The games analysis intervention

MISSING IMAGE

Material Information

Title:
The games analysis intervention a procedure to increase the peer acceptance and social adjustment of socially isolated children
Physical Description:
ix, 151 leaves : ill. ; 28 cm.
Language:
English
Creator:
Marlowe, Michael J., 1947-
Publication Date:

Subjects

Subjects / Keywords:
Games   ( lcsh )
Social adjustment   ( lcsh )
Social acceptance   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Includes bibliographical references (leaves 98-109).
Statement of Responsibility:
by Michael J. Marlowe.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000077832
notis - AAJ3126
oclc - 04868752
System ID:
AA00003497:00001

Full Text
















THE GAMES ANALYSIS INTERVENTION: A PROCEDURE
TO INCREASE THE PEER ACCEPTANCE AND SOCIAL
ADJUSTMENT OF SOCIALLY ISOLATED CHILDREN






By

MICHAEL J. MARLOWE


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY




UNIVERSITY OF FLORIDA


1978




































Digitizel arie Internet Archive
in 2010 with funding from
University of Florida, George A. Smathers Libraries with support from Lyrasis and the Sloan Foundation


http://www.archive.org/details/gamesanalysisint00marl
















ACKNOWLEDGEMENTS


Many people have offered me assistance and support during

my Ph.D. studies at the University of Florida. I would like to

take the opportunity to acknowledge these individuals in this

section.

First, I wish to thank Dr. Mary K. Dykes and Dr. William

Reid for awarding me a doctoral traineeship in 1976, so I could

begin my Ph.D. studies. Moreover, Dr. Dykes and Dr. Reid have

continued to support me throughout the course of my studies.

My doctoral committee is composed of outstanding individuals,

with varying philosophies, to whom I am indebted for their support

and guidance. My committee chairperson, Dr. Dykes, has assisted

me in my professional and personal development. Dr. Robert

Algozzine has strongly influenced my research skills and has

provided advice and support during the completion of this

dissertation. Dr. Stuart E. Schwartz has supported me through-

out my doctoral student career and has been an invaluable

reviewer of my professional activities. Dr. Ralph B. Kimbrough

has proven to be a great influence on my professional career and

has contributed to my personal growth. Dr. Harold A. Lerch has

assisted me in my educational research, both through his personal











actions and through our discussions of widely varying topics

and he has proven to be a trusted friend to my family and me.

I consider myself fortunate to have met these individuals and

to have been allowed the privilege of working with them.

Tom Ballowe and John Venn, fellow doctoral students, have

proven to be good and trusted friends during my Ph.D. studies.

They have provided me with personal and academic support as

well as intellectual stimulation.

My most sincere and heartfelt thanks are reserved for my

wife, Susan. She has been a constant source of personal support,

and she has never waivered in her faith in me. I dedicate this

dissertation to her in expression of my love for her.

Finally, I wish to thank God for allowing me the opportunity

and affording me with the strength to complete this program.

Without His assistance, it would not have been possible.
















TABLE OF CONTENTS


ACKNOWLEDGEMENTS . . iii

ABSTRACT . .. .. ....... viii

CHAPTER I

INTRODUCTION . .... .. 1

Statement of the Problem . . 7

Questions Under Investigation . 7

Definition of Terms . .... .. .. 8

Delimitations . .. .. .. 9

Limitations . . 9

Summary and Overview . .... . 10

CHAPTER II

REVIEW OF THE LITERATURE . .. 11

Socially Isolated Children . . 11

Prevalence . .... . 14

Behavioral Correlates . .... .. 15

Interventions . .. .. . 19

Sociometrics and Peer Acceptance . .... 28

Physical Activity and Social-Emotional Development 34

Games Analysis . .... 39

Student Decision Making . .. .42













Problem Solving Skills .

Assessment of Motor Skills .

Summary . .

CHAPTER III

MFTwnn ANn PRnrCFfnRF"


Subjects .

Pretraining Sociometri

Selection of Subjects


c Asses;


Materials . .

Experimental Procedures .

Experimental Design .

First Phase A .

First Phase B .

Second Phase A .

Second Phase B .

Behavioral Assessment .

Observers .

Post-Experimental Sociometric

Data . .

CHAPTER IV

RESULTS . .

Experimental Analysis of Single

Subject 1: C.F .

Subject 2: D.O. .

Subject 3: K.K .


;ment .


























Assessment
. ..

. .




. .


. .

. .











. ..


Case


. .

. .


Subjects

. .

. .

* .










Subject 4: S.S. ... 75

Subject 5: M.S. . ... .. 76

Subject 6: R.G. . ... 78

Subject 7: P.C. . . .. 80

Sociometric Assessment .... 81

Socially Isolated Games Participants ... 81

Socially Isolated Non-Participants ... 82

Moderately-Liked Games Participants ... 83

CHAPTER V

DISCUSSION .... . . 85

Discussion and Implications .... 85

Findings . .... . 86

Interpretation of Findings .. ........ 91

Problems and Limitations . .. 93

Practical Implications . 93

Suggestions for Future Research .... 95

REFERENCES . .... . 97

APPENDIX A DESCRIPTION OF SUBJECTS . .. 109

B ROSTER-AND-RATING SOCIOMETRIC TEST .. 114

C BEHAVIOR RECORDING FORM . .. 115

D GAMES SESSIONS .... .116

BIOGRAPHICAL SKETCH ...... . 149















Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of
the Requirements for the Degree of Doctor of Philosophy


THE GAMES ANALYSIS INTERVENTION: A PROCEDURE
TO INCREASE THE PEER ACCEPTANCE AND SOCIAL
ADJUSTMENT OF SOCIALLY ISOLATED CHILDREN

By

Michael J. Marlowe

August, 1978

Chairperson: Mary K. Dykes
Major Department: Special Education

The effectiveness of a therapeutic motor development program

in increasing the social adjustment and peer acceptance of seven

socially isolated fifth grade children was investigated using a

single subject experimental design. Three female and four male

children ages 10 through 11 were the subjects. The motor

development program was based on the games analysis model, a

framework from which any movement game can be analyzed and

structured by the teacher and the students to account for the

individual motoric, social, and emotional needs of the partici-

pants. The experimental condition included (a) instruction from

an instructor in the games analysis process, (b) assessment of

each child's motor skills in striking, throwing, running, catching,

and kicking, and (c) playing games designed to account for the

current stages of motor and social development of the children.


viii










Peer acceptance of the socially isolated children was

measured by a "play with" roster-and-rating sociometric instru-

ment. Social adjustment was measured through use of a behavioral

assessment tool. Two types of behavior were coded, task partici-

pation and social orientation. Data were obtained by trained

observers.

An A-B-A-B single subject reversal design was employed to

examine the experimental condition's effect on the social adjust-

ment of the children. During the baseline phases, traditional

movement games were played according to regulation rules. In the

intervention phases, the movement games played were designed and

adapted via the games analysis process by the instructor and

students. The "play with" sociometric posttest was administered

three weeks after the end of the experimental condition in order

to measure gains in peer acceptance.

Pretest-posttest assessment of the peer status of the

socially isolated children indicated that all seven children

had made median rank position gains in peer acceptance among

their classmates. Through experimental analysis of the single

case subjects it was observed that the games analysis training

resulted in a controlling effect on task participation and social

orientation. The results are discussed with regard to implications

for developing patterns of play based on individually based

activities.

















CHAPTER I

INTRODUCTION



This study concerns socially isolated children who have

become the subject of increased clinical concern in recent years.

Socially isolated children have been defined as those children

who are rejected or nonaccepted by their peers (Gottman, Gonso, &

Rasmussen, 1975; Gronlund, 1959).

One often observed characteristic of socially isolated

children is the absence of a well developed repertoire of social

behaviors (Hartup, Glazer, & Charlesworth, 1970). These

children have difficulty in learning to relate in a reciprocally

reinforcing manner with their peers (Allen, Hart, Buell, Harris,

& Wolf, 1964; O'Connor, 1969, 1972). Typically, these children

exhibit a wide range of socially deficient behaviors including

lack of functional speech (Risley & Wolf, 1967), opposition to

social requests (Wahler, 1967), and physical withdrawal from

adults and peers (Hutt & Ounsted, 1966). They are friendless

loners who are apparently unable to avail themselves of the joy

and satisfaction of social reciprocity (Kauffman, 1977).

Child development specialists have recognized the importance

of satisfying early peer contacts to children's cognitive,











emotional, and social growth (Piaget & Inhelder, 1969). Cooley

(1909) wrote that "without healthy play, especially group play,

human nature cannot rightly develop" (p. 49). More contemporary

researchers also supported this viewpoint. Jones (1967) stated

that early play experiences with peers may be as important for

human development as Harlow and Harlow (1962) found it to be for

the development of adaptive social and sexual behavior among

rhesus monkeys. McCandless and Hoyt (1961) stressed that the

normal development of preschool children is dependent upon peer

interaction. According to these authors, peer interaction

provides children with otherwise unobtainable opportunities to

rehearse important life roles. Likewise, Reese and Lipsitt

(1970) posited the importance of peer relationships to child

development. They emphasized that early peer interaction is of

profound developmental significance because it provides children

with contexts for practicing motor, language, and social skills

essential for adult functioning.

Children receiving restricted amounts of social stimulation

from peers probably experience concomitant reductions in the

physical and biological stimulation they receive (Bijou, 1966).

Developmental delays in a number of areas of personal and academic

competence are frequently associated with the socially isolated

child (Amidon & Hoffman, 1965; Ausbel, 1958; Whitman, 1970).

Socially isolated children are more likely than others to drop

out of school (Ullman, 1957), to be identified as juvenile delin-

quents (Roff, Sells, & Golden, 1972), and to experience mental











health problems (Cowen, Pederson, Rabigan, Izzo, & Trost, 1973).

Kohn and Clausen (1955) reported that the proportion of social

isolates in adult manic depressives and schizophrenics was close

to one-third, while in normal control groups the proportion was

close to zero. In a survey of research on suicide and attempted

suicide, Stengel (1971) concluded that "social isolation is the

common denominator of a number of factors correlated with a high

suicide rate" (p. 28).

Despite the potential importance of increasing children's

social reciprocity and peer acceptance, procedures for modifying

social isolation have not been generally successful (Amidon &

Hoffman, 1965; Early, 1968). In several intervention studies

(Chennault, 1967; Lilly, 1971; Rucker & Vincenzo, 1970), classroom

activities such as performing in a skit were provided to increase

isolated children's peer acceptance. Compared to control groups,

the isolated children tended to make initial gains in peer status.

When follow-up sociometric assessment was employed just several

weeks after the intervention, however, it was found that the

initial gains had not been maintained. The initial gains in peer

acceptance may have resulted from the isolated children's greater

visibility in the classroom. These interventions, however, did

not teach the children the social skills necessary to maintain

peer status when special situations were not provided. It is

important that procedures be developed which will result in more

lasting gains in isolated children's peer relationships.











Given the importance of early play experiences with peers

(Piaget & Inhelder, 1969; Loy & Ingham, 1973), instruction in

play and game skills is a variable that may be used to effect

isolates' peer acceptance. There is limited clinical evidence

which suggests that socially isolated children are deficient in

the motoric abilities and physical skills necessary to relate in

a reciprocally reinforcing manner with their peers in play and

game activities (Fraleigh, 1956; Haley, 1969; Gruber & Kirkendall,

1974). Cowell (1960) and Cowell and Ismail (1961) have conducted

a number of studies with children involving the relationship

between sociometric status and physical abilities. In general,

they found that boys who score high on physical measures enjoy

high acceptance on sociometric measures while boys who score low

on physical measures receive low acceptance on sociometric measures.

Clarke and Greene (1963) also found significant correlations between

social acceptance and motor ability items in ten year old boys.

Evidence of the relationship between social isolation and

delayed motor development has appeared in the child development

literature (Ausbel, 1958; Hartup, 1970; Whitman, 1970). Ausbel

(1958) theorized that the age at which children demonstrate

behaviors characteristic of various periods or stages of motoric

development depends, to a large degree, on the nature and extent

of their social interaction. Hartup (1970) stated, "There is

little doubt that the changes which occur in child-child inter-

actions during infancy and childhood are closely linked with

changes in sensory-motor capacities" (p. 368).





5





One intervention identified in the literature to facilitate

social interaction and the development of physical skills is the

method games analysis. Games analysis developed by Morris (1976)

is a process whereby a teacher can structure a game to account

for individual motoric differences as well as to promote specific

behavioral outcomes such as cooperation.

In games analysis, a game is divided into six categories

(players, equipment, movement pattern, organizational pattern,

limitations, and purpose) with each category consisting of a

number of components. It is possible to provide a wide range

of educational experiences by changing the components of one or

more categories dependent upon the types of experiences the

teacher desires for the children to experience. To manipulate

components and/or categories'constructively involves both the

teacher and students in a "games analysis" process. If specific

social, emotional, and motor behaviors are to be enhanced, the

games must be designed to account for the current growth and

development of the performers. For example, by allowing the

students to decide what piece of equipment to use when at bat in

a softball "type" of game, the motoric needs of each student can

be accounted for within the game structure. The highest skilled

student may choose to use a regulation softball bat and a regula-

tion softball thrown in the traditional manner. On the other

hand, the least skilled student may choose to have a large rubber

ball rolled toward him and to strike the ball with an oversized

underweighted bat. The child who was heretofore labelled as











"clumsy" may now be totally included and accepted as an equal since

all abilities are considered. Giving children the opportunity to

make decisions prevents the game structure from automatically

excluding children due to their poor motor skills.

