Wan
a ;

.1

.155?

a

53“.... r. . .
1.9,. .. “ L1,.
..8.

1‘1.

I.
A?!

 

 

 

 

 

 

 

 

 

 

I
I
(u 5.14:

 

 

 

 

. .. .

h>v 0:
«Eva!

a,“ {in

Ed. .

.m .3”..-
.1. 1 1.. .
‘flquh-Vnda. A

h ..
I.
51”.». 1

. . 5 :1 1o
arising!»
“ IAVI . fi .

.am.

.155qu
v...

1.3, 2.»...
f :4“sz

:

‘

~ .
4. «Mix
a; h

$1.11...» .
. 1.1 t.

. 5.2.
.

fix.
.r

7 .
\thu . 4
.anu.
x1)

x

a.“

.fi.
I;

w ‘ 1......
x .. _ . .= 1-
2.1.2.33. .9... 2; 31m;
5. kn. aififl‘stdtr

«main? 5.. . . ,
:9: . . m?

.. . an... ( ....

#1.? .1..

.3 1s...

Hagan“.

V 1. a:
'0‘.
vi a

. 1.
fl 1 b.
133:,-

$911

elk.

.y

: W
fifsum {.11 2

rd

 

THESIS
I

2C" 01

This is to certify that the

thesis entitled

T he An’ricoa'al Mii’tudes and The Afl'rbdos
Toward Boot Cam? of Bad Gum? lnmdbs:

A- Comfaorisom 0F kfiicaw Madam; W C(wcaeian Mm

presented by

AseAuct-s SAUéKHAMTE

has been accepted towards fulfillment
of the requirements for

ma Mme in Crimmqi Justice

QZVZ/b/WL,
Date /93/0?§///777

0-7 639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY
Michigan State
University

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

I
|

w M 694 303;.

 

 

 

 

 

 

 

 

 

 

 

 

 

woo mm.“

 

THE ANTISOCIAL ATTITUDES AND THE ATTITUDES TOWARD BOOT CAMP
OF BOOT CAMP INMATES: A COMPARISON OF AFRICAN AMERICANS AND
CAUCASIAN AMERICANS
By

Assanee Sangkhanate

A THESIS

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

MASTER OF SCIENCE

School of Criminal Justice

1 999

ABSTRACT
THE ANTTSOCIAL ATTITUDES AND THE ATTITUDES TOWARD BOOT CAMP
OF BOOT CAMP INMATES: A COMPARISON OF AFRICAN AMERICANS AND
CAUCASIAN AMERICANS
By

Assanee Sangkhanate

The present study was designed to examine the antisocial attitudes and the
attitudes toward boot camp program of boot camp participants. Specifically, the author
intended to examine if Caucasian American and African American offenders changed
their attitudes afler graduating from the program. In addition, the study was conducted to
examine if Caucasian American and African American offenders differed in their
antisocial and program attitudes before and after participating in boot camp. Data were
retrieved fi‘om the Inter-university Consortium for Political and Social Research (ICPSR)
database. Samples of were randomly selected fi'om boot camp offenders in South
Carolina. The independent variables are socio-demo graphic and criminal characteristics
of samples while scores from Antisocial Attitude Scale and Program Attitude Scale were
used as dependent variables. Results Show that Caucasian Americans became less
antisocial and more positive about the program after they participating in boot camp;
whereas, African Americans did not significantly change both attitudes over time. When
examining the interaction between race and other characteristics, most predictors such as
education, gender, and substance abuse history that were commonly used to look at
effects of programs seem applicable only to Caucasians and did not predict anything

much for Afiican Americans.

ACKNOWLEDGMENTS

I wish to sincerely thank Dr. Sheila Maxwell, the Chairperson of the Thesis
Committee. Her suggestions and assistance were invaluable. Also, I would like to
deeply thank the other committee members, Dr. Vincent Hoffman and Dr. Charles
Corley. Their critical and constructive feedback helped improve the quality of this study.

Special thanks are extended to all my teachers and friends who always had a
confidence in my firture. Last but nor least, I owe a tremendous debt of gratitude to my
parents and my sister who always provide invaluable encouragement and moral supports

throughout the cause of my graduate study.

iii

TABLE OF CONTENTS

LIST OF TABLES ............................................................................. vi
INTRODUCTION .............................................................................. 1
Statement of Problem ................................................................. 1
Purpose of Study ....................................................................... 4
LITERATURE REVIEW ...................................................................... 6
Research and Evidence on Boot Camp .............................................. 6
Boot Camp as an Alternative Sanction ..................................... 6
Boot Camp and Criminological Theory ............................................. 9
Control Theory ................................................................. 9
Military Basic Training and Deterrence Effect ............................ 9
Boot Camp, Rehabilitation Goal, and Antisocial Attitude ........... 10
Summary ............................................................................. 17
RESEARCH METHODOLOGY ........................................................... 18
Population and Sample .............................................................. 18
Data Collection and Instruments ................................................... 18
Research Setting ..................................................................... 20
Research Questions .................................................................. 21
Hypotheses ............................................................................ 22
Variables .............................................................................. 23
Independent Variables ...................................................... 23
Dependent Variables ........................................................ 23
Data Analysis Procedures ........................................................... 24
Summary .............................................................................. 25
DATA ANLYSIS AND RESULTS ........................................................ 26
Sample Characteristics ............................................................... 26
Socio-Demographjc Characteristics ....................................... 26
Criminal Characteristics .................................................... 28
Other Characteristics ........................................................ 29
The Change in Antisocial Attitude and Program Attitude of Caucasian
American and African American ................................................... 30
The Bivariate Regression for Sample Characteristics and Scores on Antisocial
Attitude and Program Attitude ...................................................... 33
Sample Characteristics and Scores on Antisocial Attitude Scale. . 34
Sample Characteristics and Scores on Program Attitude Scale ....... 37

The Average Scores on Antisocial Attitude Scale and Program Attitude Scale:

iv

A Comparison Between Caucasian American and African American. . . . . 38

Antisocial Score ............................................................. 39
Program Score ............................................................... 42
The Multiple Regression Analysis for Sample Characteristics and Scores on
Antisocial Attitude and Program Attitude Scale ................................. 42
Antisocial Attitude Model .................................................. 43
Program Attitude Model ................................................... 46
The Multiple Regression Analysis: A Comparison of Caucasian
Americans and African Americans ....................................... 47
Caucasian American and Scores on Antisocial Attitude ............... 49
Caucasian American and Scores on Program Attitude .................. 49
African American and Scores on Antisocial Attitude ................... 50
Afiican American and Scores on Program Attitude ..................... 50
DISCUSSION ................................................................................. 53
Antisocial and Program Attitude of Boot Camp Participants .................. 53
Who Benefit fi'om Boot Camp? .................................................... 59
Future Research ....................................................................... 60
Limitations of Study ................................................................. 61
APPENDIX A
Items on Jessness Antisocial Attitude Scale ...................................... 63
APPENDIX B
Items on Program Attitude Scale ................................................... 65
APPENDIX C
Reliability Analysis for Program Attitude Scale from SPSS ................... 66
BIBLIOGRAPHY ............................................................................. 70

p—g
0

LIST OF TABLES

Socio-Demographic and Criminal Characteristics of Sample ...................... 27

Paired-Sample T-Test for Mean Difference of the Scores: A Comparison

Between Timel and Time 2 ............................................................. 31
. Bivariate Regression Analysis fro Socio-Demographic and Criminal

Characteristics and Scores on Antisocial Attitude and Program Attitude ......... 35

The Average Scores on Antisocial Attitude Scale and Program Attitude

Scale: A Comparison of Caucasian American and African American ............ 40

The Independent Sample T-Test for the Average Score on Antisocial Attitude Scale
and Program Attitude Scale: A Comparison of Caucasian American and African

American ................................................................................... 41
Multiple Regression Analysis for Sample Characteristics and Scores on Antisocial
Attitude and Program Attitude Scale: An Analysis for All Sample ............... 44

Multiple Regression Analysis for Sample Characteristics and Scores on Antisocial
Attitude and Program Attitude Scale: A Comparison of Caucasian American and
African American ......................................................................... 48

vi

CHAPTER I

INTRODUCTION

_St_atement of Problem

Since the early 19805, boot camp1 has become popular among other alternative
sanctions to traditional imprisonment. The number of the prisons with boot camp
programs increased from 26 in 1990 to 65 by the end of 1995 (BJS, 1999). There are
many reasons for the growth of boot camp. First, since boot camp addresses discipline
and self control issue, politicians and policy makers believe boot camp can help them
achieve “get tough on crime” policy (MacKenzie & Piquero, 1994, p. 225). Second,
corrections officials perceive the importance of boot camp in such aspects as
rehabilitation, and recidivism reduction because they believe that therapeutic elements of
boot camp will help rehabilitate offenders. As a result, fewer offenders will be
rearrested, convicted and returned to prison. (Cowles & Castellano, 1995; MacKenzie &
Piquero, 1994, p.225).

Moreover, given the problem of prison overcrowding, almost every program
officials expect boot camp to decrease prison population. According to the Bureau of
Justice Statistics (BJS, 1998), from year-end 1990 to year-end 1998, the Nation's
incarcerated population rose by nearly 676,700 inmates, increasing at an average annual

rate of 6.0%. State, Federal, and local governments had to accommodate additional

 

' The terms “boot camp” and “shock incarceration” have been used interchangeably. In South Carolina, the
“shock nearoeration” terms were used. However, for a consistency reason, the author used only the terms
“boot camp” in the present study.

84,587 inmates per year (or the equivalent of 1,627 new inmates per week). The rate of
incarceration increased fi'om 1 in every 217 to 1 in every 149 US resident. Additionally,
by year-end 1998 the Federal prison system was estimated to be operating at 27% over
capacity, increasing since year-end 1997(19% over capacity in 1997).

The expectations from both officials and the public are gathered to form
correctional boot camp programs across the country. The goals and components of boot
camp vary from place to place. Generally, the goals of boot camp include punishment,
deterrence, rehabilitation, decreases in prison crowding, and recidivism control. The
operations of most programs, however, are almost the same (Bourque, Han, & Hill, 1996;
Bourque, Cronin, Pearson, Felker, Han, & Hill, 1996; MacKenzie & Ballow, 1989;
MacKenzie & Souryal, 1994; Parent,1989). Recently, the American Correctional
Association (ACA) has developed standards for the operation of both adult and juvenile
boot camp programs (ACA, 1995a, 1995b). These standards are expected to help
program administrators develop an effective boot camp programs as well as improve
existing programs for their clients.

As results of a variety of goals, program components and recruitment criteria of
boot camp are different from one another. Almost every program, however, targets
young, nonviolent, first-time offenders. Participants are normally housed in a military-
style environment, which are separated from regular inmates (Peters, Thomas,
Zamberlan, & Associates, 1997). The primary activities of boot camp include physical

training, labor, drill and ceremony, and summary punishmentz. Some boot camps might

 

2 “Summary punishment is an interim punishment imposed by staff for disciplinary infiactions, which
entails an on-the-spot, immediate sanction for an infraction. This punishment may include pushups, extra
chores, or another work assignment.” From Gowdy (1996), pp.23.

provide therapeutic components for participants such as education, job skills, substance
abuse counseling and treatment, and aftercare service.

There have been many researches and studies examining the impact and the
effectiveness of boot camp programs across the country (Bourque, Han, & Hill, 1996;
Bourque, Cronin, Pearson, Felker, Han, & Hill, 1996; Cowles, & Castellano, 1995;
MacKenzie& Piquero, 1994; MacKenzie, Shaw, Gowdy, 1996b; MacKenzie, & Souryal,
1994). However, few studies have focused on boot camp and racial issues. In fact, most
of the evaluation studies use race only as a control variable. This paucity on information
stands in contrast to studies that show increasing numbers of minority groups and
widespread disparities by race in the criminal justice system. Bureau of Justice Statistics
(1998) statistics show that in 1997 given the number of 100,000 US residents, 8,630
Afiican American males in their late twenties were in prisons in 1997. This figure was
relatively high when compared to the number of Hispanic males (2,730) and Caucasian
males (868). BJS statistics also Show that between 1990 and 1997 the number of
Caucasian males in prisons increased by 54% while the increases in the number of
Afi'ican American males were 61% and the number of Caucasian females and Afiican
American females were 80%. Additionally, at year-end 1997 (the latest available data),
more African American males (548,900) were incarcerated in State or Federal prisons
than Caucasian males (541,700).

Boot camp is among significant intermediate sanctions that have been used to
cope with problems in the criminal justice system. Unfortunately, the paucity of the
information on the effect of boot camp on different ethnic groups provides an unclear

picture of how participants fi‘om different ethnic groups benefit from the program. Even

though race has been included as a control variable in most criminal justice research,
none of them Specifically focus on the effect of boot camp on the antisocial attitude of
inmates from different ethnic groups. Findings from available studies on the impact of
incarceration on prison inmates across racial groups are still inconclusive. Some suggest
that the development of violent attitudes and behaviors in prison of Caucasian American
and African American did not differ (Ellis et al., 1974; Wright, 1989). Many found that
prison violence and homicide are associated with African American inmates (Cazenave
and Straus, 1990; Wolfgang et al., 1985). Whereas, some supports violence among
Caucasian Americans (Dixon & Lizotte, 1987). These studies, however, were criticized

for their research methodology problem (Harer & Steffensmeier, 1996).

