 11.70
Group 1 11.03 .372 10.62 11.70
Group 3 9.64 .338 9.29 9.97
Knee *

Total 7.16 .323 6.60 7.53
Group 1 7.30 .191 7.01 7.53
Group 3 6.95 .389 6.60 7.47
Hip emu

Total 5.96 .427 5.40 6.70
Group 1 6.22 .326 5.73 6.70
Group 3 5.57 .186 5.40 5.75
Shoulder

Total 3.64 .657 3.04 5.00
Group 1 3.91 .751 3.04 5.00
Group 3 3.24 .071 3.18 3.33
Elbow

Total 2.86 .359 2.19 3.33
Group 1 2.64 .235 2.45 3.13
Group 3 2.68 .543 2.19 3.33

 

* (m/sec)-
*p < .10, ** p < .05, **'p < .01.

138

Table D.8 Times of Occurrences of Mean Peak Resultant
Velocities Expressed as Percents of Total Time

 

Mean SD Min Max

2 Time Ankle '**

Total 62.45 4.454 53.30 67.30
Group 1 65.13 2.018 62.40 67.30
Group 3 58.43 4.090 53.30 63.30
2 Time Knee

Total 59.34 9.703 34.40 67.60
Group 1 59.57 12.686 34.40 67.60
Group 3 59.00 3.733 54.90 63.30
1 Time Hip **

Total 52.64 4.700 44.70 58.50
Group 1 55.32 3.622 50.20 58.50
Group 3 48.63 2.927 44.70 51.10
2 Time Shoulder **

Total 55.95 11.156 40.60 82.70
Group 1 60.95 11.413 48.90 82.70
Group 3 48.53 5.725 40.60 53.90
2 Time Elbow

Total 74.26 12.709 49.30 84.70
Group 1 77.20 13.745 49.30 84.70
Group 3 69.85 11.245 53.90 78.60

 

* (m/sec)-
I"p < .10, ** p < .05, ***p < .01.

139

 

 

Table D.9 Mean Peak Resultant Velocities+ of the Selected
Joints Per Quadrant
Mean SD Min Max Mean SD Min Max
Quadrant 1 Quadrant 2
Ankle Ankle
Total 5.71 1.06 4.33 7.49 8.23 .69 6.97 9.22
Group 1 6.26 1.27 4.33 7.49 6.50 .71 7.33 9.22
Group 3 5.15 .41 4.73 5.70 7.95 .62 6.97 8.65
Knee Knee
Total 4.54 .31 4.08 4.95 6.33 .27 5.82 6.59
Group 1 4.62 .32 4.09 4.88 6.39 .32 5.62 6.59
Group 3 4.46 .33 4.08 4.95 8.28 .22 5.91 6.49
Hip Hip *
Total 4.42 .33 4.02 4.98 5.96 .43 5.40 6.71
Group 1 4.45 .39 4.02 4.98 6.21 .36 5.73 6.71
Group 3 4.39 .30 4.08 4.85 5.70 .34 5.40 6.24
Shoulder Shoulder
Total 2.62 .29 2.19 3.10 3.62 .64 3.04 5.00
Group 1 2.64 .16 2.41 2.80 3.91 .81 3.04 5.00
Group 3 2.59 .40 2.19 3.10 3.32 .19 3.18 3.65
Elbow Elbow *
Total 1.45 .23 1.17 1.89 2.13 .23 1.86 2.50
Group 1 1.49 .27 1.17 1.89 2.27 .22 1.94 2.50
Group 3 1.41 . .20 1.17 1.69 2.00 .16 1.86 2.27
Quadrant 3 Quadrant
Ankle "‘ Ankle
Total 9.90 .68 9.00 11.00 2.98 .42 2.33 3.69
Group 1 10.60 .55 10.00 11.00 2.98 .52 2.33 3.69
Group 3 9.20 .45 9.00 10.00 2.97 .38 2.39 3.27
Knee Knee ‘*
Total 7.16 .32 6.60 7.53 3.50 .62 2.62 4.86
Group 1 7.31 .21 7.01 7.53 3.90 .58 3.34 4.66
Group 3 7.01 .36 6.60 7.47 3.11 .39 2.62 3.53
Total 4.86 .32 4.45 5.47 4.44 .71 2.71 5.45
Group 1 4.12 .24 4.48 5.47 4.84 .38 4.48 5.45
Group 3 4.61 .12 4.45 4.75 4.30 .78 2.71 4.7

