Marathon group counseling with imprisoned female drug abusers

MISSING IMAGE

Material Information

Title:
Marathon group counseling with imprisoned female drug abusers
Physical Description:
viii, 137 leaves : ; 28 cm.
Language:
English
Creator:
Page, Richard Collin, 1943-
Publication Date:

Subjects

Subjects / Keywords:
Drug addicts -- Rehabilitation -- United States   ( lcsh )
Women prisoners -- United States   ( lcsh )
Group counseling   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Includes bibliographical references (leaves 127-135).
Statement of Responsibility:
by Richard Collin Page.
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 - 000175827
notis - AAU2304
oclc - 03043777
System ID:
AA00003928:00001

Full Text















MARATHON GROUP COUNSELING
WITH IMPRISONED FEMALE DRUG ABUSERS





BY

RICHARD COLLIN PAGE


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


1976













ACKNOWLEDGMENTS


The author wishes to acknowledge several persons who have contributed

to the inception, development, and completion of this study:

Robert Myrick, chairperson of the doctoral committee, whose ideas,

questions, and encouragement were indispensable;

Jerald Bozarth, committee member, whose advice and support expanded

the scope of this project;

Paul Fitzgerald, committee member, who graciously consented to serve

during the final quarter of this study when much was demanded;

E.L. Tolbert, former committee member, whose other responsibilities

made his replacement on the committee necessary, but whose contributions

were still substantial;

Bruce Thomason, formerly Professor of Rehabilitation Counseling and

committee member, a teacher, friend, and mentor for many years.














TABLE OF CONTENTS
PAGE

ACKNOWLEDGMENTS .................................... .......... ii

LIST OF TABLES ................................................. v

ABSTRACT ....................................................... vi

CHAPTER I INTRODUCTION ............................. ........ 1

Need for Study ................................. 1
Purpose of the Study ........................... 2
Research Questions ............................. 3
Definitions .................................... 4
Summary of Chapter Contents .................... 5

CHAPTER II REVIEW OF RELATED LITERATURE ..................... 6

Reasons for Drug Abuse ......................... 6
Personal and Social Characteristics of Heroin
Abusers........................................ 8
Causes of Heroin Addiction ................... 8
Research on Personality Characteristics of
Heroin Users ................................ 11
Personal and Social Characteristics of Female
Heroin Addicts .............................. 14
Personal and Social Characteristics of Other
Drug Abusers .................................. 15
Amphetamine Abusers .......................... 16
Barbiturate Abusers ......................... 17
Psychedelic Drug Abusers ..................... 19
Cannabis Abusers .............................. 20
Treatment of Drug Abusers ...................... 22
Marathon Group Theory ......................... 26
Research on Marathon Groups .................... 29
Marathon Group Counseling with Drug Abusers .... 36
Marathon Group Counseling with Imprisoned
Drug Abusers ................ ................. 37
Summary of the Literature ...................... 39

CHAPTER III RESEARCH METHODOLOGY ............................ 41

Population ..................................... 41
Sample ....................................... 42
Assignment of Subjects ...................... 43








PAGE

Research Design ................................ 43
Hypotheses ..................................... 44
Instruments .................................... 52
FIRO-B Scales ................................ 53
Semantic Differential ........................ 56
Marathon Group Questionnaire ................. 57
Analysis of the Data ........................... 59
Marathon Group Strategy ........................ 60
Experimental Procedures ........................ 61

CHAPTER IV RESULTS .......................................... 64

CHAPTER V SUMMARY AND RECOMMENDATIONS ...................... 86

Summary ........................................ 86
Conclusions .................................... 87
Limitations .................................... 90
Recommendations ................................ 92

APPENDIX A ................................................... 96

APPENDIX B ..................................................... 103

APPENDIX C ..................................................... 105

APPENDIX D ..................................................... 108

APPENDIX E ..................................................... 110

APPENDIX F ..................................................... 112

APPENDIX G ..................................................... 126

REFERENCES ..................................................... 127

BIOGRAPHICAL SKETCH ........................... ............... 136














LIST OF TABLES


TABLE PAGE

1 Results of the Analysis of Variance Procedures Related
to the Main Effects of Treatment vs. Control for the
First Set of Posttest Data .............................. 66

2 Comparisons of the Means of Experimental and Control
Groups for the First Set of Posttest Data ............... 68

3 Results of the Analysis of Variance Procedures Related
to the Main Effects of Treatment vs. Control for the
Second Set of Posttest Data ............................. 69

4 Comparisons of the Means of Experimental and Control
Groups for the Second Set of Posttest Data .............. 71








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


MARATHON GROUP COUNSELING WITH
IMPRISONED FEMALE DRUG ABUSERS

By

Richard Collin Page

August, 1976


Chairman: Robert D. Myrick
Major Department: Counselor Education


The purpose of this research was to assess the impact of marathon

group counseling on changing selected attitudes of incarcerated female

drug abusers. The population studied was female inmates of a Southeastern

correctional institution who regularly abused illicit drugs before being

imprisoned. Most of these females were imprisoned because they committed

crimes (sales of drugs, armed robbery, grand larceny) to support their

drug habits. The effects of treatment on inmates who had been addicted to

heroin and on inmates who had abused drugs other than heroin (amphetamines,

barbiturates, or psychedelics) were also assessed.

Fifty-six inmates were selected randomly from the population of female

drug abusers at the correctional institution to participate in a control

or a marathon counseling group. Two pairs of group leaders were assigned

randomly to the two experimental groups.

The group leaders conducted the two marathon groups on the same day.

These groups lasted for 16 hours and were unstructured marathons in that

few exercises were used. A Marathon Group Strategy was developed to define

the type of treatment the marathon leaders would use. One goal of the

leaders was to help drug offenders develop mutually enhancing ways of








relating to peers and authority figures. A second goal was to help

group members find solutions to personal problems which contributed to

their drug abuse.

Three instruments were used to test the hypotheses of this study:

the FIRO-B, a semantic differential, and a Marathon Group Questionnaire.

The Marathon Group Questionnaire was developed specifically for use in

this research. The attitudes assessed by the different instruments related

to the feelings of the research participants toward interpersonal

relationships, counseling, self, others, drugs, the past and future, and

authority.

The research participants were administered two sets of posttests.

The first posttest administration was the day following the marathons and

the second was 4 weeks later. The results were analyzed by an analysis

of variance procedure.

There were only significant differences (j < .05) between the

control and experimental group members on three of the 29 scales of the

instruments used in this research. The marathon group participants

scored significantly lower (J < .05) than the control group participants

on the wanted affection scale of the FIRO-B on the second posttest

administration, and lower on the potency scale of The Future on a semantic

differential on the first and second posttest administrations. The

marathon group participants scored significantly higher (j < .05) on the

first posttest administration of the Confidence in Personal Involvements

Scale of the Marathon Group Questionnaire. There were no additional

significant differences (j < .05) when heroin users in the experimental

and control groups and non-heroin users in the experimental and control

groups were compared.








The results of this research were inconclusive. The differences

found between control and experimental group subjects were not numerous

enough to demonstrate the treatment was successful in changing the

attitudes of the group participants in a comprehensive manner. The

results indicated a need for further research.


viii














CHAPTER I

INTRODUCTION


Drug abuse is a major problem in American society. Large numbers of

young people experiment with drugs each year and many become addicted.

Some of these persons are apprehended by the police for violations related

to their drug abuse. Of those that are placed in jail and brought to

trial, many receive prison sentences.

Drug abusers placed in prison often learn other ways to commit crimes

because of their association with other inmates. Unfortunately, incar-

cerated drug offenders receive little help in prison to eliminate their

drug dependence. Outside the prison environment, these persons frequently

resume drug use or have further problems obeying the law. Rehabilitation

programs are needed to help offenders refrain from drug abuse once they

leave prison.

This study was designed to assess marathon group counseling as a

treatment for facilitating positive attitude and behavior changes in

incarcerated drug offenders.


Need for Study

How successful have programs been which attempted to rehabilitate

drug abusers? What impact does individual and group counseling have on

helping the addict function without drugs in society? The answers to

these questions are complex and vary. At least one expert wrote that

counseling is an ineffective method of helping addicts refrain from drug




2



abuse (Brecher, 1972). Other experts have maintained counseling can help

drug abusers develop lifestyles which are drug free (Einstein & Garitano,

1972). The degree of success of rehabilitation programs for drug users

is a debatable issue.

Rehabilitation programs in prisons have been criticized more severely

than other drug programs (Brecher, 1972; Yablonsky, 1965). One of the

problems of institutional programs is that inmates are often resistant to

change because they feel they are forced to participate. Moreover, they

tend to support one another in maintaining criminal codes of conduct

(Yablonsky, 1965). The staffs of these programs often have negative

concepts of drug abusers which prevent them from supporting rehabilitation

rather than punishment as a means of working with prisoners. Brecher (1972)

concluded that programs, such as the California Rehabilitation Center,

have very low success rates when the criterion is the number of inmates

who refrain from using drugs for the remainder of their lives.

It is probably unfair to assess the treatment outcomes of drug

rehabilitation programs solely on the basis of whether participants ever

again use drugs. Actually, there have been few attempts to determine the

kinds of attitude or behavior changes produced in imprisoned drug offenders

by different forms of treatment. Many studies (e.g., Ross, McReynolds,&

Berzins, 1974; Steinfeld, 1970) failed to utilize experimental research

designs or had other methodological flaws. These is a need for research

which can help practitioners assess their therapeutic methods.

Purpose of the Study

Although group therapy or counseling has been used with imprisoned

drug offenders, very little research has been conducted on the effects of

these groups (Hendricks, 1971; Yablonsky, 1965). Some research has








been reported on the use of marathon groups to help incarcerated addicts

develop attitudes which will enable them to remain drug free outside of

prison (Kilmann, 1974; Kilmann & Averbach, 1974; Kruschke & Stoller,

1967; Ross et al., 1974). None of these studies however, had experimental

designs and some assessed the outcomes of therapy by questionable self-

report or staff ratings (Kruschke & Stoller, 1967; Ross et al., 1974).

Thus, the effects of marathon group treatment on the attitudes and

behavior of incarcerated drug offenders are largely undetermined.

The purpose of this study was to determine the impact of group

counseling on the attitudes of incarcerated drug users. More specifically,

the effects of marathon group counseling on attitudes related to self,

others, drugs, the past and future, authority, and counseling were assessed.

An attempt was also made to determine how marathon group participation

affects members' perceptions of relationships with others and their degree

of confidence in becoming involved in personal relationships.

Research Questions

The following research questions were investigated.

1. What effect does participation in a marathon group have on

a subject's attitudes toward authority or such representatives of

authority as (a) police officers, (b) prison, (c) prison staff,

(d) counseling, (e) counselors?

2. What effect does a marathon group have on a participant's degree

of positive identification with (a) men, (b) women, and (c) others

who use drugs?

3. Is there a relationship between marathon group participation and

an inmate's level of confidence (a) in forming positive human

relationships, (b) in maintaining positive human relationships,








(c) in accepting the past and past mistakes, (d) in facing the

future realistically?

4. Does participation in a marathon group contribute to an inmate's

increased level of positive self-regard?

5. What effect, if any, does participation in a marathon group have

on an inmate's attitudes toward drugs?

6. What differences are there in the ways heroin and non-heroin

users respond to marathon group treatment?

7. If attitude changes occur, do these changes produced among the

participants of a marathon group remain stable?

Some important terms are defined as they were used in this study.

Definitions

1. Marathon group. A marathon group is a relatively unstructured

counseling group that meets for 16 continuous hours and is led by

two leaders. The leaders focus on helping members develop positive,

mutually enhancing ways of relating to one another.

2. Incarcerated drug abuser. Incarcerated drug abusers are inmates

of a correctional institution who used illicit drugs before coming

to prison and who were sentenced to prison either for possession of

drugs or for attempting to obtain money for drugs by armed robbery,

grand larceny, or sales of drugs.

3. Heroin abuser. Heroin abusers are incarcerated drug abusers who,

on a form provided to them by a drug counselor, identified their

most significant drug problem as heroin.

4. Non-heroin abuser. Non-heroin abusers are incarcerated drug

abusers who, on a form provided to them by a drug counselor,








identified their most significant drug problem as something other

than heroin, such as hallucinogenics, barbiturates, or amphetamines.

Summary of Chapter Contents

A review of the professional literature is provided in Chapter II.

This literature review includes a description of the personality charac-

teristics of heroin and non-heroin abusers and a review of drug treatment

programs. Marathon group theory, research on the use of marathon groups

with drug free populations, and a description of the ways marathon groups

have been used with drug abusers, specifically drug abusers who were inmates

of correctional institutions, are discussed.

The experimental procedures for the study are described in Chapter III,

including descriptive information on the population, sampling procedures,

experimental design, research hypotheses, instruments, the marathon

group treatment, and an outline of the steps of the study. The results

of the study are in Chapter IV. The last chapter, Chapter V, contains a

brief summary of the study and the author's conclusions and recommendations.













CHAPTER II

REVIEW OF RELATED LITERATURE


The review of the literature related to this investigation is

divided into these broad areas: (a) reasons for drug abuse, (b) personal

and social characteristics of heroin abusers, (c) personal and social

characteristics of other drug abusers, (d) treatment of drug abusers,

(e) marathon group theory, (f) research on marathon groups, (g) marathon

group counseling with drug abusers, (h) marathon group counseling with

imprisoned drug abusers, and (i) summary of the literature.


Reasons for Drug Abuse

Various investigators provide different answers regarding the reasons

people abuse different kinds of drugs. The explanations cited often

appear to depend more on the biases of a particular investigator than

on anything else.

Some experts believe certain types of people have problems which

cause them to turn to drugs as an escape. Other investigators think the

breakdown of values in modern society causes persons to turn to drugs

to find some meaning in life. Perhaps both ideas have merit.

For example, some investigators associate the increased abuse of

drugs in modern society with the problems of living in a highly mobile,

materialistic culture (Canadian Government Commission of Inquiry, 1970;

Dohner, 1972b). The members of the Canadian Government Commission of

Inquiry felt young persons abuse drugs because of their lack of ambition








and because they lose faith in reason, turning instead toward more

hedonistic lifestyles. The collapse of traditional religious values

and urban ugliness were also viewed as contributing to Canada's drug

problem.

Dohner (1972b) associated peer pressure, an attempt to gain psy-

chological support, and the tenseness of young people as an explanation

for the increasing abuse of drugs by young Americans. He felt the

general dissatisfaction of young persons with modern life caused young

Americans to abuse drugs.

At least one writer maintained that many people use drugs illegally

because human beings have an innate drive to alter normal consciousness

states (Weil, 1972). Weil (1972) considered the urge to use drugs to

produce pleasant sensations as analogous to the sex or hunger drives of

people. He hypothesized people use drugs because they become more

intuitive, are better able to accept ambivalence, and experience infinity

through drug use. According to Weil, the only way to prevent drug abuse

is to help people learn natural ways to achieve consciousness states

similar to drug induced consciousness states.

Even though some writers have suggested general theories for drug

abuse, others considered it more accurate to discuss the motivations of

abusers of certain groups of drugs such as heroin, barbiturates, psyche-

delics, and cannabis (Braucht, Brakarsh, Follingstad,& Berry, 1973;

Canadian Commission, 1970; Crowley, 1972). These various types of drugs

alter feeling states differently. Abusers may be attracted to a

particular type of drug because they gain relief from personal problems

or anxieties. Crowley (1972) maintained persons use drugs to gain relief

from stress associated with unresolved problems or inadequate personal








relationships. The personal and social characteristics of the abusers

of different types of drugs are described extensively in the literature.