Purpose of the Study

This study was designed to examine the antisocial attitude of the boot camp
participants across ethnic groups. Specially, the antisocial attitude of Caucasian
Americans and African Americans were measured and compared at the beginning of the
program and once again after they complete the program. The findings were anticipated
to clarify the differences, if any, in attitude change between Caucasian Americans and
Afiican Americans. Further, the findings should have an implication on administrative
policies of “who” would be likely to benefit from the program. Additionally, with respect
to the goals of boot camp the author expects this study to provide empirical evidence to
rehabilitation philosophy, showing if boot camp ‘Vvorks”. The review of previous studies

was presented in Chapter II. The issues of military training vs. deterrence effect,

therapeutic program vs. rehabilitation, and conformity atmosphere of boot camp vs.

control theory, and the explanation of antisocial attitude of inmates were also discussed.

CHAPTER [1

LITERATURE REVIEW

Resgarch aknd Evidence on Boot Camp

Boot Cm as an Alternative Sanction.

 

Prison overcrowding has been a significant issue in criminal justice system for a
long period of time. Among the efforts to solve this problem is to create alternative
sentences to traditional incarceration. One of them is to place offenders in the community
instead of formal institution. Community corrections such as house arrest, probation,
parole, and community service has been widely used Since. However, the goal of
reducing prison overcrowding could be unrealized because these sanctions that were
originally designed as alternatives to incarceration have actually been used for offenders
who would otherwise have received a lesser sentence (MacKenzie & Piquero, 1994).
That is when another intermediate was initiated—Boot Camp.

Boot camp has been used as an alternative to traditional imprisonment since
1980s. The rapid grth of military boot camp results fiom many reasons. Among the
most important reasons is to help reduce the number of offenders in prisons. One scholar
who has devoted so many of her studies on boot camp program is Doris Layton
MacKenzie, the University of Maryland. In one of her studies, MacKenzie and Piquero
(1994) suggested that almost everyone expect boot camp to reduce the number of
offenders in prison. Some believe that as a result of participating in boot camp program,

there will be a lower recidivism rates. Thus, fewer prisoners will return to prison and the

need for prison will be reduced. At the individual level, there are different opinions
about how they expect a lower recidivism from boot camp. Some believe boot camp will
deter offender to commit new crime, while others argue that the program will rehabilitate
offender so they will not return to criminal activities once one is released. In both of
these situations, prisoners are expected not to return to prison, which help relieve prison
overcrowding.

However, the extent that boot camp has an impact on prison overcrowding
depends on many factors. One thing that should be taken into consideration is the
selection criteria of boot camp. In order to avoid the criticism of “net widening” like
other community corrections, boot camp must strictly target only those who otherwise
would have been incarcerated in traditional prison. Furthermore, There must be a
sufficient number of eligible inmates who successfully complete the boot camp in a
shorter time than they would have been in prison (MacKenzie & Piquero, 1994).
Therefore, it should be noted that rigid revocation criteria could discourage offender not
to volunteer or will drop out and return to prison.

The popular of boot camp also results from the fact that boot camp provides both
treatment and punishment. Considering punishment perspective, politician and policy
makers support boot camp so that they can appear to be tough on crime (Mackenzie &
Piquero, 1994). Additionally, Anderson, Dyson, & Burns (1996) suggested that both
conservative and liberal legislature as well as the public also the use of boot camp as an
alternative sentence. Conservatives strongly support the use of strict discipline associated
with the idea of authority figures confronting offenders and rejecting the excuses that

they give for their participation in crime. Not only do they want to see boot camp shock

participants, but they also hope that boot camp graduates will instill in them discipline,
responsibility, and respect. Liberals, however, recognize the treatment and rehabilitation
components. Since many of boot camp graduates receive GED and learn some job skills,
they hope that boot camp will enable them to get better chances in life after they are
released. Moreover, liberals encourage the use of boot camp as many more offenders are
diverted from traditional imprisonment. They believe that spending time in prison can
ultimately change an offender to become a hardened criminal.

Originally, boot camp was designed to cope with prison overcrowding problem.
However, when the number of ofienders incarcerated in jails are gradually increasing,
jail-operated boot camp has been implemented. Although the average length of offenders
in jails is relatively short (15-16 days) compared to approximately 2-year stay of prison
inmates, jail are increasingly housing inmates who serve many months in confinement
(Austin, Jones, & Bolyard, 1996). Nevertheless, scholars suggested that Since inmates in
jails are so diversified and differs from those in prisons, goals and components of prison
boot camp may not apply or may be more difficult to achieve in jail-based boot camp
(Austin, Jones, & Bolyard, 1993; McCampbell, 1996).

Boot camp is not limited to only adult offenders. In fact, boot camp has also been
used as an alternative sanction in juvenile justice system. The juvenile boot camp
program was initiated by the Office of Juvenile Justice and Delinquency Prevention
(OJJDP) in July 1990 (Felker & Bourque, 1996). Three demonstration program were in
Cleveland, Ohio; Denver, Colorado; and Mobile, Alabama These programs were
designed by borrowing components extensively from adult boot camp. Like adult boot

camp, scholars have yet found conclusive results if boot camp “works” for juvenile

offenders (Bottcher & Isorena, 1996; Bourque, Cronin, Pearson, Felker, Han, & Mill,
1996; Cass & Kaltenecker, 1996; Cronin, 1994; Felker & Bourque, 1996; Peters,

Thomas, Zamberlan, & Associates, 1997).

Boot Cm and Criminological Theog

The present research was designed based on deterrence and rehabilitation goals of
boot camp. The main objective of this study is to examine if boot camp is successful in
changing female participants’ antisocial attitudes after placing them in a military-like
setting and participating in therapeutic programs. The researcher believes that the
changes in human behaviors depend heavily on changing their attitudes. If the deterrence
and rehabilitation goals are achieved, findings Should reveal the positive change of
participants’ attitudes. A review of control theory and the critique of boot camp under
the deterrence and rehabilitation frameworks as well as the evaluation studies of boot

camp programs are discussed.

Control Theogy.

Control theory was used as a theoretical model in initiating almost all boot camp
programs (Clark & Aziz, 1996; Anderson, Dyson, & Burns, 1998). Control theory
proposes that “nonconformity is a product of the failure of the social bond. Through the
attachment of individuals to others, conformity is assured. When such attachments fail to
develop or when they are disrupted, the internalization of legitimate norms becomes
problematic” (Clark & Aziz, 1996). It is assumed that offenders entering boot camp are

individuals whose bonds to soCiety are weak or broken. By exposing offenders to the

boot camp philosophies and practices, boot camp designers believe that it will help
strengthen and restore these bonds.

Considering Hirchi’s social control theory, the bonds to society include: (1)
attachment to others, (2) commitment to enhancing life, property, and reputation, (3)
involvement in conventional activities, and (4) belief in positive human values (Siegel &
Senna, 1997). Hirchi’s social control theory also assumed that “offenders are
opportunistic, impulsive, lack of self-control, and pursue immediate gratification rather
than postpone self-pleasure. Such criminals with low self-control normally live for the
moment.” (Anderson, Dyson, & Burns, 1998, p. 14). This concept was recognized and
applied by many boot camp programs. New York’s Shock Incarceration, for instance,
focuses on the need for offenders to strengthen their indirect controls, their internalized
controls, and their controls over opportunities for conventional activities “by promoting
responsibility for choices and stressing the consequences of their behavior” (Clark &

Aziz, 1996).

Militm Basic Training and Deterrflce Effect.

Generally, a combination of punitive and rehabilitative program elements of boot
camp is expected to serve both deterrence and rehabilitation goal. With regards to
punitive program elements, a survey of existing correctional boot camp programs
revealed commonality in the use of military drill and ceremony, military bearing and
courtesy, physical labor, physical training, and strict discipline (Morash & Rucker, 1990).
Namely, these elements in addition to incarceration itself are expected to serve as

deterrents.

IO

Specific deterrence is a high-priority goal of boot camps in many states.
Researchers found that boot camp programs in Florida, Georgia, Louisiana, Oklahoma,
South Carolina, and Texas focused on deterrence goal rather than rehabilitation
(MacKeczie & Souryal, 1994). Unfortunately, no study has focused on the examination
of boot camp and its specific deterrence effects. Previous researches on specific
deterrence of related programs, however, have not yet found promising results. Lotz,
Regoli, & Raymond (1978) claimed limited or no deterrent effect as a result of
incarceration in training school. Scholars also suggested that the boot camp itself could
not deter future crime commission; instead, the offender’s fear of the real prison is likely
to be an instrumental factor. Likewise, Zachariah (1996) pointed out that when
deterrence is set as a goal of the program, boot camp settings are usually located within
the prison, so tlmt participants can observe prison inmates and their routine prison
activities. Furthermore, boot camp staffs contribute to the offenders’ fear by scaring
them of sexual exploitation and victimization in prison.

It is unlikely that boot camp experience will lead to increased perceptions of
either the certainty or severity of punishment (MacKeczie & Souryal, 1994). In terms of
general deterrence, these researchers also found no evidence to believe that the public
will be deterred by the threat of serving time in boot camp prison. They stated that
interviews with boot camp participants showed that many participants, before entering
the program, did not think that they would have trouble meeting program requirements.

Evidences Show that the severity of punishment does not guarantee deterrence.
Tittle (1980) found the degree in which sanction could deter crimes does not necessarily

depend on harshly formal punishments. Rather, informal social sanctions are far more

effective deterrents. The researcher operationally defined the fear of infOrmal social
sanctions as the fear that derived from interpersonal and other nonlegal sources. These
fears of informal sanctions include peer fear, contingent interpersonal fear, contingent
community fear, interpersonal severity, and community severity. The researcher
explained that if people who are important to an individual are likely to disapprove a
particular behavior and an individual thinks those people will find out in case of
commission that behavior, then that possibility will have considerable deterrent effect.
Consistent to Tittle (1980), Grasmick and Bursik (1990) found that informal sanctions
enforced by parents and peers have more binding deterrent effect than laws and the threat
of formal incarceration.

Previous study also found that males and females differ little in deterrence; where
differences are apparent the researcher suggested that females should commit less
deviance (Tittle, 1980). Furthermore, sensitivity and responsiveness of an individual
toward sanction threats were found correlating to deterrence. That is, if a person is both
sensitive and responsive to sanction threats, the person are more likely to be prevented

from committed deviant behaviors.

Boot Cm. Rehabilitation Goal, and Anflsocéial Attitudes.

Most of boot camp programs provide participants rehabilitative and treatment
components. Typically, these components include academic education, vocational
education, drug and alcohol education and treatment, therapeutic training, counseling,
etc. Some programs were specifically designed for subgroup offenders such as parenting

skills for fathers and mothers. The degree in which these programs are implemented

depends very much on the main goal of boot camp. Boot camps that focus on deterrence
goal (i.e. in Florida, Georgia, Louisiana, Oklahoma, Texas, and South Carolina) might
not have a wide variety of therapeutic components as do the boot camps that emphasize
the important of rehabilitation (i.e. in New York).

Namely, it is difficult to reach a consensus on the measure of success of boot
camp since all involved parties differ in their expectations of boot camp. Lower
recidivism, prison cost saving, the reduction of prison crowding, and the crime
commission rates are ofien used as success criteria in general evaluation studies. In the
present research, the change in participants’ antisocial attitudes and program attitudes
were used as a criteria to examine if boot camp helps rehabilitate participating offenders.
Unfortunately, there are not many researches that specifically studied the boot camp
effect on participant’s attitude.

The Oregon SUMMIT program (Success Using Motivation, Morale, Intensity,
and Treatment) recognizes the importance of attitude change of their inmate clients
(Beers & Duval, 1996). SUMMIT program administrators believe in the idea that
changing thinking results in a change in behavior. Basically, the program focuses on
cognitive change. The boot camp inmates are taught to see themselves as others see them
and take responsibility for who they are what they have done. Evidences show promising
outcome of SUMMIT program. Program administrators claimed that the program has
brought intensity to the change process and enhances the inmates’ abilities to internalize
new attitudes and behaviors (Beers & Duval, 1996).

According to MacKenzie & Souryal (1993), there have been two major models

proposed to explain the antisocial of prisoner: the Deprivation Model, and the

13

Importation Model. The Deprivation Model emphasizes that the antisocial attitudes
develop in response to environmental conditions. The assertions of this model are as

follows:

“... there are unique features of prison environment that negatively influence
inmates’ behavior and attitudes (Goodstein and Wright, 1989). As a result,
inmates form a normative system often called the inmate code that enable them
to ‘reject the rejecters.’ The inmate code is reflected in prisonized attitudes that
are anti-staff and anti-prison. Such attitudes are thought to be particularly
problematic because they conflict with positive behavior and motivation in
correctional programs. In general, inmates become more prisonized as time
passes in prison and inmates who are incarcerated in facilities that emphasize
custody more than treatment have more prisonized attitudes (Field, 1981).”

(MacKenzie & Souryal, 1993)

According to Harer & Steffensmeir (1996), the deprivation predicts that
Caucasian and Afiican American were slightly different, if at all, in prison violence.
They quoted, “It is because blacks in society are more brutalized than whites that they are
more aggressive in this wider context, and it is because black and white convicts are
equally brutalized in prisons that they are equally violent.”

Contrary to the Deprivation Model, the Importation Model assumes that antisocial
attitudes reflect the experience that offenders had before entering prisons. Therefore,
inmates bring with them to prison these attitudes rather than developed in response to

deprivations in prison. The Importation asserts that Afiican American are more likely

14

than Caucasian American to be violent at prison entry (Harer & Steffensmeir, 1996).
This is because this model assumes that in the larger society the level of violence of
African American is much higher than Caucasian American.