Table D.9 ( Continued)

140

 

 

 

Mean SD Min Max Mean SD Min Max
Quadrant 3 Quadrant 4
Shoulder *** Shoulder **
Total 2.62 .78 2.02 4.59 3.01 .34 2.43 3.59
Group 1 3.09 .93 2.02 4.59 3.23 .29 2.92 3.59
Group 3 2.21 .15 2.04 2.37 2.97 .24 2.43 3.05
Elbow Elbow
Total 1.81 .71 1.40 3.79 2.57 .34 2.19 3.30
Group 1 2.11 .95 1.53 3.79 2.72 .36 2.37 3.30
Group 3 1.52 .10 1.40 1.61 2.43 .29 2.19 2.91
* (m/sec).
*p < .10, ** p < .05, ‘*'p < .01.

Table 0.10 Times of Occurrences of Mean Peak Resultant
Velocities Expressed as

141

Percents of Total Time

 

Min

 

Mean SD Max Mean SD Min Max
Quadrant 1 Quadrant 2

2 Time Ankle Time Ankle

Total 37.88 6.13 29.90 48.10 58.62 4.98 50.40 65.70
Group 1 39.30 5.50 33.50 46.70 60.68 3.81 56.10 65.70
Group 3 36.46 7.03 29.90 48.10 56.58 5.55 50.40 65.30
2 Time Knee Time Knee *

Total 42.43 6.83 35.30 53.30 57.95 4.65 50.40 65.70
Group 1 43.88 6.93 36.00 52.40 60.68 3.81 56.10 65.70
Group 3 40.98 7.20 35.30 53.30 55.22 4.65 50.40 61.40
2 Time Hip Time Hip

Total 42.97 6.41 35.30 53.30 53.15 5.36 44.70 61.40
Group 1 44.86 5.73 39.60 52.40 55.12 3.96 50.20 59.50
Group 3 41.08 7.12 35.30 53.30 51.18 6.25 44.70 61.40
2 Time Shoulder Time Shoulder

Total 42.82 6.58 35.30 53.30 52.37 6.11 40.60 61.80
Group 1 44.86 5.73 39.80 52.40 53.56 4.23 48.90 59.50
Group 3 40.78 7.37 35.30 53.30 51.18 7.73 40.60 61.80
X Time Elbow Time Elbow

Total 42.70 6.24 35.30 52.60 51.90 8.95 39.30 62.60
Group 1 44.30 6.05 39.60 52.40 53.58 5.61 48.70 62.20
Group 3 41.14 6.89 35.30 52.60 50.24 8.38 39.30 61.40 .

Quadrant 3 Quadrant 4

2 Time Ankle Time Ankle

Total 62.83 4.87 53.30 69.10 83.69 10.91 66.00 96.90
Group 1 65.10 2.27 62.40 68.00 85.10 8.47 76.70 97.30
Group 3 60.56 5.94 53.30 69.10 82.28 13.82 66.00 98.90
1 Time Knee Time Knee

Total 62.73 4.71 54.90 69.50 76.28 5.71 66.00 87.20
Group 1 64.30 3.29 60.20 68.40 77.36 1.75 74.70 78.90
Group 3 61.10 5.70 54.90 69.50 75.20 8.21 66.60 87.20
2 Time Hip Time Hip

