MSU

LIBRARIES

 

 

 

RETURNING MATERIALS:
PIace in book drop to
remove this checkout from
your record. FINES wiII
be charged if book is
returned after the date
stamped beIow.

 

 

' MWees

WM 5

MW
\ kl

 

“’5‘
H121

P “‘5‘

(:2 1

i {3117/

 

 

INCARCERATED RAPISTS' RESPONSE TO SEXUALLY
AROUSING STIMULI

By

Michael B. Kalisewicz

A DISSERTAIION

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

DOCTOR OF PHILOSOPHY

Department of Educational and Counseling Psychology

1982

ABSTRACT

INCARCERATED RAPISTS' RESPONSE TO SEXUALLY
AROUSING STIMULI

By

Michael B. Kalisewicz

As a result of increased public and institutional awareness of the
incidence of rape, an intensive search for methods of victim protection
has been in progress. Little is known about what types of resistances
are most effective. Urging a particular resistance strategy may be
advising a victim into greater danger.

Hypothesized in this study was that groups of rapists will differ
in their sexual responses as a function of their personalities and the
emotional state of a victim during a simulated rape scene. The second
hypothesis was that designated socio-behavioral variables would corre-

spond strongly (r - i .35) with levels of rapists‘ arousal in response

to simulated rape scenes.

£7 The dependent variable was rapist's sexual arousal recorded by a
plethysmorgraph wired to a penis circumference gauge. The independent
variables were audio tape simulation of a rape victim responding to a
rape in five different ways: anger, fear, submission, reasoning, and
lust. The second independent variable was the socio-behavioral

descriptive categories created for this research.

Michael B. Kalisewicz

The experimental procedure was a two step process. First, histori-
cal court and institutional documents of the subjects were evaluated by
trained judges. Subjects were then categorized on the socio-behavioral
descriptive categories. Second, the experiment was conducted with
consenting incarcerated rapists. The experimental stimuli was pre—
sented to groups of two to four subjects seated in private booths. An
erotic film was shown to provide a baseline l00% erection measure.

A baseline zero erection measure was also taken during flacidness.

AWN

Next the victim stimuli tapes were played and physiological measures

 

of a rousal were recorded.

The hypothesis of differences in rapists' sexual arousal as a
function of the victim's emotion was tested using a Multivariate
Analysis of variance. Significance was obtained for the main effects
(F=l70.5, p < .000l) but confounded by an interaction effect (F=58.68,
p < .0001). Hence, the hypotheses were not substantiated.

The second hypothesis concerning the correlation between arousal
scores and descriptive category scores was tested using a decision
rule; Pearson Product-Moment Correlation Coefficients that exceeded
i .35 were considered significant. Only two of the descriptive cate-

gories, occupational stress and aggressive style were significant.

For My Parents
Who Have Given
Immeasurably

ii

ACKNOWLEDGMENTS

My first acknowledgment goes to my advisor Norm Kagan. He supported,
pushed, taught, and advised me through this process for several years.
If one person were necessary for the completion of this project, it
would be him.

Secondly, the many people affiliated with the Michigan Department
of Corrections deserve a special thanks. The support of Mr. Charles
Anderson, Warden at the State Prison of Southern Michigan was parti-
cularly essential in gaining approval and implementation of this
project.

A special acknowledgnent to Dr. Greenbaum and Dr. Prelesnik for
their many hours of discussion and manuscript review. Dr. David
Gerbing helped immeasurably as a statistic's consultant.

Many people assisted in the development of this project. Special
thanks to Kenneth Kalisewicz fOr his electronic skills in construction
of the Penis Circumference Gauge, to Barbara Kalisewicz for her literary
skills as a script writer, and to Elaine Russ for her editorial reviews.

Finally, thanks to the many residents of the State Prison at
Southern Michigan who were willing to trust and give their time to

this researcher.

TABLE OF CONTENTS

LIST OF TABLES .
LIST OF FIGURES.
CHAPTER

I. THE PROBLEM.

Purpose

Theory. .
Hypotheses
Definition of Terms
Assumptions .
Overview .

II. REVIEW OF LITERATURE.

Introduction. . . .

Rapist Response to Sexual Cues. . . .
Classification of Rapists . . . .
Methods of Assessment of Sexual Arousal.
Summary and Conclusions . .

III. DESIGN AND METHODOLOGY .

Sample. . . .
Variables--Measures

Dependent Variable--Sexual Arousal.

Independent Variables--Victim Affect State :

Descriptive Categories. .
Procedure. . . .

Population Classification. . . .
Experimental Procedure.

iv

Page
vi

vii

_a
hkOkO LO mVVO‘QJN —-‘

N NM
00 0""

MN
NC)

0 .

35
41

43

43
44

Chapter
Design .
Experimental Hypothesis
Analysis . . . .
Summary.
IV. ANALYSIS . . . . . . . . . . . .

Research Hypothesis I .
Research Hypothesis 11.
Factors Influencing the Hypothesis.
Summary. . . . . . . . . .

V. SUMMARY, CONCLUSION, DISCUSSION, AND IMPLICATIONS
Summary.
Conclusions
Discussion. .
Subjective Observations
Implications for Future Research
APPENDICES
Appendix
A. Letter to Sample
B. Pre-Test Evaluation Sheets .
C. Population Rating Sheet .
0. Population Rating Instruction Sheet . . .
E. Reminder Letter to Raters . . . . . . .
F. Population Raters Thank You.
G. Informed Consent--Permission to Participate
H. Release From the Experiment.
I. Experimental Explanation Sheet.
J. Item Reliability Data. . .

REFERENCES. . . . . . . . . . . . . . . .

Page

75
76
78
80
81
82
83
84
85
86
108

Table

xooowosma-w

10.

LIST OF TABLES

Summary Comparison Chart Measures of Sexual Arousal .

Summary of the Demographic Characteristics of the
Qualifying and Volunteering Sample Frequency and
Percentile Distribution. . . . .

Vignette Pre-Test Results Response Distribution

Descriptive Categories With Item Description

Experimental Procedure: Stimulus Presentation Format

Latin Square: Order of Tape Presentation

Repeated Measures Design .

Repeated Measures MANOVA Summary

Correlation Values for Descriptive Categories by Arousal
Tapes. . . . . . . . . . . . . .