There are three major factors within the games analysis

process which contribute to children's social and physical develop-

ment: (1) allowing children the opportunity to share in decision

making about the structures of the games to be played, (2) instruct-

ing children in problem solving skills via games analysis strategies,

and (3) focusing on children's motor skills and designing games

and movement tasks to account for the current stages of motor

development by the children. These factors are utilized in

structuring games to promote specific social, emotional, and

motoric behaviors.

The present investigation was designed to examine the efficacy

of the games analysis intervention in increasing the peer status

and pro-social behavior of the isolated child. Peer acceptance

was measured by a "play with" roster-and-rating sociometric

instrument as designed by Roistacher (1974). Social adjustment

was measured through use of a behavioral assessment tool. Two

types of behavior were coded, task participation and social orienta-

tion. Task participation referred to whether or not the child was

participating in playground activity. Social orientation referred

to the type of peer-oriented behavior which the child was exhibiting.

There were four mutually exclusive categories to code the behavior:











(a) peer oriented/supportive, (b) uncommunicative/ignoring,

(c) uncooperative/rejecting, and (d) other, that is, behavior

in response to outside noise or distraction.

Previous researchers have neglected to assess the effective-

ness of a therapeutic motor development program in enhancing the

socially isolated child's sociometric status and social behavior.

The importance of physical activities, particularly games, as

methods of facilitating social-emotional development merits

experimental research.


Statement of the Problem


The purpose of this study was to investigate games analysis

as a method of training socially isolated children in motor and

game skills for positive changes in peer acceptance and social

adjustment. This investigation was one of the first attempts to

modify social isolation by means of a therapeutic motor develop-

ment program.


questions Under Investigation


This study was designed to answer the following questions:

1. Will the games analysis intervention increase the peer

acceptance of socially isolated children?

2. Will the games analysis intervention increase the task

participation of socially isolated children?












3. Will the games analysis intervention increase the peer

oriented/supportive behavior of socially isolated children?

4. Will the games analysis intervention decrease the

uncommunicative/ignoring behavior of socially isolated children?

5. Will the games analysis intervention decrease the

uncooperative/rejecting behavior of socially isolated children?


Definition of Terms


Games Analysis--a framework or model from which any movement

game can be analyzed, investigated, and/or modified.

Peer Acceptance--the ratings a child receives from his same

sex peers' responses to the question, "How much do you like to

play with this person at school?" as measured by a roster-and-

rating sociometric questionnaire.

Peer Oriented/Supportive Behavior--behavior toward peers on

the part of the child which is socially appropriate (looking,

touching, or talking in a neutral way toward peers) or support-

ive (smiling at or being helpful toward peers).

Socially Isolated Children--the two least accepted children

in a school classroom based on the ratings received from same-sex

classmates as measured by a roster-and-rating sociometric

questionnaire.

Uncommunicative/Ignoring Behavior--behavior toward peers

which is uncommunicative (not looking, touching, or talking to

peers) or ignoring (looking away, turned away from peers).










Uncooperative/Rejecting Behavior--behavior toward peers

which is uncooperative (bullying peers, hogging materials) or

antisocial (name calling, making mean faces at peers).


Delimitations


The sample population for this study was taken from Alachua

County in the state of Florida, and only fifth grade children parti-

cipated. The ages of the subjects ranged from 10 to 11 years.

This sample cannot be considered representative of all fifth grade

children or of all young children in the state of Florida nor in

the United States.


Limitations


At least fourlimitations of the study must be considered.

First, the subjects of this study cannot be said to be representa-

tive of all socially isolated children. There is great vari-

ability in definitions and diagnoses of the syndrome of social

isolation.

Second, the sociometric assessment used in this study does

not record actual association; it does not describe actions; and

it does not provide a picture of the actually existing group

relations in a concrete situation. Sociometry records only what

people report and has the virtues and limitations of any such











subjective data. Generalizing from a study utilizing socio-

metric instrumentation should be done with caution (Olmsted,

1959).

Third, since Alachua County is primarily a rural agrarian

environment, these findings cannot be considered representative

of children from urban industrial environments. Finally, the

researcher of the present study served as the instructor in the

games analysis training. Although a conscientious effort was

made to minimize differences in the instructor's role during

baseline and treatment phases, it may be that the instructor was

more enthusiastic and/or attentive during the treatment phases.


Summary and Overview


In Chapter I, the problem under investigation was intro-

duced and the need for such an investigation briefly discussed.

In Chapter II, the review of literature, the problem will be

discussed in more detail as it fits into the context of past

research. The actual procedures involved in the study will be

presented in Chapter III. The data obtained in the study are

presented in Chapter IV. In Chapter V the reader will find a

discussion of the data and the results of the study and recommenda-

tions for research that will further investigate the efficacy of

the games analysis intervention in increasing the peer acceptance

and social adjustment of isolated children.















CHAPTER II

REVIEW OF THE LITERATURE


There are four bodies of literature relevant to the present

study. Each will be reviewed in this chapter. The review includes

literature concerning (a) the socially isolated child, (b)

sociometrics and peer acceptance, (c) physical activity and

the social-emotional development of children, and (d) games

analysis.


Socially Isolated Children


The two most commonly used methods of identifying socially

isolated children are based upon (a) sociometric (peer ratings)

measures (Amidon, 1961; Gottman, Gonso, & Rasmussen, 1975; Marshall

& McCandless, 1957) and (b) behavioral observation measures of

social interaction frequency or duration (Allen et al., 1964;

O'Connor, 1972; Walker & Hops, 1973). These two assessment

methods have led to two distinct definitions of social isolation

in the literature. Investigators who have used behavioral

observation measures of social interaction have conceptualized

the problem of isolation as "social withdrawal," defined as low











relative frequencies of peer interaction (O'Connor, 1969, 1972).

Other investigators using sociometric measures have conceptual-

ized isolation as low levels of peer acceptance or high levels

of peer rejection (Gottman, Gonso, & Rasmussen, 1975).

It is unclear whether social isolation is a unitary con-

struct. The two conceptualizations have produced entirely

separate literatures on intervention with social isolates.

Efforts to increase the relative frequency of peer interaction

with "withdrawn" children have been relatively successful

(Evers & Schwarz, 1973; O'Connor, 1972); however, investigators

who have studied low frequencies of peer interaction have failed

initially to assess whether a low frequency of peer interaction

is a problem which should, in fact, be ameliorated. It is not

obvious that children should all interact with one another at

a specified rate or that children whose relative frequencies are

well below average are somehow at psychiatric risk (Gottman,

1977; Gottman, Gonso, & Rasmussen, 1975).

Conversely, there are data to suggest that children who

are rejected or are not accepted by their peers are at risk

and that sociometric measures are predictive of later social

functioning (Cowen et al., 1973; Roff, Sells, & Golden, 1972).

Gottman (1977) constructed a classification system for describing

socially isolated children by assessing 113 children in Head

Start classrooms using sociometric measures of acceptance and

rejection and observational measures of frequency of behavior.












There was no relationship between peer acceptance and the

relative frequency of peer interaction, suggesting that these

two measures of social isolation do not tap the same dimension.

In this study the term "socially isolated" is used as a

descriptor for children who are unpopular with peers.

Socially isolated children may be classified with respect

to a bipolar continuum that ranges from Type I, deficient

social repertoire to Type II, deficient social performance

(Strain, Cooke, & Apolloni, 1976). Type I behavior is demon-

strated by children who, in an environmental setting suitable

for the maintenance of most children's social behavior, con-

sistently exhibit a limited number of social-response behaviors.

Moreover, these social-response behaviors are emitted at low

rates and are frequently not under the stimulus control of peer

behavior. In some extreme cases, human attention does not func-

tion as a generalized conditioned reinforcer with Type I children

(Lovaas, Freitag, Kinder, Rubenstein, Schaeffer, & Simmons,

1964). The distinguishing characteristic of Type I children

is that they have not acquired the basic vocal and motor-response

behaviors that are necessary for mutually reinforcing inter-

actions with peers. Children classified under Type II, deficient

social performance, are capable of demonstrating a variety of

adaptive responses in peer interactions but rarely engage in

mutually reinforcing interactions with peers and/or engage in

mutually reinforcing interactions only (1) in the presence of











certain people such as teachers but not with peers (Hart, Reynolds,

Baer, Brawley, & Harris, 1968) or (2) with opposite sex peers but

not same sex peers (Sibley, Abbot, Stark, Bullock, & Leonhardt,

1967). Therefore, Type II children are those who have the

necessary behavioral repertoire for mutually reinforcing inter-

actions with peers, but, nonetheless are not successful in their

peer interactions. Type II children are the subjects of this

investigation.


Prevalence

Social isolation represents a major presenting symptom in

14% to 30% of cases in which children are referred for psycho-

logical services (Gilbert, 1957; Heinstein, 1969; Woody, 1969).

Gilbert (1957) found that isolate behavior was a presenting

problem in approximately 15% of the children referred to four

urban child-guidance centers. In another study, Woody (1969)

found that social isolation was a major presenting complaint in

14% of preadolescent age problem children identified through

teacher referrals. Finally, Heinstein (1969) reported that of 142

mothers surveyed in California, 30% indicated that their five year

old children had too few friends.

Social isolation may be a condition that frequently begins

in the preschool years and lasts for long periods, perhaps through-

out life (Morris, Sorokin, & Burrus, 1954). Research has shown

that children's peer-group acceptance generally stabilizes during

the preschool years (Green, 1933) with mutually congenial children











pairing off into relationships that exclude unpopular children

(Sorokin & Groves, 1950). Singer (1951) presented evidence that

indicated long-term consistency in attained levels of peer-group

acceptance. Bonney (1943) and Ausbel (1958) stated that

children not accepted by their peers in the early grades will

have considerable problems in achieving satisfying interpersonal

relationships as adults.


Behavioral Correlates

Social isolation is normal for all children in some situations.

It is generally agreed that when social isolation inhibits the

normal personality development of a child an intervention pro-

cedure is required (Kauffman, 1977; Ross, 1974). The behaviors

that make up the social isolation syndrome, as previously noted,

may occur with a varying degree of severity.

Social isolation involves behavior that keeps people at a

distance physically and emotionally (Kauffman, 1977). The un-

popular child may exhibit a lack of specific behavioral responses,

such as looking at, talking to, playing with, and touching peers

or adults (Amidon, 1961; Keller & Carlson, 1974). Usually the

child is also lacking in responsiveness to others' initiation

of social contact (Buell et al., 1968; Strain & Timm, 1974).

Numerous authorities on children's peer relationships have

suggested that the isolate or unpopular child does not engage

in mutually reinforcing interactions with peers (Blasdel, 1968;











Charlesworth & Hartup, 1967; Evers & Schwarz, 1973; Kirby &

Toler, 1970; Quiltich & Risley, 1973; Wahler, 1967). Social

reciprocity is a major antecedent of social acceptance and

rejection (Marshall & McCandless, 1957). Newcomb (1956) has

stressed the importance of positive reinforcers in the emergence

of attraction between individuals and Byrne (1961) has argued

that the proportion of positive reinforcers dispensed by one

individual to another determines attraction between them.

Hartup, Glazer, and Charlesworth (1970) conducted a study to

examine the relationship between peer reinforcement and peer

acceptance. Major findings of the study were that social accept-

ance was significantly correlated with giving positive reinforce-

ment,and rejection was significantly correlated with giving

negative reinforcement. These correlations support the role of

reinforcing interaction in the emergence of interpersonal

attraction.

Children who are unsuccessful in engaging in mutually

reinforcing interactions with peers will be seriously handicapped

in acquiring many of the complex behavioral repertoires necessary

for effective social functioning (Marshall & McCandless, 1957).

Children who are unable to relate skillfully to others are likely

to experience rejection, harassment, and generally hostile treat-

ment from peers (O'Connor, 1969, 1972). These negative experiences

would most likely reinforce interpersonal avoidance responses and

further handicap the child in developing competencies that are

socially mediated.











Bandura (1963) pointed out that learning theory predicts

that some children, especially those who have not been taught

appropriate social interaction skills and those who in the past

have been punished for attempts at social interaction, will

become social isolates. The same children may learn to counter-

agress against others who attack them, or attack unsuspecting

victims when the potential consequences of their assaults seem

likely to be favorable.

Gottman, Gonso, and Rasmussen (1975) reported that a rela-

tionship exists between peer acceptance and cognitive development.

These authors found that popular and unpopular children differed

in their knowledge in how to make friends and on a referential-

communication task. Using Piaget's stages of cognitive develop-

ment Rarden and Moan (1971) showed that peer relationships

develop in a manner similar to the development of the physical

concepts of conservation and classification. The authors

interpreted results from their study by suggesting that cognitive

and social development parallel one another and are possibly

interdependent processes. Direct support for the hypothesis of

a relationship between popularity and cognitive development comes

from a study by Goldschmid (1968) in which she found popular peers

to be more adept at conversation while other investigators have

found socially isolated children to be slower at overcoming

egocentrism (Neale, 1966) and less apt to recognize and label

emotions (Izard, 1971).