Total 73.18 5.05 60.30 78.30 75.92 5.25 66.00 84.40
Group 1 75.52 2.55 71.50 77.90 77.36 1.75 74.70 78.90
Group 3 70.84 6.94 60.30 78.30 74.48 7.33 66.00 84.40

142

Table 0.10 (Continued)

 

 

 

Mean SD Min Max Mean SD Min Max
Quadrant 3 Quadrant 4
1 Time Shoulder 2 Time Shoulder
Total 64.99 7.66 55.40 78.50 78.70 4.93 69.30 82.70
Group 1 67.80 8.35 58.90 78.50 81.18 1.86 78.30 82.70
Group 3 62.18 6.54 55.40 71.80 76.36 6.06 89.30 83.30
1 Time Elbow 2 Time Elbow *
Total 63.63 7.23 55.40 78.50 80.91 4.62 70.10 85.00
Group 1 64.18 8.68 57.00 78.50 83.36 1.13 81.90 84.70
Group 3 63.08 6.44 55.40 71.80 78.46 5.63 70.10 85.00

 

*p < .19’ #' p < .65, ***p <

.01.

143

Table 0.11 Mean Peak Resultant AccelerationS* of
Selected Joints

 

 

Mean SD Min Max
Ankle
Total 91.02 13.44 74.58 121.21
Group 1 94.78 16.58 74.58 121.21
Group 3 85.39 3.637 82.13 89.06
Knee
Total 67.67 16.88 49.85 94.69
Group 1 65.72 16.44 49.85 94.69
Group 3 70.58 19.62 53.44 89.65
Hip
Total 46.71 8.11 37.51 63.52
Group 1 49.56 9.02 39.14 63.52
Group 3 42.45 4.63 37.51 48.60
Shoulder
Total 74.89 75.08 32.92 231.72
Group 1 99.54 91.20 35.93 231.72
Group 3 37.90 3.44 32.92 40.57
Elbow
Total 51.58 20.26 37.99 106.46
Group 1 44.12 4.83 37.99 52.40
Group 3 62.77 30.24 39.87 106.46
*(m/secz).

'p < .10, '* p < .05, I""""p < .01.

144

Table D.12 Times of Occurrences of Mean Peak Resultant
Accelerations Expressed as Percents of Total

 

Time
Mean SD Min Max

1 Time Ankle

Total 62.46 4.729 57.00 68.60

Group 1 64.38 4.549 57.10 68.60

Group 3 59.58 3.756 57.00 65.00
1 Time Knee

Total 58.84 3.679 52.10 63.10

Group 1 59.73 2.226 56.60 62.70

Group 3 57.50 5.324 52.10 63.10
2 Time Hip **

Total 53.40 9.823 41.10 75.60

Group 1 59.08 8.155 53.90 75.60

Group 3 44.88 4.176 41.10 50.60
2 Time Shoulder *

Total 57.73 12.637 36.90 78.10

Group 1 63.53 11.839 48.00 78.10

Group 3 49.03 8.781 36.90 57.20
1 Time Elbow

Total 79.31 8.931 62.60 88.30

Group 1 80.08 9.296 62.60 86.40

Group 3 78.15 9.606 65.20 88.30

 

*p < .10, *' p < .05, ***p < .91.

145

Table 0.13 Mean Peak Velocities+ of the Center of Mass of
Selected Segments

 

 