Average Percent Agreement Among Raters (N = l0) on Items
in Descriptive Categories . . . . . . . .

vi

Page
26

31
4O
42
48
50
52
58

59

64

Figure

0301-th

LIST OF FIGURES

Schematic Diagram Used for Strain Gauge .
Xerox of Photograph of Mounted Strain Gauge.
Experiment Room Layout. . . . . .
Subject Arousal Score Tabulation Fonnula.
Percentile Transformation Formula .

Graphic Representation of Correlation Values for
Descriptive Categories by Arousal Tapes

Graph of Cell Means to Demonstrate Interaction Effect of
Order of Presentation and Tape Stimulus in MANOVA .

vii

Page
34
34
45
56
57

6O

62

CHAPTER I

THE PROBLEM

Rape is the fastest-growing reported crime of violence in the
United States. In Michigan, there were 11.4 incidents of rape per
hundred thousand people, in 1960. By 1977, reported rapes had increased
to 38.77 per hundred thousand people (Michigan Statistical Abstract, 1978).
Nationally, the incidence of rape has increased from 14.0 per hundred
thousand in 1967 to 29.1 per hundred thousand in 1977 (United States
Abstracts, 1978). This dramatic increase has resulted in a number of
problems, two of which were addressed in this research.

The increase of rape and the public awareness of its increase have
contributed to the development of rape crisis centers. Some of these
centers serve a dual purpose in the community, acting as advocates for
rape victims as well as recommending methods of rape prevention. The
latter includes the use of weapons and aggressive physical defense
(Medea & Thompson, 1974). In 1975, the National Crime Prevention Insti-
tute reconmended two primary methods to use in escaping rape: (l) imme-
diate resistance based on professional training in self-defense, and
(2) establishment of interpersonal contact with the rapist. The problem
with these strategies of victim defense is that they may be advising a

potential victim toward more danger. There is a growing body of single

case-study research that indicates that there is a group of aggressive
rapists who need the element of victim physical resistance in order to
complete the rape (Cohen, 1971; Rada, 1978). Therefore, the first area
investigated in this study was to detennine if present strategies of
recommended victim defense may actually increase the likelihood of the
completion of the rape act and of greater physical violence to the
victim.

The second problem addressed by this research was to test the
validity of a classification system for rapists that could be used by
mental-health professionals. As a result of the increased incidence of
rape, mental health proféssionals are increasingly called upon to make
recommendations, not only in the areas of criminal responsibility, but
also in the therapeutic treatment and assessment of rapists' readiness
for community release. Worthy of note is the fact that, in some states,
release to the community is dependent upon psychiatric-staff judgments
that the offender will present minimal risk to the safety of the com-

munity (Massachusetts, Statute, Chapter 646, 1958).

Purpose

The first purpose of this study was to detennine if rapists are
different from one another in socio-behavioral descriptive categories
and if these personality differences are related to specific types of
sexual response to the victim's emotion during rape. The second purpose
was to assess the validity of organizing the socio-behavioral variables

into a classification model.

Theory

The present research is most heavily influenced by classical
psychoanalytic theory. Emphasis is placed on the psychoanalytic view-
point not because that viewpoint is necessarily the most correct, but,
rather, because it is the only theory of normal and abnonmal sexual
development which attempts to account fOr most or all of human learning,

development and behavior. In Three Essays on the Theory_of Sexuality,

 

Freud presented the first unified theory of sexual perversion, its
origin in infant sexuality, and its vicissitudes in the early develop-

ment of the individual. In The Psychoanalytical Theory of Neurosis,

 

Fenichel (1945) describes

mental phenomenon as the result of the interaction and counter-

action of forces, that is in a dynamic way. A dynamic explana-

tion is also a genetic one, since it not only examines a

phenomena as such but the forces that brought it about as well.

It does not examine single acts; it examines the phenomena in

terms of processes, . . .

"Object Relations Theory" as described by Kernberg (1975), Masterson
(1978), and Mahler (1974), has also influenced the present research.
Specifically, object relations theory describes a pattern of "splitting"
defenses that are the result of an unresolved symbiotic relationship.
Manifest in adulthood, many rapists will have a wife whom they adore
and view as good, and a victim whom they describe as bad and, therefore,
a justifiable recipient of their sexual aggressions.

The most current published thinking on rapists' motivational
patterns is presented by Groth (1979). Consistent with analytic assump-
tions, Groth described patterns of acting-out defenses and the expression

of pent-up impulses. Three motivational forces in rape behavior that

he reviewed were anger, power, and sadism. "Angry rape" was viewed as

the expression and discharge of pent-up feelings of rage. Usually the
assault was seen as occurring in response to precipitating stressors,
most commonly being disputes or disruption in the marital relationship,
financial problems or other social situations in which frustration
results. "Angry" rapists were viewed as having a history of relation-
ships with significant females which were laden with conflict, irrita-
tion and an inability to communicate. Hence, there was believed to be
a symbolic displacement of a revengeful anger onto other females.

"Power rape" was viewed as involving the operation of sexuality as
a means of compensating for underlying feelings of inadequacy. The act
of rape, then served to express issues of mastery, strength, control,
and identity. The power rapist rarely reported precipitating stressors.
Instead, there was a build-up of internal fears concerning adequacy
which were temporarily relieved by the rape. These rapists were repeti-
tive, and their rapes frequently were planned.

Finally Groth (1979) described sadistic motivation in rape, in
which sexuality and aggression were fused into a unified psychological
experience. Anger and power were viewed as being transformed into a
sexually eroticized experience, and gratification was experience in the
abuse and torture of the victim. The instrument of abuse was the
sexual act, and the motivation was punishment and destruction. Often
there was a progression over time of increasingly assaultive offenses.

Motivation may form the basis for clinical classification. While
classification can be harmful when it is used in a manner that obscures
the individuality of the offender, it can be useful in the testing of
theoretical constructs as well as identification of etiological factors

in the commission of rape.

The most current, published clinical classification was devised by
Rada (1978) for use in both clinical and courtroom work. The system
was based on the author's personal experience in working with rapists
and, at this writing, has not yet been systematically researched. Rada
proposed five different symptomatic or characterological types based on
descriptive psychiatric categories: (1) the psychotic rapists; (2) the
situation-stress rapists; (3) the masculine-identity conflict rapists;
(4) the sadistic rapists; and (5) the sociopathic rapists. Following
is a brief discussion of each of the groups.