Socially isolated children may be low achievers in school

(Bonney, 1971) and may exhibit learning difficulties (Amidon,

1961; Amidon & Hoffman, 1965). Bryan (1974, 1976) pointed out

that the poor academic performance of the isolated child may be

a result of restrictions in sensory stimulation associated with

low interaction levels. Inadequate stimulation may result in a

failure to develop those discrimination that are necessary for

successful school performance. Harris and Sherman (1973) found

that peer-peer tutorial sessions held immediately before math

classes resulted in increased performance rates. Cobb (1970)

reported that on-task conversation about academic materials by

peers resulted in greater achievement gains than did individual

attention to task. It has been shown that children learn much

through interaction with one another. Since socially isolated

children receive little peer stimulation, their academic per-

formance may be depressed.

Other commentators have noted that socially isolated

children are likely to be retarded in motor skills (Fraleigh,

1956; Haley, 1969; Hartup, 1970). Smoot (1974) stated the

issue:

Because many motor skills and discrimination are
learned in a context of interpersonal reinforcement,
social interaction is a critical prerequisite for
much of a child's motoric development. Conversely,
the absence of social interaction probably insures
that development will be retarded. (p. 8)

McGrew (1972) in an ethological study of children's social

behavior, characterized the isolated child as physically











apprehensive, uncoordinated, and delayed in gross motor develop-

ment. Numerous individuals have conducted studies involving

the relationship between physical skills and peer acceptance

(Clarke & Greene, 1963; Coleman, 1966; Cowell, 1960; Cowell &

Ismail, 1961, 1962; Fraleigh, 1956; Gruber & Kirkendall, 1974;

Yarnall, 1966). The findings of these studies suggested the

important role physical abilities and motor skills play in peer

acceptance.

In summary, the socially isolated syndrome can be defined in

terms of the following observable events: (1) behaviors that

keep people at a distance physically and emotionally; (2)

deficient responsiveness to others initiation of social contact;

(3) failure to achieve social reciprocity; (4) delayed cognitive

development; (5) low academic performance; and (6) delayed motoric

development. The link between social isolation and developmental

deficits in a number of areas appears to be established.


Interventions

Some (Bloom, 1964) believe that environmental manipulations

have optimum impact while children are young, a time when their

behavioral repertoires are undergoing rapid expansion and refine-

ment. The most efficacious time, therefore, to influence human

social development is when children are young and first learning

to interact with peers (Apolloni & Cooke, 1975). In accordance

with this perspective, numerous attempts have been made to study

levels of interaction and degrees of peer acceptance while











experimentally altering their environment. In this section,

there is a critical review of procedures that have been

purported effective for increasing Type II socially isolated

children's friendship making ability and social interaction.

Evaluative research on operant conditioning techniques is

presented followed by a review of special event techniques

used to modify isolates' behavior and acceptance. Next, a

critical look at peer attention procedures and affective educa-

tional techniques is included, followed by a discussion of

needed research on the influence of therapeutic motor develop-

ment programs.

Operant conditioning techniques have been used extensively

to train isolated children in social skills. Behavioral shaping

and modeling procedures have been used to increase the fre-

quency of isolated children's peer interactions. In shaping

procedures, social praise or tangible rewards have been used to

increase the frequency of children's peer interactions gradually

(Allen, Hart, Buell, Harris, & Wolf, 1964; Amidon, 1961; Blasdel,

1968; Hart, Reynolds, Baer,Brawley, & Harris, 1968; Milkey, 1970;

Shores, Hester, & Strain, 1976). Although such procedures

generally increase the frequency of isolated children's peer

interactions, the children's behavior tends to return to baseline

levels once reinforcement has been terminated (O'Connor, 1972).

Case study reports have indicated that a gradual fading of

reinforcement may result in longer-lasting effects (Baer & Wolf,

1970).











In modeling procedures, socially isolated children have

viewed films of other children receiving positive experiences

when they approached other children to join an activity or

conversation (e.g., O'Connor, 1969, 1972). In the O'Connor

research, the group of socially isolated preschool children who

viewed the film were subsequently observed to increase their

peer interactions. A control group which watched a neutral

film did not change. When the author conducted a follow-up

assessment several weeks later he found that the increase in

peer interactions of the experimental group was maintained and

the control group remained unchanged. Similar results have been

found elsewhere with the same experimental film (Evers & Schwarz,

1973).

Keller and Carlson (1974) employed several videotapes in

which a different social interaction skill was featured in each

videotape. Isolated preschool children watched a series of

videotapes in which children performed positive social behaviors

with peers by either imitating, smiling and laughing, giving

tokens, or giving physical contact signifying affection. As

in the O'Connor film, the videotapes were accompanied by a

narrative sound track in which the actions of the children were

described. A control group watched neutral films. The authors

found that children who saw the modeling videotapes increased

significantly more than the control group in giving and receiving

social reinforcement and in frequency of social interaction.











Follow up results, three weeks later, indicated no significant

differences between the experimental and control conditions.

Unfortunately, none of the investigators who have studied

behavioral shaping and modeling effects have included socio-

metric measures of peer acceptance as part of the evaluation

of the intervention. These investigators have assumed that low

peer interaction rates are equivalent to being socially isolated,

by which they presume that these children are unpopular or

socially unskillful and that these children would like to, but

are actually unable to make friends with peers. It is not

obvious that children should all interact with one another at

a specified rate or that children whose relative frequencies are

well below average are somehow at psychiatric risk (Gottman,

1977).

Special event techniques is another educational procedure

that has been used with socially isolated children. In several

intervention studies environmental conditions were arranged to

provide situations conducive to social interactions. Quiltich

and Risley (1973) examined the possibility that children's social

behavior might be influenced by the nature of play materials.

Children in an urban recreational center were systematically

provided with toys designed for social or isolate play. The

authors found that the subjects spent a large percentage of time

in play under both conditions (96% with isolate toys and 98% with

social toys), but that the type of play, social or isolate,












varied dramatically as a function of the type of toy being used.

The investigators concluded that toys may be of therapeutic

value in facilitating social behavior.

Chennault (1967) used classroom skits to increase isolated

children's peer acceptance. The isolated children tended to

make initial gains in peer status, but when a sociometric assess-

ment was administered several weeks after the intervention, the

author found that the initial gains had not been maintained.

Another way to use special events to increase isolates'

play with peers was reported by Kirby and Toler (1970). Their

study attempted to increase the rate of interaction between a

five-year-old isolate boy and his nursery school classmates.

He was induced to pass out choices of candy to his classmates

and subsequently given a nickel, candy, and praise from the

teacher. By strengthening the interaction behaviors in this

particular situation it was thought that the behavior would

generalize to other situations. It was found that interaction

with classmates increased greatly during the periods he passed

out choices of candy. These changes may have been due to

increases both in his rate of initiating activities with class-

mates and to increases in his classmates rate of initiating

activities with him; however, the issue of generalization from

the training situation was not assessed empirically.

A third major intervention category with socially isolated

children has been peer attention procedures. Numerous











experimenters (Kinney, 1953; Kohn, 1966; Levison, 1971;

O'Connor, 1969, 1972; Wahler, 1967; Walker & Hops, 1973) have

investigated the influence of peer reinforcement on the fre-

quency of social responsiveness. Kohn (1966) investigated

the extent to which rate and quality of peer activity directed

toward a particular child was the function of the rate and

quality of acts that the child initiated toward others. Positive

correlations were found between (a) the rate at which a child

initiated interaction toward others and the rate at which his

peers initiated interaction toward him, (b) the proportion of

positive acts that the child initiated toward others and the

proportion of positive acts that they initiated toward him, and

(c) the proportion of positive acts that the child initiated

toward his peers and the ratio of the rate of others initiating

toward the child by rate of his initiating toward others. It was

concluded that in an interaction with peers a child manages to

evoke from them the kind of behavior that will permit him to

maintain his prevailing mode of adaptation.

Levison (1971) designed an experiment to investigate the

efficacy of peers in decreasing the social isolation of a preschool

child. It was found that as the withdrawn child becomes the

recipient of positive reinforcement from peers, the child demon-

strates an increase in both verbal interactions with others and

cooperative play. The socially isolated child, by lacking the

social skills to reward his peers, therefore, may be an infrequent











recipient of social interchange. The author suggested pairing

children who provide positive reinforcement at a relatively high

rate with those classmates who received and provided rewards at

a low rate in order to increase their cooperative play and

social skills.

In another study utilizing peer attention procedures Wahler

(1967) demonstrated that the social interaction of peers is

subject to reinforcement control of peer attention. He found

that children may control one another's behaviors in accordance

with an experimenter's instructions; for example, the control of

aggressive behavior such as throwing toys and inappropriate

shouting and running. Children's aggressive behavior increased

when it was reinforced by peers. As a result, Wahler (1967)

postulated that the socially isolated child might be most

effectively treated by techniques utilizing peer social atten-

tion contingencies.

As in the behavioral shaping and modeling studies with

isolated children, the isolated children in peer attention

studies have been identified on the basis of their low fre-

quency of peer interaction and/or teacher selection. Since no

measures of peer acceptance were collected, it is impossible

to assess whether the increased peer interactions also resulted

in gains in peer acceptance or friendships. Peer attention

procedures have produced reliable increases in training (Walker

& Hops, 1973) but have not, as yet, been shown to result in

generalized performance increases under nontraining conditions.











Affective educational strategies to assist socially isolated

children in gaining peer acceptance have been employed by

several experimenters (Amidon & Hoffman, 1963, 1965; Early,

1968; Flanders & Havumaki, 1960; Northway, 1944). Amidon and

Hoffman (1963) recommended assisting isolated children by (1)

the creation of an accepting classroom atmosphere through in-

creased teacher acceptance of all pupils, (2) group discussions

and role playing to heighten children's understanding of the

feelings associated with one another's social roles, (3)

awarding isolated children status responsibilities in the

classroom, and (4) teacher conferences with the isolated child.

Evaluative studies regarding the efficacy of these and similar

techniques, in some cases, revealed statistically significant

improvements by isolates in their classroom sociometric position

(Amidon, 1961; Moreno & Jennings, 1944; Northway, 1944). Indivi-

dual changes, however, generally were small, and no assessments

were made of the durability of the changes achieved. Furthermore,

the affective techniques employed were neither specified in

operational, repeatable terms, nor systematically evaluated

(Bonney, 1971). A number of additional evaluations of these

same tactics, moreover, resulted in nonsignificant changes for

experimental versus control subjects (Amidon & Hoffman, 1965;

Early, 1968; Mayer, Kransler, & Matthews, 1967).

While the efficacy of affective interventions with isolated

children has not been established in the literature, a number of











conclusions with implications for future research seem warranted.

First, Amidon (1961) demonstrated that teachers were more

successful in enhancing an isolate's peer popularity when they

related specific affective techniques to specific acceptance

problems. That is, affective interventions must be indivi-

dualized and tailored to meet the particular needs of each

isolated child. Secondly, Bonney (1971) noted that, although

significant changes in group performance were rare in affective

interventions, in each study, several students made exceptional

gains on sociometric choice. Bonney (1971) concluded that "in

future studies of this nature much more focus should be placed

on individuals as opposed to concentration on group data"

(p. 362). Thus, Bonney recommended using single-subject rather

than group-research design.

There have been few investigations concerning the efficacy

of therapeutic motor development programs in enhancing the social

adjustment of isolated children. A few studies (Brown, 1970;

Myrick, 1970; Wahler, 1967) have shown that social interaction

variables may be enhanced through participation in motor and play

skill training. Given the evidence (Coleman, 1966; Cowell &

Ismail, 1961, 1962; Fraleigh, 1956) which suggests a moderate

relationship between peer acceptance and physical abilities and

the findings of child developmentalists (Ausbel, 1958; Whitman,

1970) that isolated children are often retarded in motoric

development, the influence of motor training on social adjust-

ment warrants further attention.











Morris (1976) reported clinical findings which suggested

that a motor development program based on the games analysis

model may be effective in contributing to the social-emotional

development of children. In an experimental study, Marlowe,

Algozzine, Lerch, and Welch (1978) found that a motor develop-

ment program based on the games analysis model was effective

in significantly reducing the feminine play preferences of

boys identified as emotionally disturbed. These initial

findings suggest the need for further research on the effec-

tiveness of a motor development program based on the games

analysis model in influencing the social-emotional development

of children.



Sociometrics and Peer Acceptance


In dealing with affective relations within a group, the

technique of study which has probably been employed more than

any other is sociometry (Olmsted, 1959). Moreno initiated

development of sociometric measures in 1934. In sociometric

measures group members are given questionnaires asking them whom

they like or dislike, or more specifically, whom would they like

to play with, work with, sit next to, and so on. Their responses

to the questionnaire provided a picture of the group's affective

relationship among members, those who are most accepted and those

who are least accepted or rejected.