Mean so Min ‘ Max Mean so Min Max
Trunk
Vx Vv
Total. 4.05 .37 3.69 4.74 -4.09 .31 -4.73 -3.76
Groupl 4.19 .45 3.72 4.74 -4.24 .35 -4.73 -3.76
Group3 3.91 .22 3.69 4.18 -3.94 .20 -4.27 -3.78
Va It"!
Total 4.64 .33 4.31 5.25
Groupl 4.86 .32 4.39 5.25
Group3 4.45 .19 4.31 4.76
Thigh
Vx Vv
Total 5.37 .20 5.07 5.72 -2.98 4.37 -5.59 5.60
Groupl 5.42 .19 5.19 5.72 -1.00 5.74 -5.45 5.60
Group3 5.31 .20 5.07 5.53 -4.97 .36 -5.59 -4.71
Vs
Total 5.80 .34 5.39 6.31
Groupl 5.88 .37 5.39 6.31
Group3 5.73 .33 5.40 6.24
Shank
Vx Vv an.
Total 7.55 .48 6.43 8.13 7.44 .48 6.71 8.04
'Groupl 7.88 .17 7.69 8.13 7.74 .26 7.39 8.94
Group3 7.22 .48 6.43 7.62 7.14 .45 6.71 7.74
VB *I.
Total 8.56 .50 7.74 9.18
Groupl 8.94 .20 8.73 9.18
Group3 8.18 .41 7.74 8.71

146

 

 

 

Table D.13 (Continued)
Mean SD Min Max Mean SD Min Max
Foot
vx ** V? *l'
Total 9.72 .92 8.06 10.62 9.68 .95 7.91 10.90
Groupl 10.33 .48 9.55 10.82 10.40 .38 9.92 10.90
Group3 9.12 .86 8.06 9.92 6.95 .75 7.91 9.67
VR sun:
Total 11.07 .89 9.30 12.07
Groupl 11.70 .31 11.42 12.07
Group3 10.44 .84 9.30 11.53
Arm
Vx Vv
Total 2.60 .39 2.14 3.50 -.001 2.89 -3.33 2.96
Groupl 2.76 .53 2.14 3.50 .388 3.27 -3.33 2.96
Group3 2.44 .18 2.28 2.68 -.390 2.78 -2.50 2.75
Va *
Total 3.08 .84 2.58 4.70
Groupl 3.42 .77 2.69 4.70
Group3 2.74 .17 2.56 3.04
Forearm
Vx Vv
Total -.07 1.57 -1.82 1.59 1.44 1.16 -1.78 2.13
Groupl -.33 1.74 -1.82 1.59 1.17 1.66 -1.78 2.04
Group3 .19 1.53 -1.64 1.46 1.71 .26 1.43 2.13
Va
Total 1.94 .27 1.64 2.45
Groupl 2.07 .22 1.90 2.45
Group3 1.82 .28 1.64 2.32
* (m/sec)-
*p < .10, ** p < .05, ***p < .01.

Table

147

of Selected Body Segments Expressed as
Percents of Total Time

D.14 Times of Occurrences of Mean Peak Velocities+

 

 

Mean SD Min Max Mean SD Min Max
Trunk
Vx Vv
Total 55.0 4 9 45. 5 63.4 46.69 5.1 39.33 55.8
Group1 57.2 4 1 52. 5 63.4 49.04 4.9 44.3 55.8
Group3 52.7 4 9 45.5 58.9 44.2 4.4 39.3 51.3
VR snu-
Total 4.64 33 4. 31 5. 25
Group1 4.86 .32 4. 39 5. 25
Group3 4.45 .19 4. 31 4. 76
Thigh
Vx Vv
Total 56. 5 4.3 50.0 63.4 49.5 10.1 41.0 74.8
Group1 58. 3 4.6 52.5 63.4 53.9 12.6 43.9 74.8
Group3 54. 7 3.9 50.0 60.8 45.1 4.9 41.0 53.6
Va
Total 5.80 .34 5. 39 6. 31
Group1 5.88 .37 5. 39 6. 31
Group3 5.73 .33 5. 40 6. 24
Shank
vx V? I.
Total 80.1 4.6 51. 2 65.9 65.0 5.8 55.7 70.4
Group1 62. 4 3.2 58.8 65.9 66.2 5.9 56.1 70.4
Group3 57.8 4.9 51.2 64.2 63.7 5.7 55.7 70.2
VR mud
Total 8.58 .50 7. 74 9.18
Group1 8.94 .20 8. 73 9.18
Group3 8.18 .41 7. 74 8. 71