Personal and Social Characteristics of Heroin Abusers

Heroin addicts have been described as being the most difficult group

of drug abusers to rehabilitate (Brecher, 1972; Lindesmith, 1966).

Brecher (1972) wrote that heroin changes the body chemistry of the abuser

in such a way that most addicts can never remain drug free in society.

He asserted the only available cure for heroin addiction is for the

addict to substitute the lifelong use of methadone for heroin. Page and

Myrick (1975), using a semtantic differential with 85 drug abusers, found

incarcerated heroin addicts viewed drugs as being significantly (p < .05)

more potent than did incarcerated abusers of other drugs. These results

were interpreted as showing that drugs have more power over the lives of

imprisoned heroin users than the lives of imprisoned non-heroin users.

What makes persons with heroin problems such a difficult group to

rehabilitate? What makes the appeal to drugs so intense to heroin addicts?


Causes of Heroin Addiction

An attempt can be made to answer those questions by examining the

reasons heroin addicts feel compelled to use narcotic drugs. One popular

theory relating to heroin addiction emphasizes certain predisposing

personality characteristics toward addiction (Chein, 1966; Eldridge, 1963;

Isbell, 1966; Laurie, 1971; Winich, 1963). Winich (1963) emphasized one

of the principle tenants of this position when he stated, "The addict is

responding to underlying personality problems of great complexity (p.49)."

According to some theorists, addicts have personality characteristics

which attract them to heroin as an escape from reality. These theorists







think addicts were either born with certain types of character weakness

or developed character deficiencies early in life.

Winich (1963), for instance, explained heroin addiction from the

perspective of Freudian psychology. He maintained heroin addicts often

have had poor heterosexual relationships, unresolved Oedipal conflicts,

or passive oral dependent personalities before they became addicted.

These unresolved personal conflicts caused these persons to seek an

escape by using heroin. Winich further suggested addicts also learned to

use heroin as a replacement for sex by substituting drug induced feeling

states for their sex drives. Other writers maintain addicts have character

disorders or neurotic problems which predetermined their use of heroin

(Chein, 1966; Eldridge, 1963; Isbell, 1966). Laurie (1971) stated many

addicts use heroin because they had schizophrenic personalities before

they became addicted. These persons chose to use heroin as a means of

medicating themselves.

Heroin addicts have been described negatively by many writers.

Lichtenstein (1966) called heroin addicts liars and mental and moral

degenerates. He advocated hard work as the only treatment for heroin

addiction. Mills (1966) labeled heroin addicts as habitually dirty,

childishly immature, narcissistic, and irresponsible. Lichtenstein and

Mills offered no explanation for why addicts become addicts.

There were others who disagreed with these negative descriptions of

the addict's personality (Brecher, 1972; Canadian Commission, 1970;

Fort, 1966; Grinspoon & Hedblom, 1975; Lindesmith, 1966; Ray, 1972;

Schur, 1963). Fort (1966) agreed addicts often have neurotic character

traits but he felt addicts become this way as a result of their addiction

to heroin. He described drug-free addicts at the U.S. Public Health







Service Hospital in Lexington, Kentucky, as being introverted, friendly,

sensitive, and quiet. Others have gone even further in disagreeing with

the character weakness theory of addiction.

One group of theorists denied that heroin addicts have any abnormal

personality characteristics which differentiate them as a group from normal

persons. The proponents of this theory stated heroin users begin using

drugs for pleasure and continue using these drugs because of their fear of

the pain associated with withdrawal from heroin (Lindesmith, 1966;

Schur, 1963). Lindesmith (1966) supported this theory by pointing out

many addicts deny ever experiencing euphoric feeling states from heroin

use. He also stated the fear of the addict of the dreadful pain of

withdrawing from heroin causes addicts to keep using the drug.

Brecher (1972) advocated a modified version of the Lindesmith theory

of addiction. Brecher stated addicts stay addicted to heroin because

they experience anxiety, depression, and craving for the drug--the heroin

abstinence syndrome--when they don't use heroin. According to Brecher,

this syndrome probably continues to operate years after the addict has not

used drugs, causing heroin addiction to be incurable.

Grinspoon and Hedblom (1975), however, disagreed with the Lindesmith

theory of heroin addiction. They asserted the unpleasantness of the

heroin abstinence syndrome was greatly overestimated by Lindesmith and

the other advocates of his theory of addiction. Grinspoon and Hedblom

stated addicts probably display drug-seeking behaviors because they seek

to re-experience the euphoric feelings associated with their first heroin

trip, or because they want to gain a sense of identity by being a part

of the addict subculture. Persons thus become addicted to heroin in part

because they enjoy the excitement of the addict's lifestyle.








Other theories about the causes of heroin addiction have been advo-

cated. Eddy (1963) stated addicts are spiritually sick persons who cannot

find meaning in life. Still others suggested the laws in the United States,

which make addiction a crime, victimize the addict by forcing him to

illegally support his habit (Finestone, 1966; Lindesmith, 1963b). Finestone

and Lindesmith maintain the laws in America operate to drive up the price

of drugs which forces the otherwise passive addict to resort to self-

centered, antisocial, and criminal behavior to support his drug habit.

Others have maintained heroin addiction is caused by children being

reared either in an unsupportive slum family, or by the children of slums

associating with drug-using peers (Feldman, 1970; Lewis & Glaser, 1974;

Seldin, 1972). Feldman (1970) stated many slum dwellers become addicted

because the tension relaxing properties of heroin provide relief from an

otherwise violent and depressing lifestyle. Seldin (1972) stated heroin

addiction is more common among slum children reared in unsupportive homes

than among children reared in supportive homes. He wrote that children

with absent fathers and emotionally ill mothers often become heroin addicts

when they reach adulthood.

Some writers have also written there are multiple causes of heroin

addiction, which means it is fruitless to look for a single theory to

explain in a global way the reasons everyone addicted to heroin became

addicted (Canadian Commission, 1970; Ray, 1972). These theorists do

not feel all persons addicted to heroin necessarily became addicted for the

same reason.


Research on Personality Characteristics of Heroin Users

Thus,numerous theories of why people become addicted to heroin have

been proposed. Many of the theories presented so far were not formed








from the results of research; rather, they evolved from the professional

opinions and descriptions of experts who had worked to rehabilitate

heroin addicts.

Several studies have been conducted using test data as an attempt

to determine the personality characteristics of drug abusers (Berzins &

Ross, 1973; Chambers, 1971; Gendreau & Gendreau, 1973). Chambers (1971),

for example, administered the 16 PF personality test to heroin addicts

certified for treatment by the 1971 New York State Narcotic Control

Commission. He found 20% of these addicts were above average in neurotic

maladjustment, and that 29.3% were above average in their anxiety levels

as measured by the 16 PF. Data was also obtained with the 16 PF which

showed 32.7% were above average in irresponsible acting out behavior, and

30% were below average in effective behavior controls. Chambers inter-

preted this data as showing most of these addicts, being in the normal

ranges of this test, had normal personalities.

Using prisoners as subjects, Gendreau and Gendreau (1973) compared

the MMPI profiles of eight heroin addicts with nine non-heroin addict

volunteers for treatment. They also compared the profiles of 13 addicts

with 20 non-addicts who did not volunteer for treatment. No significant

differences were found in the types of profiles of these groups.

Berzins and Ross (1973) compared 600 opiate users at the Lexington

Clinical Research Center with 800 students at the University of Kentucky

on the dimensions of internal-external control as measured by Rotter's

Internal-External Locus of Control Scales. The addict group exceeded

the student group in levels of internal control at the .025 level of

significance. Since most researchers using this instrument consider it a

treatment goal to help subjects become more internally controlled, the


I








subjects of this study had scores which most researchers would probably

consider to be characteristic of adequately functioning persons. These

three studies (Berzins & Ross, 1973; Chambers, 1971; Gendreau & Gendreau,

1973) supported the hypothesis that heroin addicts have normal personalities

as measured by psychological tests.

There have also been studies related to the personality characteristics

of black and white heroin addicts (Chambers, Moffett,& Jones, 1968;

Miller, Sensenig, Stocker,& Campbell, 1973; Page & Myrick, 1975), and

male and female addicts (Miller et al., 1973; Page & Myrick, 1975). Most

of the research showed black and white and male and female heroin abusers

have major differences in their social backgrounds and personal attitudes.

Page and Myrick (1975), however, found black and white and male and female

addicts have similar attitudes. They utilized a semantic differential to

assess the attitudes of 85 heroin addicts at the Florida Correctional

Institution on 11 psychological concepts, (Women, Men, School, My

Vocational Future, Parents, Drugs I Took, Others Who Use Drugs, Past,

Future, As I See Me, and As I Would Like To Be). They found black and

white and male and female addicts to be homogeneous in their attitudes

related to these concepts.

Two studies attempted to assess the value patterns of black addicts.

Miller et al. (1973) used Rokeach's values ranking instrument along with

traditional admissions instruments to determine if the values of 284 nar-

cotic addicts admitted to the N.I.M.H. Clinical Research Center were sex

or race related. Black addicts preferred the values of a comfortable

life, equality, ambition, and intellect more than white addicts. White

addicts rated interpersonal and intrapersonal relationships higher than

black addicts. Also, Chambers et al. (1968) found 806 blacks admitted








to the Public Health Service Hospital at Lexington, Kentucky, were more

likely than non-addict black groups to have been reared in a broken home,

to have been a school dropout, and were less often legally employed. Many

of these characteristics of black addicts appear to be representative of

the lifestyles of blacks reared in slum environments.

The personal and social characteristics of male and female addicts

have also been compared. Miller et al. (1973) found male addicts valued

being ambitious, intellectual, logical, and self-controlled while female

addicts valued being clean, forgiving, happy, and having inner harmony.

Black female addicts were reared in broken homes and had intact marriages

more often than black male addicts. The findings of Miller et al.

suggest male and female addicts do not have a common set of values

emerging from drug use but instead have different, sex-related values.


Personal and Social Characteristics of Female Heroin Addicts

Since this research will be conducted with female subjects, it is

important to review in more detail the personal and social characteristics

of female heroin addicts. Female addicts have been described as being

seriously maladjusted persons who have average or higher than average

intelligence (Cuskey, Premkumar,& Sigel, 1974). These females often fail

to take proper care of their children (Cuskey et al., 1974). Yablonsky

(1965) emphasized female addicts often engage in prostitution to support

their drug habits which causes them to have considerable guilt and

confused attitudes toward men. Additionally, Yablonsky asserted female

heroin abusers are deeply sensitive persons but confused and in need of

counseling.

Raynes, Climent, Patch,and Ervin (1974) compared personal and

social information on 46 female narcotic addicts in the Correctional








Institution for Women in Framingham, Massachusetts, with the charac-

teristics of 46 voluntary female admissions at the Drug Detoxification

Unit at Boston City Hospital. They found few differences between the

two groups in terms of age, race, or religion. The prisoners had a

higher prevalence of psychiatric illnesses and used marijuana at an

earlier age than the hospital group. On the whole the two groups were

homogeneous in their social and personal characteristics.

Chambers, Hinesley, and Moldestad (1970) compared the characteristics

of 168 black and white addicts admitted between June and December, 1965

to the U.S. Public Health Service Hospital in Lexington, Kentucky. They

found black female addicts, when compared to white female addicts, were

more likely to have engaged in illegal activities, including prostitution,

and were less likely to have volunteered for treatment. Additionally,

black females more often had been reared in broken homes and were more

likely to have been labeled as having sociopathic personality disorders.

Chambers et al. (1970) found important social differences exist between

black and white female narcotics addicts.

It is thus very difficult to state with any degree of confidence the

reasons people become addicted to heroin. Writers have presented many

different versionsof what causes a person to become addicted to the

narcotic drugs. Most of the research conducted with heroin addicts

failed to support the idea that these drug abusers have personality

characteristics which predispose them toward addiction. The rehabilitation

of heroin addicts is a difficult task.


Personal and Social Characteristics of Other Drug Abusers

It is difficult to discuss in a comprehensive way the personal and

social characteristics of all drug abusers other than heroin abusers.








One reason is that the amphetamines, barbiturates, psychedelics, and

cannabis have different pharmacological effects on the people using them.

Thus, the same person may use different drugs to achieve different

consciousness states. Also, these different types of drugs vary greatly

in their addictive properties. For these reasons the personal and social

characteristics of persons abusing amphetamines, barbiturates, psychedelics,

and cannabis will be discussed separately.


Amphetamine Abusers

Amphetamine abusers have been described as being social outcasts in

the drug culture and they are among the most difficult of all types of

drug abusers to rehabilitate '(Canadian Commission, 1970; Grinspoon &

Hedblom, 1975; Kramer, Vitezslav, & Littlefield, 1967). Laurie (1971),

on the other hand, stated that the use of amphetamines among many groups

of people is a rather harmless habit. Thus, some writers describe speed

as having few ill effects while others describe speed as having disastrous

effects on its abusers. Then, what are the effects of speed abuse? What

types of personal and social characteristics do speed abusers have?

The speed abuser (amphetamine abuser) often acts in rather bizarre

ways when he uses high doses of speed. Grinspoon and Hedblom (1975)

stated libidinal energy is increased among many users and fatigue is

masked so that performance is improved on simple tasks. Most users

display paranoid symptoms, which often include auditory hallucinations,

when they use speed over a prolonged period of time (Brecher, 1972;

Grinspoon & Hedblom, 1975; Kramer, 1970; Ray, 1972; Robbins, 1970). Many

speed abusers also become violent while they are using speed; therefore,

speed users frequently become involved in drug-related murders. Persons

abusing speed commonly suffer from sleep deprivation because they often


I .








stay awake for days when using speed (Brecher, 1972; Grinspoon & Hedblom,

1975; Kramer, 1970). Additionally, Grinspoon and Hedblom (1975)

asserted the blood pressure of speed addicts rises causing many health

problems, including possible brain damage.

The amphetamines have severe effects on the emotional and physical

health of addicts who use them intravenously (Grinspoon & Hedblom, 1975).

Not all persons using amphetamines, however, become addicted to these

drugs or use them intravenously. Robbins (1970) classified speed abusers into

three types: the confirmed speed freaks, the general dope freaks, and

the occasional users. He maintained speed abuse, even when of a short

duration, has lasting effects on the personality.

Most imprisoned speed abusers have serious drug problems which

contributed to their being sentenced to prison. Confirmed speed abusers

have been described as suspicious, self-centered, lacking ambition,

hyperactive, and lacking an adequate self-concept (Canadian Commission,

1970; Carey & Mandel, 1968; Smith, 1970). These addicts are often

considered outcasts even among other groups of drug abusers because of

their compulsive and self-defeating behavior. They are frequently involved

in petty crime. Many speed abusers live in communes where violence is

common. Brecher (1972) postulated the severe depression many speed

abusers experience when they are not using speed contributes to their high

relapse rate. He also emphasized the difficulty of rehabilitating these

addicts.


Barbiturate Abusers

Literature on the personal and social characteristics of barbiturate

addicts is almost unavailable. One reason for the paucity of such

articles may be that barbiturate addicts have many of the same personal








and social characteristics as alcoholics (Brecher, 1972). There are

numerous articles on the personality characteristics of alcoholics.