If the boot camp with its strict discipline and military-like atmosphere is viewed
as a custody oriented settings, one would expect increased prisonization and antisocial
attitudes. However, since some boot camp provide therapeutic programs that are
expected to rehabilitate inmates during the course of incarceration, one might expect a
less antisocial attitudes of inmates after they graduate. Paucity of evidence makes it
difficult to determine the impact of boot camp on the inmates’ attitudes.

Findings from one available study reveal that overall, boot camp graduates
become more positive about their boot camp experience after completing the program
(MacKenzie & Souryal, 1994). On the other hand, prison inmate as a comparison group
either did not change or be more negative about their traditional imprisonment.
Interestingly, both boot camp participants and prison inmate did not differ in term of their
antisocial attitude. Both became less antisocial. The researchers also found that time
devoted to rehabilitation, program rigor, and voluntariness related to greater reductions in
antisocial attitude.

In support to Deprivation Model, Morash & Rucker (1990) argued that boot camp
provides setting conducive to high level of unpredictability and contrived stress rather
than promoting rehabilitation. They suggested that boot camp environments are not likely
to encourage prosocial behaviors. Instead, the antisocial behavior is increased when

program staffs provide aggressive models for behavior. Similarly, MacKenzie and

15

Souryal (1994) also pointed out boot camp programs could not maximize their treatment
potential due to some limitations especially the authoritarian atmosphere of boot camp.

Evidences also show that individuals response distinctively to prison environment
and treatment program when race difference is addressed. Baldwin’s theory of Black
personality explained the unique development of norms, attitude, and personality of
Afiican American (Roberts, 1996). In her study, Roberts (1996) cited that factors that
affect Black identity include parental socialization, quality of interracial interaction, and
value oriented. Apparently, the roles of these factors are relatively different from other
ethnic group. The study indicates that the message transmitted from Afi'ican American
parents to their children about their heritage will definitely impact on the nature of their
racial identification. Mothers in Black families play an important role in the socio-
emotional environment. When considering the following dimensions with Caucasian
Americans: (1) harmony with nature vs. control over nature, (2) spiritualism vs.
materialism, (3) collectivism vs. individualism, (4) strong vs. weak religious orientation,
and (5) interdependence vs. separateness, African Americans score significantly higher
across the first dimensions (Robert, 1996).

Although the effect of boot camp on each individual might be different, Gowdy
(1996a) suggested that any subgroup of offenders can be targeted for participation in a
treatment boot camp as long as they share a particular problem that can be altered
through the use of treatment intervention provided by the program. To achieve this
treatment model, carefirl attention should be given to the design and all aspects of boot

camp programs for women and other minority groups.

16

Summm

Considering deterrence and rehabilitation goal of boot camp, no empirical
evidence has not yet found the promising result that boot camp programs provide more
effective mechanism than traditional imprisonment or any other community sanctions.
Rather, the strict discipline and aggressive environment were found related to an
increased stress of participants. The only vivid positive outcome, according to boot camp
participants, is the fact that they become more physically fit as a result of basic training.

As far as the antisocial attitudes, two models have been proposed to explain these
attitudes—the Deprivation Model and Importation Model. The Deprivation Model
proposed that the antisocial attitudes are the result of inmates’ response to prison
environment. On the other hand, the Importation Model argued that these attitudes
reflect the experiences offenders had before they entering prison. Therefore, offenders
import antisocial attitudes into prison.

The impact of boot camp on inmates’ antisocial attitudes has not yet been cleared.
Available evidence provides inconclusive answer. Previous research shows both positive
and negative effect of boot camp on participants. Emotionally, boot camps inmates
benefit differently from the program given their socio-demographic and criminal

background. Physically, most of the participants report the positive outcomes.

17

CHAPTER III

RESEARCH METHODOLOGY

Population and Sample3

The population of this study were convicted offenders in South Carolina who
were participating in boot camp program when the data were collected. It should be
noted that boot camp samples were randomly selected twice. The first boot camp
samples were drawn from those who participating in the program when it was being
administered by Probation, Parole, and Pardon Services in 1989. The second samples
were randomly selected firom those who participating the program when it was under the

Department of Corrections in 1991.

Data CoHecticmnd Instruments

The data used in this research were retrieved fi'om the Inter-university Consortium
for Political and Social Research (ICPSR) database’, which was originally collected by
MacKenzie & Souryal (1994). MacKenzie and Souryal (1994) collected data fi'om both
official records and inmates self-report questionnaires. Self-report questionnaires were

conducted at the beginning of their period of incarceration and once again at

 

3 For additional detail on data collection and sampling method, see:

MacKenzie, D. L. & Souryal, C. (1993). Inmate attitude ch_a_nge durinL incarceration: A
comparison of boot camp and traditional m’son (A final report to the National Institute of Justice). College
Park, MD: University of Maryland. And

MacKenzie, D. L., & Souryal, C. (1994). Multisite evaluéation of shock incarceration (Report No.
NCJ 142462). Washington, DC: US. Government Printing Office.

" Study No. 6986. Abstract, codebook, and data are available on-line on ICPSR web site.

18

approximately 90 days later. Researchers also conducted interviews with correctional
officers, boot camp inmates and parole agents supervising boot camp graduates. The
interviews helped verify the validity and reliability of self-report questionnaires.

A “record data” instrument was used to collect data on age, race, sex offense type,
sentence length, and prior adult felony arrests and convictions. The self-report
questionnaires consisted of Inmate Self-Report History and Inmate Self-Report Attitude.
Data collected by Inmate Self-Report History included employment, education, drug and
alcohol use. Meanwhile, Inmate Self-Report Attitude was used to measure changes in
attitudes over time. It was administered at the beginning of boot camp and again at
approximately 90 days later. Inmate Self-Report Attitude consisted of two scales: the
Jesness Antisocial Attitude Scale (See Appendix A) and Program Attitude Scale (See
Appendix B). Both Program Attitude Scale and Jesness’s Scale have 5 attributes like
Likert’s scale, ranging from “strongly disagree” to “strong agree.” The scores from both
scales were used as dependent variables of the study. Since Jesness Antisocial Attitude
Scale was developed and tested by previous researchers, its validity and reliability were
not tested again in this study. According to previous study, the Cronbach’s Alpha of
Jesness Antisocial Attitude Scale was .73 for South Carolina samples indicating that this
scale was quite reliable (MacKenzie & SouryaL 1993).

The reliability of Program Attitude Scale, however, was tested in this study (See
Appendix C). The Cronbach’s Alpha for Time 1 was .77 and Time 2 was .81. The
values of Cronbach’s Alpha were relatively large for both Time 1 and Time 2, which

means the scale was reliable.

Research Setting,

Boot camp in South Carolina is an alternative sentence to prison incarceration.
According to MacKenzie and Souryal (1994), there are two phases of the boot camp
administration in South Carolina During the first couple years, the boot camp had been
run by the Probation, Parole, and Pardon Services. Judges was the authority figures who
sentence offenders directly to the program. However, after the legislative changes in
1990, boot camp program has been technically under the supervision of the South
Carolina Department of Corrections (SCDC). The intent of this change was not to
change the operation of the program, but to alter the selection of boot camp participants.
The new legislation empowered the Department of Corrections to select participants fi'om
offenders sentenced to the SCDC. Apparently, the control over the placement offender
was transferred from judges to SCDC.

Eligibility criteria for boot camp that remains unchanged after the program was

transferred were as follows:

“(1) eligible for parole in 2 years, or if sentenced, convicted of an offense that
carries a sentence of at least five years (or returned for probation violation);

(2) offenders convicted of violent offenses...such as homicide or criminal sexual
conduct are ineligible; however, offenders convicted of offenses that are violent
in nature such as assault and battery, but are n_ot classified as violent by the
Omnibus Crime Control Act are still considered eligible;

(3) physically and mentally capable of participation;

(4) no previous incarceration in a state correctional facility or Shock probation/

incarceration program. . .and

20

(5) sentence that does not specifically prohibit shock incarceration.”

(MacKenzie and Souryal, 1994)

After the legislative change, age limit was changed from 17-24 years of age to 25
years or younger. For the “old” boot camp, boot camp participation and exit were
involuntary. However, according to the virtue of “new” boot camp, offenders who are
sentenced directly to SCDC and qualified for the program must voluntarily agree to
participate in boot camp. Offenders, though, have the right to drop out voluntarily. Upon
the graduation, offenders received moderate community supervision.

Boot camp in South Carolina targeted both male and female offenders who
convicted of nonviolent offenses. The main goal of the program was deterrence. The
setting was located in larger prison, while separate program was developed for female
inmates. The program length was 90 days. Even though the main goal of South Carolina
boot camp’s focus on deterrence rather than rehabilitation, offenders had to participate in
therapeutic programs. Offenders in South Carolina spent 4 hours every day in
educational programs and 3 hours each week in drug education. However, drug
treatment program was not among the high priority goals. Accordingly, offenders spent

little time in treatment.

Research Questions
Research questions of the present study are:
1. Are Caucasian Americans and African Americans different in terms of their

antisocial attitude before and after participating in boot camp?

21

2. Do Caucasian Americans and African Americans change their antisocial
attitude after graduating in boot camp?

3. Are Caucasian Americans and African Americans different in terms of their
attitude toward boot camp before and after participating in boot camp?

4. Do Caucasian Americans and African Americans change their attitude toward

boot camp after graduating from boot camp?

Hyp_otheses

The hypotheses of this study are as follows:

Hypothesis 1: After participating in boot camp, Caucasian Americans will
positively change their antisocial attitude.

Hypothesis 2: After participating in boot camp, Caucasian Americans will
positively change their attitude toward boot camp.

Hypothesis 3: After participating in boot camp, African Americans will positively
change their antisocial attitude.

Hypothesis 4: After participating in boot camp, African Americans will positively
change their attitude toward boot camp.

Hypothesis 5: Before participating in boot camp, the antisocial attitude of
Caucasian Americans and African Americans are different.

Hypothesis 6: After participating in boot camp, the antisocial attitude of
Caucasian Americans and Afiican Americans are not different.

Hypothesis 7: Before participating in boot camp, the attitude toward boot camp of

Caucasian Americans and African Americans are different.

22

Hypothesis 8: After participating in boot camp, the attitude toward boot camp of

Caucasian Americans and Afi'ican Americans are not different.

Variables

Independent Variables.

1. Socio-demographic Characteristics: race, status of sample, age, education,
employment, income, and military experience. All but education and employment were
collected from official data.

2. Criminal Characteristics: type of crime, prior adult arrest record, prior adult
conviction record, and length of sentence

Type of crime includes violent (robbery, assault and battery), property crime
(burglary, theft, and larceny), drugs, and other.

3. Other Characteristics: drugs and alcohol abuse index

Demndent Variables.

The dependent variables are samples’ scores from Jesness Antisocial Attitude
Scale and Program Attitude Scale. The Jesness Antisocial Attitude Scale measures
antisocial attitudes. Specifically, this scale was used to measure attitudes towards police
and authorities, level of maturity, and degree of social deviance. The Jesness Antisocial
Scale consisted of 30 items. The scale has been found to be associated with recidivism
and short-term change in behavior (MacKenzie and Souryal, 1994). For the
interpretation, the higher the Antisocial Social Attitude score, the more antisocial the

samples are.

23

Program Attitude Scale, on the other hand, consisted of 12 items that measured
the degree to which offenders expected their period of incarceration to motivate them to
change in a positive manner, and offenders’ believes that the boot camp will help them
make positive changes. The higher the Program Attitude Score, the more positive about

the program.

Data Analysis Procedures

To test the hypotheses, the author employed Bivariate and Multivariate
Regression as well as Independent Sample T-Test. It should be noted that missing values
in the data file were excluded from the analysis. Chapter IV presents the results of data
analysis. First, socio-demographic and criminal characteristics of Caucasian Americans
and African Americans were compared and discussed. Then, a paired sample t-test was
conducted to examine if Caucasian American and African American offenders changed
their antisocial attitudes and their attitudes about boot camp after completing the
program. Next, bivariate regression analysis was conducted to examine the relationship
between each socio-demographic and criminal characteristics and scores on Antisocial
Attitude Scale and Program Attitude Scale. This was conducted to examine if race was a
significant predictor of antisocial attitudes and program attitudes. Subsequent to the
bivariate regression analysis, an independent sample t-test was conducted to compare the
attitudes of Caucasian American and African American and to verify the influence of
race and antisocial and program attitude. Lastly, multiple regression analysis was
conducted to examine the relationship between race and scores on both scales when

controlling for other predictors.

24

mum

Data used in this study was retrieved from ICPSR web site. Samples were
randomly selected twice from convicted offenders who were participating in boot camp
program in South Carolina. Independent variables include socio-demographic and
criminal background of samples, while the dependent variables were scores from
Antisocial Attitude Scale and Program Attitude Scale. Both bivariate and multivariate
analysis were used to analyzed data. Results of the analysis were presented in Chapter

IV.