Table D.14 (Continued)

148

 

 

 

Mean SD Min Max Mean SD Min Max
Foot
Vx Vv
Total 60.1 4.7 51.2 65.9 66.1 4.6 54.9 70.0
Group1 62.3 3.4 58.4 65.9 67.6 2.3 64.3 70.0
Group3 57.9 5.1 51.2 64.5 64.5 5.9 54.9 69.4
Va
Total 11.07 .89 9.30 12.07
Group1 11.70 .31 11.42 12.07
Group3 10.44 .84 9.30 11.53
Arm
Vx Vv
Total 57.8 5.7 46.7 65.2 62.0 19.6 39.1 83.3
Group1 57.2 5.8 49.3 65.2 68.3 18.4 46.6 83.3
Group3 52.3 5.7 46.7 59.3 55.8 20.8 38.1 80.8
Va ’
Total 3.08 64 2.58 4.70
Group1 3.42 77 2.69 4.70
Group3 2.74 .17 2.58 3.04
Forearm
Vx Vv
Total 71.2 14.5 51.5 88.3 77.1 10.7 49.3 85.6
Group1 74.0 15.2 54.2 87.3 76.1 15.0 49.3 83.6
Group3 68.4 14.9 51.5 88.3 78.2 5.6 70.1 85.6
Total 1.94 .27 1.64 2.45
Group1 2.07 .22 1.90 2.45
Group3 1.82 .28 1.64 2.32
+ m/sec.
*p < .10, ** p < .05, ***p < .01.

149

Table 0.15 Mean Peak Accelerations+ of the Selected Body

 

Segments
Group 1 Group 3 Total Group
Segment Min Max Min Max Min Max

 

Horizontal Acceleration

Trunk -33.73 40.88 -30.80 34.24 -33.73 40.88
Thigh -37.15 44.76 -49.46 37.18 -49.46 44.76
Shank -63.44 -56.47 -76.57 -51.64 -76.57 -51.64
Foot -94.83 110.03 -78.72 -65.11 -94.63 110.03
Arm -27.40 107.89 -39.52 23.33 -39.52 107.89
F.Arm -30.98 21.17 -85.33 25.80 -85.33 25.80

Vertical Acceleration

Trunk"I 27.33 83.18 -29.70 38.36 -29.70 83.18
Thigh 32.25 67.06 38.21 52.95 32.25 67.06
Shank 59.50 98.57 51.39 67.22 51.39 98.57
Foot ‘* 87.47 99.03 34.47 85.12 34.47 99.03
Arm -33.18 119.48 -39.21 39.01 -39.21 119.48
F.Arm -41.18 37.42 -49.44 37.15 -49.44 37.42

Resultant Acceleration

Trunk 30.48 84.88 31.90 40.68 30.48 84.88
Thigh 34.78 68.07 38.85 57.78 34.78 68.07
Shank 61.44 104.61 57.76 78.88 57.76 104.61
Foot ** 89.20 122.19 72.71 95.32 72.71 122.19
Arm 34.29 134.95 26.45 42.19 26.45 134.95
F.Arm 36.89 42.47 25.62 93.07 25.62 93.07

 

 

* (m/sec2)-
*p < .10, ** p < .05, '**p < .01.