The psychotic rapists were defined as those borderline personalities
who may decompensate into psychosis-~functiona1 or organic--under
stressful conditions. The acutely psychotic rapist would very often be
unable to give a lucid account of the rape event, often being only aware
of an overwhelming rage and anger at the time of the rape, and unable
to give any other motivation for the crime. Rape by a psychotic rapist
may be a particularly terrifying event fer the victim, because neither
the victim nor the rapist is in control of the situation.

Rada described the situational stress rapist as presenting an
immediate pre-rape history of severe situational stress occurring in a
general downhill course in their lives. This rapist rarely had a
history of sadistic or violent masturbatory fantasies or sexual devia-
tions. Prior to the situation stress, there may have been no history
of obvious emotional abnormalities. At the time of the rape, however,
most of these rapists were suffering from a moderate to severe agitated
depression. Although he may have physically subdued his victim, this

type of rapist rarely used excessive violence.

The masculine-identity conflict rapists shared in common an actual
or felt deficiency in their masculine roles, developing a characterolog-
ical style of dealing with this role-identity conflict. Doubts about
masculinity and fear of passivity led not to rationalization and
repression, but rather to denial and projection. They were among the
most dangerous and violent rapists since they often used more force
than was necessary to complete the rape. Their rapes are usually
planned.

The sadistic rapist was seen as carefully planning his rape and
would often demand that the victim perfOrm a variety of huniliating
activities. He appeared to derive greater satisfaction from his ritualis-
tic degrading of the victim than from the actual intercourse.

The sociopathic rapist engaged in a variety of criminal activities
and had frequent encounters with the law. Rape was frequently just one
of his antisocial aggressive and sexual acts. Usually, the rape was
impulsive. These rapists often described the development of an inde-
finable psychic tension that was only subsequently relieved by some

kind of antisocial activity.

Hypotheses
As a result of the review of the literature, the following hypothe-
ses were developed:

1. Groups of rapists will differ in their sexual responses as a
function of their personalities and the emotional state of a
"victim" during a simulated rape scene.

2. Designated socio-behavioral variables will correlate strongly
(r = $.35) with levels of rapists arousal in response to

simulated rape scenes.

Definition of Tenns

 

Special terms used in this study are defined as follows:

Rape: The act of forcibly having or attempting sexual intercourse

with a woman without her consent.

Rapist: An incarcerated male sexual felon who at the time of the
rape was at least 21 years old and who presently is less than 50

years old.

Victim: A female who at the time of the rape was at least 16
years old and who was forcibly and without her consent involved in

sexual intercourse or its attempt.

Assumptions
There were five assumptions which were made in the design of this

research.

Assumption one: The magnitude or degree of a descriptive trait

 

exists on a continuum, from marginal existence to the domination
of the personality. The present state of the art does not allow
for numerical descriptions of the magnitude of a descriptive trait
in an individual. However, one can make relative inter-group

comparisons.

Assumption two: Descriptive traits are heterogeneous. Inclusion

 

of a descriptive trait is not done at the exclusion of other

descriptive traits.

Assumption three: The existence of a descriptive trait can be

 

verified by its manifest behavior.

Assumption four: Subjects can descriptively be categorized by

 

the dominance of one trait over all other traits.

Assumption five: Rapists in Michigan are similar to rapists else-

 

where.

Overview
In Chapter II, methods of measurement of sexual arousal and previous
studies are reviewed. Chapter III contains the experimental procedure,
research design, and methodology used in collecting and analyzing the
data. In Chapter IV, the data transformations, factors influencing
the hypothesis, and test of the hypothesis are described. In Chapter V,

the summary, conclusion, and discussion are presented.

CHAPTER II

REVIEW OF LITERATURE

Introduction

 

The review of literature in this chapter is focused on the following
areas relevant to this study: (1) previous studies of rapists' response
to sexual cues, (2) classification systems, (3) methods of measuring
sexual arousal in men, and (4) implications of the literature and con-

clusions.

Rapist Response to Sexual Cues

 

Preliminary studies which measures sexual arousal of rapists
employed a variety of techniques to discriminate rapists from non-
rapists. No studies could be found which answered the question of
whether rapists are different in their arousal as a function of the
victims' emotion.

Karacan (1974) reviewed nocturnal penile tumescence, sleep-EEG, and
eye movement patterns in non-rapists, rapist prisoners, and non-rapist
prisoners. While a variety of EEG changes were noted, no differences
were found in erection measures between rapists and non-rapists.

Kercher and Walker (1973) presented rapists and non-rapist prisoners
with a variety of sexual (but not exclusively rape) cues while recording

their erections and galvanic skin responses (GSR). Once again, no

10

differences in erections were noted, although rapists did show greater
GSR changes while watching slides of sexual activity. There were three
serious limitations of the methodology. First the authors were attempting
to differentiate rapists from non-rapists using only heterosexual cues.
Presenting a variety of rape versus non-rape cues may have added to the
discrimination of rapists from non-rapists in the study. Second, slides
normally provide rather weak sexual stimuli when presented for only ten
seconds, which may not allow sufficient opportunity for erections to
occur. Last, the changes in GSR may have been related to other arousal
states rather than sexual arousal. The authors themselves noted these
possible confounding variables.

Abel et a1. (1977) presented a three-part study in which the com-
ponents of rapists' sexual arousal were investigated. Using a single-
case experimental design with a group of unincarcerated rapists and a
group of non-rapists, they investigated whether erection measures could
separate the more dangerous rapists from the less dangerous ones. In
study one, 20 male subjects were referred to an outpatient facility for
evaluation of their deviant sexual arousal. Rapists were diagnosed on
the basis of (1) their self-report, (2) their description of their
victim's response, and (3) collaborative history from police reports.

Of the 20 subjects, 13 were diagnosed as rapists and seven as non-rapists.
Two of the subjects were private patients and the additional 18 were
subjects from part of a research study to evaluate sexual-arousal

patterns in rapists. No difference was found in the results of the
measurements between the research subjects and the private patients,

therefore the results were combined for analysis.