25

CHAPTER IV

DATA ANALYSIS AND RESULTS

Sample Characteristics

 

Socio-Demographic Characteristics.
Of all 242 samples in this study, 97 (41.1%) were Caucasian American (CA) and
145 (59.9%) were African American (AA). Chi-square analysis and independent sample
t-test were used to examine the difference in socio-demographic background between CA
and AA. According to Table 1, CA and AA were Significantly different fiom each other
in terms of their status, employment, age, and education. The majority of CA (62.9%,
=61) were old boot camp graduates, while approximately 66% (n=95) of AA were new
boot camp graduates who were included in the study after the boot camp program was
transferred to the Department of Corrections of South Carolina. At the time of crime
commission, 70.1% (n=68) of CA and 54.5% (n=79) reported that they were employed.
Apparently, the percentage of AA who did not have a job was much higher than of CA.
Further, the average age at prison entry of CA were 20.15 years, and significantly
younger than AA (20.98 years). Given their educational background, the average years
in school of CA and AA were 9.73 and 10.41 years respectively. It should be noted that
the number of years in school did not represent the intellectual capability of the sample.
Rather, it simply measured the length of years the samples had spent in educational

system.

26

Table 1
Socio-Demogr_aphic and Criminal Characteristics of Samples

 

 

 

 

Caucasian American (CA) Afi'ican American (AA)
Variables n % n %
Status of Sample"
"Old" boot camp graduates 61 62.90 50 34.50
"New" boot camp graduate; 36 37.10 95 65.50
Gender
Male 79 81.40 127 87.60
Female 18 18.60 18 12.40
Military Experience
No 93 95.90 139 95.90
Yes 4 4.10 6 4.10
Employment Status“
No 29 29.90 66 45.50
Yes 68 70.10 79 54.50
Type of Present Crime“
Violent 7 7.40 16 11.00
Property 56 58.90 55 37.90
Drugs 15 15.80 51 35.20
Other 17 17.90 23 15.90
Prior Adult Arrest
No 32 33.30 32 22.10
Yes 65 67.00 1 13 77.90
Prior Adult Conviction
No 51 52.60 72 49.70
Yes 46 47.40 73 50.30
Alcohol Use Index
Lesser Problems 53 55.20 94 65.70
1 14 14.60 22 15.40
2 8 8.30 8 5.60
3 11 11.50 10 7.00
4 7 7.30 4 2.80
Greater Problems 3 3.10 5 3.50
Drug Use Index“
Lesser Problems 52 53.60 108 75.00
1 15 15.50 10 6.90
2 9 9.30 9 6.30
3 5 5.20 11 7.60
4 9 9.30 3 2.10
5 5 5.20 3 2.10
Greater Problems 2 2.10 - -
Education (Years in School)"
Mean (S.D.) 9.73 (1.52) 10.41 (1.46)
Estimated Monthly Income
Mean (S.D.) 587.01 (565.52) 579.01 (563.56)
Age at Prison Entry (Y ears)”
Mean (S.D.) 20.15 (2.08) 20.98 (2.52)
Length of Sentence Months)"
Mean (S.D.) 36.41 (24.30) 28.47 (23.55)

 

Note. "‘ CA is significantly different fi'om AA (Chi-square significant at p < .05).
** CA is significantly different fi'om AA (T-Test significant at p < .05).

27

However, statistics did not Show the significant difi‘erence between the two
samples in terms of their gender, military experience, and income. Table 1 shows that the
majority of both CA and AA were males (81.4% of CA and 87.6% of AA). Additionally,
almost 96% of both CA and AA reported that they did not have military experience.
Their incomes were also not significantly different from each other; the average monthly

income of CA and AA were about 587 and 579 dollars respectively.

Criminal CMcteristics.

Criminal characteristics include type of crime, prior adult arrest, prior adult
conviction, and length of sentence. These data were retrieved from the official data by
the original study. Chi-square analysis and independent sample t-test were used to
examine the difference in criminal characteristics of Caucasian American and African
American.

Table 1 shows that CA and AA were significantly different fiom each other in
terms of their type of crime and length of sentence. Almost 60% (n=56) of CA were
convicted of property offense, while 15.8% (n=15) and 7.4% (n=7) were convicted of
drugs and violent crime respectively. The percentage of property offense of CA were
obviously higher than that of AA (37.90%, n=55). However, when taken into
consideration violent crime and drug crime, the percentage of AA was much higher than
of CA. About 35.2% (n=51) of AA were convicted of drugs crime while only 15.8%
(n=15) of CA with this same offense. Further, 11% (n=16) of AA were convicted of

violent crime while 7.4% (n=7) found guilty with this violent offense.

28

When comparing the type of crime, CA were convicted of more relatively serious
offenses than those of AA. One would expect to see harsher punishment for CA.
However, it was not the case in this study. Interestingly, the average length of sentence
of CA (36.41 months) were significantly lower than of AA (months).

Given prior adult arrest and prior adult conviction background, evidence did not
Show the significant difference between CA and AA. The majority of both CA (67%,
n=65) and AA (77.9%, n=113) had previous arrest record. As far as the conviction,
52.6% (n=51) of CA did not have previous conviction records. On the other hand, the
percentage of AA who had and did not have prior conviction record were as very close as

49.7% (n=72) and 50.30% (n=73).

Other Chflcteristics.

Other characteristics include drug use and alcohol use index. Independent Sample
T-Test was conducted to examine if CA and AA were different in terms of their drug use
and alcohol use index. The index ranges from Lesser Problems, 1, 2, 3, 4 to Greater
Problems. The result in Table 1 shows that CA and AA were significantly different from
each other only in term of drug use. Slightly more than half of CA (53.6%, n=52)
reported that they had lesser problem. Meanwhile as large as 75% (n=108) of AA

claimed to have lesser problem.

29

The Change in Antisogal Attitude an_d Program Attitude of
Qau_cas_ian American (CA) a_n_d African American (AA)

The results in Table 1 reveals that CA and AA were different in some
characteristics. These findings were going to be taken into consideration when
comparing their antisocial attitudes and program attitudes. Yet, before comparing the
difference between CA and AA in terms of their antisocial and program attitudes, a
paired sample t-test were conducted to examine if CA and AA had any change in their
antisocial attitudes and program attitudes after participating in boot camp. Separate
models were examined for CA and AA. In each model, the scores fi'om Time 1 and Time
2 were measured and compared to determine the change in their antisocial and program
attitudes overtime. The results were presented in Table 3 and used to test Hypothesis 1,
2, 3, and 4. Overall, the samples were less antisocial after participating in boot camp.
The average antisocial score significantly decreased fi'om 14.02 to 13.31 (p = .029).
However, even though the average program score of general sample increased at the end
of boot camp, their mean difference was not significant (p = .529). The results from the

analysis of CA and AA were as follows:

30

as .1. 83?. 535 .8 a a: u 83?. 383.2 8a a o
a u 833 838$ é .13 u 83?. .2828... é a ..

m _N u 83?. Seem 8. : .mNN .1. 83?. 388:5 é a .
nova _.

 

0:328:58.

536 mm. need we.» 3.6 3.5 503. 3.5 2.5V :56 e2 36.- 26 N06. 5o.n 3.2 86 3.2 qaomfi

 

9:895

39.: an Nem.~- 3.6 Sufi—1 510 3.3 8.5 3.: 89.: we 554m and .55.— $4. 8.2 3.9 8.3 528230

 

emmd in one: 8.5 3w. 8.5 8.: no.5 50.5w 3:... 3H mafia 36 .15.: 5Q». 3.2 $4. 8.3 $385.3.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

623.6 8.35 d.m :82 Hm :82 :8:an BEES d.m .52 dd :82
em a. a Hm 5.: N 85 BE... .5 e a NE .52 N SE. 3.5 035%
2.5.5. 8.5.5 8.5:... 3855.

 

 

 

 

N 2.5 ea 2 age 8833 ESE O < ”888 85% SSSSEQ as: ea afié 8. m .5
N 83¢

31

 

Hypothesis 1: After participating in boot camp, Caucasian Americans will
positively change their antisocial attitude.

As shown in Table 3, the average antisocial score of CA at Time 1 was

14.80. The average score decreased to 13.01 at Time 2. The difference between these

two means was 1.79 (p = .000). The results imply that there was a significant change in

antisocial attitude of CA afier participating in boot camp; that is, CA has become less

antisocial. The Hypothesis 1 was then supported.

Hypothesis 2: After participating in boot camp, Caucasian Americans will
positively change their attitude toward boot camp.

When considering the average program score, the mean difference of -1.58 (p =
.024) indicate that the average program score had significantly increased fiom 48.46 at
Time 1 to 50.04 at Time 2. The statistics reveal that the attitude toward boot camp of CA
had positively changed over time. They became more positive about boot camp after

they had completed the program. Therefore, Hypothesis 2 was supported.

Hypothesis 3: After participating in boot camp, African Americans will positively
change their antisocial attitude.

The average antisocial score of AA at Time 1 and Time 2 were 13.49 and 13.51
respectively. The mean difference was —0.02. Apparently, there was a very minimal
change in their average score. Given that the t-test was not significant (p = .961), the null

hypothesis that the two means were equal cannot be rejected. That is, the average

32

antisocial score of AA was not significantly different over time. Consequently,

Hypothesis 3 was not supported.

Hypothesis 4: After participating in boot camp African Americans will positively
change their attitude toward boot camp.

At time 1, the average program score of AA was 47.1]. It decreased to 46.57 at
Time 2. AA seemed to be less positive after they finished the program. However, this
change was not statistically significant since the significance level of t (p = .487) was
greater than .05. The null hypothesis that the two means were equal cannot be rejected,
indicating that the average program score of CA did not significantly different.

Hypothesis 4, therefore, was not supported.

The Bivariate Regession of Sample Characteristics and Scores
on Antisocial Attitude an_d Prom Attitude

Given the findings in Table 2, samples generally had become less antisocial after
graduating from the program. However, they did not significantly change program
attitude over time. In addition, separated paired sample t-test revealed that CA became
less antisocial and more positive about the program while AA had no significant change
in their antisocial and program attitudes. These findings imply that race seems to be a
significant predictor of antisocial attitudes and program attitudes. Thus, bivariate
regression analysis was conducted to examine if race really relates to those attitudes.
Further, to make sure that the study did not overlook the effects of other characteristics,

all other socio-demographic and criminal background were also examined. Since race,

33

status of sample, gender, military experience, employment, type of crime, prior adult
arrest, and prior adult conviction are qualitative variables, they were recoded into dummy
variables. For race, Caucasian American was coded 1, and Afiican American was coded
0. For status of sample, new boot camp graduate was value 1, and old boot camp
graduate was coded 0. For gender, male was coded 1, and female was coded 0. For
military experience, employment, prior adult arrest, and prior adult conviction, yes was
coded 1, and no was coded 0. Finally, type of crime was recreated into three dummy
variables: offense 1, offense 2, and offense 3. All attributes of type of crime then were
recoded to be compared to property crime. That is, for offense 1, violent crime was
coded 1, property crime was coded 0; for offense 2, drugs crime was coded 1, property
crime was coded 0; and for offense 3, other was coded 1, and property crime was coded

0. The result of bivariate regression analysis was shown in Table 3.

Sample Characteristics and Scores on Antisocial Attitude Scale.

According to Table 3, race was not a significant predictor at both Time 1 and
Time 2. This indicates that being CA or AA does not affect score on Antisocial Attitude
Scale. Rather, gender, alcohol use, drug use, and education significantly related to
antisocial attitude score at the beginning of boot camp program. After the sample
completed boot camp, status of sample, gender, offense 3, and alcohol use were the
significant predictors. Gender and alcohol use were the only two predictors that

significantly related to the scores both at the beginning of the program and after they

graduated.

34

Table3

Bivariate Regression Analysis for Socio-Demogr_aphic and Criminal Chmteristics and

Scores on Antisocial Attitude an_d Program AniLude

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Antisocial Attitude Score Program Attitude Score
Predictors Time 1 Time 2 Time 1 Time 2

B 3.5. R2 B s.5. R2 B 3.5. 52 B 5.5. 5’
Race 1 .16 .62 .02 -.52 .66 .00 l .27 .97 .00 3.16* .96 .05
(Caucasian, 1)
Status of Sample .51 .61 .00 L40" .64 .02 -1.7 .96 .00 -3.72 .94 .07
(New, 1)
Gender -l.86* .84 .02 -2.86* .89 .04 1 .46 1.34 .01 520* 1 .3 1 .07
(Female, 1)
Military -.92 1.52 .00 .30 1.63 .00 -2.29 2.39 .00 -3. 14 2.42 .01
(Yes, 1)
Employment .77 .62 .01 . 12 .66 .00 -.74 .98 .00 - 1 .84 .98 .02
(Yes, 1)
Offensel -.50 1.04 .00 -1.21 1.10 .01 -.57 1.63 .00 .13 1.64 .00
(Violent, l)“
OffenseZ .91 .68 .01 .32 .73 .00 -.40 1 .07 .00 -2.60* 1.07 .03
(Drug, 1)“
Ofimse3 1.52 .81 .01 2.08* 8.62 .02 -.87 1.28 .00 -.58 1.30 .00
(other, 1)"
Prior Arrest .62 .69 .00 .29 .73 .00 .64 1.08 .00 -.93 1.09 .00
(Yes, 1)
Prior Conviction -.22 .61 .00 -.65 .65 .00 .03 .95 .00 -.04 .97 .00
(Yes, 1)
Alcohol Use .66* .21 .04 .48* .22 .02 .04 .33 .00 -.27 .34 .00
Drug Use .76" .20 .06 .36 .22 .01 .32 .32 .04 .50 .33 .01
Education -.48* .20 .02 -.37 .22 .01 .70* .32 .02 .02 .32 .00
Monthly Income .00 .00 .00 .00 .00 .00 .00 .00 .01 .00 .00 .00
Age .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00 .00
Length of Sentence -.01 .01 .00 -.03 .01 .02 .00 .20 .00 .04 .02 .02
* p<.05
.. Compared to Property Offense

35

 

Time I

As shown in Table 3, females (B = -1.86) had less antisocial score than males,
meaning females were less antisocial at the beginning of boot camp. Being a female (R2
= 02), however, can predict only 2% of variation of the antisocial score. Further, the
positive relationship between alcohol use (B = .66) and antisocial score implies that the
samples who had more serious problems with alcohol were more antisocial. Likewise,
those who reported having more serious problem with drug use (B = .76) had a tendency
to be more antisocial. About 4% and 6% of variation of antisocial score can be
predictable by alcohol use (R2 = .04) and drug use (R2 = .06) respectively. On the other
hand, the negative relationship between education and antisocial score (B= —48) indicates
the fact that the higher the education the sample had, the less antisocial. Education (R2 =

.02) can predict 2% of variation of antisocial score.