Table D.16 Mean Peak Joint Reactio

150

n Forces

 

 

Mean SD Min Max
Elbow
Vertical Force‘
Total Group -1.56 1.33 -2.09 2.22
Group 1 -1.15 1.89 -2.09 2.22
Group 3 -1.98 0.04 -2.02 -1.94
Horizontal Force+
Total Group -0.74 1.11 -1.44 1.36
Group 1 -0.24 1.46 -1.44 1.36
Group 3 -1.24 0.09 -1.39 -1.16
Moment (Nm)
Total -54.11 55.28 -97.80 96.80
Group 1 -36.90 76.78 -86.00 96.80
Group 3 -69.32 20.03 -97.80 ~53.20
Shoulder
Vertical Force*
Total Group 1.62 1.28 -2.01 2.17
Group 1 1.25 1.82 -2.01 2.17
Group 3 2.00 0.05 1.95 2.08
Horizontal Force+
Total -1.28 0.09 -1.41 -1.14
Group 1 -1.31 0.11 -1.41 -1.14
Group 3 -1.25 0.07 -1.35 -1.16
Moment (Nm)
Total 97.04 27.24 50.80 134.30
Group 110.16 17.79 92.00 134.30
Group 3 83.92 30.38 50.80 124.30

 

* Normalized by body weight.

*p(

.10,

**p(

.05,

I'"'""p < .01.

151

Table D.17 Times of Occurrences of Mean Peak Joint Forces
and Moments Expressed as Percents of Total Time

 

 

Mean SD Min Max
Elbow

2 Vertical Force

Total 54.81 5.225 45.10 62.60
Group 1 56.86 3.328 53.20 61.60
Group 3 53.76 2.823 45.10 62.60
2 Horizontal Force

Total 47.12 6.280 39.10 56.00
Group 1 50.22 4.485 45.80 56.00
Group 3 44.02 6.677 39.10 55.10
X Moment

Total 49.48 5.971 40.10 58.50
Group 1 51.74 5.617 46.80 58.50
Group 3 47.22 5.991 40.10 56.60

Shoulder

2 Vertical Force

Total 55.43 5.940 44.20 63.04
Group 1 57.46 4.693 53.20 63.40
Group 3 53.40 6.860 44.20 62.60
2 Horizontal Force ,
Total 46.72 6.111 38.60 58.30
Group 1 49.88 4.943 44.80 56.30
Group 3 43.56 5.885 38.60 53.20
X Moment

Total 57.50 5.310 48. 0 84.80
Group 1 59.18 5.511 52. 0 64.80
Group 3 ~ 55.82 5.100 48.90 63.00

 

I"p < .10, ** p < .05. '**p < .01.

152

Table D.18 Joint Reaction Forces and Moments of the

Shoulder Per Quadrant

 

 

Total Group 1 Group 3 Total Group 1 Group 3
Group Group
Quadrant 1 Quadrant 2
Vertical Force+
Mean -0.56 -0.59 -0.53 2.02 2.05 1.98
SD 0.07 0.04 0.08 0.09 0.11 0.07
Min -0.45 -0.55 -0.45 1.91 1.92 1.91
Max 0.66 0.61 -0.66 2.17 2.17 2.06
Horizontal Force
IV
Mean -0.52 -0.44 -0.60 -1.28 -1.31 -1.25
SD 0.13 0.13 0.07 0.09 0.12 0.07
Min -0.67 - .65 -0.67 -1.41 -1.41 -1.35
Max 0.30 0.30 -0.51 -1.35 -1.35 -1.16
Moment (Nm)
Mean -24.74 -20.80 -28.68 69.62 108.86 30.38
SD 26.84 41.26 11.39 69.45 19.40 81.40
Min -54.70 -54.70 -43.20 -78.00 86.10 86.10
Max 51.10 51.10 -16.30 134.30 134.30 124.10
Quadrant 3 Quadrant 4
Vertical Force+
Mean 1.77 1.08 1.75 0.60 0.59 -0.60
SD 0.23 0.11 0.33 0.06 0.09 0.02
Min 1.21 1.68 1.21 0.49 -0.49 -0.57
Max 2.04 1.20 2.04 -0.71 -0.71 -0.62
Horizontal Force*

Mean 0.74 0.97 0.51 0.29 0.48 0.11
SD 0.58 0.11 0.78 0.69 0.60 0.69
Min -0.88 9.82 -0.88 -0.67 -0.59 -0.67
Max 1.13 1.13 0.87 0.88 0.88 0.66