Another reason for a lack of information may be that many barbiturate

addicts abuse more than one drug, using barbiturates and alcohol or

barbiturates and speed simultaneously (Brecher, 1972). Only a relatively

small number of persons use barbiturates intravenously because of the

severe physiological and psychological effects of these drugs.

The symptoms of barbiturate abuse have been associated with appre-

hension, muscular weakness, tremors, repeated vomiting, twitches,

delirium tremens, panic states, and brain damage. Death can result from

effects associated with the withdrawal syndrome (Chambers, Brill, &

Inciardi, 1972). The abuse of barbiturates at high doses also produces

marked behavioral changes in its abusers such as confusion, irritability,

fighting, and an unclean physical appearance (Essig, 1970). Barbiturate

abuse has also been linked to the increased violence and high number of

attempted or successful suicides among its abusers. Barbiturate use

often contributes to an addict's initial depressed feeling state (Brecher,

1972; Canadian Commission, 1970; Chambers et al., 1972). The continual

abuse of barbiturates thus has some pronounced physiological, behavioral,

and psychological effects on its abusers.

Many persons experiment with barbiturates without becoming addicted

to them. Wesson and Smith (1972) identified three different types of

barbiturate abuse as chronic intoxication, episodic intoxication, and

intravenous barbiturate use. Since incarcerated barbiturate addicts used

these drugs frequently before imprisonment, the personal and social

characteristics of barbiturate abusers seriously addicted to these drugs

will be described.








Many barbiturate addicts have been described as being middle aged

and coming from middle class backgrounds. These addicts often obtain

their drugs from legal sources, such as doctors (Chambers et al., 1972;

Laurie, 1971). Whitlock (1970) maintained barbiturate abusers usually

have a very disturbed and almost non-existent sex life. Many addicts show

lifelong patterns of personality disorders. Most addicts are isolated

and lonely individuals which has contributed to their proneness to physical

injury and abuse (Whitlock, 1970). Several writers indicated that many

addicts attempt suicide, or are severely depressed persons (Brecher, 1972;

Canadian Commission, 1970; Laurie, 1971; Whitlock, 1970). Many barbiturate

addicts have severe medical problems such as weight loss and brain or

liver damage (Brecher, 1972; Canadian Commission, 1970; Chambers et al.,

1972). A higher percentage of women, primarily housewives, than men also

appear to abuse the barbiturates (Cooperstock, 1971).


Psychedelic Drug Abusers

The effects of psychedelic.drugs on the user are different in some

important ways from the effects of the opiates, amphetamines, and bar-

biturates. The person using hallucinogenic drugs does not become

physically addicted to these drugs, and there is no tendency to use these

drugs in increasing quantities in a pattern similar to heroin (Fort, 1974).

Most experienced users of drugs such as LSD tend to use these drugs in

decreasing amounts (Ray, 1972). For these and other reasons, the persons

using psychedelics often have personal and social characteristics

different from the abusers of other drugs.

Experts have asserted it is very difficult to predict how the

psychedelic drugs will affect an individual user (Brecher, 1972; Canadian

Commission, 1970). The effects of these drugs on the user appear to vary








with the personality of the user, the setting in which the drug is

used, the mood and emotional set of the user, the people present when

the drug is used, and the dosage used (Brecher, 1972; Canadian Commission,

1970; Fort, 1974; Ray, 1972). Some people have experienced what they

describe as highly pleasant feeling states while on LSD. Others have

had psychotic adverse reactions which included paranoid delusions and

hallucinations (Canadian Commission, 1970). For these reasons, it is

difficult to generalize about how LSD affects the feelings of individuals

using it.

Most LSD users probably do not take large quantities of this drug

over long periods of time (Brecher, 1972; Canadian Commission, 1970).

There are abusers, however, who do take large amounts of LSD regularly

for periods of years (Blacker, 1970; Cheek, Newell,& Sarett, 1970; Jones,

1973; Smith, 1970). Cheek et al. (1970) indicated such persons gain an

increased sense of intimacy by using LSD in groups. Smith (1970) main-

tained a psychedelic syndrome occurs among heavy users which includes a

belief in magic and mental telepathy. An overt passivity and repression

of anger also occurs among many users (Blacker, 1970; Laurie, 1971;

Smith, 1970). Jones (1973) asserted heavy users become present oriented

and often fail to have insight into their own motivations and behavior or

the motivations and behaviors of others. They also become alienated from

the traditional values in society, causing them to drop out of society.

Some abusers also appear to have elevated MMPI scores on scales designed

to measure depression, paranoia, and schizophrenia (Kendall & Pittel, 1971).


Cannabis Abusers

Unlike the literature on the psychedelics, at this time there are a

number of books and articles relating to marijuana and to the personal







and social characteristics of marijuana users (Becher, 1944; Grinspoon, 1971;

National Commission on Marihuana and Drug Abuse, 1972). The effects of

marijuana are unlike the effects of most other drugs, although the use of

high doses of marijuana may have certain perceptual effects which have

similarities to the perceptual effects produced by some of the psychedelics.

Persons do not become physically addicted to marijuana although a very small

percentage of users may use marijuana more than once daily for periods of

years (Grinspoon, 1971; National Commission on Marihuana and Drug Abuse, 1972).

The effects of marijuana on all types of users is still a debatable topic.

The reasons people use marijuana are probably determined by a number

of complex psychological and social factors (Grinspoon, 1971; National

Commission on Marihuana and Drug Abuse, 1972). Recreation (Becher, 1944)

or an escape from boredom (Schoenfeld, 1944; Wallace, 1944) have been

suggested as reasons many people use marijuana. It is probably more

meaningful to consider the motivations of different types of marijuana

users than to make general statements about the reasons people smoke

marijuana. Thus, several experts have grouped marijuana users by the

amount of marijuana they smoke (Becher, 1944; Grinspoon, 1971; National

Commission on Marihuana and Drug Abuse, 1972). Grinspoon (1971) discussed

the personal and social characteristics of occasional users, social users,

and pot heads. The characteristics of pot heads or heavy users will be

discussed as this type of abuser is more frequently imprisoned.

The parents of heavy users of marijuana have been characterized as

smoking marijuana or drinking alcohol with a greater frequency than most

parents (Hockman & Brill, 1973). Marijuana use has been described as

the most important activity in the heavy marijuana abuser's life

(Grinspoon, 1971; National Commission on Marihuana and Drug Abuse, 1972;








Ray, 1972). Heavy users of marijuana tend to be irresponsible, non-

conforming, often resistant to authority, and tend to lose interest in

everything except smoking marijuana. Pot heads often display paranoid-

like attitudes and rebel against the establishment (Grinspoon, 1971;

National Commission on Marihuana and Drug Abuse, 1972). Grinspoon (1971)

stated heavy users often become involved in illegal activities.

Much conflicting information has been published about marijuana abuse.

Kolansky and Moore (1972) wrote that marijuana use may result in structural

changes in cerebral cells in the brain. They saw 13 adults between the

ages of 20 and 41, all of whom had smoked cannabis from 3-10 times per

week for 13 months to six years. They noted a clinical syndrome in all

of these subjects which included disturbed awareness of self, apathy,

confusion, and poor reality testing. Therefore, these clinicians

postulated marijuana use may result in permanent brain damage. These

conclusions, however, were questionable because these doctors based their

opinions on only a few subjects. The conclusions have not been supported

by other research investigations (Becher, 1944; Canadian Commission, 1970;

Grinspoon, 1971; National Commission on Marihuana and Drug Abuse, 1972).

Treatment of Drug Abusers

Although writers have disagreed on the causes of drug abuse, most

agree the rehabilitation of persons with serious drug problems is a very

difficult task. Different kinds of treatment programs have been attempted

with varying degrees of success.

A number of investigators have attempted to identify the components

of a successful treatment program with drug addicts (Brecher, 1972;

Dohner, 1972a; Einstein & Garitano, 1972; Warner, 1973). Warner (1973)

maintained drug rehabilitation programs should focus on changing the








behavior and attitudes of persons with drug problems rather than

attempting to provide information to these persons. Einstein and

Garitano (1972) stressed that persons involved with the rehabilitation

of drug abusers should communicate to their clients the belief that

addiction can be cured. Others have stressed professional or peer

counselors can help addicts find alternatives to drug use (Brecher, 1972;

Dohner, 1972a). Brecher (1972) suggested sensitivity training, encounter

therapy, Yoga, and Transcendental Meditation may help addicts achieve

alternate states of consciousness in a drug free manner.

Meditation has been suggested as an especially suitable alternative

to drugs for drug addicts. Meditation tends to provide natural highs

which lead to personal growth or personal insights. This can help addicts

learn to relax and cope with tension (Campbell, 1974; Payne, 1973; Weil,

1972).

Other writers advocate individual counseling, claiming it can help

addicts resolve personal problems through an expression of repressed

feelings of anger and guilt (Jaffe, 1973; Myers, Myers, Tapp,& Tapp, 1972).

Ketai (1973) asserted a special counseling method, peer observed psycho-

therapy, is effective with drug abusers. When this approach is used, a

professional therapist first works individually with one client while

other clients watch what he does. Once the therapist has finished, he

encourages the observers to provide feedback to the client.

Other writers have stressed traditional group psychotherapy as a

method to help addicts overcome drug problems. For example, Fort (1955)

and Thorpe and Smith (1955) maintained that at least 10 sessions of

regular group therapy helps addicts discover the reasons they used drugs







in the past and, subsequently, decreases the likelihood that these addicts

will abuse drugs in the future.

The federal government has supported many different types of

rehabilitation programs for drug abusers. The Task Force on Federal

Heroin Addiction Programs (1974) evaluated the degree of success of

several different types of programs using federal funds. This Task Force

rated detoxification programs and daycare treatment programs as being

rather ineffective as a means of helping addicts reduce the degree to

which they were addicted to heroin. Residential treatment centers, such

as Synanon, were rated as an effective means of helping addicts remain

drug free once they returned to society from these programs. The Task

Force on Federal Heroin Addiction Programs also advocated methadone

maintenance programs be used as a temporary way to help addicts withdraw

from heroin. The members opposed programs which encouraged addicts to

use methadone indefinitely. This Task Force rated drug counseling and

drug education programs for drug addicts in prison settings as ineffectual.

Lindesmith (1963a) and Brecher (1972) also criticized the number and

quality of drug rehabilitation programs in the penal systems in the

United States. Lindesmith (1963a) asserted drug abusers were punished

in prisons rather than rehabilitated. Brecher (1972) stated drug

rehabilitation programs in prisons failed to rehabilitate drug addicts.

There appears to be a dearth of research on treatment programs for

inmates in correctional settings. Evaluation of the existing programs

in correctional settings is difficult because of the meager number of

articles in the professional literature on these programs. The available

literature indicates that a few programs conducted by professional staff

or by ex-addicts have been developed.







Professional therapists have used covert sensitization to help

addicts learn to refrain from drug use. Covert sensitization is a

behavior modification technique in which a therapist presents an adverse

stimulus while a client is engaging in a behavior the therapist and client

wish to extinguish. For example, therapists used covert sensitization

with imprisoned heroin addicts by presenting undesirable stimuli, such

as rats crawling over a dead body, while addicts imagined a scene

associated with the abuse of heroin (Steinfeld, 1970; Steinfeld, Rautio,

Rice,& Egan, 1974a).

Hendricks (1971) used family counseling in groups to help addicts

at the California Rehabilitation Center develop positive family rela-

tionships. Family members came to the California Rehabilitation Center to

participate in these groups. Hendricks claimed this treatment method

helped to rehabilitate heroin addicts although he presented no data to

support this claim.

Ex-addicts and ex-prisoners have also conducted rehabilitation

programs in prisons to rehabilitate drug addicts. Synanon programs have

been started in several prisons by former heroin abusers. These programs

were based on the program of Synanon House in California where addicts

help one another resolve drug problems in groups without the aid of

professional therapists. The proponents of Synanon in prisons demand

their incarcerated members act responsibly by cooperating with prison

staff and by giving up prisoner codes of behavior. Members failing to

change their behavior are confronted. Those members failing to conform

are asked to leave the program.. Yablonsky (1965) reported that a Synanon

program was so successful at the Nevada State Prison that prison officials

granted the program leaders an honor camp for 20 participants.







Marathon Group Theory

It is apparent that more research is needed to determine the types

of programs which will produce positive attitudinal and behavioral

changes among incarcerated drug offenders. A few writers have maintained

that marathon groups offer an effective way of helping imprisoned drug

offenders to rehabilitate themselves (Kruschke & Stoller, 1967; Yablonsky,

1965). Before discussing the ways marathons can help incarcerated drug

offenders, the actual theory and operation of marathons will be reviewed.

Mintz (1971) wrote that marathon groups can be divided into two

major types, structured marathons and unstructured marathons. Schutz

(1967b, 1971, 1973) wrote about structured group sessions which he called

encounter groups. The leaders of encounter groups use various group

exercises to facilitate interactions among group members. Persons joining

these groups agree to follow certain principles of encounter, including

open and honest expression, concentrating on feelings, staying in the

here and now, assuming responsibility for one's own actions, attempting

to express feelings by using the body, and others. The encounter group

leader structures the encounter groups by using exercises and the principles

of encounter to help people openly express their feelings to other members.

Schutz (1971) advocated other principles of groups which have had a

major impact on the sensitivity group movement. He wrote that people

have three interpersonal need areas which are inclusion, control, and

affection. Encounter groups progress through the inclusion, control,

and affection stages of group interaction. The FIRO-B was designed by

Schutz (1967a) to assess these three interpersonal need areas. Addi-

tionally, he wrote members must work through emotional as well as

physical blocks in groups in order to resolve personal problems. He








stressed the importance of group leaders using such therapeutic techniques

as Kundalini Yoga, Rolfing, Bioenergetic exercises, Tai-Chi Chuan, and

others in encounter groups to help members work through physical and

emotional blocks.

Several proponents of marathons have stressed the therapeutic

potential of prolonged groups which have an absence of structured exer-

cises (Bach, 1966, 1967a, 1967b, 1967c; Coulson, 1970; Mintz, 1971;

Rogers, 1970; Stoller, 1968a, 1968b). Leaders in these groups generally

rely on the spontaneous interactions of members with one another to

provide the material for group discussion. Rogers is recognized as one

of the most influential of the advocates of prolonged groups, which are

at least 24 hour long group sessions. Coulson (1970) also consistently

stressed the therapeutic potential of Rogerian encounter groups.

Rogers (1970) and Coulson (1970) asserted their encounter groups

evolve in growth-enhancing directions for group members when the group

leader allows members to freely express themselves in a permissive group

setting. They stressed the group leader often expresses constructive

feedback directly to group members because the leader interacts with his

members as a real person rather than as a transference figure. Rogers

observed his encounter groups tend to begin with a milling around stage,

and then move toward a stage where the description of past feelings occurs.

He stressed the development of other stages which focus on the expression

of negative feelings, confrontations among members, the expression of

personal feelings of closeness among members, and others. He suggested

these group stages develop by a natural evolutionary process, rather than

because the leader forces their development.







Bach (1966) and Stoller (1968a, 1968b) conducted pioneer work on

marathons when they led groups and formulated one of the first compre-

hensive theories to account for the operation of marathons. Both of these

writers also conducted some of the first research on marathon group

outcomes (Bach, 1967a, 1967b, 1967c; Kruschke & Stoller, 1967). The

concepts Bach (1966) and Stoller (1968a, 1968b) formulated have influenced

subsequent practitioners in many settings. Mintz (1971) also contributed

to the literature by advocating a theory of the operation of marathons

similar to the theory of Bach and Stoller.