11

Subjects were presented vivid two-minute taped descriptions. The
first description was of mutually enjoyable intercourse. The second
tape was a description of the same scene, but clearly depicted the
patient as forcing himself on the victim. Patients were instructed
before each audio tape description to "listen closely, try to visualize
in your mind's eye exactly what is being described, try not to get
ahead of the description but just try to see what is being described."

A repeated measure design was used for stimulus presentation, and
the results for each individual subject as well as each group were
recorded for data analysis. The data was subjected to a three-way
analysis of variance (subject group x stimulus content x measurement
method). Analysis yielded significant main effects of the stimulus
content and a significant interaction of group by stimulus, with no
other main effects of interaction.

The authors noted that, using the repeated-measure design, the
responses were relatively stable across the design. An additional
interesting aspect was that a longitudinal interpretation of the data
yielded information that indicated "that simple removal from society
without specific treatments for arousal to rape cues does not greatly
effect such arousal to rape stimuli." Comparing rapists who had been
incarcerated in a range of 11 months through 23 years supported the
assumption of the existence of a character trait related to sexual
aggression.

Finally, the authors attempted to discriminate rapists in terms of
frequency of rape and extent of injury to the victim. To control for
individual variability and to develop a single measure of arousal to

rape cues, mean-percent erection to rape cues was divided by the

12

mean-percent erection to mutually enjoyable intercourse cues, yielding

a "rape index" as follows:

Rape Index = X % erect1on to rape cues

X % erection to mutually enjoyable intercourse
A score of 1.0 indicates that arousal to rape exceeded arousal of
mutual intercourse. Three findings were yielded from this index.
First, a rape-index score greater than 0.5 identified most rapists and
excluded most non-rapists. Second, the index score were highest for
four rapists who had raped at the highest frequencies. Finally, rapists
with the highest rape index scores also most clearly enjoyed injuring
their victims.

In study two, the authors investigated the relationship of rapists'
arousal to rape cues and aggressive cues. Nine of the original 13
rapists were selected for participation in this part of the study. They
were presented two audio-tape descriptions. The first, a non-sexual
aggression stimulus, described the subject physically assaulting a
female. The second audio stimulus described the patient raping the
same victim in the same situation as the first aggression scene. Measure-
ment was similar to that described in study one. A repeated-measures
design was used for stimulus presentation. The mean-percent erection to
pure aggressive scenes was highly correlated with the mean-percent
erection to rape scenes (r = .98; p < .01), indicating clearly that,
for these rapists, a commonality existed between response to aggressive
and sexual cues. Additionally, two rapists were unable to respond to
either the pure aggression or the pure rape cues, but responded highly

to the aggressive rape stimulus. Interestingly, these were sadistic

13

rapists who greatly enjoyed torturing their victims. The need for
aggression as part of a sexual cue was consistent with these rapists'
sadistic history. The authors concluded that "if aggression is a
necessary component of sexual interaction to generate erections, the
likelihood of aggression directed at the rape victim is quite high."

In study three, the authors investigated rapists' arousal as a

 

function of victim age. Previous research has demonstrated that the
rapists' preferred age of the sexual object can be detennined by pre-
senting slides of nudes of varying ages while recording his erection
(Freund, 1965). The same nine rapists that were used in study two were
used for study three. Only three of the rapists reported a preference
for very old or very young victims. Stimulus presentation was four
two-minute audio descriptions of rape developed for each rapist with
the age of the victim varying in each description. Scenes were pre-
sented randomly on two separate occasions. The mean-percent erection
in response to each age victim for the two presentations constituted
the data. Looking at the data individually, rapists' peak erections
occurred in response to those victims whose ages they had reported as
most erotic to themselves.

In another set of studies (Abel et a1., 1977), rapists (including
homosexuals and pediophiliacs) were clearly descriminated from non-
rapists by the magnitude of their reaction in response to hetersexual
as opposed to rape-specific sexual cues. Using penis circumference
gauges to measure arousal, rapists were found to respond equally to
rape-specific and hetersexual cues, while non-rapists were found to
respond only to hetersexual cues. Furthermore a "rape index" developed

by the three authors (score = mean-percent erection to rape cues :

14

mean-percent erection to mutually enjoyable intercourse cues) was found
to descriminate rapists along frequency-of—rape and extent—of-victim-
injury continua. In investigating the behavior of rapists who had been
incarcerated for periods of time ranging from 3 months to 12 years, the
authors found evidence that suggested that the proclivity towards rape
is unaffected by simple removal from society. Finally, use of audio-
tape stimulus and erection measures yielded information concerning
choice of victim age that was consistent with rapist history.

There are a number of problems with the Abel et a1. (1977) study.
The first two of these were noted by the authors. First, the number
of rapists measured is relatively small (E = 13 rapists and E = 7 non-
rapists). Second, the data was obtained from cooperative rapists. The
above two concerns certainly are considered threats to external validity
(Campbell & Stanley, 1963). Also, the subjects were not homogeneous
for age of assailant, age of victim, or sex of victim. Inclusion of
homosexuals and pedophiliacs would confound the internal validity in
making conclusions about rapists as defined in the present research.
Finally, rape cues used by Abel et al. failed to provide information
concerning etiological factors, such as victim response, that may con-

tribute to rapist violence.

Classification of Rapists
Most of the theory and research in the area of abnormal personality
classification has been based on psychoanalytical theory. Early classi-
fication systems evolved around analyses of pathological aims or objects.
Guttmacher and Weihofen (1952) observed that force rape has several
basic motivational patterns. They discerned that there is a rapist

whose assault is the explosive expression of pent-up sexual impulses,

15

and that there is a propensity for this to occur in individuals with
strong, latent homosexual personality components. These were classified
as the true sex offenders. Another type that was labeled was the
sadistic rapist, many of whom were found to have their deep-seated
hatred focused particularly on women. The third type of rapist, para-
docially, was not primarily a sex offender, but an aggressive antisocial
criminal. The authors likened the third type to the soldiers of a
conquering army who are out to pillage and rob.