Lindzey and Bryne (1968) summarized Moreno's (1934) require-

ments of a sociometric test as follows:

1. The limits of the group should be indicated to the

subjects. For example, choices could be limited to members of

a specific classroom.

2. The subjects should be permitted an unlimited number

of choices.

3. The subjects should be asked to indicate the individuals

whom they choose or reject in terms of a specific criterion.

Each sociometric choice should be made with a meaningful activity

in mind; for example, "name the students you would like to play

with."

4. Results of the sociometric question should be used to

restructure the group. Subjects should be told that their input

will be used in making such decisions.

5. The subjects should be permitted to make their choices

privately.

6. The questions used should be gauged to the level of

the understanding of the sample.

Lindzey and Byrne (1968) commented on the above mentioned

requirements as follows.

The requirements outlined above identify the socio-
metric measure in a more or less pure form, and are
generally in agreement with Moreno's definition.
However, relatively few studies in this area meet all
the requirements. For example, the technique as used
today seldom involves the promise of restructuring
the group One of the more frequent modifica-
tions involves specifying the number of choices the
individual is required to make. (p. 455)












Thus, sociometric studies are not uniform in their method-

ology. This poses a problem in assessing the reliability and

validity of the technique; however, since pure applications of

sociometric testing are the exception, rather than the rule,

and since the variations of the pure form seem to yield similar

results; they will all be considered together and the reliability

and validity of sociometric testing (in its broad sense) will be

said to apply to the sociometric procedure used in this study.

In fact, most of the studies in which the reliability and

validity of sociometric tests have been addressed have been

variations of the pure form (Gronlund, 1959). Concerning the

reliability of sociometric testing, a point made by Gronlund

(1959) needs to be considered.

Perfect consistency from one test to another is neither
expected nor desirable, owing to the dynamic nature of
social relations. Revealing actual changes in social
relations is as important a requirement of the socio-
metric test as providing results that are constant
enough to have predictive value Thus, when
applied to sociometric testing, the various coefficients
of reliability refer to the consistency of choice
behavior rather than to the characteristic of
the test itself. (p. 119)

Thus, the changes in sociometric results do not necessarily

reflect testing error; to a great extent they reflect actual

changes in the social preferences of the subjects. With this

point in mind, the internal consistency and test-retest reli-

ability of sociometric tests will be discussed.

Grossman and Wrighter (1948) determined the internal con-

sistency of their sociometric testing with four classrooms of











sixth graders. They randomly divided each group in half and

then correlated the sociometric status of each subject, as

rated by one half of the class, with his status in the other

half of the class. They reported coefficients of internal

consistency from .93 to .97. Bass and White (1950) and

Ricciuti and French (1951), using college students as subjects,

reported internal reliability coefficients of .90. Ausbel,

Schiff, and Gasser (1952) reported coefficients from .54 to

.86 for third, fifth, and seventh graders and coefficients of

.89 and .90 for 11th and 12th graders.

Test-retest reliability has been determined by correlating

the sociometric status of subjects on one sociometric test with

their status on another test at a later date. Witryol and

Thompson (1953) and Thompson and Powell (1951) studied the

consistency of sociometric choices made by sixth graders at

intervals of one week, four weeks, and six weeks. Witryol and

Thompson (1953) reported reliability coefficients ranging from

.60 to .90, and Thompson and Powell (1951) reported correla-

tions ranging from .89 to .92. In both studies, the coefficients

tended to decrease after two months.

Using longer time intervals, Byrd (1951) found a reliability

coefficient of .89 with fourth graders selecting partners for a

play after a two-month interval, and Gronlund (1955) reported

an average reliability coefficient of .75 for fourth, fifth,

and sixth graders over a four-month interval. Bonney (1943)











administered a sociometric test, an IQ test, and an achievement

test to a group of second graders for four consecutive years.

The correlations of students sociometric status from one year

to the next ranged from .67 to .84. The sociometric status of

the children in this study was as consistent as their IQ and

achievement scores over the four year period.

In studying the consistency of sociometric choices of high

school students Northway (1947) reported coefficients of .90

for a one week interval and .60 for a one-year interval.

Jennings (1950) investigated the sociometric choices of

adolescent girls (12 to 16 years old) over time and found a

correlation of .96 after four days and a correlation of .65

after eight months.

Thus, the results of sociometric testing are relatively

consistent. They are almost as reliable as typical intelligence

and achievement tests, as Bonney (1943) has demonstrated, and

more reliable than most attitude/personality measures.

As with reliability, the concept of validity, as it is

typically applied to testing and measurement, needs to be

qualified in its application to sociometric tests. If socio-

metric tests are supposed to measure merely social choice, then

they are by definition valid (Pepinsky, 1949). However, if they

are to measure actual social relationship, they will certainly

fall short of this expectation because, as Gronlund (1959) has

mentioned,











An individual's actual associations are influenced
by environmental limitations, personal inhibitions,
lack of reciprocal feelings on the part of the
desired associates, and other related factors, as
much as they are by his preferences. Thus actual
association can be expected to show some variation
from the desired associations indicated in socio-
metric choices. (p. 159)

Although sociometric choices do not correspond perfectly

with actual associations, there is a great deal of overlap.

This overlap, sociometric choices that are also actual associa-

tions, is considered evidence of the validity of sociometric

testing for this study because the first hypothesis purports to

measure peer acceptance. Studies in which the validity of socio-

metric testing have been investigated have related sociometric

status and sociometric choices with other measures of popularity

and friendship choices (Bonney & Powell, 1953; Gronlund, 1955,

1956).

Concerning sociometric status, the observations of teachers

and independent investigators correspond rather closely to the

results of a sociometric test. Bonney and Powell (1953) found

that sociometrically high first and second graders participated

more frequently in cooperative group activities and associated

with more children than did sociometrically low children.

Newsletter, Feldstein, and Newcomb (1938) found a correlation

of .76 between the sociometric status and the camp counselor

ratings of popularity for 30 adolescent boys. Gronlund (1951)

had 40 sixth grade teachers rank their students according to

popularity; the average correlation between the teachers'












rankings and peers' rankings was .60. Gronlund (1955, 1956)

obtained similar results in two other studies.

The individual choices of a subject are more complex and

variable (Gronlund, 1959) than sociometric status. Therefore,

individual choices are more difficult to observe and less likely

to correspond closely with sociometric results. Biehler (1954)

compared the first sociometric choice of kindergarten children

with their observed play companions. About 74% of the chosen

companions actually played with the children who chose them.

Gage, Leavitt, and Stone (1955) asked 103 fourth, fifth, and

sixth grade teachers to predict how each of their students

would respond to a sociometric item that asked for the student

to list five children in their room whom they would most prefer

as classmates if the class were divided into two groups. The

average correlation between the teacher's prediction and the

sociometric results was .48. Considering that observations

are subject to error and that some aspects of friendship cannot

be observed, the above-mentioned evidence suggests that socio-

metric results correspond rather closely to actual friendships

and peer acceptance.


Physical Activity and Social-Emotional Development


Few persons would deny that physical activities, including

organized sports, have the "potential" for influencing social-

emotional growth in children. Participation in physical











activities--including games, play, and sport--provides the

opportunity for considerable social interaction under a wide

range of situations. Helanko (1957) has suggested that physical

activities have been developed by society for the specific

purpose of developing social-emotional competence in young

people.

The importance of physical activities, particularly games,

as methods of facilitating social-emotional development has

received attention in several cross-cultural studies. Roberts,

Arth, & Bush (1959) suggested that games of strategy were related

to mastery of the social system, games of physical skills were

related to mastery of the physical environment, and games of

chance were related to familiarity with the supernatural.

Roberts and Sutton-Smith (1962) used these game categories in

their analysis of the child rearing practices of 111 societies.

Their findings in part confirm Roberts, Arth, & Bush's (1959)

earlier proposal and in part supplement or replace it with

evidence of additional motivational theme. Basically, they

found that

1. societies that stress obedience training emphasized

games of strategy;

2. societies that emphasize responsibility training

stressed games of chance; and

3. societies that emphasize achievement stressed games

of physical skill.

These findings were then used to predict game preferences among











segments of our society that could be differentiated on these

three child-rearing dimensions. Sutton-Smith, Roberts, and

Kozelka (1963) confirmed the prediction within our society for

boys versus girls, and among adults who differed by sex, educa-

tion, and occupation. Thus the combined within and cross

cultural findings indicate a positive relationship between

these games and children's social-emotional development.

Sutton-Smith et al. (1963) did not provide any insight into

how these activities are utilized in developing social-emotional

competence.

Experimental investigations dealing with the contributions

of physical activity to social-emotional development have been

reviewed by Cowell (1960) and Layman (1960, 1970). Although

Cowell sought evidence to show that physical activity contributed

to social-emotional development, he was only able, at best, to

establish that a moderate relationship exists between several

social variables and participation in physical activity.

Layman (1970), relying on similar evidence as reviewed by

Cowell, attempted to substantiate six major propositions.

1. Engaging in sports promotes physical fitness;
physical fitness is associated with good emotional
health and a lack of fitness with poor emotional
health.

2. The acquisition of motor skills involved in sports
contributes toward meeting the basic needs of
safety and esteem in young children of both sexes
and in young men from the early grades through
college years.

3. Supervised play presents potentialities for promot-
ing emotional health and preventing delinquency.











4. when play, recreation, and athletic
activities are planned with individual needs
in mind, they may be very valuable means of
improving emotional health among emotionally
disturbed children.

5. Play and sport supply outlets for the expression
of emotion in approved activities which is con-
ducive to the development and maintenance of
emotional health.

6. Competitive sports, if properly used, may enhance
emotional health and the acquisition of desirable
personality traits. (p. 45)

The research evidence presented by Cowell (1960) and Layman

(1960, 1970) does not substantiate these generalizations.

Experimental studies in the physical education literature are

largely correlational. Researchers have only suggested that a

relationship exists between social-emotional development and

physical activity, but they have not shown that participation in

physical activity "causes" social-emotional growth (Stevenson,

1977). Physical educators preoccupation with justifying that

physical activities "do" develop social-emotional variables has

diverted them from examining "how" physical activity may facilitate

social-emotional variables.

After reviewing the available experimental evidence,

Fraleigh (1956) also identified a moderate relationship between

physical activity and social adjustment (sociometric measures

of acceptance), but he avoided interpreting a cause and effect

relationship. Fraleigh did suggest, however, that the relation-

ship may be a spiraling circular one.











The better adjusted tend to participate in more
social and competitive play because they have
relatively higher levels of physical skills and
because of their high skill levels they gain
favorable self-evaluations in addition to higher
status recognition from their peer groups. On
the other hand, the more poorly adjusted tend to
participate in more individualized and less com-
petitive play because of relatively lower levels
of physical skill. This lower level of skill
leaves the more poorly adjusted in a less advanta-
geous position in terms of gaining desirable
evaluations of self and high status recognition
from the peer group. (p. 271)

How does one enter the circular relationship hypothesized

by Fraleigh (1956)? One way may be to improve motor skills.

Although evidence is not conclusive, numerous studies have

shown that in some cases improvement in motor skills by

participation in motor development programs or other forms

of physical activity has resulted in increased peer acceptance

and social skills (Allen et al., 1964; Brown, 1970; Buell et

al., 1968; Galvin & Witt, 1969; Gump & Sutton-Smith, 1955;

Haley, 1970;Johnston et al., 1966; Myrick, 1970). Brown (1970)

concluded that a six week motor program improved preschool

males' ability to acquire rapport, to communicate, and to accept

responsibility. Gump and Sutton-Smith (1955) utilized thera-

peutic play techniques in helping shy students become more

aggressive and to help aggressive children become more controlled.

Galvin and Witt (1969) found that a sociorecreation program for

conduct-disordered boys assisted in the development of social

skills and rechanneling deviant behavior. Haley (1970) employed












a movement training program with withdrawn emotionally disturbed

children which resulted in increases in their interaction levels.

The development of social-emotional skills from participa-

tion in physical activity programs has not always been observed

(Olson, 1968). Further research, however giving careful atten-

tion to the type of physical activity, the instructional

techniques used, and the specific social-emotional skills to

be developed, should aid in clarifying these ambiguities.


Games Analysis


Games analysis is a method by which any movement game can be

analyzed and investigated in the sense that game structure

promotes specific outcome behaviors. Morris (1976) using the

games analysis model, divides the structure of a game into six

categories; players, equipment, organizational pattern, movement

pattern, limitations, and purpose. Each category consists of a

number of components. Table 1 contains an illustration of a

variety of components selected from a few games by Morris (1976).

It is possible to provide a wide range of educational experiences

by changing the components of one or more categories dependent

upon what the teacher wants to occur during the game experience.

To manipulate components and/or categories constructively

involves both the teacher and students in a "games analysis

process."