Time 2

Status of sample, gender, offense 3, and alcohol use were significant predictors at
Time 3. For the status of sample, new boot camp graduates (B = 1.40) had a higher score
than old boot camp graduates, meaning new boot camp graduates were more likely to be
antisocial after completing boot camp. Only 2% of variation of the antisocial attitude
score can be predictable by status of sample (R2 = .02).

For gender, females (B = -2.86) again had lower score than males at Time 2. This
implies that females were less likely than males to have an antisocial attitude after they
graduated from boot camp. Gender (R2 = .04) as a single variable can explain only 4% of

variation of the antisocial score.

36

Offense 3 was a dummy variable of type of crime and had two categories
(property = 0, other = 1). In this study, “other” referred to any crimes other than violent,
property, and drugs. Table 2 shows that offense 3 (B = 2.08) was a significant predictor
of antisocial score, indicating that samples convicted of offense 3 were more antisocial
than those convicted of property crime. Predictability of offense 3 (R2 = .02) was 2% of
variation of antisocial score.

Alcohol use (B = .48) was also a significant predictor of antisocial score at Time
2. Samples who reported more serious problem of alcohol were more likely to be
antisocial even after completing the program. Even though alcohol use was a significant
predictor for both Time 1 and Time 2, the predictability was somewhat decreased. At
Time 1, alcohol use can predict 4% (R2 = .04) of variation of antisocial score, while it

decreased to 2% (R2 = .02) at Time 2.

SmeIe Ch_ar_acteristics and Score on Prpgram Attitude Scafi

For the program score, race was a significant predictor at Time 2. Table 3 shows
that CA compared to AA were more positive about boot camp program after they
graduated. Being CA can predict 5% of variation of program score. In addition, other
predictors that significantly related to program score were gender, offense 2, and
education.

Time I

From Table 3, at the beginning of boot camp, education was the only predictor
that significantly related to program attitude score. The positive relationship between

education and program score (B = .70) indicates that the higher the education, the more

37

positive about boot camp. The predictability of eduCation was 2% of variation of
program score.

Time 2

At the end of boot camp, race, gender, and offense 2 were significant predictors of
program attitude score. Table 2 reveals that females’ program score (B = 5.20) were
higher than males at Time 2. That is, females were more positive about boot camp after
completing the program. Predictability of gender at Time 2 was 7% (R2 = .07), indicating
that when the samples were females, one can predict 7% of variation of program score.

Offense 2 was also a significant predictor of program score at Time 2. Offense 2
was a dummy variable of type of crime, representing drug crime in comparison to
property crime (drug crime = 1, and property crime = 0). As shown in Table 2, the
negative relationship between offense 2 and program score (B = -2.60) revealed that the
samples convicted of drug crime had less score than those convicted of property crime.
That is, compared to those convicted of property crime, samples convicted of drug
offense graduated from boot camp with less positive attitude about the program. Further,

offense 2 (R2 = .03) can predict 3% of variation of program score at Time 2.

_T_h_e Average Scores on An‘tisocial Attitude Scale and Program Attitude Scale:
A Comison Between Caucasian American (CA) an_d African Ameri9can (AA)

From Table 3, the results of bivariate regression analysis indicate that race was a
significant predictor only of program attitudes at Time 2. To verify these findings, four
separate models of independent sample t-test were conducted, comparing the average

scores on both scales of CA and AA at Time 1 and Time 2. The results were presented in

38

Table 4 and 5. Levene’s test for equality of variance (F value) in Table 5 was used to
determine if the standard deviation of the two samples were equal. F value helps
determine which t-test value should be used; if the F value was not significant, one would

assume the equal variance of the two samples and vice versa.

Antisocial Score.

At Time 1, the average antisocial score of AA was 13.61 while the average of CA
was 14.77. Given these average scores, CA seems more antisocial than AA at the
beginning of the program. However, statistics in Table 5 does not support this
anticipation. Given the significant F value of Antisocial Scale at Time 1 (F=5.610, p =
.019), the t value of —.1963 (p = .051) was assumed. Since the significance level of t was
greater than .05, the null hypothesis that the two means were equal cannot be rejected.
This implies that the average scores of CA and AA at Time 1 were not different.

Table 4 shows that the average antisocial score of AA and CA at Time 2 were 13.52 and
13.01 respectively. As shown in Table 5, the F value of .395 (p = .530) was not
significant; therefore, t value of .780 (p = .436) was assumed. Given the significant of t
(p > .05), the null hypothesis that the means of the two means were equal cannot be

rejected. This implies that the average score of CA and AA were not different.

39

Table 4

_The Average Scores on Antisocial Attitude Scale and Program Attitude Scale: A

Comarison of Caucasian American (CA; anAAfiican American (AA)

 

 

 

 

 

 

Scales Race N Mean SD. SE.

Antisocial Scale: Time 1 AA 144 13.61 5.01 0.42
CA 97 14.77 4.14 0.42

Antisocial Scale: Time 2 AA 141 13.52 5.06 0.43
CA 95 13.01 4.84 0.50

Program Scale: Time 1 AA 135 47.00 7.31 0.63
CA 94 48.27 7.02 0.72

Program Scale: Time 2 AA 136 46.70 7.52 0.64
CA 92 49.86 6.47 0.67

 

 

 

 

 

 

Eot_e One might notice that the mean scores of samples in this table are different

fi'om ones presented in Table 2. This results from the difference in the number of
samples included in the analysis. In Table 2, each sample must have scores at both

Time 1 and Time 2 so that they can be compared in the analysis. Samples who had a
score only Time 1 or Time 2 were excluded from the analysis. On the other hand, in
Table 4, the analysis foucused on the difference between CA and AA, not the difference
between Time 1 and Time 2. Therefore, those who had a score only for Time 1 or Time 2
were included in the analysis.

40

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.Nm... 8...“. NS... 8.8. .8... eNN.m.N e35- .8383 8.. 883.3 .35
SN... N3“- .3... 8.3. .8... eNN 8N5 RN... 3... .8388 83.3., .35 m 8.3 8.8m 8335
“Ne... $3. 23... 8N..- 3.... SNSN CNN... .833 8e 83:3 .35

5... $3- e3... SN... .2... NNN :N..- E... as... .8388 883.8.» .35 _ on: 8.8m 8335
8.... Ed. 3...... :3 N3... SEEN SN... 8538 .3 80353 .35

3.... 9.5.? 8e... :2. e9... 3N one... on... an... .833 80353 .35 N 08.... 8.8m 308.22.
82. eNNN- New... N2..- .8... NSeNN m8... 833 8.. 83.3 .35

m3... NNNN- 3e... N2... o8... 3N New..- 8...... 2e.“ 83.3 823.3 .35 . 03:. 8.8m 308.23.
895 .633

=85 2.. 8 .388. 8855 83.55 825$ 8 . .5 a
8.8330 .83 .md 582 3m
83.5 .8 8.3....
382 8 3.35 8.. .8...

 

com 80% $23.64

 

 

 

 

m 033.

41

Program Score

At time 1, the average program score of AA as shown in Table 4 was 47.00 while
CA had an average score of 48.27. Apparently, CA had a higher average score than AA,
indicating that at beginning of the program, CA were more likely to be positive about
boot camp. However, evidence in Table 5 does not show the significant difference
between the two means. Considering the F value of .084 (p = .773), the equal variances
of the two samples were assumed. Given the t value of —1.311 with the significance of
0.191 (p > .05), the null hypothesis was accepted. That is, the average program scores of
CA and AA were not different.

At Time 2, the average program scores of AA and CA in Table 4 were 46.70 and
49.86 respectively. Again, CA tended to be more positive about boot camp since they
had a higher average score. As shown in Table 5, the significant of F (F = 1.185, p =
.277) was greater than .05, the equal variances of the two samples were assumed. The t
value of —3.290 (p = .001) with the significance level of less than .05 results in the
rejection of the null hypothesis that the two sample means were equal. Statistically, the

attitudes toward boot camp of CA and AA at Time 2 were significantly different.

The Multiple Regression Analysis for Sample Characteristics and
Scores on Antisogial Attitude Scale___ and Prcgram Scale

Table 3, 4 and 5 indicate that the results of bivarate regression and independent
sample t-test were consistent. That is, race was a significant predictor only for program
attitude at Time 2. However, when conducting a bivariate regression, the author focused

only on the effect of race on antisocial and program attitudes without controlling other

42

predictors. Thus, it was questionable if the effect of race was the same when controlling
for other predictors. Therefore, to verify the effect of race when controlling for other
predictors, four separate models of multiple regression analysis were conducted—
Antisocial Attitude Time 1, Antisocial Attitude Time 2, Program Attitude Tirnel, and
Program Attitude Time 2. Predictors in each multiple regression model were race, status
of sample, gender, military experience, employment, offense 1, offense 2, offense 3, prior
adult arrest, prior adult conviction, alcohol use, drug use, education, income, age, and

sentence length. Table 6 presents the results of multiple regression analysis.

Antisocial Attitude Model.

As presented in Table 6, R2 statistics shows the predictability of each model. For
the antisocial score, when all predictors were included together in one model, they can
predict 25.4% (R2 = .254) of variation of antisocial score in Time 1 model and 21.7% (R2
= .217) of variation of antisocial score in Time 2 model. When considering each
predictor in the models, race was not a significant predictor in both Time 1 and Time 2

model.

43

Table 6
Multiple Regression Analysis for Sample Charaflristics and Scores on An‘tisociaLAttitud
and Program Alitude Scale: An Analysis for All Sample

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Antisocial Attitude Program Attitude
PREDICI‘ORS Time 1 Time 2 Time 1 Time 2

B SE. B SE. B SE. B S.E.
(Constant) 20.33 3.89 20.1 1 4.39 33.83 6.59 41.96 5.52
RACE (Caucasian, I) 1.42 0.82 -0.70 0.90 2.34 1.35 2.13 1.16
SAMPLE (New, I) 1.93 1.06 1.53 1.18 0.43 1.79 -2.18 1.53
GENDER (Female,1) -2.71* 1.18 471* 1.30 1.68 1.91 4.05“ 1.64
MILITARY (Yes, I) -2.21 1.74 -0.72 1.91 -4.59 3.05 -5.85* 2.61
EMPLOYED (Yes, 1) 0.74 1.10 -0.07 1.21 -0.83 1.86 -1.04 1.55
OFFENSE l(Violent, I )" 0.01 1.23 -0. 12 1.35 -0.51 2.00 -0.51 1.79
OFFENSEZ (Drugs, 1)“ 1.83* 0.89 -0.02 0.99 -0.50 1.48 -2.15 1.25
OFFENSE3 (Other, I)" 2.34" 1.04 3.04* 1.15 -1.91 1.71 -2.78 1.50
ARREST (Yes, I) 0.57 1.01 0.46 1.13 349* 1.69 1.15 1.42
CONVICI‘(Yes, 1) -0.21 0.92 -0.81 1.01 -2.12 1.53 -0.69 1.33
ALCOHOL 0.05 0.27 -0.09 0.30 0.26 0.44 0.01 0.38
DRUG 0.70* 0.26 0.54 0.28 0.06 0.43 0.25 0.36
EDUCATION -0.62* 0.25 -0.45 0.28 1.08* 0.41 0.09 0.36
INCOME 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
AGE -0.15 0.18 -0.10 0.20 0.17 0.31 0.33 0.26
SENTENCE 0.01 0.02 -0.01 0.02 -0.03 0.03 ~0.01 0.03

R2 0.254 0.217 0.134 0.223
n

"' p < .05
** compared to property crime

 

Antisocial Attitude Time I

Hypothesis 5: Before participating in boot camp, the antisocial attitude of
Caucasian Americans and African Americans are diflerent.

In Time 1 model, race was not a significant factor of antisocial attitudes. That is,
given their socio-demographic and criminal characteristics, CA and AA were not
different in term of their antisocial attitudes at the beginning of the program. Therefore,
Hypothesis 5 was not supported Instead, gender (B = -2.71), education (B = -.62),
offense2 (B = 1.83), offense3 (B = 2.34), and drug use (B = .70) were significant

predictors of antisocial attitudes at the beginning of the program.

Antisocial Attitude Time 2

Hypothesis 6: After participating in boot camp, the antisocial attitude of
Caucasian Americans and African Americans are not diflerent.

Antisocial Attitude Time 2 model can predict 21.7% of variation of antisocial
score. The relation of race and antisocial attitudes in this model was not significant when
controlling for other predictors. This finding implies that after participating in the
program, the antisocial attitudes of CA and AA were not different. Thus, Hypothesis 6
was not supported. In this same model, however, gender (B = -4.71) and offense3 (B =

3.04) were significant predictors.