Bach (1966), Mintz (1971), and Stoller (1968a, 1968b) advocated the

use of unstructured marathons lasting from one to three days. During

these groups, the leader openly provides feedback to group members so that

other members will follow his example in providing feedback on the behavior

of group members. The leader and other members also are expected to

listen carefully when feedback is given to them. Members cannot escape

the consequences of their behavior during the group because of the length

of the group. Members receive feedback which encourages these persons

to experiment with new, more appropriate ways of relating to people. The

group members learn to accept responsibility for their own actions.

These persons also learn to communicate honestly and openly with others,

and to express anger as well as feelings of love and concern to others.

Stoller (1968b) indicated marathon groups differ from time-limited

groups in a number of ways. He maintained members become highly involved

with each other during marathons; that tension mounts in these groups as

members risk exposing their feelings to others. Stoller asserted the

high degree of involvement of the members enables the participants of

marathons, more than the participants of many short groups, to learn how








to care for others. The time perspective of the marathon members is

altered causing time to move by quickly during these meetings. Stoller

maintained marathon group therapy differs from regular group therapy

because of the extended time format, and the unstructured nature of

marathons.

The stages of marathon groups have also been described by Stoller

(1968a, 1968b). He viewed marathons as progressing through these stages

by a natural evolution which occurs in all unstructured marathons. During

the first stage, the members relate in a formalized style. They tell their

stories to other members and receive feedback concerning how they affect

one another. Stage two involves a more personal type of sharing. In

this stage, members receive feedback on the openness and honesty with

which others feel they relate. The focus shifts from what is presented

to the manner of presentation. Behavior within the group becomes the

major concern as group members are helped to relate in an open, honest,

and caring manner. Stage three is one of urgency and elation. During

this stage, positive feelings are expressed as members learn how to care

for one another. The sense of urgency and elation come from the reali-

zation that mutual concern and care are possible.


Research on Marathon Groups

Many researchers have performed studies to determine if marathon

groups are effective in producing attitudinal and behavioral changes

among the members of different populations. Most of the research done

on marathon treatment with these different groups tends to support the

contention these groups have positive effects on the participants. Much

of the research performed on marathons has been plagued by experimental

design weaknesses and often by a lack of researcher objectivity








(Diamond & Shapiro, 1975; Kroeker, 1974; Marks, Conry,& Foster, 1973;

Treppa & Fricke, 1972). These weaknesses make it difficult to assess

the outcomes of research on marathon groups or to arrive at any conclusions

concerning the utility of this form of treatment with different populations.

Diamond and Shapiro (1975) criticized much of this research because

researchers failed to specify exactly what marathon group treatment

entailed. Other writers pointed out that many marathon group researchers

failed to use suitable instruments to assess outcomes, did not assign

their subjects randomly to control and experimental groups, and/or used

improper statistical techniques to analyze their data (Diamond & Shapiro,

1975; Marks et al., 1973). Marks et al., (1973) also asserted most mara-

thon group researchers used only one instrument rather than multiple

instruments to assess outcomes. Perhaps the most common criticism of the

designs used in most of these studies involved the failure of the re-

searchers to handle adequately pretest treatment interactions (Marks, et

al., 1973; Treppa & Fricke, 1972). Several writers advocated that

persons directing research projects on marathon groups should avoid the

use of protests in their research designs (Kroeker, 1974; Marks, et al.,

1973; Treppa & Fricke, 1972).

Lieberman, Yalom, and Miles (1973a, 1973b) conducted what many persons

consider to be the most comprehensive research investigation ever

performed on encounter groups, the Stanford project. These researchers

compared and contrasted the effectiveness of 10 group methods: National

Training Laboratory groups, Gestalt groups, Transactional Analytic groups,

Esalan Eclectic groups, National Training Laboratory groups--Western

style, Synanon groups, psychodrama groups, marathons in George Bach style,

psychoanalytically oriented groups,and encounter group (leaderless) tapes.







Experienced group leaders were used to conduct 17 groups which contained

a combined total of 137 Stanford University students. The students were

assigned randomly to treatment groups. A control group was also selected

consisting of 69 students who could not attend the encounter groups

because of schedule conflicts. Seventy-eight measures were used to

assess outcomes; results were combined to assess five broad areas of

change. Data was collected before the groups were held, two weeks after

the groups ended, and again six to eight months later.

The Stanford project produced some rather startling results. One

third of the participants in the treatment groups changed in positive ways,

one third changed in negative ways, and one third showed no change. No

particular type of group appeared to be more or less effective than any

of the others in producing changes in members. However, leaders who

moderately stimulated group members or who exerted a moderate amount of

executive behavior in their groups were more effective than leaders

without these behaviors. Perhaps the most startling section of the

Stanford project involved Lieberman et al. (1973a) finding that 9% of the

179 members of these groups completing training sustained what they

described as serious psychological damage six to eight months after

these groups were held.

The Lieberman, Yalom, and Miles study produced some results which

seriously challenge the effectiveness of encounter groups in helping

people. The reports that people were seriously hurt in these groups con-

tributed to skepticism about encounter groups. Recently Rowan (1975),

Schutz (1975), and Smith (1975) criticized the research design and degree

of objectivity of the researchers of the Stanford project. The criticisms

of these writers were multifaceted and wide ranging. Miles (1975) and








Lieberman (1975) responded to these critiques but failed to answer the

underlying arguments of Schutz, Rowan, and Smith.

Schutz (1975) gave several reasons why the comparisons made of the

outcomes of the different types of encounter groups in the Stanford

project were unjustified. The leaders in the various groups were not

all equally qualified to lead groups, and the lengths of the meetings were

different for different groups. Comparisons between the control groups

and experimental groups were unjustified because the members were not

assigned randomly to experimental and control groups. Both Schutz and

Rowan stated that all of the comparisons of the Stanford study were com-

promised because the members of the experimental groups were warned before

they entered treatment that they might incur psychological damage as a

result of group participation. Therefore, the members may have entered

these groups expecting to be hurt and this may have contributed to the

negative group outcomes.

The Stanford project has also been criticized for other reasons.

Rowan (1975) asserted the outcome measures used for this research were

too complex, making the entire project unwieldy. Others stated it was

inappropriate to use students as subjects when comparing the overall

utility of different group methods because of the psychological instability

of the student population. Schutz (1975), Smith (1975), and Rowan (1975)

also claimed the Stanford project defined encounter group casualties

in a sensational and prejudicial manner. It was contended that the fact

Lieberman et al. (1973a) assigned subjects to a casualty status on the

basis of interviews they conducted also contributed to the lack of ob-

jectivity of this project. The Stanford project offers an illustration

of some of the major difficulties involved in conducting sound research

on group therapy.








At least two major reviews of the research related to marathon

groups, or sensitivity groups, have been undertaken (Diamond & Shapiro,

1975; Gibb, 1971). Diamond and Shapiro reviewed the results of over

10 research projects on marathon groups conducted by students and

teachers of the University of Hawaii during a 42 month period. These

groups were conducted with many different populations including teachers,

prison inmates, correctional workers, police, students, nurses, and para-

professional workers. Diamond and Shapiro pointed out that many of the

faults of the Stanford project were eliminated by the research procedures

with these marathons. For instance, all of these research projects

contained group goals stated in operational terms. Everyone was pre- and

posttested and experimental research designs were used. The participants

in the great majority of these marathons showed positive behavior or

attitude changes which included improvements in self-concept, the de-

polarization of attitudes, and increased hypnotic suggestibility.

Gibb (1971) reviewed the research related to what he called human

relations training. He defined human relations training groups as groups

which focus on the here and now, on personal growth, on the available

interpersonal data, on promoting interactions between members, and on

encouraging members to try new behaviors in the group. Such groups are

conducted with normal group members rather than with emotionally disturbed

group members. Gibb stated that the research on these groups shows

human relations training can help members develop positive feelings toward

self, more positive perceptions of others, improved feeling management, and

enhanced directionality of motivation. He stated the available evidence

shows the dangers of sensitivity training have been vastly overrated.







Seven studies (Foulds, Guinan,& Warehime, 1974; Guinan & Foulds, 1970;

Jones & Medvene, 1975; Kimball & Gelso, 1974; King, Payne, & Mclntire,

1973; Weissman, Seldman, & Ritter, 1971; Young & Jacobson, 1970) studied

the effects of marathon groups on university students. All of this research

indicated marathon groups helped the students develop more positive

outlooks toward themselves and others. However, the results obtained in

these studies were weakened by some research design deficiencies. For

instance, in these seven studies only two sets of researchers assigned

subjects randomly to control and experimental groups (Jones & Medvene,

1975; Kimball & Gelso, 1974). Many of these researchers also failed to

define adequately the types of marathons used in their studies, making

replication of the studies impossible. It is therefore difficult to know

how to evaluate the results of this research.

In spite of the limitations, these studies provide some evidence

that marathon group treatment can bring about positive attitude changes

among student populations. Research has shown that students participating

in marathon groups changed more than members of control groups in the

following ways: increased levels of self-actualization (Guinan & Foulds,

1970; Kimball & Gelso, 1974; Young & Jacobson, 1970), increased self-

acceptance (Guinan & Foulds, 1970; King et al., 1973), gains in self-

actualization by participants with average or above-average ego strength

(Jones & Medvene, 1975), and increased self-understanding, self-confidence,

and increased understanding of others (Guinan & Foulds, 1970). Addi-

tionally, student marathon group participants changed more than the members

of control groups by achieving lower levels of dogmatism and a more open

belief system (Foulds et al., 1974), and by increasing in tendencies to

prefer complexity, novelty, and ambiguity (Weissman et al., 1971).








The effects of marathon groups on the attitudes and behavior patterns

of other populations have also been examined, although not as thoroughly

as with college students. Harrison'(1966) found marathon group treatment

produced an increased proportion of concepts dealing with feelings and

emotions in the speech patterns of 115 businessmen participating in NTL

groups. This research, however, failed to utilize a control group.

Posthuma and Posthuma (1973) compared the effects of a marathon

group, 10 three-hour sessions of a placebo group, and a control group.

Encounter group treatment produced more positive changes on the Edwards

Personal Preference Schedule and the Behavioral Change Index among church

members than did 10 three-hour sessions of a placebo group or a control

group composed of students from a community college. This research was

also characterized by design weaknesses, including the failure to assign

subjects randomly to the control group, the placebo group, and the marathon

group.

Chambers and Ficek (1970) studied the effects of marathon groups on

the attitudes of female juvenile delinquents in a residential training

school. They reported marathon group treatment increased the numbers of

positive entries in the diaries of the marathon participants. This

research had several design weaknesses, including the failure of the

researchers to assign subjects randomly to the control or marathon group.

The available research on marathon groups supports the contention

that marathon group treatment can produce positive personality changes

in a variety of populations. Most of this research, however, has been

compromised by design weaknesses.








Marathon Group Counseling with Drug Abusers

Marathon group counseling has been used successfully to help

individuals belonging to different populations develop interpersonally

and intrapersonally. There have been very few reports in the literature,

however, concerning the use of marathons with drug abusers. Most of what

has been reported has related to the use of day-long Synanon groups led

by the staff of drug houses (24-hour residential programs) for the residents

of their programs. Synanon groups have special characteristics which make

them different from the marathons led by Stoller, Bach, Schutz, or Rogers.

Many writers have described Synanon groups as being an unconven-

tional and radical approach to group treatment (Kramer, 1963; Ruitenbeek,

1970; Volkman & Cressey, 1966; Yablonsky, 1965). These groups stress

confrontation and their leaders use such methods as attack therapy, or

exaggerated statements or ridicule to help members overcome their drug

dependency. Leaders encourage members to confront other members to help

these members recognize their irresponsible and manipulative behavior

patterns (Ruitenbeek, 1970; Yablonsky, 1965). These groups are led by

former addicts rather than by professional group leaders. The advocates

of Synanon groups think drug addicts need to be resocialized. Yablonsky

(1965), for instance, stated drug addicts need to learn to conform to a

healthy group norm in order to stop using drugs. He also stressed that

the members of Synanon groups need to gain feelings of belonging and

respect from other group members in order to gain the strength to overcome

their drug dependence.

Densen-Gerber (1973), the director of Odyssey House, described one

or two day long marathons she led for the women residents of Odyssey House

(a 24-hour residential drug treatment program). The purpose of these








groups was to help the participants develop positive female identities.

Densen-Gerber was the primary therapist of these groups, although

different members of her staff served as co-therapists. She stressed

that group leaders were to interact with group members as real persons

rather than as transference figures. Densen-Gerber stipulated that the

members of her groups were to be completely honest with each other although

she also stressed the importance of offering hope and acceptance to

honest members. Many themes emerged during these groups which often

included the discussion of the members' feelings toward their children,

men, sex, parents, religion, and race. Densen-Gerber asserted the members

of these groups were often able to work through feelings of guilt by

talking with other group members. She supported her claim that marathon

groups have positive effects on their participants from her own experiences

as a group leader and not from the results of research.


Marathon Group Counseling with Imprisoned Drug Abusers

Only a few articles are available in professional journals relating

to the use of marathon groups with incarcerated drug offenders. Unlike

the leaders of marathons in drug house settings (24-hour residential

programs), the therapists of marathons with imprisoned drug addicts have

generally been the professional staff of universities or the staff of

prisons. Much of the research on this type of treatment in prison

settings has been undermined by improper instrumentation or the lack of

experimental or quasi-experimental research designs. This overall

unavailability of sound research in this area points out the need for

further studies to help determine whether marathons produce positive

counseling outcomes with imprisoned drug offenders.


-^ I








Marathon groups have been utilized as a treatment modality at the

California Rehabilitation Center, the prison to which drug offenders

are sentenced in California (Kilmann, 1974; Kilmann & Averbach, 1974;

Kruschke & Stoller, 1967). In one research project, Kilmann and Averbach

(1974) assigned 84 female narcotic addicts at this institution to two

non-directive and two directive marathon groups. A control group was

also used, although the authors said nothing about assigning subjects

randomly to experimental and control groups. The subjects receiving non-

directive therapy decreased in levels of A-Trait anxiety while those

persons receiving directive therapy increased in levels of A-Trait anxiety

(Kilmann & Averbach, 1974). The marathon group participants, to a

greater degree than the control group subjects (2 < .05), shifted toward

being more externally controlled rather than internally controlled on

Rotter's Locus of Control Scale. Kilmann (1974) interpreted the shift

toward greater external control for the experimental subjects as involving

a positive change for this population.

Kruschke and Stoller (1967) led a 20-hour marathon over a two day

period with 10 volunteer drug addicts incarcerated at the California

Rehabilitation Center. They identified group stages in which the members

played psychologist, used the hot seat approach, told personal things

about themselves, shifted from a discussion of past mistakes to emphasizing

future successes, and expressed feelings of personal closeness for each

other. Kruschke and Stoller maintained the group helped its members gain

an increased understanding of the needs of staff, show a more positive

outlook in their dormitories, and spend less time in prison than the

average inmate of the California Rehabilitation Center. Kruschke and

Stoller made no systematic attempt to assess the outcomes of their group








but instead were more concerned in providing their impressions of how

marathons operate with incarcerated drug abusers.