In one of the first statistical studies concerning classification
of rapists, Brancle, Ellis, and Doobar (1952) identified four major
social-legal classes of rapists. The first was the normal sex offender,
who was said to be a reasonably well-adjusted individual who participated
in sex acts which were legally not permissible but were not abnormal or
pathological (e.g., adultery or coitus with a somewhat underaged partner).
This group also included individuals who occasionally, and under the
influence of alcohol, engage in abnormal sex acts, but who did not
habitually or exclusively derive satisfaction from such acts. The
second class was the sexually deviant but psychiatrically non-deviant
offenders who were defined as individuals who regularly or frequently
engaged in abnormal sex acts, e.g., homosexuality and pedophilia, but
who remained sufficiently well-integrated and emotionally stable to
pursue their aberrant behavior without getting themselves into trouble
with society. The third class was the sexually and psychiatrically
deviant offenders, who were defined as individuals who engaged in abnormal
acts and who did so in a repetitive, compulsive, or otherwise emotionally
disturbed manner. The final class was made up of sexually non-deviant

but psychiatrically deviant offenders, who were defined as individuals

16

who engaged in nonmal sex acts, but who did so in bizarre and nonintegra-
tive ways--e.g., exhibitionism, walking around naked or masturbating in
public--that were socially repulsive and sooner or later got them into
official difficulties (Brancle, Ellis, & Doobar, 1952). The range of
object and aim in this type of classification system was so broad as to
be of little assistance in understanding personality differences within
object-similar groups.

The number of factor-analytic studies of rapists on personality
inventories has been limited. The primary problem has been reliability,
because rapists intentionally misrepresent themselves due to fear of
reprisal from being identified as pathological (Abel, et al., 1977).
However, in one such study, Howell (1954) used the Edward's Personal
Preference Schedule to provide the factors for an analysis of one
hundred male rapists ranging in age from 18-60. Two sets of rapists
were identified: (1) men whose problems with impotency had achieved
pathological proportions and (2) men who were antisocially oriented.
The frame of reference of the latter was such that he felt that if his
female victim were degraded by force, she; rather than he, would be the
frightened subject who would be unable to complete the act while he was
dominant. The antisocial rapist, on the other hand, used the rape act
as a generalized retribution of assault. Having been previously and
regularly rejected in the past, he vented his wrath by exaggeratedly
humilitating an immediately available female.

Kobb (1953) also attempted to understand the rapist in terms of
the relationship between his character and his symptoms, and identified
two classes of rapists. The first he designated as "conconant." This

type of rapist was seen as an antisocial psychopath for whom rape was a

l7

manifestation of a larger characterological disorder of antisocial
acting out. Thus the rape was "conconant" with, although an exaggerated
ferm of, his characteristic mode of defense. The second class of
rapists was the "dissonant“ type. He was seen as an overly compliant
individual who rarely asked for what he wanted because he felt that he
did not deserve satisfaction from others, and for whom the rape repre-
sented a breakthrough of alien impulses. The dissonant rapist chacter-
istically felt guilty over the rape, and was very much concerned about
the well-being of his victims.

Another study in which character structure was used as the basis
of classification and understanding was conducted with 200 sex offenders
at the Sing Sing Prison (Hammer & Glueck, 1957). Using both psychiatric
appraisal and psychological examination, four factors were identified
which were believed to contribute to the acting out of sexually mala-
daptive behavior. The first was psychodynamic pattern consisting of
(a) fear of sex contact with the adult female, (b) Oedipal involvement,
and (c) feelings of genital inadequacy. The second factor was a schizoid-
schizophrenic continuum. The range of schizophrenic-like adjustment of
these men clustered around the clinical phenomenon commonly described
as incipient schizophrenia. The writers used this term to mean that a
person is, at present adjusting on a borderline level; the psychosis is
not frank. The subject's overt behavior, while frequently showing
indications of noticeable disturbance, does not reveal the hallucina-
tions, delusions, or bizarre actions associated with overt and definite
schizophrenia. The third factor was the control of impulses. The vast
majority of rapists and pedophilias have ego pictures of rigid control

by weak control mechanisms but control so inflexible and so tightly

18

spread that occasionally there is a breakthrough of impulses. Only the
incest group harbors a predominance of people who consistently exhibit
inadequate control. The fourth factor was called "concrete orienta-
tions." These people were conspicuously nonverbal in their integrations,
and on psychological tests were superficial and showed little capacity
for introspective thinking.

Specific analytically oriented papers have identified the Oedipal
and Electra complexes as the basis for sexual-acting out behavior
(DeRiver, 1958). Three classifications of rape were defined, each
classification proposed to be the result of a lack of resolution in the
complexes. The first class consisted of cases that occurred on an
aggression-to-sadism continuum. The second consisted of antisocial
acting out, wherein the sexual impulses and gratification were thought
to be the approximate cause. The third classification described a lust
murder. The authors attributed this to the power of lust and sex which
nullified the process of reasoning. The rape act would lead to a sexual
murder in which the attacker, from accidental conditions, would possibly
commit a sadistic crime such as murder.

In a reworking of early analytical aim theory (Guttmacher &
Weihofen, 1952), Cohen et al. (1971) identified a triparate clinical
classification of rape. The classification consisted of (l) rape-
aggressive aim, (2) rape-sexual aim, and (3) rape-sex-aggression defusion.
In the rape-aggressive aim, the degree of violence varied from simple
assault to brutal vicious attacks resulting, on occasion, in the victim's
death. The woman appeared to be the victim of the offender's destructive
wishes, and he, in fact, described his emotional state as anger. The

victims were almost always complete strangers. The anger that was

19

encountered was clearly a displacement of intense rage on a substitute
object. In the rape-sexual aim group, the act of rape was clearly
motivated by sexual wishes. Aggression materialized primarily in the
service of the sexual aim. The degree of aggressive behavior varied,
but there was a relative absence of violence, and the act lacked any
characteristic brutality. Finally, the rape-sex-aggression defusion
rapists revealed the presence of a strong sadistic component. They
possessed no inherent ability to experience sexual excitement without
some degree of violence. The degree of sadism varied. The extreme
degree was seen in lust-murders, wherein excessive brutality and mutila-
tion occurred before, during, and after the murder. This group is
relatively rare. In many ways these men are similar to psychopathic
characters. Other indications of impulsive behavior and a general lack
of control are expressed in stealing, truancy, running away, and lying.
This group shows the greatest degree of paranoid features, and, under
certain conditions, these are of psychotic proportion.