153

Table D.18 (Continued)

 

Total Group 1 Group 3 Total Group 1 Group 3
Group Group

 

Quadrant 3 Quadrant 4
Moment (Nm)

Mean 94.97 106.00 83.94 -21.13 -27.18 -15.08
SD 25.88 16.67 30.42 21.09 26.15 15.03
Min 50.80 85.50 50.80 -50.10 -50.10 -26.40
Max 124.70 124.70 124.40 17.00 17.00 11.20

 

+Normalized by body weight.
*p < .10, ** p < .05, ‘**p < .01.

154

Table D.19 Times of Occurrences of Joint Reaction Forces
and Moments of the Shoulder Per Quadrant
Expressed as Percents of Total Time

 

Total Group 1 Group 3 Total Group 1 Group 3
Group Group

 

Quadrant 1 Quadrant 2

2 Vertical Force

Mean 10.06 11.00 9.12 54.79 56.80 52.74
SD 5.09 4.53 5.96 5.82 3.84 7.14
Min 1.20 3.80 1.20 44.20 53.20 44.20
Max 15.90 15.90 15.30 64.00 61.60 64.00

1 Horizontal Force

Mean 41.08 43.30 38.86 46.63 49.36 43.90.
SD 7.30 7.16 7.51 6.06 4.35 6.59
Min 32.20 36.20 32.20 38.60 44.80 36.60
Max 51.60 51.30 51.60 56.30 56.30 54.90
X Moment

**

Mean 13.69 18.70 8.68 55.18 57.88 52.48
SD 7.81 7.40 4.48 6.61 4.90 7.50
Min 4.70 9.70 4.70 44.50 52.00 44.50
Max 28.50 28.50 15.30 64.40 64.00 64.40

Quadrant 3 Quadrant 4

2 Vertical Force

Mean 58.29 59.48 57.12 91.99 89.72 94.26
SD 3.96 3.75 4.22 4.74 2.38 5.68
Min 54.00 55.60 54.00 86.80 88.10 86.80
Max 64.80 64.80 64.40 98.70 93.90 98.70

1 Horizontal Force

Mean 68.95 71.16 66.74 78.70 79.08 78.32
SD 6.20 4.09 7.58 7.80 6.33 10.15
Min 57.10 65.80 57.10 64.80 73.50 64.80

Max 77.80 77.10 77.80 90.50 90.90 90.50

155

Table D.19 (Continued)

 

 

Total Group 1 Group 3 Total Group 1 Group 3
Group Group
Quadrant 3 Quadrant 4
X Moment

Mean 57.83 59.56 59.10 92.11 91.90 92.32
SD 4.85 3.73 5.81 5.41 2.62 5.41
Min 48.90 55.60 46.90 82.80 88.60 82.80
Max 64.80 64.90 64.40 98.80 94.40 98.80

 

*p < .10, ** p < .05, *'*p < .01.

156

Table 0.20 Joint Reaction Forces and Moments of the Elbow
Per Quadrant

 

Total Group 1 Group 3 Total Group 1 Group 3
Group Group

 

Quadrant 1 Quadrant 2

Vertical Force*

Mean -0.31 -0.12 -0.49 -1.57 -1.16 -1.98
SD 0.47 0.63 0.07 1.33 1.89 0.04
Min -0.63 -0.63 -0.58 -2.09 -2.09 -2.02
Max 0.58 0.58 -0.43 2.22 2.22 -1.94

Horizontal Force+

Mean -0.42 -0.23 -0.61 -0.72 -0.24 -1.21
SD 0.39 0.48 0.14 1.10 1.46 0.11
Min -0.76 -.54 -0.76 -1.44 -1.44 -l.39
Max 0.62 0.62 -0.46 1.36 -1.11 -1.11

Moment (Nm)