Ross, et al. (1974) compared the effectiveness of marathon groups

and conventional group therapy for, female narcotic addicts committed to

an NIMH facility under the Narcotic Rehabilitation Act of 1966. The

subjects in the marathon groups decreased in their tendency to view the

criminal subculture positively while the subjects in the traditional groups

increased in this same tendency. Both marathon and traditional group

treatment reduced the number of neurotic complaints of participants. But

this research study had definite weaknesses; six subjects were assigned in

a non-random manner to each of the treatment groups.


Summary of the Literature

The reasons persons chose to abuse drugs are complicated and are

probably determined by a combination of factors. The abusers of

different drugs, (i.e., opiates, amphetamines, barbiturates, psychedelics,

and cannabis) often have different personal motives for abusing drugs.

It is therefore more accurate to discuss the personal and social charac-

teristics of specific groups of drug abusers than to discuss the

characteristics of drug abusers as a whole.

Heroin abusers have been viewed as being among the most difficult

groups of drug addicts to rehabilitate. The abusers of the amphetamines

and the barbiturates also have major drug problems and are prone to use

drugs for very long periods of time. LSD and marijuana users do not

become addicted to these drugs and these users do not necessarily use

these drugs in increasing amounts. Therefore, the abusers of LSD and

marijuana often have quite different personal and social characteristics

than the abusers of opiates, amphetamines, and barbiturates.







There have been many types of programs designed to rehabilitate

drug abusers, including methadone maintenance programs, drug houses, and

group counseling. There has not, however, been much written in the pro-

fessional literature about programs or methods designed to rehabilitate

incarcerated drug abusers. Marathon groups have been used successfully

to help bring about positive attitude changes among student populations,

businessmen, and young females in adjustment centers. Marathon groups have

also been used as a treatment method in drug houses.

Since marathons have been used with reported success with other

populations, or with drug addicts in drug houses, this form of treatment

might also be used successfully to help incarcerated drug abusers. The

types of personal changes marathon group treatment can bring about among

incarcerated drug offenders has not been determined and this investigation

is designed to further explore this question.













CHAPTER III

RESEARCH METHODOLOGY


An experimental research design was used to assess the effects of

16-hour marathon groups in changing the attitudes of incarcerated drug

offenders. The subjects for this study were female drug addicts. A

marathon group strategy was developed by the drug counselor to function as

a treatment. The remainder of this chapter is devoted to a more detailed

explanation of the research methodology used in this study.


Population

The population was composed of female inmates of the Florida

Correctional Institution, Lowell, Florida, between March-April 1976, who

were serving sentences for drug or drug related problems and who, prior

to their imprisonment, regularly abused illicit drugs. These women came

from all areas of Florida. At least 80% were between the ages of 15 and

35. Of this population, about 60% were black and 40% were white.

Approximately 50% of these females were heroin addicts and approximately

50% abused other drugs (amphetamines, barbiturates, psychedelics, or

cannabis). Alcoholics were excluded from participation in this study.

Most of this population had been convicted of possession of illicit drugs,

sales of drugs, grand larceny, armed robbery, or shoplifting. All of

these persons indicated they had had serious drug problems, prior to their

imprisonment, which contributed in some way to their being incarcerated.

J








The population of the institution at the time of this study was

approximately 450 female inmates. About 150 of these inmates were

serving sentences for drug or drug related problems and had regularly

abused illegal drugs prior to their imprisonment. Most of the inmates

comprising the population used for this study had sentences ranging from

6 months to 10 years. All of these inmates were assigned to work areas,

educational or vocational programs in the institution, and all lived in

dormitory settings. Some of them had received some form of counseling

prior to their imprisonment, usually with drug counselors in jails, drug

houses, or methadone programs, or perhaps with other professional counselors,

psychologists, or psychiatrists. Some were also participating in

individual or group counseling at the time of this study. The prior

counseling experience of these inmates was a serious limitation to this

study but if these females were to serve as subjects for this research,

it was a limitation which could not be overcome.


Sample

At the time of this study approximately 106 inmates of the 150 inmates

had requested of the drug counselor that they be included in group coun-

seling. The drug abuse counselor interviewed 82 inmates randomly selected

from the list of 106 inmates who had requested to participate in groups

and identified 56 participants for this study. Twenty-six inmates declined

to participate in a marathon group even though they had previously

requested to be placed in drug groups. Some who refused to participate

expected visitors on the weekend when the marathon would be conducted and

were unwilling to give up seeing these visitors. All of the inmates who

were interviewed were asked to give up visitation privileges during the

time the groups met. Only volunteers were used. The experimenter








conducted all precounseling interviews, asking the subjects if they

wished to participate in the study. Anyone who wished to participate

was allowed to participate.


Assignment of Subjects

Using a table of random numbers, the subjects were assigned randomly

to one of two experimental groups or to a control group. Fourteen

subjects were assigned to each of the two experimental groups and 28

subjects were assigned to a control group.


Research Design

The Posttest Only Control Group Design (Campbell & Stanley, 1966)

was used to test the hypotheses of this study. Instead of using only one

set of posttests, however, all criterion instruments were delivered twice

after the marathon groups were completed. The first set of posttests was

delivered the day after the marathons were conducted. This set of tests

ascertained the attitudinal changes of the marathon participants directly

following the marathons. The second set of posttests was delivered to

all the experimental and control group participants 1 month (4 weeks)

following the first set of posttests. The second set of comparisons was

designed to test the stability of any changes that may have resulted from

the marathon group experience. A time period of 1 month was used between

the two sets of posttests because inmates continually arrive at and leave

the institution. The design used in this study may be diagrammed:

R X 01 02
R 01 02

R random selection
X experimental treatment
01 first posttest
02 second posttest







The Posttest Only Control Group Design has both strengths and

weaknesses. Campbell and Stanley (1966) stated that this design controls

for these factors which often jeopardize the internal validity of research:

history, maturation, testing, instrumentation, regression, selection,

mortality, and the interaction of selection and maturation. The Posttest

Only Control Group Design also controls for a factor often limiting the

external validity of research, the interaction of testing and X. This

design, however, does not always control for these factors related to the

external validity of research: the interaction of selection and the

experimental treatment, and reactive arrangements.

The Posttest Only Control Group Design controls for weaknesses

characterizing much of the past research on marathon group outcomes. Much

of the previous research on marathon groups inadequately controlled for

the confounding influences of testing and the treatment (Kroeker, 1974;

Marks, Conry, & Foster, 1973; Treppa & Fricke, 1972). When subjects of

marathons take protests, the short time period between protests and post-

tests means many subjects may remember how they responded on the protests

when they take the posttests, or the pretest may sensitize participants

to the types of changes the researcher wants to bring about by his

treatment. The Posttest Only Control Group Design eliminates protests

in order to control for these confounding influences.

Hypotheses

This research assessed 19 hypotheses related to the effects of

marathon group treatment on the interpersonal and intrapersonal attitudes

of incarcerated female drug abusers. The study assessed attitudes relating

to self, others, drugs, the past and the future, authority, and counseling.

The attitudes of heroin and non-heroin users were assessed. The following

null hypotheses were tested:








Ho:l There will be no significant differences between experimental

and control group subjects on the variable of expressed behavior in the

area of inclusion, as measured by the FIRO-B.

Sub Ho:la There will be no significant differences between

heroin users in the experimental and control groups on the variable

of expressed behavior in the area of inclusion, as measured by the

FIRO-B.

Sub Ho:lb There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of expressed behavior in the area of inclusion, as measured

by the FIRO-B.

Ho:2 There will be no significant differences between experimental

and control group subjects on the variable of wanted behavior in the

area of inclusion, as measured by the FIRO-B.

Sub Ho:2a There will be no significant differences between

heroin users in the experimental and control groups on the variable

of wanted behavior in the area of inclusion, as measured by the

FIRO-B.

Sub Ho:2b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of wanted behavior in. the area of inclusion, as measured

by the FIRO-B.

Ho:3 There will be no significant differences between experimental

and control group subjects on the variable of expressed behavior in the

area of control, as measured by the FIRO-B.

Sub Ho:3a There will be no significant differences between

heroin users in the experimental and control groups on the variable

of expressed behavior in the area of control, as measured by the FIRO-B.








Sub Ho:3b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of expressed behavior in the area of control, as measured

by the FIRO-B.

Ho:4 There will be no significant differences between experimental

and control group subjects on the variable of wanted behavior in the area

of control, as measured by the FIRO-B.

Sub Ho:4a There will be no significant differences between

heroin users in the experimental and control groups on the variable

of wanted behavior in the area of control, as measured by the FIRO-B.

Sub Ho:4b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of wanted behavior in the area of control, as measured by

the FIRO-B.

Ho:5 There will be no significant differences between experimental

and control group subjects on the variable of expressed behavior in the

area of affection, as measured by the FIRO-B.

Sub Ho:5a There will be no significant differences between

heroin users in the experimental and control groups on the variable of ex-

pressed behavior in the area of affection, as measured by the FIRO-B.

Sub Ho:5b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of expressed behavior in the area of affection, as measured

on the FIRO-B.

Ho:6 There will be no significant differences between experimental

and control group subjects on the variable of wanted behavior in the

area of affection, as measured by the FIRO-B.








Sub Ho:6a There will be no significant differences between

heroin users in the experimental and control groups on the variable

of wanted behavior in the area of affection, as measured by the

FIRO-B.

Sub Ho:6b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of wanted behavior in the area of affection, as measured

by the FIRO-B.

Ho:7 There will be no significant differences between experimental

and control group subjects on the variable Counseling, as measured by the

evaluative and potency scales of a semantic differential.

Sub Ho:7a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Counseling, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:7b There will be no significant differences between

non-heroin users in the experimental and control groups on the variable

Counseling, as measured by the evaluative and potency scales of a

semantic differential.

Ho:8 There will be no significant differences between experimental

and control group subjects on the variable Authority, as measured by the

evaluative and potency scales of a semantic differential.

Sub Ho:8a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Authority, as measured by the evaluative and potency scales of a

semantic differential.








Sub Ho:8b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Authority, as measured by the evaluative and potency scales

of a semantic differential.

Ho:9 There will be no significant differences between experimental

and control group subjects on the variable Drugs I Took, as measured by

the evaluative and potency scales of a semantic differential.

Sub Ho:9a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Drugs I Took, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:9b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Drugs I Took, as measured by the evaluative and potency

scales of a semantic differential.

Ho:10 There will be no significant differences between experimental

and control group subjects on the variable Others Who Use Drugs, as

measured by the evaluative and potency scales of a semantic differential.

Sub Ho:lOa There will be no significant differences between

heroin users in the experimental and control groups on the variable

Others Who Use Drugs, as measured by the evaluative and potency scales

of a semantic differential.

Sub Ho:lOb There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Others Who Use Drugs, as measured by the evaluative and

potency scales of a semantic differential.


V -








Ho:ll There will be no significant differences between experimental

and control group subjects on the variable Parents, as measured by the

evaluative and potency scales of a semantic differential.

Sub H4o:lla There will be no significant differences between

heroin users in the experimental and control groups on the variable

Parents, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:llb There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Parents, as measured by the evaluative and potency scales

of a semantic differential.

Ho:12 There will be no significant differences between experimental

and control group subjects on the variable Women, as measured by the

evaluative and potency scales of a semantic differential.

Sub Ho:12a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Women, as measured by the evaluative and potency scales of a semantic

differential.

Sub Ho:12b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Women, as measured by the evaluative and potency scales of

a semantic differential.

Ho:13 There will be no significant differences between experimental

and control group subjects on the variable Men, as measured by the

evaluative and potency scales of a semantic differential.

Sub Ho:13a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Men, as measured by the evaluative and potency scales of a semantic

differential.








Sub Ho:13b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Men, as measured by the evaluative and potency scales of

a semantic differential.

Ho:14 There will be no significant differences between experimental

and control group subjects on the variable The Future, as measured by the

evaluative and potency scales of a semantic differential.

Sub Ho:14a There will be no significant differences between

heroin users in the experimental and control groups on the variable

The Future, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:14b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable The Future, as measured by the evaluative and potency scales

of a semantic differential.

Ho:15 There will be no significant differences between experimental

and control group subjects on the variable The Past, as measured by the

evaluative and potency scales of a semantic differential.

Sub Ho:15a There will be no significant differences between

heroin users in the experimental and control groups on the variable

The Past, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:15b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable The Past, as measured by the evaluative and potency scales

of a semantic differential.








Ho:16 There will be no significant differences between experimental

and control group subjects on the variable My Real Self, as measured by

the evaluative and potency scales of a semantic differential.

Sub Ho:16a There will be no significant differences between

heroin users in the experimental and control groups on the variable

My Real Self, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:16b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable My Real Self, as measured by the evaluative and potency

scales of a semantic differential.

Ho:17 There will be no significant differences between experimental

and control group subjects on the variable Confidence in Personal In-

volvements, as measured by the Marathon Group Questionnaire.

Sub Ho:17a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Confidence in Personal Involvements, as measured by the Marathon

Group Questionnaire.

Sub Ho:17b There will be no significant differences between

non-heroin users in the experimental and control groups on the variable

Confidence in Personal Involvements, as measured by the Marathon

Group Questionnaire.

Ho:18 There will be no significant differences between experimental

and control group subjects on the variable Counseling Readiness, as

measured by the Marathon Group Questionnaire.

Sub Ho:18a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Counseling Readiness, as measured by the Marathon Group Questionnaire.








Sub Ho:18b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Counseling Readiness, as measured by the Marathon Group

Questionnaire.

Ho:19 There will be no significant differences between experimental

and control group subjects on the variable Self Confidence, as measured by

the Marathon Group Questionnaire.

Sub Ho:19a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Self Confidence, as measured by the Marathon Group Questionnaire.

Sub Ho:19b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Self Confidence, as measured by the Marathon Group

Questionnaire.


Instruments

Two sets of posttests were administered to all of the experimental

and control group subjects. The first set of posttests was given 1 day

after the marathon groups were held. The second set of posttests was

given to subjects still in the institution at the beginning of the 5th

week following the week of the first posttest administration. Fifteen of

the inmates who took the first set of tests were unavailable to take the

second set of tests because they had left the institution. Three instru-

ments were used at each of the two posttest administrations to assess

outcomes: (1) the FIRO-B, (2) a semantic differential, and (3) Marathon

Group Questionnaire. A general description of each of these instruments

follows.








FIRO-B Scales

The FIRO-B (Consulting Psychologists Press) was developed by

William Schutz to measure how individuals relate to other people. The

FIRO-B, a Guttman scale, assesses how subjects behave toward other persons,

rather than how they feel toward other persons. This instrument has

six scales, each of which measures either the behavior an individual

expresses toward others or the behavior he wants them to express toward

him. The instrument also measures one of three interpersonal dimensions,

the inclusion, affection, and control areas. Thus, six sets of scores

can be obtained from the FIRO-B including expressed behavior in the

control, affection, and inclusion areas and wanted behaviors in the

control, affection, and inclusion areas. These six scales of the FIRO-B
I A C I A C
are labeled the e e e w w and w scales.