In a summary paper on analytic thinking in the area of rape, Arthur
Hyatt Williams (1965) described rape as existing on a continuum of
aggression. Rape/murder was seen as being on one end of the continuum
of behavior and minor assaults at the other end of the scale.

The aggressive sexual actions are viewed as a result of the

conflict between impaling forces and controlling forces. The

impaling forces are seen as destructive forces derived from
pre-genital phases of emotional development which remain un-
integrated and unmitigated by effective counteracting forces

or else remain split off or incapsulated under pressure within

the personality. It is these forces which cause the individual

to be either a potential or actual sexual criminal. These

destructive forces are of course instinctual drives. The

counter forces or controlling forces stem from fear of internal

or external punishment; and also in some cases concern for the

victim in fact, compassion or even love of the object. Whether

a mild or serious sexual crime is committed depends on the
resultant of impaling and controlling forces. (p. 211)

20

Basically, analytic assumptions view the ego as the negotiator between
instinctual urges and environmental demands. Crimes of sexual violence
are viewed as the breakdown of ego control and the resultant acting out
of primitive instinctual drives.

In an early study attempting to identify problems in the emotional
expression of adult sex offenders, two operationally defined types of
rapists were classified (Lopez, 1970). The first of these was the
"characterological offender," which included men whose offenses were
essentially consistent with their basic life styles. The second group
included all other rapists, and was labeled the "decompensatory offen-
ders." The results of the study indicated that characterological
offender to be more responsive with regard to palm sweat, less accurate
in his perception of at least some aspect of the affect in others, and
less intent with regard to the experience of unpleasant affect libido
and an overall variety and strength of affect than is the decompensatory
offender. The sample was 34 inmates committed to the Center for Diagnosis
and Treatment of Sexually Dangerous Persons in Massachusetts.

Another area of rapist research has been in interpersonal motiva-
tions. Efforts have been made to classify rapists as pseudo-sexual
offenders in which the primary motive is not one of physical gratifi-
cation (Groth & Burgess, 1977). These authors discount the antisocial
motivation of rape, saying that sexual motivation cannot be viewed as
the motivating force in any fonn of rape, and that the literature indi-
cates the availability of socially acceptable sexual gratification at
the time of sexual rape for most rapists. In a study investigating
2,133 offenders and 92 rape victims during the years 1972 to 1973 in

Massachusetts, a dichotomous emotional intent of rape was identified.

21

At one end of the dichotomy was anger, and at the other end of the
dichotomy was power rape. Both issues of power and anger became cast
in the sexual arena. It is in sexual relationships that these men most
clearly feel themselves to be victimized when confronted with their own
inadequacies. This study described the actions of the assailant as
being their only way to relate to women, and stated that they had not
developed the social skills necessary for lighter forms of interactions
with females. Interestingly, while denying the sexual gratification
aim in rape, these authors did recognize a transient reaction to extra-
ordinary stresses. A second reason suggested was a result of a more
endogenous state of affairs in which the offenders' psychological
resources are developmentally insufficient to cope with the excessive
and increasing demands of life. This study acknowledged that the
situational rapist had conflict-free areas of psychological functioning,
whereas the character-disordered rapist had fewer or no conflict-free
areas of functioning.

The previous review of classification systems have been based on
motivation (Cohen, 1971), object choice (Brancle, 1952), id impulses
(Williams, 1965), aggression (DeRiver, 1958), and psychodynamic patterns
(Hammer, 1957). Each system has its advantages and drawbacks, depending
upon the purpose for and setting in which it is used. All of these

systems are, to some extent, arbitrary and artificial.

Methods of Assessment of Sexual Arousal
A discussion of sexual arousal can be broken down into three main
modes of assessment: (a) self-report measures, (b) non-penile physio-
logical measures (EEG, heart, respiration, GSR, pupillary eye movement),

and (c) penile tumescence measures.

22

In the area of self-report, rapists appeared to consistently under-
report their sexual response to deviant sexual stimuli. Kerchert and
Walker (1973) presented 28 rapists and 28 non-rapist prisoners a variety
of sexual stimuli while recording their erections, GSR, and subjective
ratings (Thorne's Sex Inventory). No significant differences were noted
in penile tumescence. There was a significantly heightened GSR for
rapists, and rapists consistently underestimated their sexual response
to the sexual stimuli. There were a number of factors which may have
contributed to the lack of differences in erection measure. First, the
subjects were not presented rape cues. Second, the visual stimulus of
slides was a rather weak sexual stimuli as their presentation period
was only 20 seconds, which may not have provided sufficient opportunity
for erection to occur. Most important to the present discussion, how-
ever was the underrating of rapists' self-report of their own arousal.

Another study, conducted by Abel et a1. (1977) which compared 13
non-incarcerated rapists to 7 non-rapists, found results that were
supportive of previous research, i.e., that rapists consistently under-
estimate their sexual arousal in a self-report mode. Thus serious
validity problems arise from the use of rapists' self-report. Even
though studies indicate rapists reliably underestimate their sexual
arousal, it is impossible at this time to quantify the underestimation
and make correction factors for it. Therefore, self-report appears to
be the least useful of sexual arousal measures.

The second area of sexual arousal has been defined as non-penile-
tumescence physiological measures. This includes the Galvanic Skin
Response, heart rate, respiration rate, electroencephalograph activity,

and pupillary eye movement. A number of studies (Masters & Johnson, 1966;

23

Karacan, 1974) show that all of these measures correlate very highly
with sexual arousal. A serious problem with these measures is that
they are not specific to sexual arousal, but also highly correlate with
general arousal states.

These responses are not specific to sexual arousal but may

reflect orientating to novelty or emotions other than sexual

arousal. Measures of penile erection have been developed,

and have proven useful as specific sexual arousal indices:

(Zuckerman, 1971)

Using any of these physiological measures with incarcerated subjects
adds serious validity problems to the conclusions. That is, it might
be expected that presenting a group of sex offenders with sexual stimuli
may be cause for general anxiety arousal as well as an arousal to the
independent variable. Hence, use of these physiological measures
threatens the external validity of a study.