40







a) S- 0- S-
> O *.-
4 O ) (A 3'
0 E 4- > c U
L- 0 o- .- o i
a) U E S- -0 ) 4-' E (o
a0 O O c V) (A >
0 4-' 4-' 0*
a) 0 > C Q a) S.- *
(A tU o 0 n I 0 0- 0 O
0 0 S-- ) o0 E 0- S-
a- a) 0- ) >)t- a) o t (.o
S-. a) -0 4-) > U U
3 oS- > o (a a>)
EO a) o n C C 04-)' a) C.L
C 0 *r- -0 C *4 r- Q. 0 4- 4-) 0 (
*r- L > r- O 0 *r- O i- a
3 .rCl ( .-- E > C, S- (LaU a S- E C a
S-*r- O -- 0 -C > r- 0 > 4-
O O a) L S- *o O a) S- a) ,--
4-) 4-m (n C o U c U 0 (l 03 a 4-


4-'
4-) l L >
(A (D (A C s.- S S- S-
m' C C ,- a r 0 a w
4- 0 4- 3 >, *
C: *r- C L M Ls o OI .-C
a) 4-3 C a)- 4 a)
C ( *r *r4- 0 O E S-- *r- L

E E ~ C 4-) 3 C.
S 3 > > o 4- -
u -J 0 3 E3 E a) E- S -- -0 (- a)
*o n In 0 o E E E v-


C0 0- C ca) 0 C 0 4-- '
-C-) -- L- 4-) 00 c E.0(A


(A


-0 C E ) C a)
0n O- a i- E -- I.
0 *r- l 0 r-
S( 4-) 0 0
ra ) N4-) 0L U U a1) a0
I- c- (A S- 0 -a -
C- E ) c a) a 0 C: c a)
a) s O -E 0- i- c C S- C
SU 3 :. a) O C E -r ro c"-.
0 S- C S- 3 a *)1 3 C*
C O ( r- 0 O r- 0 4- > S*- S- O
U a a) 0 4-)
(A
*--

>- 4-- Cn C 0) n
c cn cn o
CC E C ) C L *r C C C C 0
a) L C C C -r C C *r- *0 *r- C
C E 1) *r- a *r- 0 *r- 3 C *-
< I4 C 0. a a0 s O U ( i- c-
>-> Ec E *r- U S r- c-.
SO 3 3 0o 0 -n -C co 3 0
a) S- -)L ( 4-) U +4- S-
E


(A-
4+-' C
0 *'- V)
C
a) 0 ( 4-3
E (A *r- C *r- Q.
,.a (A a ( (n -- (A *( O >
U > (a a) a) -0 -- *-
4- 0 0 C 3 00-0
Cy 0fa 4-) 0 0 (0 0 CO
LU 00 ( 0) CC S- -0 3 -B -.


S-E
i- a) S.- S- a)r s-
A fo cr 3 (a2 -0 a) a
S- 3 0 O- E4-)L-
a) -0 4- 4-' 3 E (
> E E E E C o C :3 r
3 > ca n3 to 3 0 C r C C4- C-.
a- G- a) a) a) a) E to Ca O 0 cr
C- ~. C a E c 0
*-r CC ) O (n -) o0O a)000











A number of rationales have been offered for having games

as part of the physical education program in elementary schools

(Kirchner, 1974; Mauldon & Redfern, 1970; Mosston, 1966). The

following are the most quoted reasons for including games in a

school setting:

1. Games promote the socialization process.

2. Games aid in the development of motor skills.

3. Games help develop emotional understanding between

and within children.

Not one of these game purposes will be served in a game unless

the game is structured to promote the specific behavior. Thus

it is important to understand that the design of every game

played promotes and elicits resultant behaviors, and if certain

outcomes are desired from a game then the game must be structured

accordingly and specific behaviors observed and recorded (Morris,

1976).

There are three major factors which underly the games

analysis process: (1) allowing children the opportunity to share

in decision making about the structure of the games to be played,

(2) instructing children in problem solving skills via games

analysis strategies, and (3) focusing on children's motor skills

and designing games and movement tasks to account for the current

stages of motor development by the children. Each of these major

factors will be reviewed and their role in the games analysis

process will be delineated.











Student Decision Making

Numerous authors have reported in the physical education

literature that participating in the planning of an activity

is a strong factor contributing to the overall enjoyment

(Anderson, 1966; Mosston, 1966; Nutting, 1973). Martinek,

Zaichkowsky, and Cheffers (1976) found that allowing children

to participate in decision-making concerning a physical activity

program had a positive effect on the development of motor skills

and self-concept. Mancini, Cheffers, and Zaichkowsky (1975)

concluded that allowing children to participate in decision-

making in a human movement program facilitated peer interaction

and interaction with the teacher.

According to Piaget (1965), Sutton-Smith (1971), and Loy

and Ingham (1973) children's ability to make decisions concerning

game structure both emotionally and socially undergoes a develop-

mental process. Helanko (1957) developed a table of game

socialization stages. Children five to six years of age regard

game rules as sacred and absolute and exhibit a great deal of

egocentric behavior. Thus to try to teach group decision-making

skills concerning the design of a game is difficult. By ages

seven to nine, children begin to exhibit cooperative behavior,

game rules need not be absolute, and game decision-making skills

can be introduced. By the ages of 10' to 12, children are

capable of handling complex social interactions involving

cooperation and competition. Rules are regarded as relative.












According to Helanko (1957), this is the age to teach decision-

making skills to deal with social and emotional reactions and

conflicts that develop due to game design.

In the games analysis process students are given an

opportunity to make decisions when the teacher allows the stu-

dents to change the structure of the game to meet the motoric,

emotional, or social needs of an individual youngster or a

group of youngsters. By allowing the youngsters to change or

manipulate one or more of the components within a given

category or categories, the decision-making process contri-

butes to the children feeling that they have some control over

their environment. Decision-making allows children a feeling

of security with the game world (Morris, 1976; Mosston, 1966).

Morris (1976) has developed a hierarchy for decision-making

responsibility in students.

Initially the teacher controls the game environment
by making all of the decisions about the game for all
of the youngsters. When game structure decisions are
gradually relinquished to the youngsters in a pre-
scribed manner the youngsters learn how complex game
design is, how one game category relates to another,
and how to deal with the responsibility that accompanies
any decision making process .. At all grade levels,
3-12, it is suggested that the students first be
allowed to alter the type of movement employed during
the game. Next allow them to change the number of
players per team; follow this by allowing them to
alter, adjust, or change the equipment or change the
type of equipment being used. (pp. 19-20)

Giving children the opportunity to make decisions prevents

the game structure from automatically excluding children due to











movement or coordination problems. Morris (1976), Cratty (1967,

1969, 1970), Smoot (1974), and Mosston (1966) have commented

that children of all ages compensate for movement and motoric

difficulties with inappropriate social behavior such as social

withdrawal, aggression, or clowning. Utilizing the decision-

making process in games analysis the teacher can intervene by

asking the child or the group to change the game. Morris (1976)

reports that both decision-making strategies have been employed

with children from grade three to grade 12 with a high degree

of success as measured by improved emotional control and socially

appropriate behavior.

Problem-Solving Skills

Problem solving is considered the highest level of inter-

personal development (Spivack & Shure, 1974). Some recent

studies have focused attention on the relationship between

interpersonal problem solving and social adjustment, although

they are few in number. Shure and Spivack (1972) and Shure,

Spivack, and Jaeger (1971) examined how specific aspects of

problem solving ability relate to behavioral adjustment among

five year old children. In the results of these studies it was

delineated that cognitive problem-solving skills were distinctly

superior among those judged adjusted in social interaction.

Those children judged socially inhibited were not able to think

through ways to solve typical problems successfully. These











studies support the hypothesis that one should be able to enhance

the social adjustment of young children if one can enhance their

ability to see a human problem, their appreciation of different

ways of handling it, and their sensitivity to the potential

consequences of what they do.

Interpersonal problems are constantly developing in

physical education settings when youngsters are participating

in games. Children must constantly deal with the analysis of

the other teams' weaknesses or strengths. Children must be able

to adapt to any change within their play environment.

Interpersonal problem solving through games analysis requires

both physical and cognitive involvement. According to Morris

(1976) a major consideration is that the games analysis approach

not only can provide opportunities for individuals to seek

solutions to stated problems, but more important, involves the

student in the process of asking questions and defining problems.

It is this unique characteristic of developing and cultivating

the ability to independently discover and design new problems

which is the essence of games analysis.

Morris (1976) pointed out that by utilizing problem solving

all children can experience some degree of success within the

game situation. Incorrect responses are not criticized but

serve as a contribution to the basic learning situation. Each

child is allowed to make decisions concerning game design based

upon self-evaluation, knowledge, skills, and attitude. In problem

solving the focus is not upon accepted standards of movement but











rather on the ability to manipulate game components to accommodate

the needs of an individual child or a group of children. Through

creative game design students begin to understand the relation-

ships among categories within a game structure. The students

soon realize why it is necessary to have rules, how a change in

one category affects the design of another category, how within

a game there is a place for everyone's abilities, and, most

importantly, how differences in physical abilities make one feel

good or bad about oneself (Morris, 1976).


Assessment of Motor Skills

Assessment is used for diagnosis and prescription of movement

behavior. An assessment tool provides the teacher with informa-

tion which identifies why a child is behaving in a certain way,

and from this the teacher should be able to provide the children

with helpful information as to how they can improve their motor

performance.

A simple assessment tool called a movement profile sheet

indicates what each child's ability is in certain motor behaviors.

Developed by Morris (1976) the movement profile is based on a

task and factor analysis format. Table 2 contains a sample motor

profile as provided by Morris (1976) which he employed to analyze

and demonstrate the striking ability a student has displayed

while considering the interaction of two factors--striking

implement and size of the object being struck. This profile

sheet indicates to the teacher which of the two factors under











investigation influenced this particular child's striking

ability. By observation, the striking implement seems to be

the more limiting factor of the two because the student was

able to strike all the sizes of the object but could only strike

with a hand or paddle. This information is important for the

teacher because it explains why a child demonstrates a particular

kind of motor performance. Thus, in a game design this particular

child should be allowed to use a hand or a paddle to strike. It

is highly probable if he/she uses a bat he/she will not be

successful.


Table 2

Movement Profile #1


S1

Il Succeeded on
1-8-76

12 Tried but failed
1-8-76

13 Tried but failed
1-8-76


S2
Succeeded on
1-9-76

Succeeded on
1-10-76

Tried but failed
1-10-76


S3
Succeeded on
1-17-76

Succeeded on
1-22-76

Tried but failed
1-22-76


S4
Succeeded on
1-23-76

Failed on
1-23-76

Failed on
1-23-76

Failed on
1-23-76


S = size of object
S1 = 12" balloon
S2 = 10' plastic beach ball
S3 = 8" rubber ball
S4 = 4" rubber ball


I = striking implement
Il = hand
12 = paddle
13 = racquet
14 = whiffle ball


(p. 56)











When assessing a motor skill, the teacher needs to be

cognizant of all the factor interactions. Morris (1976)

identifies the following factors that affect the motor

behaviors of striking, catching, kicking, and throwing.

1. Size of object.

2. Color of object.

3. Trajectory of object.

4. Weight of object.

5. Speed of object.

6. Color of background.

7. Anticipation location.

8. Texture of object.

9. Illumination level.

10. Illumination type.

11. Object direction in flight relative to performer.

12. Speed of performance by student.

Movement profiles may be easily designed in order to assess the

various factor interactions. Morris (1976) provides examples of

movement profiles on kicking and the interaction of size of object

and weight of object, and on catching and the interaction of

texture of object with angle of trajection. These appear in

Tables 3 and 4.

The purpose of the movement profiles is to determine each

child's current movement status, to assess the factors that most

affect each child's performance, and to offer direction for the















Table 3

Movement Profile #2




Kicking Profile Sheet


Size of Object (S)


Weight of Object (W)


Easy-------------------------- Difficult


6 oz. ball
8 oz. ball
12 oz. ball
16 oz. ball
largest ball
large ball
small ball
smallest ball


- 18" diameter
- 14" diameter
- 12" diameter
- 8" diameter


1 S2 S3 S4
Easy

W1


W2


W3

W4
Difficult
Difficult


(p. 64)















Table 4

Movement Profile #3




Catching Profile Sheet


Texture of Object (T)


Angle of Trajection (A)


Easy----------------Difficult


fleece ball
nerf ball
nylon ball
whiffle ball
horizontal plane
vertical plane
ball travels in arc


(p. 64)


T1 T2 T3 T4

A1


A2

A3











teaching of the motor behavior. The profile concept is for

individual analysis rather than individual comparison to group

standing and the teacher uses the profile sheet to assess the

children's progress.

Morris (1976) pointed out that games analysis can incor-

porate the task analysis concept quite easily. A teacher can

design a series of games that allows for progressive improvement

in specified motor behaviors by focusing upon one, two, three,

or more factors that affect the said motor behaviors. Table 5

illustrates how to apply the motor profile information to the

games analysis grid. By referring to the children's profile

sheets the teacher can design games appropriate to their movement

needs and skills.

From the profile sheets information, the teacher can

develop lead-up games which are games that lead into the tradi-

tional game. If the teacher is going to teach children the game

of football, the teacher should not expect the children to be

able to play the game efficiently on the very first day. Rather

the teacher develops over a period of days the skills necessary

to play football. Thus, the teacher instructs the children in

lead-up games that promote the development of the skills

necessary to play football.

First, the teacher decides on the terminal game objective

in terms of skills requirement and concept involvement. The

teacher then writes behavioral objectives for the day's lesson.