Schutz (1967a, 1967b, 1971, 1973) defined the interpersonal dimensions

of inclusion, control, and affection in The FIRO Scales, the manual to

the FIRO tests, and in several books related to groups. In The FIRO Scales

he defined these interpersonal dimensions in the following way:

I. The interpersonal need for inclusion is the need
to establish and maintain a satisfactory relationship
with people with respect to interaction and association....
C. The interpersonal need for control is the need to
establish and maintain a satisfactory relationship with
people with respect to control and power. Control
behavior refers to the decision-making process between
people....
A. The interpersonal need for affection is the need to
establish and maintain a satisfactory relationship with
others with respect to love and affection.... (Schutz,
1967a, pp. 4-5)

The FIRO Scales also includes descriptions of what the six scales of the

FIRO-B measure:








e I make efforts to include other people in my
activities and to get them to include me in theirs....
eC I try to exert control and influence over things.
I take charge of things and tell other people what to
do.
e I make efforts to become close to people....
w I want people to include me in their activities
and to invite me to belong, even if I do not make
an effort to be included.
C
w I want others to control and influence me....
A I want others to express friendly and affectionate
feelings toward me and to try to become close to me.
(Schutz, 1967a, p. 5)

These six scales were used to test hypotheses one through six.

The FIRO-B scales have a moderately high reliability. As is

appropriate for a Guttman scale, the manual reported reliability statistics

on the reproducibility of the scales rather than the usual split-half

method. Approximately 1,500 college students and Air Force personnel

were used to compute the reproducibility of the different scales of the

FIRO-B. The e w1, eC, w, eA, and wA scales all showed .94 reproduc-

ibility. Approximately 120 Harvard University students were used to compute

the test-retest reliabilities for the different scales of the FIRO-B.

The test-retest reliabilities of the different scales of the FIRO-B
C I
ranged from .71 for the w scale to .82 for the e scale.

The intercorrelations of the different scales of the FIRO-B were

determined from a sample of 1,340 subjects. Schutz stated significant

correlations exist between the e and w scales for Inclusion and Affection,

and that smaller but statistically significant correlations exist between

the I and A scales. Schutz warned the FIRO-B contains non-independent

scales, although he recommended maintaining all of the scales.

The FIRO Scales contains information on the content validity and

concurrent validity of the FIRO-B. Schutz (1967a) stated that the content








validity of these scales is adequate because all legitimate cumulative

scales have adequate content validity. Research on the concurrent validity

of the FIRO-B has attempted to determine the relevance of the inclusion,

control, and affection scales for the possible psychiatric classification

of patients and to relate the dimensions of inclusion, control, and

affection to such areas as scientific creativity and differences among

occupational groups. The FIRO Scales presents much information showing

how members of different occupational groups scored on the different scales

of the FIRO-B. Schutz (1967a) maintained that this research indicated

striking differences exist between the ways the members of different

occupational groups respond on the different scales of the FIRO-B, and

that these differences are consistent with occupational stereotypes.

Bloxom (1972), a contributor to The Seventh Mental Measurements

Yearbook, reviewed the research related to the FIRO-B, and commented on

the general adequacy of the instrument. He rated all the scales of the

FIRO-B as being high in internal consistency, and stated that the test-

retest reliabilities of the different scales were adequate. Bloxom

maintained the non-independence of the different scales of the FIRO-B

should not prevent researchers from using all the scales. He suggested

that the validity of the FIRO-B suggests its scales are related to non-test

interpersonal behavior. He also stated these scale scores are highly

correlated with rated creativity, the diagnosis of schizophrenia, and other

factors. Bloxom stated of the FIRO-B, "Its subscales show a sufficient

degree of relationships to interpersonal behavior and to personality

measures to merit its use in research." (Bloxom, 1972, p. 170).








Semantic Differential

The semantic differential was also used to assess the attitude

changes of the participants of this research project. The semantic

differential was first developed by Osgood, Suci, and Tannenbaum (1957)

to measure the meaning of different concepts. They employed opposite

adjective pairs to assess how subjects regard a concept, person, or thing.

Isaac (1971) stated the semantic differential has three elements which

include a concept to be evaluated, a series of undefined scale positions,

and polar adjective pairs. Osgood et al. (1957) factor analyzed 76 pairs

of opposite adjective pairs to produce the principal factors, the activity,

evaluative, and potency scales. Any of the responses a subject makes when

rating an adjective pair can be grouped into an evaluative, potency, or

activity category. This instrument has been used extensively in many

settings and with many populations.

Page and Myrick (1975) used a semantic differential to assess the

counseling needs of 85 inmates of the Florida Correctional Institution

who were imprisoned for drug or drug related charges. The results obtained

by Page and Myrick influenced how concepts were selected for assessment by

the semantic differential in this research. Appendix B presents the mean

scores for the total population and for heroin and non-heroin users in

the evaluative, activity, and potency categories for each of the concepts

assessed by Page and Myrick. The higher numerical ratings in Appendix B

represent stronger evaluative, activity, and potency loadings. The

significant differences in the ways heroin and non-heroin users responded

to the concept Drug I Took and to the concept Others Who Use Drugs

suggested the use of a factorial design to compare the effects of marathon

group treatment on heroin and non-heroin users.








Marathon Group Questionnaire

The Marathon Group Questionnaire developed by this investigator

was the third instrument used in this study (Appendix C). This instrument

was developed to assess some of the types of behavior changes that Bach,

Rogers, and Stoller thought occurred among the participants of marathon

groups. Questions on this instrument were written to assess how respondents

feel about confrontation, the discussion of feelings, prison officials,

counselors and counseling, the development of interpersonal relationships,

trust of others, responsibility for self, self-assertion, drugs, and peers

in the drug culture.

In 1975, this writer administered this instrument to 107 male and

female inmates of the Florida Correctional Institution in order to perform

a factor analytic validity study. These respondents included drug offenders

as well as inmates imprisoned for other offenses, such as murder. The

responses of these inmates were factor analyzed with an Orthogonal Varimax

Rotation using the S.P.S.S. Package (Nie, Hull, Jenkins, Steinbrenner,

& Bent, 1975). The computer was programmed to force the appearance of

three factors.

Three principal factors were obtained from the results of the factor

analysis. Factor 1 accounted for 46.8% of the variance, Factor 2 for

37.1% of the variance, and Factor 3 for 16.1% of the total variance of

the factors on this questionnaire. A decision was made that questions

correlating .40 or higher with any of these three factors were to be used

to assess outcomes associated with this factor. If a particular question

correlated .40 or higher with more than one factor, this question was to

be used to assess outcomes relating to the factor with which the correlation

was highest. The Varimax Rotated Factor Matrix obtained from the








statistical printout of this factor analysis is presented as Appendix D.

The means and standard deviations obtained from the statistical analysis

of the responses of 106 inmates to each question are presented as

Appendix E.

It was found that the factors to be used for assessment in this

study each contained six to nine questions. The following questions were

used to measure the attitudes of respondents on each of the three factors:

Factor 1, Questions 7, 9, 10, 11, 13, 15, 29, 32; Factor 2, Questions 4, 5,

8, 14, 17, 24, 25, 38, 39; and Factor 3, Questions 3, 16, 33, 35, 37, and

40. These factors were named by attempting to describe what each of the

questions correlating .40 or higher with a factor measured in common.

The first factor was called Confidence in Personal Involvements, the

second factor was Counseling Readiness, and the third factor was Self

Confidence. These three factors of the Marathon Group Questionnaire were

used to test hypotheses 17-19 in this research.

The test-retest reliabilities of the Marathon Group Questionnaire

and its three factors were determined by delivering this instrument twice

with a 1 week lapse between testinqs to 26 male and female inmates of the

Florida Correctional Institution. The test-retest reliability for the

total score of the Marathon Group Questionnaire was .90. The test-retest

reliabilities of the different scales of this instrument were as follows:

.86 for the Confidence in Personal Involvements scale; .84 for the

Counseling Readiness scale; and .82 for the Self Confidence scale. The

test-retest reliabilities obtained for the total score and for the

different scales of the Marathon Group Questionnaire were very high.

Further reliability studies need to be performed with this questionnaire

to substantiate these findings. Even though this instrument showed








a high test-retest reliability, the scores obtained for individuals

on this test should be interpreted with caution because of the limited

number of reliability and validity studies which have been performed on

the Marathon Group Questionnaire.


Analysis of the Data

The scores of the subjects on the evaluative and potency scales of

the semantic differential for each concept and on the scales of the FIRO-B

and the Marathon Group Questionnaire were used for analysis. The semantic

differential for each concept was scored by assigning numbers from one

through seven for each adjective pair so that higher numerical ratings

represented stronger evaluative or potency loadings.

Each subject's FIRO-B was scored by key to obtain raw scores for the
I I C C A A
e w e w e and w scales. The range of possible scores for a

subject on each of these scales was from zero through nine, nine being the

highest score and zero the lowest score a subject could obtain.

Each subject taking the Marathon Group Questionnaire placed a

number from one through five beside each question. The scorer reversed

the numbers placed by each subject beside questions 3, 4, 5, 8, 14, 17, 24,

25, 33, 35, 38, and 39 so that one was scored as five, five was scored as

one and three remained the same for these questions. The numbers placed by

the subjects beside questions 7, 9, 10, 11, 13, 15, 16, 29, 32, 37, and 40

were scored as the subject recorded them. The scores on questions 7, 9, 10,

11, 13, 15, 29, and 32 were summed to obtain a total raw score for each

participant on the Confidence in Personal Involvements Scale. The scores

for questions 4, 5, 8, 14, 17, 24, 25, 38, and 39 were summed to obtain

a total raw score for each subject on the Counseling Readiness Scale and

the scores for questions 3, 16, 33, 35, 37, and 40 were summed to obtain a








total raw score for each subject on the Self Confidence Scale. The range of

possible scores for each individual on the Confidence in Personal

Involvements Scale was from 8-40, the range for the Counseling Readiness

Scale was 9-45, and the range for the Self Confidence Scale was 6-30..

The higher scores for these scales were associated with positive marathon

group outcomes and the lower scales were associated with negative outcomes.

Fifty-eight 2x2 analysis of variance procedures were used to assess

the hypotheses and sub-hypotheses of this study, 29 2x2 ANOVA's were per-

formed on data from the first posttests, and an equal number were performed

on data from the second posttests. The analysis of variance procedures

used for the two sets of posttests were as follows: The hypotheses and

sub-hypotheses pertaining to each scale of the Marathon Group Questionnaire

and the FIRO-B were each assessed by a 2x2 ANOVA. The hypotheses and

sub-hypotheses pertaining to each concept of the semantic differential

were each assessed by one 2x2 ANOVA for each of the two scales of the

semantic differential, the evaluative and potency scales. The first

variable for each 2x2 ANOVA represented group (control or experimental) and

the second variable represented drug of abuse (heroin or non-heroin use).

Marathon Group Strategy

A Marathon Group Strategy was developed to define the type of

treatment the group leaders conducting the marathons would use. The

leaders of these groups adhered as closely as possible to the process

outlined in this strategy. One goal of the leaders was to help drug

offenders develop mutually enhancing ways of relating to peers and

authority figures. A second goal was to help group members find solutions

to personal problems which contributed to their drug abuse.








The Marathon Group Strategy contains very few exercises in order to

enable the marathon groups to evolve in a spontaneous and natural manner.

The Marathon Group Strategy .includes descriptions of the ground rules,

member behaviors, leader behaviors, and group stages of marathon groups.

The ground rules section of this. strategy is very general and was designed

to help promote constructive interactions among group members. The

leader behaviors section includes a discussion of when group leaders

should use empathy responses and confrontation responses; how leaders

should respond to hostility, drug oriented discussions, and transference

reactions by group members; and how to identify group themes. The group

stages sections includes discussion of the following stages: relaxation,

hostility or projection, beginning of intimacy, beginning of feedback,

discussion of problems, provision of feedback, modifying behaviors,

elation or relaxation, and ending stage. This Marathon Group Strategy

is presented as Appendix F.


Experimental Procedures

The procedures of this investigation were carried out in the

following order:

1. Three marathon group leaders were trained by the researcher to

conduct marathon groups. The researcher served as the fourth leader

of the experimental groups. The education of the three trained

leaders was 1) an Ed.D. in counselor education, 2) an Ed.S. in

counselor education, and 3) a B.A. in history with post graduate

courses in counseling. The four prospective leaders were employees

of the Florida Correctional Institution and all at the time of this

study were or had been staff members of a therapeutic community for

drug abusers at the institution. Thus, all of these group leaders








had professional experience in working with incarcerated female drug

abusers prior to the time these groups were conducted.

2. Each of these prospective group leaders led at least one 16-hour

marathon with the investigator prior to conducting the research groups.

These pre-treatment groups were all conducted with incarcerated

female drug abusers at Florida Correctional Institution who were

not a part of the study.

3. Two other leaders were asked to serve as stand-by leaders in case

one of the trained leaders was unable to attend the treatment groups

due to sickness or other causes. Neither of the stand-by leaders

had led a marathon group with the investigator prior to the time the

treatment groups were conducted. All of the prospective leaders

and stand-by leaders read the Marathon Group Strategy and discussed

it before the date of the experimental groups.

4. Eligible subjects were assigned randomly to one of the two

experimental groups or to the control group. A table of random

numbers was used to assign subjects to groups.

5. The flip of a coin was the means of randomly assigning one pair

of group leaders to each of the experimental groups.

6. One of the trained group leaders (the leader with the Ed.D. degree)

was sick the day of the marathons. One of the substitute leaders

co-led a group with the investigator. This substitute leader had an

Ed.S. degree in counselor education and was also a staff member of

the institution where he was working as a drug counselor at the

therapeutic community within the institution.

7. The 16-hour marathons were conducted simultaneously at the

Florida Correctional Institution.








8. Three sets of tests, a semantic differential, the FIRO-B, and

the Marathon Group Questionnaire were delivered to all of the subjects

the day following the marathon treatment. The test instruments were

scored and the subjects' raw scores, corresponding to each of the

scales to be used in this research, were recorded.

9. The same instruments were delivered 4 weeks following the date of

the first testing to the subjects still remaining in the institution.

Forty-one subjects responded to the instruments at that time.

Fifteen subjects did not receive the second set of posttests because

they were no longer at the institution.

10. The second set of posttests was scored and each subject's raw

scores, corresponding to each of the scales of the instruments to

be used in this research, was recorded.

11. Approximately 1 month after the second set of posttests were

delivered, interested participants of the control group were given

the opportunity to participate in a marathon group.

12. Followup counseling was available to all of the research parti-

cipants.

With the exceptions noted the experimental procedures were carried

out in accordance with the research design. The results are presented in

the following chapter.














CHAPTER IV

RESULTS


The population used for this study was incarcerated female drug

offenders of the Florida Correctional Institution who, prior to their

imprisonment, had serious drug problems which contributed to their

imprisonment. All subjects in the control and experimental groups were

volunteers who had requested counseling. Subjects were placed randomly

in their respective groups through the use of a table of random numbers.

Group leaders were also randomly assigned, one pair to each of the two

counseling groups. The group leaders conducted the counseling groups on

the same day. Leaders followed a marathon group strategy developed by

the researcher.

Posttests were administered to the control group subjects and the

marathon group subjects. The instruments included a marathon group

questionnaire, the FIRO-B, and a semantic differential. All instruments

were delivered to the 56 control and experimental group subjects the day

following the marathons. These instruments were administered again to

those subjects remaining in the institution 4 weeks following the first

posttest administration.

Fifty-eight 2 x 2 ANOVA's were performed to assess the outcomes of

this study. The data obtained from each 2 x 2 ANOVA presented information

on the main effects of treatment vs. control, heroin vs. non-heroin users,

and the interaction effects of heroin and non-heroin users with the

experimental and control groups. This data was analyzed first by








determining the significance of any interaction effects and then by

determining whether the control and experimental group means differed

significantly (p < .05).