The third method of assessment, penile tumescence, has been used
as a means of measuring sexual arousal for approximately 15 years.
Early work in the area by Freund (1965, 1967) was used primarily for
the diagnosis of homosexuality and pedophilia. There have been a
number of mechanical devices for the measurement of penile tumescence
(Freund, 1965; Bancroft et al., 1966; Barlow et al., 1970; Bancroft
et al., 1966). Commonly, these instruments were wired through a
plethysmograph or an EEG recording machine. Following is a discussion
of the studies that have used penile tumescence as a dependent measure.
Included is a discussion of problems with this methodology.

Bancroft et al. (1966) described a treatment in which the strain
gauge was used in a case of pedophilia. In a conditioning experience,

the subject received electrical shock whenever responding to pedophilia

stimuli. The author noted that there was almost a direct relationship

24

between the increase in the diameter of the loop of the strain gauge and
the increase in microamps registered. Additionally, the author stated
that the strain gauge can be put easily into position by the subject
himself. Movement artifacts are easily identifiable in that movement
produces sharp deflections, while an increase in tumescence is a more
gradual sustained deflection.

It has been reported that the subject's ability to control erections
may bias the results of experiments that use the penile gauges. How-
ever, Laws and Ruben (1969) have shown that subjects' ability to gen-
erate erections voluntarily is much poorer than their ability to suppress
erections. Therefore, when erections are obtained, one can be reasonably
confident that the cues presented were responsible for generating the
erection. To circumvent the problem of voluntary suppression of
erections, it was recommended that appropriate experimental designs to
determine the reliability of erotic cues be used. Specifically, a
counter balance design has been recommended.

Another aspect of sexual arousal which has been demonstrated to
influence the response is stimulus modality. Abel et a1. (1975) com-
pared video tapes, slides or pictures, and audio-tape stimulus to 20
male homosexual subjects. The stimulus modalities were compared to
three arousal conditions. Video tapes generated the highest level of
arousal, audio tapes the lowest level of arousal, and slides an inter-
mediate level. The participants displayed substantial voluntary sup-
pression ability with the use of both tapes and slides. Interestingly,
the audio tapes effected the least amount of suppression. The authors
offered two hypotheses. One, it may be an inherent characteristic of

an auditory stimulus that it competes more readily with other thoughts

25

or images, enhancing the incompatibility of suppression of erection
responses by mental means. Second, suppression occurs only when arousal
is significantly high. Since the erections to audio tapes were con-
sistently lower than to video tapes, the suppression effect was minimized.

Available dependent measures to assess sexual arousal are sum-
marized in Table 1. Each measure is listed and the major weakness of
using that measure is also listed. It appears that measures of penile
tumescence are the most valid means of assessing sexual arousal.
Additionally, there was discussion of the research that used penile
tumescence measures. Three main characteristics of the instrument were
assessed. First, the instruments were determined to be easily self-
administered and movement artifacts are clearly distinguishable from
erection motion (Bancroft, 1966). Second, the suppression effect was
discussed (Laws & Ruben, 1969) and it appeared that audio tape stimulus
was the least vulnerable to suppression effect. Third, stimulus modali-
ties (Abel, et al., 1975) were compared and it was established that
visual films invoked the strongest sexual response, slides and pictures
an intermediate level and audio tape descriptions the least sexual

response.

Summary and Conclusions
Discussed in the review of literature were prior studies that have
investigated components of rapists' sexual arousal, classification
systems, and methods of measuring arousal. There are numerous classi-
fication systems based on motivations, object choice, id impulses,
aggression, psychodynamic patterns, and psychiatric symptoms that have

been reviewed. One common problem with all of these suggested

26

TABLE 1

Summary Comparison Chart Measures of Sexual Arousal

 

Self-Report Validity problem: Sex offenders
underestimate sexual arousal.

Galvanic Skin Response Validity problems in that one

Heart must infer sexual arousal--these

Respiration dependent measures also respond

Pupillary Eye Movement to other arousal states, i.e.,
fear, anger, anxiety, etc.

Penile Tumescence Can conclude sexual arousal
including subliminal sexual
arousal.

 

classification systems is that there is minimal research to support any
of the systems. The purpose of classifications is not the pigeon holing
of individuals into typological exclusive categories. Rather, classi-
fications could be helpful in identifying etiological factors in rape.
Additionally, it seems inherent in decision making to classify and

label types of behavior (Rabin, 1968). Therefore, there does presently
exist a need to generate research that would create better classifica-
tion models.

Because the measurement of sexual arousal has taken on considerable
specificity in the past 15 years with the development of various
erection measures, self-reports are found to be invalid as rapists tend
to underestimate their sexual arousal. This is to be expected when
presenting sexual stimuli to incarcerated sexual offenders. Other
physiological measures, while highly correlated with sexual arousal
state, also highly correlate with other general stages of arousal.
Sexual arousal then becomes difficult to infer because the anxiety or

tension in an experimental situation may confound the results. Present

27

methods of measuring erections as an estimate of sexual arousal appear
to be the most scientifically sound. Problems of suppression effects
with erection measures can be accounted for by use of counter-balance
designs in research.

Other problems, such as movement artifacts, can easily be accounted
for by differences in tracing patterns on an oscioloscope. Mechanical
strain gauges at present are inexpensive, and self-administered, and
their measurement have been shown to be reliable and highly correlated
with observation and self-reports of sexual arousal.

Preliminary studies have shown that rapists can be discriminated
from non-rapists on the basis of rape-specific cues and erection
measures. Additionally, evidence has been noted that indicates the
possibility that rape may, in fact, be the result of a character trait.
Further discriminations have been made on the basis of the frequency
of rape, age of the victim, and injury to the victim. Research has
begun the process of identifying etiological factors which, at some
point, may be used in the prediction of rape and subsequent early inter-
vention as a means of prevention. One such area of research that has
not yet been investigated is a rapist-by-victim interaction based on

type of rapist and the victim's emotional state.

CHAPTER III

DESIGN AND METHODOLOGY

This research was designed to answer two questions. First, is
there a relationship between a rapist's sexual arousal and his victim's
emotional response? Second, does the descriptive classification system
proposed by Rada (1978) differentiate types of rapists?

In this chapter the process of identifying the population is
described. Independent variables of victim affective state and
descriptive categories are operationalized. The experimental situation
tested the rapist's sexual response as a function of his descriptive
classification and of the victim's affect state. A counterbalanced
quasi-experimental design (Campbell & Stanley, 1963) was used. The
independent variable victim affect was presented in a repeated measure
design using a Latin Square format.