The teacher then writes the game into the grid. Next, the
















Un
4-' c
(d *r-
() u

o c-i
0
S 0 C V

O II
>






0
(0 aTC C7 ---------------




4 a)

*r~

4-
E Wa
F- C)






E
fu ai a) i a a)
C a) E cJ E E co E c
a 0< < rc: 3 < c rZ <
o 0 HI V) I f)1 1, ) I
4- E I I I I I I I
0) i 0) 1 0) 1 0) 1 I0

S | *r- S C *r* C *r- .C .C
+J (JU U -Y U -M U (U
(d 4-- 4+-' 04-' u4-a u0-'
0
a 4-

S uE i-f I- I I n I I
(0 3 S 3 3 3
,- *r r c- i c 0 -c co






4.- D c D)CU CU U DC (
*r-
S0 1 00 0 m
C
C V I VL) I ) I L) uL)

4 C F= I I I I I I







4-'
S C
so E
U C:





C)D












a)
E m LO
ro
CD











teacher plans how and when to teach specific motor skills and

game concepts. Morris (1976) provides an example of stated

behavioral objectives relative to psycho-motor development

(motor) and cognitive development. In addition, one can also

state behavioral objectives relative to affective development.

Unit Football Skills and Concepts
Objectives Motor:
Develop ability to pass the football a distance
of 10 yards with accuracy; hand the ball off without
fumbling the ball; center the ball with accuracy a
distance of 5 feet; punt the football with accuracy
a distance of 15 yards; catch a thrown football 10
feet, five of ten attempts; display ability to run
with the ball; demonstrate ability to screen a
block.
Objectives Cognitive:
Students will understand via performance the
following concepts: line of scrimmage, scoring
system, five running plays, five pass patterns,
rules of touch football, defensive positioning,
offensive positioning. (p. 28).

Objectives Affective:
Students will participate in "huddle" strategy;
will cooperate in deciding play patterns and position
duties.

Then, the teacher begins to teach the motor skills and game

concepts via some form of game to promote interest or move

toward the terminal objective--regulation touch football. At

this point, games analysis assists the teacher in designing new

games or altering existing games to meet the motoric and

cognitive needs of the students. Morris (1976) provided an

example of a lead-up game to regulation touch football based

on the preceding stated objectives (see Table 6).














E c
L C u
QLOO OW (A (-
o 0 t I-- *i,- C: a n C
) -- Q- 4- -" a Ct-
a 0a D 30 3-
0- > S- S- E -
SL a -J-- 0 a) 0 (U -J

O w-.e S- O O *r-"= C\i
0 CL U U 4-)
4-'
C O
-E 0 0 0 L
E a) cD u
Cn (A S- 4- ) '4- 1 4- c 4- c
C : .- a) ai ( a ) Ln > >- ) ) S- S-+4-
O E--4--. 3 I C .O S- 3
r- S- C- ) 4-

.J 4-) O C & C C
ro> I- S -)- -04-) i )f-- (a 0




0 c ( -04- La C -- 4-4- 0 E 0-4-) 0 C -
0 r c 0 oOO r_






E U 4 S0 ) CC -4 S



4-) Q E C: c **

CO 4- a P r o
C CCL o O So 4- =O O-









r -C ( E 4-O


(- r 4-
Co L *r- 4- e



a)
E E E
S0 *4-)

n a u c
0)








E c .-
CD <





S I -4





c:
CL
E -0 (



LU 4







C(
E) E










ta)
L3 4-
0 .











a)
E Cn
(0 C3
) 00
4-

a) >,



P C -
to a)











In summary, games analysis is predicated on the notion that

teachers need to understand the relationships between game

designs and behavioral outcomes if they want each child to

develop to his or to her capacity within a game environment.

If specific social, emotional, and motoric behaviors are to be

developed, then the game must be structured to promote the

specific behaviors. Games analysis is a process the teacher

can utilize in structuring games.



Summary


After a review of the literature that is relevant to the

question of the efficacy of the games analysis method in

increasing the acceptance and adjustment of socially isolated

children, a number of conclusions logically follow. First,

while Morris (1976) has presented anecdotal reports of the

efficacy of the games analysis method in promoting the social

development of socially isolated children, there are no studies

which have employed even minimal standards of experimental

control and rigor in investigating the games analysis method

as a treatment for socially isolated children. Second, a study

needs to be undertaken that will experimentally investigate the

efficacy of the games analysis method in promoting the social

development of socially isolated children. To meet this need

this study was undertaken to experimentally explore the efficacy






56




of the games analysis method in increasing the acceptance and

adjustment of socially isolated children.
















CHAPTER III

METHOD AND PROCEDURES


In Chapter III, the methods and procedures used in the

study will be presented. This chapter is divided into three

sections; description of subjects and their selection,

description of materials, and description of procedures,

including the experimental design and instrumentation.



Subjects

The subjects for this study were 14 fifth grade public

elementary school students. Nine of the subjects were ten

years old and five subjects were 11 years of age. Eight of

the subjects were males of whom four were black, three were

white, and one was American Indian. Six of the subjects were

female of whom three were black and three were white. (See

Appendix A for a further description of the subjects.)


Pretraining Sociometric Assessment

A roster-and-rating sociometric questionnaire was adminis-

tered in each of three fifth grade classrooms by the researcher











who introduced himself as someone who was interested in learning

about how children play together in school. Children rated

each other in answer to the question, "How much do you like to

play with _at school?" The rating scale consisted

of five points with 1 being "I don't like to" and 5 being "I

like to a lot." Children were taught how to use the scale and

examples of play situations in school were suggested. Children

were provided with five point scales with faces above each

point ranging from a frown to a happy face. (See Appendix B

for a sample questionnaire.)

The roster-and-rating sociometric questionnaire has been

found to correlate highly with peer-nomination measures (Justman

& Wrightstone, 1951; Young, 1947); it has two particularly

attractive features compared with the peer nomination socio-

metric measure. First, the classroom roster decreases the like-

lihood of a person not being chosen because he or she was

momentarily forgotten. Second, the method provides an indica-

tion of the child's acceptance by all of the group members

since each child is rated by all of his or her classmates.

The "play with" sociometric questionnaire was administered

to assess the peer acceptance of the fifth grade children. One

measure was obtained for each child from the sociometric

questionnaire; a "play" rating based on the average rating

received from other children. For the purposes of this study

measures were based on same-sex ratings or nominations since











fifth grade children typically give low ratings to opposite-

sex peers (Criswell, 1939).


Selection of Subjects

Selection of least preferred children (social isolates)

and seven other participants was based on the mean ratings

received from classmates on the "play with" sociometric measure.

The two least preferred children of each sex in each of the

three fifth grade classrooms were grouped together for a total

of 12 children. Seven of these children were chosen randomly

for the study by shuffling cards containing their names. Moder-

ately preferred peers were selected to form a second group for

the study. Seven of the third through eighth most frequently

preferred children of the same sex as the isolated children were

randomly chosen from the three classrooms as participants for

the games sessions by shuffling cards containing their names.

Selection of subjects resulted in seven moderately preferred

children and seven least preferred children.



Materials


The variety of developmental motor equipment used in the

study was included to reflect the individual motor needs of the

children. Equipment used by the children in various games played

were:

1. Hula hoops,

2. Bean bags,











3. Playground balls (various diameters),

4. Soccer balls,

5. Volleyballs, nets, and ropes,

6. Beachballs (various diameters),

7. Nerfballs,

8, Plastic balls,

9. Softballs (various colors),

10. Whiffle balls,

11. Softball bats, bases, and gloves,

12. Plastic bats,

13. Tennis rackets,

14. Batting tee,

15. Bowling pins, and

16. Basketballs and basketball goals.

The actual extent to which these were used will be described in

detail later.



Experimental Procedures


Experimental Design

The experimental design for this study was a variant of the

A-B-A design (Sulzer & Mayer, 1972). It is described as an

A-B-A-B single subject reversal design (Holland & Skinner, 1961)

with two replications for each subject. The design phases are

summarized as follows:











A = traditional games played by children (i.e., baseline)

B = games played by children designed via games analysis

model (i.e., intervention)

The data were collected according to the following

schedule:

Number of Experimental Sessions (Days)

Phase Sessions

Al 1- 5

B 6 14

A2 15 18

B2 19 -22

The data collection involved a period of five weeks.

This design was selected for use in this study for three major

reasons; (1) it facilitates direct measurement over time of speci-

fied observable behaviors (Sulzer & Mayer, 1972), (2) it allows

the researcher to demonstrate experimental control if the behavior

changes only at the point when the treatment is introduced

(Risley & Baer, 1973), and (3) reversal strategies can be

employed with an individual subject (Sidman, 1960).


First Phase A

The children played traditional games such as softball, kick-

ball, basketball, and volleyball according to regulation rules.

The effects of various factors (e.g., size of object) on the

children's movement skills were observed by the instructor. (See

Appendix D for a further description of the games played.)











First Phase B

The children were introduced to and trained in the games

analysis model. They were instructed in the decision making

and problem solving skills needed to adapt and design games via

the games analysis process. Utilizing his observations of the

motor skills of the children, the instructor designed games for

the children to play which accounted for their current stages

of motor development. The games played were then adapted and

changed by the instructor and/or students during the games

sessions to meet the specific social, emotional,and/or motor

needs of an individual or a group of individuals. (See Appendix

D for a further description of the games played.)


Second Phase A

The children played traditional games according to regulation

rules. Softball, volleyball, kickball, and basketball were played

(see Appendix D).


Second Phase B

The children played games designed via the games analysis

model. The designs of the games reflected the individual needs

of the children (see Appendix D).


Behavioral Assessment

The playground behavior of each socially isolated child was

observed and coded every tenth second for 18 consecutive ten

second intervals on a daily basis for 22 games sessions. Two











types of behavior were coded each time datawere collected,

task participation and social orientation. Task participation

referred to whether or not the child was participating in the

activity. Since 18 observations were made per games session,

scores could range from zero to 18. Social orientation referred

to the type of social behavior which the child exhibited every

tenth second. The observer could select one of four mutually

exclusive categories to code the behavior; (a) peer oriented/sup-

portive, that is, behavior toward peers which was socially

appropriate (looking, touching, or talking in a neutral way

toward playground members) or supportive (smiling at or being

helpful toward playground members); (b) uncommunicative/ignoring,

that is, behavior toward playground members which was uncommunica-

tive (not looking, touching, or talking to playground members) or

ignoring (looking away, turned away from playground members);

(c) uncooperative/rejecting behavior, that is, behavior toward

playground members which was uncooperative (bullying playground

members, hogging materials) or antisocial (name-calling, making

mean faces at playground members); and (d) other, that is,

behavior in response to outside noise or distraction. For each

games session, the frequency total for each category could

range from zero to 18. (See Appendix C.)

The frequencies obtained in this study were evaluated

according to experimental criterion (Risley, 1970). The

experimental criterion refers to a comparison between behavior












during the intervention with what it would be if the intervention

had not been implemented.

The following procedures were employed to satisfy the ex-

perimental criterion. Baseline data were collected and served

as a basis for determining the present level of behavior and

predicting what behavior would be without the intervention.

The experimental method was introduced to reveal a change from

this projected performance. The reliability of a finding accord-

ing to experimental criteria was achieved by "replicating" the

baseline level of performance and a different level of performance

during intervention, as shown in the A-B-A-B design employed in

this study.

In practice, there were a few ways in which results clearly

met the experimental criterion. First, performance during the

treatment intervention when plotted could not overlap with per-

formance during baseline. The data points of baseline could not

extend to the levels achieved by the data points during the inter-

vention. If this non-overlapping design were replicated over time

with a given subject (A-B-A-B design), there would be little

question that the results were reliable (Risley, 1970). Second,

a more common criterion for experimental evaluation was related

to divergent slopes in baseline and treatment phases but was

less stringent than completely non-overlapping distributions.

This criterion emphasized the trends or slopes in each phase.

Usually the baseline phase showed a relatively stable performance

rate with no particular trend. When treatment was implemented,











usually a definite trend was evident indicating that behavior

was changing from baseline. If baseline conditions were

reinstated, the trend was likely to be in the opposite direc-

tion of the intervention. By alternating baseline and

experimental phases, systematic changes in trend comprised

strong evidence for the experimental reliability of the effect

(Risley, 1970).


Observers

The data collectors were two female graduate students in

the Department of Special Education of the College of Education

at the University of Florida. The data collectors were trained

to be behavioral observers by the researcher during a pre-

experimental training period. The specific skills the observers

acquired during the pre-experimental training sessions included

(1) a thorough understanding of the operational definitions of

the dependent variables, (2) accurate observations of the

dependent variables, (3) facility with the recording form, and

(4) facility with the stopwatch.

Interrater reliability was based on the percentage of total

agreements out of the total observations for each of the pre-

experimental games sessions during which the observers recorded

data on children. At each tenth-second observation point agree-

ment between observers only counted when both assessments of

participation and social orientation behavior were in agreement.