45

P_rogram Attitude Model.

Given R2 in Program Attitude model shown in Table 6, the predictability of the
models was 13.4% (R2 = .134) at Time 1 and 22.3% (R2 = .223) at Time 2. It means that
when including all predictors together in the same model, they can predict 13.4% of
variation of program score at Time 1, and 22.3% of variation of program attitude at Time

2.

Program Attitude Time I

Hypothesis 7: Before participating in boot camp, the attitude toward boot camp of
Caucasian Americans and African Americans are different.

In Program Attitude Time 1 model, race was not a significant predictor of
program attitude when controlling for other predictors, meaning CA and AA were not
different in terms of their program attitude at the beginning of the program. Therefore,
Hypothesis 7 was not supported by this finding. Only prior adult arrest (B = 3.49) and
education (1.08) were significant with program attitude in this model. Program Attitude

Time 1 model can predict 13.4% of variation of program score.

Program Attitude Time 2

Hypothesis 8: After participating in boot camp, the attitude toward boot camp of
Caucasian Americans and A fiican Americans are not different.

In Program Attitude Time 2 model, the relation between race and program
attitudes was not significant when controlling for other predictors. That is, when

controlling for their socio-demographic and criminal background, the attitudes about boot

46

camp of CA and AA were not significantly different. This finding is contrary to the
finding from bivariate regression analysis presented in Table 2. The effect of race on
program attitude at Time 2 disappeared when controlling for other predictors. Therefore,
Hypothesis 8 was not supported. Gender (B = 4.05) and military (B= -5.85), on the other
hand, were significantly related to program attitude in this model. Predictability of this

model were 22.3% (R2 = .223) of variation of program attitude.

The Multiple Regression Anglysis : A Compgrison of Cayggsgm Americans and

African America_n§.

From Table 6, multiple regression analysis revealed that race had no significant
effect on both antisocial and program attitude scales when controlling for other
predictors. Most previous research using race as a predictor stopped at this stage of the
analysis in their examination of the effects of race on program outcomes. This might lead
one to assume that race is not a significant predictor of program outcomes. However, it is
possible that race may interact with other characteristics such as type of offense or
employment; such an effect cannot be detected in models such as the one presented in
Table 6. Therefore, even though the results fiom Table 6 indicate a non-significant effect
of race when controlling for other predictors, another multiple regression models were
conducted to examine if the relation between predictors and antisocial and program
attitudes of CA and AA were different. Four multiple regression models were conducted
for each ethnic group--Antisocial Attitude Time 1, Antisocial Attitude Time 2, Program
Attitude Time 1, and Program Attitude Time 2. All socio-demographic and criminal

characteristics were included the models. The results were presented in Table 7.

47

2...... atone-a 8 venues i.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3v...

2. .. a 8 fl .3 R a

1.... .8... N8... N8... 3.... 3.... N3... 3.... NM.
2.... 8... 2.... 5.... mm... 8.... a... 8... 3... 8...- .o... 8.... 8... 8... 8... 8...- 82.32..
3... R... 9.... t... on... N... a... 8.....- 9.... 8... 3... 3....- m... 8.... .N... .2.- .9.
8... 8... 8... .8... 8... 8... 2.... .8... 8... 8... 8... 8... 8... 8... 8... 8... .282.
3... 8...- ...o ..u.. a... .3...- ...... 3...- ...o N... n... .2... 3... .8...- om... 9....- 20:522.
3... 2... S... 8...- ...o a... .2. 9.... 3... .2. m... 3...- R... .2. 3... .9... 03.2
2... .N... .2. 2.... 3... 3... mm... 3... S... o...- m... 3... 3... .2...- 3... .2.- .0203...
88 3.. SN 3... 3... 8.”- ..I .8..- .2 .2...- ME 9..”- 3... 3.. m... .2 2.825528
.3 .2- .2 3.. b... m2 8.. 8.. a. .3... 5.... .3... 3.. 3... 3.. 2.... 2.82355.
.2 .2- 2.” 3...- N... .2 h... m... .2 N3- 3. .3.- .E .3... .2 3.” .....§..o.m.mzm.n.-..o
:... .2. ..... 2.. 3.. 3.... .2 2.8 3.. .2...- Sa .8... m... .N. .N. a... 12.335.332.30
3. .5...- NE NE 8.. S..- m... s... .2 3a- .2 a..- 3.. .2 m... N...- :2..E...>..mmzm.&o
.2 o... a... 9...- a... 3...- ... .... N2 N2. 8. .2- m... a.” a... 2.... ....8>.2m>o.§m_
R... 8... a... .m..- .2. 3.. .2 .5- m... x...- 8. .3... EN 2....- ..2 .m..- 2 8.5.5.53...
8.. ..... 8.. o... .3 .3... NE ...N- z... o: 8.” E... n... .3...- mm. .2...- 2.2.5.5595
m2 2...- ...N t... a... 3...- m... 3.. ....N .2- 8.” :3- 3. .2 3.. .2 2 382......25
3... 2.2. v... :3 3.. fl... 5.. a... .3 2.... R... 8.... 3.. 3.8 8.. 3.2 23.38.
.m... m m... m .m... m 4.... m m... m m. m .m... m s... m
45.; . 8...: N 2:: . as; N may . 81.1.: N .15: . 2:...

8.3..=<ammm...lul 33.5.2833... .3...<§§.| 8.3.2.2325 9.08.8...

7.52.2... 25...“... 252323.525

 

 

 

 

.- 05....-

48

glucasian American and Scores on Antisocial Attitude.

Table 7 shows that the predictability of Time 1 model was 42.4% (R2 = .424) of
variation of antisocial score. Gender (B = -3.00) and drug use (B = .63) were significant
predictors. It means that Caucasian females who had less serious drug problem scored
lower on antisocial scale than Caucasian males or females who had more serious drug
problem.

Time 2 Model, on the other hand, can predict 39.2% (R2 = .392) of variation of
antisocial score. In this model, gender (B = -4.96), offense3 (B = 4.65), alcohol use (B =
-.76), and education (B = -0.81) were significantly related to antisocial score. These
statistics reveal that Caucasian females with higher education and more, convicted of
property crime, and even though had more serious alcohol problem scored higher on
antisocial score than Caucasian males or females who convicted of crime other than

drugs and violent, had a lower education, and had less serious alcohol problem.

@ucasiafin Americmd Scores on Program Attitude.

According to Table 7, Program Attitude Time 1 model can predict 19.0% (R2 =
.190) of variation of program score. Two predictors were significant: arrest (B = 5.67)
and education (B = 1.40). It means that Caucasian with higher education and had been
arrested before scored higher on program score than Caucasian who had lower education
and never been arrested. In other words, at the beginning of boot camp Caucasian with
higher education and had been arrested before were more positive about the pro gram.

When considering Time 2 Model, offense2 (B = -4.12), arrest (B = 6.39), and

convict (B = -4.93) were significantly related to program score. The predictability of

49

Time 2 model was at 33.9% (R2 = 33.9), meaning Time 2 model can predict 33.9% of
variation of program score. The implication of this model is that Caucasian convicted of
drug crime, had no prior arrest record, and had prior conviction scored lower than
Caucasian convicted property crime, had prior arrest record, and had no prior conviction
record. That is, Caucasian convicted of drug crime, had no prior arrest record and had

prior conviction record were less positive about boot camp after graduation.

African Americh Scores on Antisocial Attitude.

Table 7 shows that even though the Antisocial Attitude Time 1 model can predict
23.2% (R2 = 23.2) of variation of antisocial score, none of predictors was significantly
related to antisocial score.

In Time 2 model, gender (B = -4.76) was the only predictor that related to
antisocial score. The negative relationship means that AA females who scored lower on
antisocial score than AA males, indicating that AA females were less antisocial. This

model predicts 23.2% (R2 = .232) of variation of antisocial score.

Afiican Americafld Scores on Proggam Attitude

Program Attitude Time 1 model in Table 7 reveals that only education (B = 1.21)
was significant with program score. The positive relationship implies that AA who had
higher education scored higher on Program Attitude Scale than AA who had lower
education. In other words, AA with higher education were more positive about boot

camp. Given R2 of .19, Time 1 model predicts 19.0% of variation of program score.

50

Additionally, as shown in Table 7, Time 2 model reveals the predictability of 21.4%.

However, none of predictors I the model significantly related to program score.

Multiple regression analysis in Table 6 revealed that race had no significant effect
on both antisocial and program attitudes when controlling for other predictors. As
mentioned earlier, most previous research using race as a predictor stopped at this stage
of their analysis. This led one to assume that race was not a significant predictor of
program outcomes. In the present study, however, when examining the interaction
between race and other predictors, the results provided more informative explanation
about the effect of race on antisocial attitudes and program attitudes. As presented in
Table 7, the results indicate that it was possible that race could interact with other
characteristics such as gender and education and then affected samples’ antisocial and
program attitudes. Such an effect could not be detected in models such as the one
presented in Table 6.

In contrast with Table 6, the results in Table 7 revealed race potentially affected
antisocial and program attitudes of samples when it interacted with other socio-
demographic and criminal characteristics. Apparently, CA and AA were different in
terms of significant predictors of their antisocial and program attitudes. Most of the
significant predictors in Table 6 remained significant in Table 7 for CA, whereas almost
all the significant predictors of antisocial and program attitudes of AA disappeared.
Overall, traditional predictors commonly used to look at the effects of the program were
applicable only to Caucasian Americans and did not say any thing much for African

American.

51

Gender and education were the only significant predictors of antisocial and
program attitudes of AA. When race interacted with gender, AA females were less
antisocial than AA males after participating in boot camp. This result was contrary to
previous studies that boot camp was not appropriate for females to develop their positive
attitudes. As for education, AA who had higher education were more positive about boot
camp at the beginning of the program than those with lower or no education. In other
words, the higher the education, the more positive about boot camp. After participating
in boot camp, however, the average scores of Afiican Americans on Program Attitude
Scales decreased. Even though this change was not significant, it implies that boot camp
did negatively affect some African Americans. Given these findings in terms of program
implication, it is essential for boot camp administrators to recognize the difference in
significant predictors of CA and AA antisocial and program attitudes. Results from
Table 7 suggest that CA were more receptive to the program and sensitive to traditional
predictors. The unchanged of attitudes of AA as indicated in this study may result from
the fact that boot camp failed to address the different needs of AA. The implications of

these findings were thoroughly discussed in Chapter V.

52

CHAPTER V

DISCUSSION

Antisocial and Prom Attitude of Boot Camp Participants

 

This study was designed to examine the antisocial attitude of boot camp
participants across ethnic groups. The author hypothesized that both Caucasian
Americans and African Americans should positively change their attitude after
participating in boot camp, and that the attitude of the two samples should be different
given their socio—demographic and criminal background. As mentioned in Chapter II,
there have been two major paradigms proposed to explain antisocial attitude of prisoners:
Deprivation Model, and the. Importation Model (MacKenzie & Souryal, 1993). The
Deprivation Model emphasizes that the antisocial attitudes develop in response to
environmental conditions. According to Deprivation Model, the difference between
Caucasian American and African in term of their antisocial attitude in prison should not
be minimal since they both are equally brutalized by prisons. Meanwhile, the
Importation Model asserts that antisocial attitudes reflect the experience that offenders
had before entering prisons. Prisoners “imported” these attitudes into prison rather than
developed in response to deprivation atmosphere of prison. Given the reality in larger
society, African American should be more violent or antisocial than Caucasian American.
If the boot camp with its strict discipline and military-like atmosphere is viewed as a
custody oriented settings, one would expect increased prisonization and antisocial

attitudes. And there will not be much different between Caucasian American and African

53

American. On the other hand, some boot camp that provide therapeutic programs might
expect less antisocial attitudes of inmates after they graduate.

When taken into consideration the components of South Carolina boot camp, the
main goal of program was deterrence. However, there had been an increased focus on
“reimbilitative” activities such as education and pre-release counseling. As a
requirement, offenders placed in boot camp had to participate in these programs. In fact,
offenders in South Carolina spent 4 hours every day in educational programs and 3 hours
each week in drug education. Despite, since drug treatment program was not among the
high priority goals, offenders spent little time in treatment. Given the overall program
elements, the less antisocial attitude and the more positive about the program of South
Carolina participants were expected in this study.

Given the difference between antisocial score at the beginning and at the end of
the program, the results of this study show that general samples became less antisocial at
the end of boot camp, even though their attitude toward boot camp program and staff did
not significantly change over time. These findings are somewhat consistent with the
study of MacKenzie and Souryal (1993). The evaluation of 8 boot camp programs
insinuate that boot camp graduates became less antisocial and more positive about the
program and the staff at the end of the program.

When examining Caucasian American and African American samples in separate
models, evidences from the present study show that the antisocial attitude and program
attitude of Caucasian offenders significantly changed overtime while African American
did not significantly change theirs. Caucasian American became less antisocial and more

positive about the program. These findings support previous study showing that some

54

rehabilitation programs were successful with some offenders (Andrews et. al., 1990;
Antonowitz & Ross, 1994; Gendreau & Ross, 1987), and that prison may have some
positive effects on particular group of inmates (Goodstien and Wright, 1989).