Only one of the 58 2 x 2 ANOVA's which was performed produced a

significant (2 < .05) interaction effect. On the first set of posttests,

the expressed inclusion (e ) scale of the FIRO-B produced an F ratio

which was significant (p < .05) for the interaction effects of the type of

treatment with the drug used. This signified that the scores of heroin

users in the control group might differ significantly (2 < .05) from the

scores of heroin users in the experimental group or that the scores of

non-heroin users in the control group might differ significantly from the

scores of non-heroin users in the control group for this scale. However,

when a Scheff6 multiple comparisons test was used to compare the scores of

heroin users in the experimental and control groups and non-heroin users in

the experimental and control groups, no significant differences (P < .05)

between means were obtained.

The F ratios for the first 28 ANOVA's (related to the first set of

posttests) comparing the main effect of the significance of the difference

between the means of the control group members and the means of the

marathon group members on the scales of all the instruments used for this

research are presented in Table 1. The mean scores of the experimental

and control groups for the scales on the first set of posttests are

presented in Table 2.

The F ratios for the second 29 ANOVA's (related to the second set of

posttests) comparing the main effects of the significance of the differences

between the means of the control group members and the means of the

marathon group members on these same scales are presented in Table 3.


















r-.d-:oI m mm Cj Cior- co co
M) LO C\i 910-f.o>mf) LO r-.
U- a) M C') '- -*- t.0 C) r-.
CO oOOOrO- COor--
oooooooooC


rO LCO OJ OMM COOO C') OO t
M C M C00LO O U '- OO

~oC ; (\; 6 C -


(Z
C 4-
.- t1



0 Q

S0)
4-J



Q) 4-
4- 0




o4-
wV)
- 0.-



OLL
0


S-
L)n -

.'- 0
s- s-












o o
0 0
( 4-






0 0


Co\ CMt r, CM r o k









U) LO U) U) Ln U) L) LO LO M
_ -: 1-. r--:-., -- O r _.
,,.O rm L m m ,,. --r-OO N'


U3-

0 0
0) 0



*U O
0)
0) 00a
4-) =3 4--

*r V0 4 (
> C 0 o 0 0
S c _- 0 0 0) 0)
:3 4- 3 3- )



( 4) -l' C3 S- E C M 0) 0
0 0 S- -l (0 0 0)-C >,
>?O oL3 i-0


Um M. O U) m m)


4-
0
E

O)S
00
0/) 0
51-0)

0) Li1
S-C21


0)30)30)3



















Co C\J m

k o oo
000 L


CJC 0 -00


C)C; C) C)


I' ko-


Coaoon
C)C)
4'<


LI-
A
-0
0
S-
L-








4- c
0 S
V)












4-
0
>













Ql-o
1)
4)
OS-a
r0
ED
a 0














a) Li.
0


0Zco
CM 'i 'Za-
M Lf. lC
* en ^i
N -


OJ C) In M CD C) Z1-

- CoD o M o C

LO C) O t- C D- C


LnInOM LnL MLn Lnm L
-O M LO M- -O MiO O


)
4-)
U








0a) S -
~0

-0)
a)





CL n 0 nS
*-- C7)



0 -') U)
+) 0





OWS- .. 4-))4-'
S0)D ) a
00% 0


4- ) .^ ..C S.

3 4 3.. -C-'SC
0 3 S-+- t
C_?cCC o 0


4-

S- CL
=3 4-)



0 w_ 0- >
-3C ) -F-
3 S


m C) C)
o;00C

*C
00


"m r-^ Lnm C) CDLO Mm"j
c00D C O- MO m
"-OI-CDOOOCM MO "Q
M-oo oC CJO O CDm-::t C


C)'- mz o
Go C) ko











in in LO
Lr)0OLf
1-:_:


V)
4-4

(a





0 0)
01









00
CQ-




C .-
--





- C
-a) 0


Sj 0-
C0 (0 0)
*r- 01 )


OO
C) C),-


oIt *









TABLE 2 Comparisons of
Control Groups


the Means of Experimental and
for the First Set of Posttest Data


Experimental
Group Means


e (expressed inclusion)
w (wanted inclusion)
ec (expressed control)
C
w (wanted control)
e (expressed affection)
w (wanted affection)

Semantic Differential


4.3571
2.9286
2.2857
2.7143
3.2500
3.2500


Control Group
Means


3.8214
2.7857
2.8214
2.8929
3.5357
3.2857


Evaluative Scales
Counseling
Authority
Drugs I Took
Others Who Use Drugs
Parents
Women
Men
The Future
The Past
My Real Self

Potency Scales (6
Counseling
Authority
Drugs I Took
Others Who Use Drugs
Parents


(6 adjective pairs)










adjective pairs)


Women
Men
The Future
The Past
My Real Self

Marathon Group Questionnaire


CIPI Scale
CRS Scale
SC Scale


FIRO-B


34.9643
24.4643
24.2857
20.9643
32.5357
31.0357
28.0357
36.1786
26.3571
35.6429


26.7143
31.0357
29.7500
27.2143
25.4643
22.1429
26.0357
24.0714
25.5714
22.9643


32.3929
23.0714
22.8571
19.0714
31.9286
28.1071
27.5714
36.3214
25.3571
33.3214


26.1071
30.1429
28.2143
26.5000
25.4643
23.2143
26.3929
27.3214
24.5357
22.7500


25.6071
33.3929
23.8214


28.1071
34.5714
24.7857




















con J ,-fC-M L ,' CD0,

OOOOOOOOO


'- '# J O


" ,J c-C) o k0 mO r-f-- m

r- o cm Id-,:to',-0 t om w
,- o cO 0 CD \ ,-- M ,- o









SCD ,on M r- kooi













C .C, 'C) C ) i C) C)-C)


-c

0
0
4-)







U)
o0





.-
0


o o
>- -


U 4-)



4-

) 0
oS-
L- i-
O0


4-U 4-)
4-


















C) to0


0
4-












U)


D) L
C m


S4-

0)


C) 00 0) C)
- o-Co S- LI)


(A 0 (A S- = Co U-0- (-)
0) a) a)
U)OOQ-cEC L-OC)
(-) CCM CDCL 9 X F


o o C
. d- ,^-





/U












LL- cO ro r-. c) oo oo c- cj tDo --
0 Or-- W M)rC C\J M C:
0 .O M M .. r M ) t .0.
-: 0000000000 000










L -C







4- 0 C)?t- M -C 0 r- 0 r- C
o- O0O M COD 0- C 0 0 0 o
.O-- 0 K* aC\j r-0
E4 co. rUcN OC m m C m
=0 0 0- r-






4-
0


V) 0 C. D C) .
S_ a) ---------- 0-
a) a)
0) Li-









S-


CL


(U









M) a cl
a 4- 0




(.0 S-
a 34a)




03 0) a c
co = 4-) M = S
o o S- 3


3 (- 00CC a) M m a)

< U)0 C LJa -4 0









TABLE 4 Comparisons of the Means of Experimental and
Control Groups for the Second Set of Posttest
Data

Experimental Control Group
Group Means Means

FIRO-B

eI (expressed inclusion) 4.2381 4.3000
wI (wanted inclusion) 2.5714 3.1000
ee (expressed control) 2.6667 2.7000
wC (wanted control) 2.5238 3.6500
eA (expressed affection) 3.1905 3.3500
wA (wanted affection) 2.0000 3.7500

Semantic Differential

Evaluative Scales (6 adjective pairs)
Counseling 31.1429 31.5000
Authority 23.7143 24.0000
Drugs I Took 23.0000 20.6000
Others Who Use Drugs 21.7143 18.2000
Parents 32.8095 30.2500
Women 29.5714 28.6000
Men 28.1429 25.1000
The Future 34.4286 35.3500
The Past 25.2381 23.9000
My Real Self 33.0952 33.1500

Potency Scales (6 adjective pairs)
Counseling 25.9524 26.9000
Authority 28.6667 30.1500
Drugs I Took 29.2381 31.7500
Others Who Use Drugs 27.3810 25.7000
Parents 24.4286 26.8500
Women 24.2857 22.2500
Men 26.2857 26.1500
The Future 23.3333 26.5000
The Past 24.9524 25.3000
My Real Self 24.9048 22.4000

Marathon Group Questionnaire

CIPI Scale 27.8095 25.6000
CRS Scale 34.7619 33.8500
SC Scale 24.3333 23.9500








The mean scores of the experimental and control groups for the scales

on the second set of posttests are presented in Table 4.

The first hypothesis stated:

Ho:l There will be no significant differences between experimental

and control group subjects on the variable of expressed behavior in the

area of inclusion, as measured by the FIRO-B.

The two related sub-hypotheses were:

Sub Ho:la There will be no significant differences between

heroin users in the experimental and control groups on the variable

of expressed behavior in the area of inclusion, as measured by the

FIRO-B.

Sub Ho:lb There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of expressed behavior in the area of inclusion, as measured

by the FIRO-B.

There were no significant differences (p < .05) between the control and

experimental group subjects, heroin users in the control and experimental

groups, or non-heroin users in the control and experimental groups on

the e (expressed inclusion) scale of the FIRO-B. No significant dif-

ferences were found either on the first posttest or second posttest ad-

ministrations. Thus Ho:l, Ho:la, and Ho:lb were not rejected (2 < .05).

The second hypothesis stated:

Ho:2 There will be no significant differences between the experimental

and control group subjects on the variable of wanted behavior in the area

of inclusion, as measured by the FIRO-B.

The two related sub-hypotheses were:

Sub Ho:2a There will be no significant differences between








heroin users in the experimental and control groups on the variable

of wanted behavior in the area of inclusion, as measured by the

FIRO-B.

Sub Ho:2b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of wanted behavior in the area of inclusion, as measured

by the FIRO-B.

Ho:2, Ho:2a, and Ho:2b were not rejected either on the first or second

posttest administrations.

The third hypothesis stated:

Ho:3 There will be no significant differences between experimental

and control group subjects on the variable of expressed behavior in the

area of control, as measured by the FIRO-B.

The two related sub-hypotheses were:

Sub Ho:3a There will be no significant differences between

heroin users in the experimental and control groups on the variable

of expressed behavior in the area of control, as measured by the

FIRO-B.

Sub Ho:3b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of expressed behavior in the area of control, as measured

by the FIRO-B.

Ho:3, Ho:3a, and Ho:3b were not rejected either on the first or second

posttest administrations.

The fourth hypothesis stated:

Ho:4 There will be no significant differences between experimental and

control group subjects on the variable of wanted behavior in the area of

control, as measured by the FIRO-B.








The two related sub-hypotheses were:

Sub Ho:4a There will be no significant differences between

heroin users in the experimental and control groups on the variable

of wanted behavior in the area of control, as measured by the FIRO-B.

Sub Ho:4b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of wanted behavior in the area of control, as measured

by the FIRO-B.

On the second posttest administration, the control group subjects tended

to score higher than the experimental group subjects on the wC (wanted

control) scale (Table 3). The control group subjects also scored

slightly higher on the first posttest administration on the wC scale of

the FIRO-B. However, none of these differences were significant (p < .05).

Therefore, Ho:4, Ho:4a, and Ho:4b were not rejected on either of the

posttest administrations.

The fifth hypothesis stated:

Ho:5 There will be no significant differences between experimental

and control group subjects on the variable of expressed behavior in the

area of affection, as measured by the FIRO-B.

The two related sub-hypotheses were:

Sub Ho:5a There will be no significant differences between

heroin users in the experimental and control groups on the variable

of expressed behavior in the area of affection, as measured by the

FIRO-B.

Sub Ho:5b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of expressed behavior in the area of affection, as measured

by the FIRO-B.









Ho:5, Ho:5a, and Ho:5b were not rejected on either of the posttest

administrations.

The sixth hypothesis stated:

Ho:6 There will be no significant differences between experimental

and control group subjects on the variable of wanted behavior in the

area of affection, as measured by the FIRO-B.

The two related sub-hypotheses were:

Sub Ho:6a There will be no significant differences between

heroin users in the experimental and control group on the variable

of wanted behavior in the area of affection, as measured by the

FIRO-B.

Sub Ho:6b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable of wanted behavior in the area of affection, as measured

by the FIRO-B.

On the first posttest administration, the treatment group subjects scored

only slightly lower on the wA (wanted affection) scale of the FIRO-B than

the control group subjects, but Ho:6, Ho:6a, and Ho:6b were not rejected

on the first posttest administration. However, on the second posttest

administration, the treatment group subjects did score significantly

lower (p < .05) than the control group subjects on the wA scale. Thus,

Ho:6, Ho:6a, and Ho:6b were rejected on the second posttest administration.

The seventh hypothesis stated:

Ho:7 There will be no significant differences between experimental

and control group subjects on the variable Counseling, as measured by

the evaluative and potency scales of a semantic differential.








The two related sub-hypotheses were:

Sub Ho:7a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Counseling, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:7b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Counseling, as measured by the evaluative and potency scales

of a semantic differential.

Ho:7, Ho:7a, and Ho:7b were not rejected either on the first or second

posttest administrations.

The eighth hypothesis stated:

Ho:8 There will be no significant differences between experimental

and control group subjects on the variable Authority, as measured by the

evaluative and potency scales of a semantic differential.

The two related sub-hypotheses were:

Sub Ho:8a There will be no significant differences between

heroin users -in the experimental and control groups on the variable

Authority, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:8b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Authority, as measured by the evaluative and potency scales

of a semantic differential.

Ho:8, Ho:8a, and Ho:8b were not rejected either on the first or second

posttest administrations.








The ninth hypothesis stated:

Ho:9 There will be no significant differences between experimental

and control group subjects on the variable Drugs I Took, as measured by

the evaluative and potency scales of a semantic differential.

The two related sub-hypotheses were:

Sub Ho:9a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Drugs I Took, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:9b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Drugs I Took, as measured by the evaluative and potency

scales of a semantic differential.

Ho:9, Ho:9a, and Ho:9b were not rejected either on the first or second

posttest administrations.

The tenth hypothesis stated:

Ho:10 There will be no significant differences between experimental

and control group subjects on the variable Others Who Use Drugs, as

measured by the evaluative and potency scales of a semantic differential.

The two related sub-hypotheses were:

Sub Ho:10a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Others Who Use Drugs, as measured by the evaluative and potency

scales of a semantic differential.

Sub Ho:O1b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Others Who Use Drugs, as measured by the evaluative and

potency scales of a semantic differential.








There was a slight tendency for the marathon group participants to

evaluate Others Who Use Drugs higher than the control group participants

on the first posttest administration. On the second posttest administration,

the experimental group members evaluated Others Who Use Drugs even higher

when their scores were compared with the scores of the control group

participants. The experimental group members also had a slight tendency

to regard the potency of Others Who Use Drugs as being higher than did

the control group members on both the first posttest administration and

on the second posttest administration. However, Ho:10, Ho:10a, and Ho:10b

were not rejected because none of these differences between the scores of

the marathon group participants and the control group participants were

significant (p < .05).

The eleventh hypothesis stated:

Ho:11 There will be no significant differences between experimental

and control group subjects on the variable Parents, as measured by the

evaluative and potency scales of a semantic differential.

The two related sub-hypotheses were:

Sub Ho:lla There will be no significant differences between

heroin users in the experimental and control groups on the variable

Parents, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:llb There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Parents, as measured by the evaluative and potency scales

of a semantic differential.