The minimum requirement for observer reliability was 86%, as

suggested by Risley (1970). The two observers were trained

to a range of 92%-100% per games session prior to the experi-

ment. The observers were not informed of the peer status of

the game participants. One graduate student served as the

principle observer of all the children's game sessions and

the other graduate student was employed to provide random

reliability checks on two different occasions distributed

throughout the study. Their percentage of agreements for each

play session ranged from 98% to 100% with an average reli-

ability of 99%. The observers' naivete as to the peer

acceptance status of the children was verified by a post-

experimental check.


Post-Experimental Sociometric Assessment

The experimental phase lasted for a period of approximately

five weeks. The interval between the end of the experimental

phase and the administration of the post-experimental assessment

was 20 days. On the 21st day the "play with" sociometric

questionnaire was again administered in each of the three fifth

grade classrooms by the same person who administered the pre-

test.


Data

Two types of data were used in this study. First, frequency,

the number of occurrences of a given behavior emitted by an





67




individual, was the datum for the behaviors of "task participa-

tion," "peer-oriented/supportive," "uncommunicative/iqnorinq,"

and "uncooperative/rejectinq." Second, sociometric play rating

scores, the average "play with" rating received from same-sex

classmates, was the datum for peer acceptance.















CHAPTER IV

RESULTS


In Chapter IV, the results of the study will be presented.

This chapter is divided into two sections; presentation of

the results of the experimental analysis of the social behaviors

of the socially isolated children during the games sessions and

presentation of the pretest and posttest sociometric ratings of

(1) socially isolated children who participated in the study, (2)

socially isolated children who did not participate in the study,

and (3) moderately-liked children who participated in the study.



Experimental Analysis of Single Case Subjects


In this section the social behavioral data collected on the

socially isolated children will be presented. Phase means and

ranges for task participation and social orientation behaviors

during both baseline and intervention phases are delineated.

The data points of these subjects are analyzed according to

experimental criterion.


Subject 1: C.F.

During the initial baseline phase (Al) C.F.'s task partici-

pation ranged from 14-16 for the 18 ten-second interval recordings











per observation period with a phase mean of 14.5. Supportive

behaviors ranged from 12-15 with a phase mean of 12.8. Uncom-

municative behaviors ranged from 2-4 with a phase mean of 2.8.

Rejecting behaviors ranged from 1-4 with a phase mean of 2.4.

When treatment was implemented (BI) task participation and

supportive behaviors increased (means = 17.4 and 17.3, ranges =

14-18 and 14-18). Uncommunicative behaviors decreased (mean =

0.7, range = 0-4), while no rejecting behaviors were recorded

during the B1 phase.

When baseline conditions were reinstituted (A2) task partici-

pation and supportive behaviors decreased (means = 13.50 and 13.25,

ranges = 12-15 and 12-16). Uncommunicative and rejecting behaviors

increased (means = 3.25 and 1.5, ranges = 2-4 and 2-3).

Reinstatement of treatment (B2) led to task participation

and supportive behaviors increasing (means = 18.0 and 17.7, ranges

= 18-18 and 17-18). Uncommunicative behaviors decreased from the

first day under phase B2 (mean = 0.25, range = 0-1). No rejecting

behaviors were recorded during this phase.

Through visual inspection of C.F.'s daily data in Figure 1,

an upward trend was evident in task participation and supportive

behaviors while a downward trend was evident in uncommunicative

behaviors during phases Bl and B2. At no time did C.F.'s data

points during Al and A2 for rejecting behaviors extend to the

levels of the data points during B1 and B2. These systematic

changes in trend are evidence of the reliability of the treatment

effect.










(Subject C.F.)


0-
1 5


(1Ti 31 t
(B31)


10
Session


Paci(; l i ne
(AZ)


I ieaitmerit
(B2)
'O'^-
b *t /


15 20
15 20


Task participation o-
Supportive behaviorO-O
Uncommunicative behaviork-l
Rejecting behavior 0-D

Figure 1

Occurrences of Various Pupil 3ehaviors
In Relation to Games Played











Subject 2: D.O.

As presented in Figure 2, D.O.'s task participation during

Al ranged from 11-14 with a phase mean of 12.4. Supportive

behaviors ranged from 11-15 with a phase mean of 12.6. Uncom-

municative behaviors ranged from 0-5 with a phase mean of 2.6.

Rejecting behaviors ranged from 0-6 with a phase mean of 3.0.

Implementation of treatment (B1) resulted in task partici-

pation and supportive behaviors increasing (means = 17.5 and

17.7, ranges = 16-18 and 17-18). Uncommunicative behaviors

decreased (mean = 0.33, range = 0-1) while no rejecting behaviors

were recorded throughout the nine day B1 phase.

Withdrawal of treatment (A2) led to task participation and

supportive behaviors decreasing (means = 15.0 and 15.75, ranges =

12-18 and 14-18). Under A2 conditions uncommunicative and

rejecting behaviors increased (means = 1.7 and 1.0, ranges =

0-3 and 0-2).

Reinstatement of treatment (B2) resulted in task participa-

tion and supportive behaviors increasing (means = 18.0 and 18.0,

ranges = 18-18 and 18-18). No uncommunicative nor rejecting

behaviors were recorded throughout the B2 phase.

Through visual inspection of D.O.'s daily data in Figure 2,

under B1 and B2, an upward trend in task participation and

supportive behaviors was evident while a downward trend was

evident in uncommunicative and rejecting behaviors. These

systematic changes in trend suggest the experimental reliability

of the treatment effect.








Basel ine
(Al)


Trea
(


(Subject D.O.)
18- 8 <

17-

16-

15- 0

14- ,

13-6

12-

11-*

10-

U 9-
C
8-
QJ
^L 7-

6-

5- 7 '

4- d
3- \

2-

1- / v "

I I I I I I I II
1 5 10

Task participation -4
Supportive behavior 0-O
Uncommunicative behavior H-
Rejecting behavior [--D]


Ltent
BI)

*s ,


Session


Basel ine
(A2)


Treatment
(B2)


S I I I I I 2I
15 20


Figure 2


Occurrences of Various Pupil Behaviors
In Relation to Games Played











Subject 3: K.K.

During phase Al K.K.'s task participation ranged from 11-13

with a phase mean of 12.0. Supportive behaviors ranged from

10-12 with a phase mean of 10.8. Uncommunicative behaviors

ranged from 4-7 with a phase mean of 5.6. Rejecting behaviors

ranged from 0-6 with a phase mean of 1.2. Implementation of

treatment in B1 resulted in task participation and supportive

behaviors increasing (means = 17.1 and 17.1, ranges = 16-18 and

16-18). Uncommunicative behaviors decreased (mean = 0.9, range

= 0-1) while no rejecting behaviors were recorded under B1

conditions.

Removal of treatment in A2 resulted in K.K.'s task participa-

tion and supportive behaviors declining (means = 10.25 and 9.0).

Uncommunicative and rejecting behaviors increased (means = 8.5

and 0.75, ranges = 6-13 and 0-2).

Reinstitution of treatment in B2 led to task participation

and supportive behaviors increasing (means = 18.0 and 18.0,

ranges = 18-18 and 18-18). No uncommunicative behaviors were

recorded under B2 conditions, while rejecting behaviors decreased

(mean = 0.25, range = 0-1).

As presented in Figure 3, at no time did K.K.'s data points

during Al and A2 for the measures of task participation and

supportive and uncommunicative behaviors extend to the levels

of the data points during B1 and B2. These non-overlapping

distributions are evidence of the treatment's effect on these

dependent measures. Although K.K. emitted a low frequency of











(Subject K.K.)
18-

17-

16-

15-

14-

13-

12-



10- 0-0
9- I


8-

u 7-

6-rn-
5- O

4-

3- .



1- 0- 0A ,D -/
o- b-D D-D----D-
I i l t i I T T ~I--T -- i T I T
I 5 10 15
Session
Task participation --*
Supportive behaviorO -0
Uncommunicative behavior U
Rejecting behavior F --U

Figure 3

Occurrences of Various Pupil Behaviors
In Relation to Games Played











rejecting behavior during Al and A2, divergent slopes in base-

line and treatment phases were present which indicate the

controlling effect of the games analysis treatment on this

measure.


Subject 4: S.S.

Observations of task participation during phase Al ranged

from 13-17 with a phase mean of 15.0. Supportive behaviors

ranged from 11-13 with a phase mean of 12.4. Under Al conditions,

uncommunicative behaviors ranged from 5-7 with a phase mean of

5.6. No rejecting behaviors were recorded during the Al phase.

Introduction of treatment (Bl) resulted in task participation

and supportive behaviors increasing (means = 17.7 and 17.5,

ranges = 17-18 and 17-18). Uncommunicative behaviors decreased

(mean = 0.44, range = 0-1) while no rejecting behaviors were

recorded during the B1 phase.

Return to baseline conditions in phase A2 resulted in task

participation and supportive behaviors dropping (means = 14.75

and 14.75, ranges = 13-18 and 13-18). Uncommunicative behaviors

rose (mean = 3.75, range = 0.5), while no rejecting behaviors

were recorded during the A2 phase.

Return to treatment in phase B2 resulted in task participa-

tion and supportive behaviors rising (means = 18.0 and 18.0,

ranges = 18-18 and 18-18). No uncommunicative behaviors were

exhibited during phase B2. As in the preceding three phases, no

rejecting behaviors were exhibited.












As presented in Figure 4, during phases B1 and B2 there

was an upward trend in S.S.'s task participation and supportive

behaviors and a downward trend in her uncommunicative behaviors.

These systematic changes in trend suggest the reliability of

the treatment's effect. Because S.S. did not emit any rejecting

behaviors during the experiment, no trend was evident for these

behaviors.


Subject 5: M.S.

Observations of M.S.'s task participation during phase Al

ranged from 16-18 with a phase mean of 17.6. Supportive

behaviors ranged from 16-17 with a phase mean of 16.6. No

uncommunicative behaviors were exhibited, while rejecting

behaviors ranged from 1-2 with a phase mean of 1.4.

Implementation of treatment resulted in task participation

and supportive behaviors increasing (means = 17.9 and 17.3,

ranges = 17-18 and 17-18). As in phase Al, no uncommunicative

behaviors were exhibited during phase Bl. Rejecting behaviors

decreased (mean = 0.11, range = 0-1).

Return to baseline conditions in phase A2 led to task

participation and supportive behaviors decreasing (means = 17.5

and 15.75, ranges = 16-18 and 15-16). Only one uncommunicative

behavior was exhibited during phase A2 while rejecting behaviors

increased (mean = 2.0, range = 2-2).

Reinstitution of treatment (B2) resulted in task participa-

tion and supportive behaviors increasing (means = 18.0 and 18.0),










(Subject S.S.)
18-

17-

16-

15-


14-


Basel ine
(Al)











o-d L



6-0
















Sn-iu


Tr-ea t.ient
(B1)










/O


I I I I I I I I I I I I


Basel ine
(A2)

































-D i


Treatment
(B2)


I I I I I I
15 20


Session
Task participation *-0
Supportive behavior 0-0
Uncommunicative behavior --
Rejecting behavior i---D

Figure 4

Occurrences of Various Pupil Behaviors
In Relation to Games Played











ranges = 18-18 and 18-18). No uncommunicative nor rejecting

behaviors were exhibited during phase B2.

As presented in Figure 5, during phases B1 and B2, there

was an upward trend in task participation and supportive

behaviors; however, because his data points were recorded at

such high levels for these dependent measures throughout both

sets of A and B phases, no conclusion as to the experimental

reliability of the treatment's effect can be made. Since only

one uncommunicative behavior was recorded during the experiment

no interpretation of the data pattern for this dependent measure

could be made. The data points for rejecting behaviors at base-

line did not extend to the levels of the data points during

intervention for both sets of A and B phases. Thus, application

and withdrawal of the treatment suggested a controlling effect

on rejecting behaviors.


Subject 6: R.G.

It was not possible to evaluate the results of R.G.'s data

points due to his infrequent presence at the games sessions.

His lack of attendance was attributable to two factors. First,

R.G. was remiss in returning his informed consent form for

participation in the study and did not receive clearance for

participation until the fourth day of the Al phase. Hence, it

was not possible to record R.G.'s social behavior during the

first baseline phase. Secondly, R.G. was absent from school










(Subject M.S.)
18-

17-

16-

15-

14-


Baseline
(Al)





0-0


E-q

-u-f
t*- DR


Trea tment
(B1)


1 ft-^-i~r- ~ r^


U

K 8R !!!!.


I I I I I I I I I I
1 5 10


Task participation "--
Supportive behavior 0-0
Uncommunicative behavior I--n
Rejecting behavior U -I]


Session


Baseline
(A2)


E -0-O


Treatment
(B2)


I I I I I I
15 20


Figure 5

Occurrences of Various Pupil Behaviors
In Relation to Games Played











for seven consecutive days during the third and fourth weeks

of the study. As a result, no behavioral data were collected

on R.G. for the last three sessions of B1 and for the entire

A2 phase. These combined absences resulted in having only one

data point of R.G.'s social behaviors during the two baseline

phases. Therefore, it was not possible to evaluate the treat-

ment's effect on R.G.'s social interactions.

Despite his absences, R.G. was present for ten of the 13

sessions in which the games played were adapted and designed

according to the games analysis model. R.G. expressed enjoyment

in playing the games and his means for task participation and

supportive behaviors during the ten sessions were 17.0 and 16.7.