Since Caucasian American did change both attitudes over time while African
American did not, one would expect that race might be a significant predictor of both
antisocial attitude and program attitude; that is, Caucasian American and African
American should be different in terms of their antisocial attitude and program attitude.
However, when comparing the antisocial attitude and program attitude of Caucasian
American and Afiican, the results of multiple regression analysis indicate that their
attitudes were n_ot significantly different at any time given their socio-demographic and
criminal background. These results were inconsistent with the assertion of Importation
Model that African American should enter prison with more violent attitude was not
supported.

Even though the statistics did not show the significant effects of race on antisocial
attitudes and program attitudes for general samples, more in-depth examinations of
Caucasian Americans and Afiican Americans yield different results. That is, when race
interacted with other socio-demographic and criminal characteristics, the antisocial
attitudes and program attitudes of Caucasian Americans were somewhat different from
African American. Apparently, traditional predictors commonly used in assessing
programs and seem applicable only to Caucasians and did not really say anything for
African Americans. The antisocial and program attitudes of Caucasian Americans were
affected by gender, offense2, offense3, drug use, prior adult record, prior adult

conviction, and education. Whereas, almost all socio-demographic and criminal

55

characteristics of Afi'ican Americans were not significantly related to their antisocial and
program attitudes. Only gender was significant with African Americans’ 5 antisocial
attitudes at Time 2, and education was significant with their program attitudes at Time 1.
There are some explanations for these findings. Given the results of samples’
characteristics analysis in the present study, it is possible that African Americans need to
have higher qualifications than Caucasian Americans in order to be qualified for boot
camp. For instance, they were older than Caucasian Americans, and had a higher
education. They also had shorter length of sentence. This might make boot camp more
punitive to African Americans than Caucasian Americans. Unsurprisingly, after they
completed the program African Americans did not significantly change their attitudes
about boot camp while Caucasian Americans became more positive.

Furthermore, the difference in the attitudes of Afi'ican Americans and Caucasian
Americans might result from the difference in their value orientation as discussed in
Chapter II. Compared to Caucasian Americans, African Americans have a more
communal and interdependent lifestyle, and presents stronger religious orientation
(Robert, 1996). Previous research also suggest that high rates of violence among Afi'ican
Americans might stem from a subcultural value system, and the roots of this subcultural
value might be socio-historical conditions of economic hardship and social isolation
(Harer, 1996). According to Harer (1996), “blacks are more likely than whites to
perceive situations or environments as being aggressive or dangerous...[and] these
expectations of dangerousness lead to aggressive or violent responses aimed at protecting

self or preventing retaliation.”

56

Given the significance of education as a predictor of program attitudes when
interacting with race, African American who had higher education were more positive
about boot camp at the beginning of the program than those with lower or no education.
This might be because African Americans with a higher education were more exposed to
disciplines, rules, and norms of the large society. Thus, the education system might
contribute to their receptive attitudes. These findings suggest that the higher the
education, the more positive about boot camp. After participating in boot camp,
however, the average scores of Afi'ican Americans on Program Attitude Scales decreased.
Even though this change was not significant, it implies that boot camp did negatively
affect some African Americans. Therefore, in term of program implication, when
designing rehabilitative elements for boot camp, it is essential for boot camp
administrators to prioritize educational program that addressing the difference between
Afiican Americans and Caucasian Americans.

When race interacted with gender, results were consistent with previous studies
that male and female offenders adjusted differently in boot camp. Afier participating in
boot camp, both Caucasian and African American females became less antisocial than
males. Namely, findings were contrary to previous studies that expecting females to be
more negative and more stressfhl than males. Morash & Rucker, (1990) suggested that
the environment of boot camp is not healthy for women to develop or improve their
attitudes and behaviors. Male-dominant atmosphere and the emphasis on the masculinity
cause women more stress. One possible explanation for a positive attitude of females in
boot camp might be the relationship between female inmates and staffs. Previous study

found that thoughtful, professional male drill instructors have been very effective with

57

women in the program (Clark, 1996). Evidences show a positive feedback from female
boot camp graduates for their male drill instructor. In addition, female staff who are
confident and clear about who they are as women are invaluable role models for female
offenders.

Program administrators should consider gender difference issues along with race
difference when designing the program for boot camp inmates. Researchers suggested
that current eligibility criteria of boot camp, which originally were developed for men,
limit the number of women in boot camp. Moreover, most of studies about boot camp
addressed that the need to recognize the women’s needs in rehabilitative programs that
was overlooked by many programs in the past (Clark, 1996; Gowdy, 1996a; MacKenzie,
Ellis, Simpson, & Skroban, 1996). Since the majority of female boot camp participants
are mothers and have abusive relationships, Clark (1996) addressed that any boot camp
program for women must focus on parenting, issues of abuse, and domestic violence.
Health and nutrition are also equally needed. These programs have the potential to build
self-esteem in women and help them to take a positive control of their lives.

When collecting the data for this study, MacKenzie and Souryal (1993) suggest
that none of offenders thought that the boot camp was easy given the living conditions,
strict discipline, and daily activities. Therefore, given the explanation of the Deprivation
Model, the attitudes of samples should be more negative rather than positive. However,
the present study found that Caucasian American significantly developed positive attitude
toward program and staff after completing the program Why so? Beside the explanation
of Goodstien and Wright (1983) that prison might have some positive influence on some

inmates, this might result from a Hawthorne effect since South Carolina boot camp was a

58

highly visible program and the participants knew they were being studied MacKenzie and
Souryal (1993). When the data were collected, the researchers noticed that the program
had committed staff who were vested in making boot camp ‘Vvorks”. Interviews with
samples revealed that staff were viewed as helpfiil and caring. This might be a possibility
of why inmates develop positive toward program. Yet, if a positive program attitude was
a result of Hawthorne effect, African American should have also had a positive program

attitude. More in-depth study in the firture should provide more informative answer.

Who Benefit from Boot Car_np;?

Even though the results reveal that the antisocial attitude of Caucasian and
African American were not significantly different, there was a significant change in the
attitudes of Caucasian American overtime. That is, Caucasians became less antisocial
and more positive about the program. This might lead one to assume that Caucasian
Americans benefit more from boot camp than African American. However, findings
fi'om the present study did not provide a clear answer whether or not these changes
resulted from boot camp participation. Since original data were not originally collected
for this study, some predictors that might have been used to examine the effect of the
program were not found. Such predictors include race and gender of instructors, the
relationship between offenders and instructor, and the number and types of treatment
elements.

It should be noted that antisocial attitude per se is not necessarily a major factor
that produces criminal behaviors (Andrews et. a1, 1990). Being more antisocial alone

does not mean that one will commit more crime or even has criminal behaviors. Rather,

59

as presented in this study, antisocial attitude were associated other criminogenic needs of
offenders such as drug and alcohol use. When examining the interaction between race
and substance abuse history, these predictors significantly related the antisocial attitudes
of Caucasian offenders. These factors somehow encourage less antisocial individual to
commit criminal activities just to support their habit. Failure to address these
criminogenic needs in the program may results in the fact that the program does not help
rehabilitate offenders at all. For example, most of the samples in this study reported
having either drug or alcohol problem or both. Unfortunately, the primary goal of South
Carolina boot camp was deterrence and did not focus on drug treatment program Given
this scenario, it was no surprise that general offenders did not change their attitudse

toward program.

Future Research
The present study leads to some issues regarding boot camp program and boot
camp participants. First of all, since the present study focused only on the antisocial
attitude and program attitude of Caucasian and Afiican American offenders in boot camp,
the results do not provide the information on the level of antisocial attitude compared to
those who were sentenced to other programs. Future research might address this issue by
comparing the attitude of Caucasian and African American boot camp participants with
those who are prison inmates, or probationers, or parolees.
Next, although results show that Caucasian American who graduated from boot
camp became less antisocial and more positive about the program while African

American did not change their attitude at all, there was no statistical evidence showing

60

what causes this change. Since it was not in the scope of this study, it would be
premature to conclude that the change in antisocial of Caucasian resulted from the boot
camp per se. Multivariate analysis controlling for potential predictors such as boot camp
program elements, family background, and religion scale should provide more
informative picture. Those potential predictors included. As stated in Chapter 11, family
background and religion beliefs of African Americans were quite different from
Caucasian Americans. These might contribute to their antisocial attitudes and attitudes.
As for program elements, the firture should address the issues of eligibility for
boot camp, the relationship between offenders and staff, race of staff, and types and
characteristics of therapeutic elements. Future research should examine if African
Americans and Caucasian Americans differ in terms of their attitudes and behaviors when
participating in particular programs within boot camp such as educational program and

drug treatment.

Limitations of the Study

 

The use of secondary data has some limitations. Apparently, the author has no
control over the data collection process. Also, since the data were not originally collected
for this research purpose, some information that might help improve the quality of this
study was not found such as the number of treatment programs in boot camp,
characteristics of treatment program, and characteristics of program staff.

The generalizability is another problem of this research. The data used in this
study was collected from particular settings with particular population. As discussed in

Chapter I and II, boot camp programs vary from place to place especially in terms of

61

goals, focus of treatment, type of offender, setting environments, and participants’
background. Therefore, findings of this study may or may not apply to boot camp
graduates in general. One should use findings fi‘om this study cautiously when applying

them to offenders from other settings.

62

APPENDIX A

10.

11.

12.

13.

14.

15.

l6.

17.

APPENDIX A

Items on Jessness Antisocial Attitude Scale

. When you’ re in trouble, it’s best to keep quiet about it.

I get into a lot of fights.

If the police don’t like you, they will try to get for anything.

Women seem more fi'iendly than men.

Police stick their noses into a lot of things that none of their business.

I always like to hang around the same bunch of friends.

I hardly ever get a fair break.

A lot strange things happen to me.

If someone in your family gets into trouble, it’s better for you to stick together than to
tell the police.

It often seems like something bad happens when I’m trying my best to do what is
right.

Most people in authority are bossy and overbearing.

It seems like wherever I am, I’d rather be somewhere lese.

I think that boys fourteen years old are old enough to smoke.

It makes me mad that some crooks get off free.

Police usually treat you dirty.

I often fell lonesome and sad.

A lot time I do think that my family me I shouldn’t.

63

18. A lot of people say bad things behind my back.

19. It seems like people keep expecting me to get into some kind of trouble.
20. Other people are happier than I am.

21. Police and judges will tell you one thing and do another.

22. It doesn’t seem wrong to steal from crooked storeowners.

23. My life at home is always happy.

24. Most police will try to help you.

25. Nobody seems to understand me or howl feel.

26. I don’t mind lying if I’m in trouble.

27. I think my mother should have been stricter than she was about a lot of things.
28. I feel alone even when there are other poeole around me.

29. Things don’t seem real to me.

30. I think there is something wrong with my mind.

APPENDIX B

8.

9.

APPENDIX B

Items on Pmram Attitude Scale

. There is nothing in this place that will help me.

This place will not help me get a job.

I am tough enough to handle this place.

This experience will not change me.

This place will help me learn self-discipline.
The guard put on a big show, but that is all it is.
This place will never help me in any way.

I will learn things about myself here.

I am becoming a better person here.

10. The programs in this place will never help me in anyway.

11. I am becoming more mature here.

12. Because of my experience here, I will probably not get in trouble again.

65

APPENDIX C

APPENDIX C

Reliability Analysis for Program Attitude Scale from SPSS

RELIABILITY ANALYSIS TIMEl- SCALE (AL
PHA)
Mean Std Dev Cases
1. RE_P1 4.0218 1.2047 229.0
2. RE_P2 3.3974 1.2405 229.0
3. P3 4.2533 .9396 229.0
4. RE_P4 4.0742 1.1989 229.0
5. P5 4.1528 1.1271 229.0
6. RE_P6 3.1266 1.3786 229.0
7. RE_P7 4.1921 1.0504 229.0
8. P8 4.0000 1.1002 229.0
9. P9 4.0262 .9819 229.0
10. RE_P10 4.1572 .9831 229.0
11. P11 3.9476 1.0458 229.0
12. P12 4.1703 1.1401 229.0
Correlation Matrix
RE_P1 RE_P2 P3 RE_P4 P5
RE_P1 1.0000
RE_P2 .3552 1.0000
P3 -.0010 .0676 1.0000
RE_P4 .3815 .2484 .1117 1.0000
P5 .0169 -.0217 .2697 .1603 1.0000
RE_P6 -.OO43 .1295 .0293 .2543 .1202
RE_P7 .3572 .2037 .2304 .5215 .2641
P8 .0728 .1414 .2291 .2527 .3218
P9 .1107 .0706 .2590 .3448 .3927
RE_P10 .3156 .1787 .0802 .4850 .1682
P11 .1854 .1378 .2769 .2060 .3603
P12 .0196 .0450 .0947 .1287 .2868
RE_P6 RE_P7 98 P9 RE_P10
RE_P6 1.0000
RE_P7 .2345 1.0000
P8 .1417 .2201 1.0000
P9 .1628 .3523 .5115 1.0000
RE_P10 .2377 .6163 .1784 .3501 1.0000
P11 .1811 .2727 .5222 .5779 .2683
P12 .1369 .1557 .3217 .4074 .1638

66

P11 P12

P11 1.0000

P12 .3827 1.0000

N of Cases = 229.0

Statistics for Scale Mean Variance Std Dev N of Variables
47.5197 51.8647 7.2017 12

Item Means Mean Minimum Maximum Range Max/Min Variance

3.9600 3.1266 4.2533 1.1266 1.3603 .1177

Item-total Statistics

Scale Scale Corrected Squared Alpha

Mean Variance Item- Multiple if

if Item if Item Total Correlation Item

Deleted Deleted Correlation Deleted
RE_P1 43.4978 45.5932 .2963 .3018 .7702
RE_P2 44.1223 45.9762 .2586 .1721 .7751
P3 43.2664 47.6261 .2587 .1601 .7712
RE_P4 43.4454 42.2569 .5242 .4036 .7438
P5 43.3668 44.9789 .3714 .2449 .7613
RE_P6 44.3930 45.0992 .2627 .1346 .7776
RE_P7 43.3275 42.7738 .5814 .5002 .7396
P8 43.5197 43.6981 .4782 .3691 .7499
P9 43.4934 43.2598 .5915 .4931 .7401
RE_P10 43.3624 44.1531 .5163 .4469 .7474
P11 43.5721 43.0617 .5617 .4656 .7417
P12 43.3493 45.2897 .3438 .2233 .7643
Reliability Coefficients 12 items
Alpha = .7729 Standardized item alpha = .7830

67

R E L I A B I L I T Y

H A)

l—’
OKDCDQONU‘IAUJNH

H H
N H
. .