Mean -8.78 -11.62 -5.95 -54.90 -38.89 -69.31
SD 17.64 23.03 12.23 55.27 76.76 20.03
Min -32.20 -32.20 -17.35 -97.80 -85.99 -97.60
Max 27.89 27.89 7.47 96.80 96.80 -53.16
Quadrant 3 Quadrant 4

Vertical Force*
**

Mean -1.48 -1.10 -1.85 -0.53 -0.49 -0.57
SD 1.18 1.66 0.13 0.06 0.05 0.02
Min -2.05 -2.05 -2.01 -0.60 -0.54 -0.60
Max 1.86 1.86 -1.72 -0.41 -0.41 -0.55

Horizontal Force+

Mean 0.90 0.95 0.86 0.43 0.48 0.39
SD 0.10 0.11 0.07 0.42 0.49 0.39
Min 0.76 0.78 0.79 -0.37 -0.37 -0.30
Max 1.08 0.97 1.08 0.80 0.80 0.62

157

Table 0.20 (Continued)

 

 

Total Group 1 Group 3 Total Group 1 Group 3
Group Group
Quadrant 3 Quadrant 4
Moment (Nm)

Mean 26.03 31.69 20.36 8.83 8.85 9.10
SD 9.56 6.57 9.07 12.71 14.48 12.39
Min 9.10 23.39 9.10 -14.88 -14.88 -10.00
Max 40.66 40.66 30.59 23.43 23.43 22.06

 

* Normalized by bOdy weight.
I"p < .10, ** p < .05, I""""p < .01.

158

Table D.21 Times of Occurrences of Joint Reaction Forces
and Moments of the Elbow Per Quadrant Expressed
as Percents of Total Time

 

Total Group 1 Group 3 Total Group 1 Group 3

 

Quadrant 1 Quadrant 2

1 Vertical Force

Mean 9.28 11.76 4.80 55.42 57.32 53.52
SD 6.03 5.36 4.77 6.14 3.79 7.64
Min .50 3.10 .50 45.10 53.20 45.10
Max 16.90 1.90 12.10 66.40 61.60 66.40
2 Horizontal Force
Mean 40.69 42.90 38.46 47.11 50.20 44.02
SD 6.67 6.50 6.76 6.27 4.46 6.68
Min 32.20 36.20 32.20 39.10 45.60 39.10
Max 50.70 50.70 49.80 56.00 56.00 55.10
X Moments
Mean 21.57 18.96 24.18 49.64 51.74 47.54
SD 11.79 12.14 12.18 5.79 5.62 5.73
Min 1.50 1.50 8.10 40.80 46.80 40.80
Max 34.90 33.50 34.90 58.50 58.50 56.60
Quadrant 3 Quadrant 4
2 Vertical Force
Mean 57.74 60.08 55.40 93.33 92.52 94.14
SD 5.16 4.40 5.19 4.92 3.48 6.39
Min 48.50 55.80 48.50 86.20 89.30 86.20
Max 64.80 64.80 63.00 99.90 97.30 99.90
2 Horizontal Force
.
Mean 69.39 72.44 86.34 77.76 79.90 75.62
SD 5.74 3.55 6.20 7.70 5.46 9.59
Min 57.10 69.40 57.10 64.40 74.50 64.40

Max 77.80 77.80 73.60 90.50 86.60 90.50

159

Table D.21 (Continued)

 

Total Group 1 Group 3 Total Group 1 Group 3

 

 

Quadrant 3 Quadrant 4
X Moment

$¥

Mean 66.28 70.12 62.44 81.79 82.00 81.50
SD 5.59 2.45 5.24 9.68 6.32 11.90
Min 56.20 68.20 56.20 72.50 76.30 72.50
Max 73.80 73.80 89.10 99.40 96.50 99.40
'p < .10, ** p < .05, I""""p < .01.

 

"I7'l'ilfiiflflifllfllfiiflifl’fl'lfiflflr