His means for uncommunicative behavior and rejecting behavior

were recorded at 0.4 and 0.9 for the ten sessions.


Subject 7: P.C.

P.C. fractured her foot in a bicycle accident during the

first week of the study and was not able to participate in the

games from sessions three to 19. Therefore, it was not possible

to determine the treatment's effect on P.C.'s social behavior.

Despite her injury, P.C. chose to come to the games sessions

and watch from the sidelines. The instructor encouraged her to

participate in the group's decision-making and problem-solving

processes during the treatment phases and P.C. appeared to enjoy

doinq so. Hence, it was felt by the instructor that P.C.'s












"sideline" attendance was beneficial in promoting her under-

standing of game designs and behavioral outcomes.



Sociometric Assessment


Socially Isolated Games Participants

Pre- and post-sociogram data were analyzed in terms of mean

ratings on the "play with" scale and the child's median rank

among classmates of the same sex on the "play with" scale.

Sociogram mean ratings could range from a low of 1.0 to a high

of 5.0. The socially isolated children's mean ratings and

median rank on the pre-and post-intervention sociometric assess-

ment are presented below.

Pre-Median Post-Median
Pre In- Post In- Rank Among Rank Among
terven- terven- Same-Sex Same-Sex
Subject tion X tion X Net Classmates Classmates

C.F. 1.94 2.31 +.37 16th 14th
D.O. 2.78 2.92 +.14 16th 14th
K.K. 2.43 3.29 +.86 17th 10th
S.S. 2.45 3.38 +.83 14th 8th
M.S. 2.70 3.18 +.48 15th 9th
R.G. 1.65 2.16 +.51 20th 19th
P.C. 2.33 2.94 +.61 17th 15th

An analysis of the results of the posttest sociometric

ratings indicated that all seven isolated children gained in peer

acceptance. The isolated children received higher ratings from

the classmates they interacted with in the games analysis sessions

as well as classmates who did not participate in the games

analysis sessions. More importantly, all seven of the isolated











children increased in their median rank among classmates of the

same sex and only one subject, R.G., met the definition of being

an isolated child at posttest (the two least accepted members

of each sex in a classroom).


Socially Isolated Non-Participants

Pre- and post-sociogram data of the socially isolated

children who did not participate in the study were also analyzed

in terms of the mean ratings on the "play with" scale and the

child's median rank among classmates of the same sex on the

"play with scale." The results are presented below.

Pre-Median Post-Median
Pre In- Post In- Rank Among Rank Among
terven- terven- Same-Sex Same-Sex
Subject tion X tion X Net Classmates Classmates

V.B. 1.92 2.28 +.34 17th 16th
D.T. 2.62 2.71 +.09 16th 17th
R.J. 1.85 2.11 +.26 19th 20th
R.R. 2.13 2.47 +.34 18th 18th
T.B. 2.81 2.46 -.35 13th 14th

An analysis of the posttest sociometric ratings indicated

that four of the five socially isolated children received higher

mean ratings from same-sex classmates at posttest; however, four

of these children, D.T., R.J., T.B., and R.R., were ranked at

posttest as the least accepted children among same-sex peers in

their classrooms. V.B. increased one rank among his same-sex

classmates, at posttest, but his median rank position still met

the definition of his being a socially isolated child. Hence,

all five socially isolated children who did not participate in

the games sessions remained socially isolated children at posttest.











Moderately-Preferred Games Participants

Pre- and post-sociogram data on the moderately preferred

children, who participated in the games sessions, were also

analyzed in terms of mean ratings on the "play with" scale and

the child's median rank among classmates of the same sex on the

"play with" scale. The results are presented below.

Pre-Median Post-Median
Pre In- Post In- Rank Among Rank Among
terven- terven- Net Same-Sex Same Sex-
Subject tion X tion X Classmates Classmates

M.Z. 3.94 4.00 +.06 3rd 1st
A.C. 3.50 3.72 +.22 7th 7th
D.T. 3.50 3.83 +.33 7th 5th
C.M. 4.57 4.18 -.39 5th 5th
S.J. 4.00 4.21 +.21 8th 5th
T.N. 3.66 3.73 +.07 6th 6th
G.H. 4.07 3.84 -.23 3rd 7th

An analysis of the posttest sociometric ratings indicated

that five of the seven moderately preferred children gained in

mean ratings at posttest. Three of the children gained in their

median rank among same-sex classmates, three children remained

at the same median rank, and one child dropped in her median rank.

Six of the seven moderately preferred children met the definition

of being a moderately preferred child at posttest (the third

through eighth highest children of the same sex in a classroom),

while one child, M.Z., was the highest rated child among same sex

peers in his classroom at posttest.

In summary, the analysis of the pre- and post-sociogram data

on the three groups of children indicated that all of the games











analysis trained socially isolated children gained in their

median rank among same-sex classmates. These gains ranged from

one to seven median rank positions with an average gain of 3.5

positions. Of the socially isolated children who did not receive

training, one child gained one rank position, three children

dropped one rank position, and one child retained the same rank

position. The effect of the treatment on the moderately preferred

children was ambiguous. Three of the children gained in their

rank positions, three of the children retained the same rank

position, and one child dropped in her rank position. No child

in the moderately preferred group gained or loss more than four

positions.
















CHAPTER V

DISCUSSION


Children who are socially isolated from their peers have

become the subject of increased clinical concern in recent

years (Strain, Cooke, & Apolloni, 1976); however, procedures

for modifying social isolation have not been generally success-

ful (Lilly, 1971; Rucker & Vincenzo, 1970). Evidence of a

relationship between social isolation and deficient play and

game skills has appeared in child development literature

(Ausbel, 1958; Whitman, 1970). If the games analysis treatment

were proven to be an effective treatment, there would be

important educational, as well as psychological, implications.

For these reasons this study was designed to examine the efficacy

of the games analysis treatment in promoting positive changes

in isolated children's peer acceptance and social adjustment.



Discussion and Implications


Discussions of three major areas are presented in the

following sections. The first area contains a presentation of

a summary discussion of the findings of the study and an inter-

pretation of those findings. The second major area is a











discussion of the problem and limitations that were encountered

during the course of the study. The third area includes a dis-

cussion of the implications of the study for the practitioner.


Findings

The analysis of the data that were collected relative to

the five experimental questions of the study resulted in five

major findings. The five experimental questions are listed

below with a subsequent presentation of the data relevant to

each question and the conclusions those data allow to be drawn.

1. Will the games analysis intervention increase the

peer acceptance of socially isolated children?

The results on the analysis of the "play with" sociometric

ratings at posttest indicated that each of the seven isolated

children made positive gains in peer acceptance. The positive

gains achieved ranged +.14 to +.86 with the group's mean gain

being +.54. More importantly, each of the seven isolated child-

ren achieved sociometric gains in their median ranks among

same-sex classmates. The median rank gains achieved ranged

from a gain of one rank to a gain of seven ranks with the group

averaging again of 3.5 rank positions. As a result of these

gains only one of the isolated children at posttest met the

definition of being a socially isolated child (the two least

accepted children of each sex in a classroom).

All five socially isolated children who did not receive

games analysis training remained socially isolated children at











the sociometric posttest. Four of these children received the

lowest median rank among their same-sex classmates, while the

other child received the second lowest median rank among his

same-sex classmates.

Given the above data, it may be concluded that the games

analysis intervention increased the peer acceptance of the

socially isolated children in this study. The positive changes

in sociometric mean ratings and median ranks for the socially

isolated children who received games analysis training suggest

the treatment's effectiveness.

2. Will the games analysis intervention increase the

task participation of socially isolated children?

The application and withdrawal of the games analysis treat-

ment resulted in a controlling effect on task participation for

four of the five isolated subjects. Its effect on the fifth

subject was not reliable due to the subject's high frequency of

task participation throughout both baseline and intervention

phases.

For C.F. the frequencies of task participation were 20%

and 31% higher in Bl and B2 than in the immediately preceding

non-treatment phases. D.O.'s rates of task participation

were 41% and 20% higher in B1 and B2 than in the immediately

preceding non-treatment phases. K.K.'s frequencies of task

participation increased 41% and 75% in B1 and B2 when compared

to the immediately preceding non-treatment phases. For S.S. the











frequencies of task participation were 18% and 22% higher in B1

and B2 than in the immediately preceding Al and A2 phases.

For M.S. the application and withdrawal of the treatment's

effect on task participation was not clearly demonstrated due

to M.S.'s high frequency of task participation during the non-

treatment phases. Institution of treatment resulted in task

participation increasing 2% and 3% in B1 and B2 when compared

to the immediately preceding baseline phases.

Hence, the conclusion may be drawn that the games analysis

intervention is, indeed, effective in increasing the task partici-

pation of socially isolated children in this study. The positive

changes in the frequencies of task participation for four of the

five subjects are evidence of the reliability of the treatment's

effect.

3. Will the games analysis intervention increase the

peer oriented/supportive behavior of socially isolated children?

The application and withdrawal of the games analysis treat-

ment resulted in a controlling effect on supportive behaviors

for all five of the isolated children. C.F.'s frequencies of

supportive behavior were 36% and 26% higher in B1 and B2 than in

the immediately preceding non-treatment phases. D.O.'s fre-

quencies of supportive behavior were 40% and 18% higher in Bl

and B2 than in the immediately preceding non-treatment phases.

K.K.'s frequencies of supportive behavior were 58% and 94% higher

in Bl and B2 than in immediately preceding non-treatment phases.

S.S.'s frequencies of supportive behavior were 41% and 22% higher











in B1 and B2 than in the immediately preceding baseline phases.

Finally, M.S.'s frequencies of supportive behaviors were 4% and

14% higher in B1 and B2 than in the preceding non-treatment

phases.

Hence, it may be concluded that the games analysis inter-

vention is, indeed, effective in increasing the supportive

behavior of socially isolated children in this study. The

positive changes in the frequencies of supportive behavior for

all five subjects are evidence of the reliability of the treat-

ment's effect.

4. Will the games analysis intervention decrease the

uncommunicative/ignoring behavior of socially isolated children?

The implementation and removal of the games analysis treat-

ment resulted in a controlling effect on uncommunicative behavior

for four of the five isolated subjects. Its effect on the fifth

subject was not demonstrated due to the subject not emitting any

uncommunicative behavior during both sets of A and B phases.

For C.F. frequencies of uncommunicative behavior were 75%

and 92% lower in B1 and B2 than in the immediately preceding

nontreatment phases. D.O.'s emission of uncommunicative behavior

was 87% and 170% lower in B1 and B2 than in the immediately pre-

ceding baseline phases. K.K.'s emission of uncommunicative

behavior was decreased by implementation of treatment. Her

frequencies of uncommunicative behavior were 560% and 910% lower

in B1 and B2 than in the immediately preceding non-treatment











phases. For S.S. frequencies of uncommunicative behavior were

92% and 325% lower in B1 and B2 than in the immediately preceding

nontreatment phases.

No interpretation of the games analysis intervention's effect

on M.S.'s frequency of uncommunicative behavior could be made due

to his low emission of uncommunicative behavior. Only one un-

communicative behavior was recorded during the four phases of

the study.

Hence, the conclusion may be drawn that the games analysis

intervention is, indeed, effective in decreasing the uncommunica-

tive behavior of socially isolated children in this study. The

changes in the frequencies of uncommunicative behavior are

evidence of the reliability of the treatment's effect.

5. Will the games analysis intervention decrease the un-

cooperative/rejecting behavior of socially isolated children?

The application and removal of the games analysis treatment

clearly led to a controlling effect on rejecting behavior for four

of the five isolated subjects. Its effect on the fifth subject

was not demonstrated due to the subject not emitting any rejecting

behavior throughout the four phases of the study.

For C.F. frequencies of rejecting behavior decreased 240%

and 150% in B1 and B2 when compared to the immediately preceding

non-treatment phases. D.O.'s frequencies of rejecting behavior

were 300% and 100% lower in B1 and B2 than in the immediately

preceding non-treatment phases. K.K.'s frequencies of rejecting











behavior decreased 120% and 66% during B1 and B2 when compared to

the immediately preceding non-treatment phases. M.S.'s frequencies

of rejecting behavior were 97% and 200% lower in B1 and B2 than

in the immediately preceding baseline phases. No interpretation

of the treatment's effect on S.S.'s rate of rejecting behavior

was possible as she did not exhibit any rejecting behavior

throughout the four phases of the study.

Hence, it may be concluded that the games analysis treatment

is, indeed, effective in decreasing the rejecting behavior of

isolated children in this study. The positive changes in the

frequencies of rejecting behavior suggest reliable treatment

effects for four of the five subjects.


Interpretation of Findings

These data allow several interpretations of the findings to

be made. First, the presence of a functional relationship between

institution of the games analysis treatment and an increase in

pro-social behaviors and a decrease in negative social behaviors

was established for the subjects of this study. The magnitude

of the changes in the dependent measures upon institution of the

games analysis treatment is evidence for the acceptance of this

conclusion. In addition to the magnitude of change, the consistency

of change provides additional evidence with which to support the

existence of a functional relationship. In essence, this study

provides ten replications of the experimental condition, i.e.,

designing and adapting the games the children played via the games