RE_P1
RE_P2
P3
RE_P4
P5
RE_P6
RE_P7
P8
P9

RE_P10

P11
P12

Correlation Matrix

RE_P1
RE_P2
P3
RE_P4
P5
RE_P6
RE_P7
P8

P9
RE_P10
P11
P12

RE_P6
RE_P7
P8

P9
RE_P10
P11
P12

P11
P12

N of Cases =

RE_P1

1.0000
.3552
-.0010
.3815
.0169
-.0043
.3572
.0728
.1107
.3156
.1854
.0196

RE_P6

1.0000
.2345
.1417
.1628
.2377
.1811
.1369

P11

1.0000
.3827

229.0

A N A L Y S I S TIME 2 -

bwbbbbwbbbwb

Mean

.0218
.3974
.2533
.0742
.1528
.1266
.1921
.0000
.0262
.1572
.9476
.1703

RE_P2

1.0000
.0676
.2484

-.0217
.1295
.2037
.1414
.0706
.1787
.1378
.0450

RE_P7

1.0000
.2201
.3523
.6163
.2727
.1557

P12

1.0000

68

Std Dev

H'H

F’H‘h‘k‘k’

H

P3

1.0000
.1117
.2697
.0293
.2304
.2291
.2590
.0802
.2769
.0947

P8

1.0000
.5115
.1784
.5222
.3217

.2047
.2405
.9396
.1989
.1271
.3786
.0504
.1002
.9819
.9831
.0458
1.1401

S C A L E

Cases

229.
229.
229.
229.
229.
229.
229.
229.
229.
229.
229.
229.

RE_P4

1.0000
.1603
.2543
.5215
.2527
.3448
.4850
.2060
.1287

P9

1.0000
.3501
.5779
.4074

OOOOOOOOOOOO

(A L P

P5

1.0000
.1202
.2641
.3218
.3927
.1682
.3603
.2868

RE_P10

1.0000
.2683
.1638

Statistics for
Scale

Item Means

N of

Mean Variance Std Dev Variables

47.5197 51.8647 7.2017 12

Mean Minimum Maximum Range Max/Min
3.9600 3.1266 4.2533 1.1266 1.3603

Item-total Statistics

Scale Scale Corrected Squared
Mean Variance Item- Multiple
if Item if Item Total Correlation
Deleted Deleted Correlation
RE_P1 43.4978 45.5932 .2963 .3018
RE_P2 44.1223 45.9762 .2586 .1721
P3 43.2664 47.6261 .2587 .1601
RE_P4 43.4454 42.2569 .5242 .4036
P5 43.3668 44.9789 .3714 .2449
RE_P6 44.3930 45.0992 .2627 .1346
RE_P7 43.3275 42.7738 .5814 .5002
P8 43.5197 43.6981 .4782 .3691
P9 43.4934 43.2598 .5915 .4931
RE_P10 43.3624 44.1531 .5163 .4469
P11 43.5721 43.0617 .5617 .4656
P12 43.3493 45.2897 .3438 .2233
Reliability Coefficients 12 items
Alpha = .7729 Standardized item alpha = .7830

69

Variance

.1177

Alpha
if

Item
Deleted

.7702
.7751
.7712
.7438
.7613
.7776
.7396
.7499
.7401
.7474
.7417
.7643

BIBLIOGRAPHY

BIBLIOGRAPHY

Anderson, J. F., Dyson, L., & Burns. J. C. (1998). BOOMSZ an intermediate
sanction. Lanham, MD: University Press of America.

Andrews, D. A., Zinger, I., Hoge, R. D., Bonta, J., Gendreau, P., & Cullen, F. T.
(1990). Does correctional treatment works? A clinically-relevant and psychologically
informed meta-analysis. Criminology, 28, 369-404.

American Correctional Association. (1995a). Standards for adult correctional
boot camp programs. Laurel, MD: The Association.

American Correctional Association. (1995b). Standards for juvenile correctional
W. Laurel, MD: The Association.

Antonowitz, D. H., & Ross, R. R. (1994). Essential components of successful
rehabilitation programs for offenders. International Journ)al of Therapy and Comparative

Criminology, 38 (2), 97-104.

Austin, J., Jones, M., & Bolyard, M. (1993). The gmw_mg° use of jail boot camps;
the current state of art (Report No. NCJ 143708). Washington, DC: National Institute of
Justice.

Austin, J., Jones, M., & Bolyard, M. (1996). A survey of jail-operated boot camps
and guidelines for their implementation. In D. L. MacKenzie, & E. E. Hebert (Eds),
Correctional boot (ms; tough intermediate sanction (Report No. NCJ 157639, pp.
119-134 ). Washington, DC: US. Government Printing Office.

Beers, B & Duval, C. (1996) The Oregon Summit Program. In A. Fins, M. Kelly,
& M. Selby (Eds), Juvenile and adult boot mp5 (pp.39-59). Upper Marlboro, MD:
Graphic Communications.

Bottcher, J. & Isorena, T. (1996). First year evaluation of the California youth
authority boot camp. In D. L. MacKenzie, &. E. E. Hebert (Eds), Correctional boot
camps: a tough intermediate sanction (Report No. NCJ 157639, pp. 159-178).
Washington, DC: US. Government Printing Office.

Bourque, B. B., Cronin, R. C., Pearson, F. R., Felker, D. B., Han, M., & Hill, S.

M. (1996). Boot cm for juvenile offenders: an m1p° lementation evaluation of three
demonstration programs (Report No. NCJ 157316). Washington, DC: US. Government

Printing Office.

 

Bourque, B. B., Han, M., & Hill, S. M. (1996). A national survey of aftercare
provisions for boot amp graduates (Report No. NCJ 157664). Washington, DC: National
Institute of Justice.

70

Bureau of Justice Statistics. (1998). Probation and parole in the United State 1998
[On-line]. Available: http://www.ojp.usdoj.gov/bjs/pub/press/ppus98.pr

Cass, E. S. & Kaltenecker, N. (1996). The development and operation of juvenile
boot camps in Florida. In D. L. MacKenzie, & E. E. Hebert (Eds), Correctional boot
mp5: a tough intermediate sanction (Report No. NCJ 157639, pp. 179-190).
Washington, DC: US. Government Printing Office.

Clark, C. L. (1996). Sisters are doing it for themselves: Women in correctional

boot camp. In A. Fins, M. Kelly, & M. Selby (Eds), Juvenile and adults boot mp5 (pp.
309-320). Upper Marlboro, MD: Graphic Communications.

Clarck, C. L. & Aziz, D. W. (1996). Shock incarceration in New York State:
Philosophy, results, and limitations. In D. L. MacKenzie, & E. E. Hebert (Eds),
Correctional boot camm tough intermediate sanction (Report No. NCJ 157639, pp. 39-
68). Washington, DC: US. Government Printing Office.

Cowles, E. L., & Castellano, T. C. (1995). Boot cm drug treatment and
aftercare gervention: an evaluation review (Report No. NCJ 153918). Washington, DC:
Washington, DC: National Institute of Justice.

Cronin, R. C. (1994). Boot camps for adult and juvenile offenders: overview a_na
up_d_ate (Report No. NCJ 149175). Washington, DC: National Institute of Justice.

Ellis, D., Grasmick, H. G., & Gilman, B. (1974). Violence in prison: A
sociological analysis. American Journal of Sociology, 80, 15-43.

Feld, B. C. (1981). A comparative analysis of organization mm and inmate
subcultures in institutions for juvenile offenders. Crime and Delinquency, 27, 336-363.

Felker, D. B. & Bourque, B. B. (1996). The Development of boot camps in the
juvenile justice system: Implementation of three demonstration programs. In D. L.
MacKenzie, & E. E. Hebert (Eds), Correctional boot caalps: a tough intermediate
sanction (Report No. NCJ 157639, pp. 143-158). Washington, DC: US. Government
Printing Office. '

Gendreau, P., & Ross, R. R. (1987). Revivification of rehabilitation: Evidence
from the 1980s Justice Merl; 4. 349-408.

Goodstein, L. & Wright, K N. (1989). Inmate adjustment to prison. In L.
Goodstein & D. L. MacKenzie (Eds), The American Prison Issue in Research aad
Policy. New York, NY: Plenum Press.

Gowdy, V. B. (1996a). A critical look at boot camps for women. In A. Fins, M.

Kelly, & M. Selby (Eds), Juvenile and adults boot camps (pp. 297-307). Upper
Marlboro, MD: Graphic Communications.

7]

Gowdy, V. B. (1996b). Historical Perspective. In D. L. MacKenzie, & E. E.
Hebert (Eds), Correctional boot mtomgh intermediate sanction (Report No. NCJ
157639, pp. 1-38 ). Washington, DC: US. Government Printing Office.

Grasmik, H. & Bursick, R (1990). Conscious, significant others and rational
choice: Extending the deterrence model. Law and Society Review. 24. 837-861.

Harer, M. D., & Steffensmeier, D. J. (1996). Race and prison violence.
Criminolog)’, 34(3), 323-343.

Lotz, R., Regoli, R, & Raymond, R. (1978). Delinquency and special deterrence.
Criminology, 15(4), 539-547.

MacKenzie, D. L., Elis, L. A., Simpson, S. S., & Skroban, S. B. (1996). Boot
camp as an alternative for women. In D. L. MacKenzie, & E. E. Hebert (Eds),
Correctional boot caarps: a tough intermediate sanction (Report No. NCJ 157639, pp.
233-244). Washington, DC: US. Government Printing Office.

MacKenzie, D. L., & Ballow, D. B. (1989). Shock incarceration prom in state

correctional jurisdictions: an umte ( Report No. NCJ 120287). Washington, DC:
National Institute of Justice.

MacKenzie, D. L. & Souryal, C. (1993). Inmate attitude change durmg'
incarceration: A corrmarison of boot caalp and traditional prison (A final report to the
National Institute of Justice). College Park, MD: University of Maryland.

MacKenzie, D. L., & Piquero, A. (1994). The impact of schock incarceration
program on prison crowding. Crime Ed Delinquency, 40, 222-249.

MacKenzie, D. L., & Sourya, C. (1994). Multisite evaluation of shock
incarceration (Report No. NCJ 142462). Washington, DC: US. Government Printing
Office.

MacKenzie, D. L., Shaw, J. W., & Gowdy, V. B. (1993). An evaluation of shock
in_carceration in Louisiana (Report No. NCJ 140567). Washington, DC: National Institute
of Justice.

McCampbell, S. W. (1996). A jail boot camp program: Broward Sheriff 8 Oflice.
In A. Fins, M. Kelly, & M. Selby (Eds), Juvenile and adult boot mags (pp.77-91).
Upper Marlboro, MD: Graphic Communication.

Morash, M., & Rucker, L. (1990). A critical look at the idea of boot camp as a
correctional reform. Crime and Delinquency. 36. 204-222.

72

Morash, M., Bynum, T. S., & Koons, B. A. (1998). Women offenders:
Programming needs afll promising approaches [On-line]. Available:
http://www.ncjrs.org/txtfiles/ l 71 668.txt

Parent, D. G. (1989). Shock incarceration: An overview of existing proggams.
Washington, DC: National Institute of Justice.

Peters, M., Thomas, D., Zamberland, C., & Associates, C. (1997). Boot caps for
juvenile of offenders: WOW (Report No. NCJ 164258). Washington, DC:
Office of Juvenile Justice and Delinquency Prevention.

Roberts, V. J. (1996). A exploratory analysis of Baldwin’s Theory of Black
Personality Development. Unpublished master’s thesis, Michigan State University.

Siegel, L. J. & Senna, J. J. (1997) Juvenile delinquency: Theory, practice, and law
(6m ed.). St. Paul, MN: West Publishing.

Tittle, C. R. (1980). Sanctions a_nd social deviance. New York: Praeger
Publishers.

Wordes, M., & Jones, S. M. (1998). Trends in juvenile detention and steps toward
reform Crime and Delinquency, 44, 544-560.

Wright, K. N. (1989). Race and economic marginality in explaining prison
adjustment. Journal of Research in Crime and Delinquency, 26, 67-89.

Zachariah, J. K. (1996). An overview of boot camp goals, components, and
Results. In D. L. MacKenzie, & E. E. Hebert (Eds), Correcmal boot camps: a tough
i_n_termediate satiation (Report No. NCJ 157639, pp. 17-38). Washington, DC: US.
Government Printing Office.

73

   

1113111111 jjlllllllflllljll