There was no particular trend apparent when the evaluative subscales of

the control and marathon group participants and the potency subscales of








the control and marathon group participants were compared on the concept

Parents on the first posttest administration. However, on the second

posttest administration, the marathon group participants tended to score

higher on the evaluative scale but lower on the potency scale than did

the control group members on Parents. None of these differences were

significant (p < .05). Therefore, Ho:ll, Ho:lla, and Ho:llb were not

rejected.

The twelfth hypothesis stated:

Ho:12 There will be no significant differences between experimental

and control group subjects on the variable Women, as measured by the

evaluative and potency scales of a semantic differential.

The two related sub-hypotheses were:

Sub Ho:12a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Women, as measured by the evaluative and potency scales of a semantic

differential.

Sub Ho:12b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Women, as measured by the evaluative and potency scales of

a semantic differential.

The marathon group participants tended to evaluate Women higher than

the control group participants on both the first posttest administra-

tion and on the second posttest administration. The marathon group

participants regarded the potency of Women as being lower than the

control group participants on the first posttest administration but

higher than the control group participants on the second posttest

administration. Since none of these differences were significant (jp < .05)

Ho:12, Ho:12a, and Ho:12b were not rejected.








The thirteenth hypothesis stated:

Ho:13 There will be no significant differences between experimental

and control group subjects on the variable Men, as measured by the

evaluative and potency scales of a semantic differential.

The two related sub-hypotheses were:

Sub Ho:13a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Men, as measured by the evaluative and potency scales of a semantic

differential.

Sub Ho:13b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Men, as measured by the evaluative and potency scales of a

semantic differential.

Ho:13, Ho:13a, and Ho:13b were not rejected either on the first or second

posttest administrations.

The fourteenth hypothesis stated:

Ho:14 There will be no significant differences between experimental

and control group subjects on the variable The Future, as measured by the

evaluative and potency scales of a semantic differential.

The two related sub-hypotheses were:

Sub Ho:14a There will be no significant differences between

heroin users in the experimental and control groups on the variable

The Future, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:14b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable The Future, as measured by the evaluative and potency scales

of a semantic differential.








There were no significant differences (2 < .05) between the ways control

and experimental group subjects scored on the evaluative scales on either

of the posttest administrations for The Future. Ho:14, Ho:14a, and

Ho:14b were not rejected as these hypotheses related to the evaluative

scales on either of the posttest administrations. On the other hand,

there were significant differences (p < .05) between the ways experimental

and control group subjects responded to the potency scale of The Future

on the first posttest administration and on the second posttest

administration. Therefore Ho:14, Ho:14a, and Ho:14b were rejected on the

first and second posttest administrations as these hypotheses related to

the potency scale of The Future.

The fifteenth hypothesis stated:

Ho:15 There will be no significant differences between the experimental

and control group subjects on the variable The Past, as measured by the

evaluative and potency scales of a semantic differential.

The two related sub-hypotheses were:

Sub Ho:15a There will be no significant differences between

heroin users in the experimental and control groups on the variable

The Past, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:15b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable The Past, as measured by the evaluative and potency scales

of a semantic differential.

Ho:15, Ho:15a, and Ho:15b were not rejected either on the first or second

posttest administrations.








The sixteenth hypothesis stated:

Ho;16 There will be no significant differences between experimental

and control group subjects on the variable My Real Self, as measured by

the evaluative and potency scales of a semantic differential.

The two related sub-hypotheses were:

Sub Ho:16a There will be no significant differences between

heroin users in the experimental and control groups on the variable

My Real Self, as measured by the evaluative and potency scales of a

semantic differential.

Sub Ho:16b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable My Real Self, as measured by the evaluative and potency

scales of a semantic differential.

The experimental group subjects tended to score higher than the control

group subjects on My Real Self on the first posttesting on the evaluative

scale. These results were not maintained on the second set of posttests.

On the other hand, there were almost no differences between the scores of

the experimental and control group subjects on the first posttest

administration on the potency scale but on the second posttest

administration the experimental group subjects tended to score higher

than the control group subjects. None of the differences, however, were

significant (p < .05). Therefore Ho:16, Ho:16a, and Ho:16b were not

rejected either on the first or second posttest administrations.

The seventeenth hypothesis stated:

Ho:17 There will be no significant differences between the

experimental and control group subjects on the variable Confidence in

Personal Involvements, as measured by the Marathon Group Questionnaire.








The two related sub-hypotheses were:

Sub Ho:17a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Confidence in Personal Involvements, as measured by the Marathon

Group Questionnaire.

Sub Ho:17b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Confidence in Personal Involvements, as measured by the

Marathon Group Questionnaire.

On the first posttest administration, the experimental group subjects

scored significantly higher (p < .05) than the control group subjects on

the CIPI scale of the Marathon Group Questionnaire. These significant

differences between the scores of the control and experimental subjects,

however, were not maintained on the second set of posttests, although the

experimental group subjects did score higher than the control group

subjects on the second posttest. Ho:17, 17a, and 17b were rejected on

the first posttest administration but failed to be rejected on the second

posttest administration.

The eighteenth hypothesis stated:

1Ho:18 There will be no significant differences between experimental

and control group subjects on the variable Counseling Readiness, as

measured by the Marathon Group Questionnaire.

The two related sub-hypotheses were:

Sub Ho:18a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Counseling Readiness, as measured by the Marathon Group Questionnaire.








Sub Ho:18b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Counseling Readiness, as measured by the Marathon Group

Questionnaire.

There was a very slight tendency for the experimental group subjects to

score higher than the control group subjects on the CRS Scale of the

Marathon Group Questionnaire, both on the first posttest administration

and on the second posttest administration. None of these differences

approached significance (p < .05). Therefore Ho:18, Ho:18a, and Ho:18b

were not rejected on either of the posttest administrations.

The nineteenth hypothesis stated:

Ho:19 There will be no significant differences between experimental

and control group subjects on the variable Self Confidence, as measured

by the Marathon Group Questionnaire.

The two related sub-hypotheses were:

Sub Ho:19a There will be no significant differences between

heroin users in the experimental and control groups on the variable

Self Confidence, as measured by the Marathon Group Questionnaire.

Sub Ho:19b There will be no significant differences between

non-heroin users in the experimental and control groups on the

variable Self Confidence, as measured by the Marathon Group Ques-

tionnaire.

There was a slight tendency on the SC Scale, similar to the other scales

of the Marathon Group Questionnaire, for the experimental group subjects

to score higher than the control group subjects on both the first posttest

and the second posttest. Since none of these differences approached

significance (p <.05) Ho:19, Ho:19a, and Ho:19b were not rejected either

on the first or second posttest administrations.




85



The null hypotheses and null sub-hypotheses which were rejected

included Ho:6, Ho:6a, and Ho:6b on the second posttest administration,

Ho:14, Ho:14a, and Ho:14b on the first and second posttest administrations,

and Ho:17, Ho:17a, and Ho:17b on the first posttest administration.

A discussion of the implications of this research follows in Chapter V.














CHAPTER V

SUMMARY AND RECOMMENDATIONS


Summary

The purpose of this research was to assess the impact of marathon

group counseling on changing selected attitudes of incarcerated female

drug abusers. The population studied was female inmates of the Florida

Correctional Institution who had major drug problems before being

imprisoned. Most of these females were imprisoned because they committed

crimes (sales of drugs, armed robbery, grand larceny) to support their

drug habits. The effects of treatment on inmates who had been addicted

to heroin and on inmates who had abused illicit drugs other than heroin

were also assessed.

Two pairs of group leaders conducted two counseling groups on the

same day. These groups lasted for 16 hours and were called marathons.

The treatment guidelines followed by the group leaders comprise the

Marathon Group Strategy (Appendix F).

Fifty-six inmates were selected randomly to participate in the

control group and experimental groups. Two pairs of group leaders were

assigned randomly to the two experimental groups. Three sets of

instruments were used to test the hypotheses of this study: the FIRO-B,

a semantic differential, and the Marathon Group Questionnaire. The

Posttest Only Control Group Design (Campbell & Stanley, 1966) was used

to assess the outcomes of this study. The research participants were








administered two posttests; the first posttest administration was the

day following the marathons and the second was given 4 weeks later. The

results were analyzed by an analysis of variance procedure.

The findings of this research were undramatic. The marathon group

participants scored significantly lower (p < .05) than the control group

participants on the wA (wanted affection) scale of the FIRO-B on the

second posttest administration, and lower on the potency scale of

The Future on a semantic differential on the first and second posttest

administrations. The marathon group participants scored significantly

higher (p < .05) on the first posttest administration of the Confidence in

Personal Involvements Scale of the Marathon Group Questionnaire. -There

were no significant differences (2 < .05) between the ways the control and

experimental group participants scored on any of the other scales of the

instruments used for this research.

Conclusions

The purpose of this research was to assess the impact of marathon

group counseling on changing certain key attitudes of incarcerated drug

abusers. The attitudes which were assessed related to the feelings of

the research participants towards interpersonal relationships, counseling,

self, others, drugs, the past and future, and authority. The FIRO-B,

the Marathon Group Questionnaire, and a semantic differential were the

instruments which were used to assess the attitudes of the research

participants in these areas. The types of attitude changes which occurred

as measured by each of these instruments are discussed below with the object

of drawing conclusions, based on this study, about the effectiveness of

marathon groups as a treatment modality with incarcerated drug abusers.








The FIRO-B assesses how persons form and maintain relationships in

the areas of inclusion, control, and affection. It was difficult to

discern any major differences in the ways the marathon group members and

the control group members responded to most of the scales of the FIRO-B

on either the first or second set of posttests. There were no apparent

trends when the scores of the marathon and control group participants

were compared. Thus, the conclusion must be made that either the FIRO-B

failed to assess the manner in which the group participants changed their

attitudes in the inclusion, control, and affection areas or that the

marathon groups failed to change the attitudes of the participants in

these areas.

The one possible exception to the lack of differences in the attitudes

of the control and marathon group participants on the scales of the FIRO-B

was on the affection scale. On the second posttest administration, the

control group participants scored significantly higher (p < .05) than the

experimental group participants on the wanted affection scale. These

results may be interpreted as showing that the wanted affection of the

marathon group participants was reduced because of their participation

in the marathons. The marathons may have helped to meet the affection

needs of the participants.

The Marathon Group Questionnaire was developed by this writer to

assess the impact of marathon groups on attitudes of imprisoned drug abusers.

The Marathon Group Questionnaire has three scales which were labeled

Confidence in Personal Involvements, Counseling Readiness, and Self

Confidence. This questionnaire was designed so that higher values

associated with each scale (CIPI Scale, CRS Scale,and SC Scale) were also

associated with positive marathon group outcomes.








The marathon group participants obtained higher scores on the CIPI

Scale, the CRS Scale, and the SC Scale on both posttest administrations

than did the control group participants. Although this trend existed,

most of the differences between the scores of the marathon group parti-

cipants and the control group participants were not significant.

The marathon group participants scored significantly higher (Jp< .05)

on the Confidence in Personal Involvements Scale on the first posttest

administration than did the members of the control group. Thus, the day

after the marathons, the group participants appeared to be more confident

in their ability to relate to others in a meaningful way than the control

group participants. On the second posttest administration, the marathon

group participants still scored higher on this scale than the control group

participants but the differences were no longer significant (P < .05).

Therefore, it might be concluded that the marathon group participants

did not maintain the same level of confidence in their abilities to

relate to others after 1 month as they had 1 day following a marathon

group experience.

A semantic differential was used to assess the attitudes of the

subjects toward the following concepts: Women, Men, Parents, Authority,

The Past, The Future, Counseling, Drugs I Took, Others Who Use Drugs,

and My Real Self. There were greater differences between the experimental

and control group participants on the potency scales than on the

evaluative scales on most of the concepts evaluated by the semantic

differential. The only scale of the semantic differential on which there

were significant differences between the control group members and the

marathon group members was the potency scale for the concept The Future.








On both posttest administrations, marathon group participants

scored significantly lower (j < .05) on the potency scale for the con-

cept The Future than the control group participants. Whatever differences

existed between the control and marathon group members on this scale were

maintained over a 1 month time period. The meaning of these differences

is somewhat elusive. One possible interpretation is that the marathon

group participants became more realistic in their perceptions of The

Future which contributed to their regarding The Future as being a somewhat

less potent force in their lives. Since correctional offenders often

negate or deny the difficulties they may have in adjusting to society,

these changes may constitute positive changes for this population.

Limitations

This research had several different types of limitations which may

very possibly have confounded the results. One of the major limitations

was that most of the persons who participated in this research as control

or marathon group participants had previously received some form of

individual or group counseling at the Florida Correctional Institution.

Therefore, the effects of the marathon groups had to be powerful enough

to produce results over and above any benefits the control and marathon

group members had previously received from participating in other forms

of counseling at F.C.I.

It is possible that many of the research participants already had

changed many of their attitudes on a number of scales before they

participated in the marathon groups due to their previous counseling

involvements. For instance, in this research both control group and

marathon group participants evaluated counseling highly on a semantic








differential. The control group participants obtained an average score

of between 5 and 6 points (out of 7 possible points) on the evaluative

scales on the concept Counseling. Both the control and experimental group

subjects had low evaluative scores for Drugs I Took and Others Who Use

Drugs. Different results might have been obtained if subjects had been

selected randomly into control and marathon groups from a population of

imprisoned drug offenders whose experience with counseling was limited

rather than extensive.

Another possible limitation of this research involved the instruments

used to assess outcomes. Many questionnaires commonly used to assess

counseling outcomes are rather primitive. The instruments used to assess

the attitudes of the experimental and control group subjects in this

research were viewed as among the best available to assess counseling

group outcomes, but these instruments had many limitations.

The FIRO-B is a self report inventory which can be faked. The

different scales of the instrument are not independent. The major scales

of inclusion, affection, and control may not be sensitive to the types of

changes the group members should have been expected to make from parti-

cipating in a marathon group.

The semantic differential also presents problems; it provided a lot

of information but it was difficult to interpret the meaning of certain

changes. The control and experimental subjects also appeared to be a

little confused about the format of the semantic differential. When they

were administered this instrument, they did not seem to readily understand

the meaning of many of the adjective pairs.

The Marathon Group Questionnaire also has several limitations.

The instrument had not been used prior to this study. The instrument is








also a self report scale which can be faked. On the other hand, one of

the positive aspects of this instrument was that the changes on the scales

of the instrument were easily interpreted. The scales of the instrument

measured the types of changes this researcher wished to bring about by

the marathon groups.

Another problem affecting the scores on all the instruments was that

the inmates of correctional institutions often do not enjoy being tested.

A few subjects rushed through the questionnaires when they were administered

and seemed bored while taking them. Some subjects had difficulty in

reading or understanding the questionnaires, especially the semantic

differential. However, it did appear that when most of the subjects took

these instruments they answered the questions thoughtfully and carefully.

One further possible limitation of this research was that the

instruments which were used did not measure specific factors. The different

concepts on the semantic differential were very general. The Marathon

Group Questionnaire assessed factors different from both the FIRO-B and

the semantic differential. It probably would have been more relevant to

have choose fewer specific variables, such as perceptions of interpersonal

relationships, and to have used more instruments to assess more specific

factors.


Recommendations

A number of recommendations for further research are suggested. First,

instruments which assess specific types of counseling outcomes should be

developed. For instance, instruments of high reliability and validity

need to be developed which assess such varied therapeutic outcomes as

how a person feels toward drugs, interpersonal relationships, family

relationships, self, and others. More specifically, the Marathon Group