A statement of the research hypothesis is made. Finally, the

organization and analysis of the data is detailed.

Sample

The population was drawn from the incarcerated population of sex
offenders at the State Prison of Southern Michigan at Jackson, Michigan,

the largest walled prison in the world. An estimation made of the

28

29

incarcerated sex offender population in Michigan in the summer of 1978
established that there were approximately 800 sex offenders, incarcerated,
with 700 of them incarcerated at the State Prison in Jackson, Michigan.
In compliance with state statutes and departmental policy, it is re-
quired that sex offenders be incarcerated near or in a major walled
institution.

The population of sex offenders incarcerated at Jackson traverses
all socioeconomic classes of the State of Michigan. Michigan can be
considered representative of a national sample, in that the continuum
of rural to inner-city culture exists within the state's boundaries.

Identification of the sex offenders at Jackson, Michigan was made
through the use of the Michigan Department of Corrections Data Pro-
cessing Center. Felons were identified within the system by the charges
under which they were incarcerated.

Additionally, it was possible to identify prison residents who
were not currently incarcerated on a sex offense, but who had previously
been incarcerated on a sex offense. The Department generated a computer
list of all sex offenders at the State Prison of Southern Michigan who
were committed as sex offenders or who have a history of sex offenses.
Because the experimental design was a within-rapist group experiment,
non-rapists were not included.

Once the population list was generated from the Department, it was
screened for two restrictions. The first was that the victims of the
sex offense were female victims and at least 16 years old at the time
of the rape. The age of 16 has been chosen as a developmental mean age
when females might physically pass as adults. In this fashion, identi-

fied pedophilias and homosexuals were not included in this population.

30

The second population qualifier was that the rapist himself was
then at least 21 years old and no older than 50 years. Juvenile sex
offenders make up less than 2% of the population of sex offenders in the
state and are housed at other institutions. The age of 50 as the cut-off
for the rapists was due to physiological changes in the aging man. Males
beyond 50 have a tendency towards an increasing delay in their sexual
reaction time (Master & Johnson, 1970; Kaplan, 1974). In the present
research, delay in reaction time could have been scored as a nonresponse
to the independent variables, which would have confounded the results.

From the 700 available sex offenders, 62 were randomly selected as
the "Qualifying" sample. Of the 62 qualifiers, 29 chose to participate
and became the "Volunteer" sample. Demographic data for each members of
the sample were collected and presented in a percentile and frequency
distribution format (see Table 2). Direct comparison of the Qualifying
Sample and Volunteering Sample is possible with the demographic data.
The Volunteering Sample was representative of the Qualifying Sample on
all of the demographic characteristics except Occupation. The Qualifying
Sample had 26% with clerical occupations and 3% professional. The
Volunteering Sample had 18% clerical and 0% professional. The difference
seems to have been made up in the Unskilled occupation; in that the
Qualifying Sample had 35% unskilled and the Volunteering Sample had 52%
unskilled. This difference may be a result of the weekend days that this
research was conducted. Saturday and Sunday prison industries are not
operating. Hence there was a greater availability of free time for the
unskilled workers. Clerical or professional residents normally hold
clerical jobs in the prison and most of these jobs operate on a seven

day a week basis. Hence clerical people would have had to leave their

31

TABLE 2

Sunmary of the Demographic Characteristics of
the Qualifying and Volunteering Sample
Frequency and Percentile Distribution

 

Qualifying Sample Volunteer Sample
(N=62) (N=29)

 

Percent Frequency Percent Frequency

 

 

 

 

 

 

Race: Black 34% 21 26% 8
White 63 39 74 21
Other __3 _2_ __Q ;_'_
100% 62 100% 29
Educational Level: Grade 1-7 39 24 26 7
8-10 26 17 33 10
11-12 27 18 33 10
13+ 5 ._3 __;L ._2
100 62 100 29
Occupation: Professional 3 2 0
Clerical 26 16 18
Service Worker 0 O
Farming 0 0
Fishing 0 O
Forestry 0 0
Skilled Trade 18 11 10 3
Unskilled Trade 35 22 52 15
Structural Worker 1 1 0
Student 1 1 0
None 15 _9 _fl _6_
1% 52 100 29
Marital: Single 53 33 59 17
Married 32 20 30 9
Unknown __1_5_ _62 __l_l _3
100 62 100 29
Opiate DrugyUsage: Episodic 6 4 7 2
Addicted 2 l 3 1
Unknown ._22 .51 ._91 .26
100 62 00 29
Addicting Drug Usage: Episodic 0 0
Addicted 2 1 4 1
Unknown _98 fl _9_6_ 28
100 62 100 29
Alcohol: Episodic 10 6 7 2
Addicted 10 6 ll 3
Unknown 80 ‘59 ._82 .24
TE 62 100 29
Psychiatric History: Yes 29 18 26 8
N0 71 _4_4 .15 a
100 62 100 29

 

32

job assignment to participate in this research. Difficulties with
supervisors and/or finding replacements may have precluded the clerical
occupations from volunteering. All data was described in group form,
thereby guaranteeing the confidentiality of the participants. The
demographic data was previously gathered by the Michigan Department of
Corrections, either through interview or document review, and made avail-
able through the Data Processing Center with the identification of each
sex offender.

Variables--Measures

 

In this section the operationalization of the variables is described.
In the next section "Procedure," the measures are described as they were

used in the experimental situation.

Dependent Variable--Sexual Arousal

The review of the literature has shown that erection measures, from
a validity perspective, are the measures of choice for inferring sexual
arousal. A basic knowledge of the physiology of the penis is necessary
to understand the circumference gauge used in the present research.
Helen Singer Kaplan (1974) described the physiology of the penis

. . . the two dorsal cylinders, the corpora, cavernosa, which
are specifically adapted for erection, consist of exquisitely
engineered tiny cavers or compartments and a network of spe-
cialized blood vessels. The tiny cavers are collapsed and the
blood flows quietly through the penis when it is in a flacid
state. During erection, however, these caverns are distended
with blood which pours rapidly into the penis through the widely
opened penile blood vessels and at the same time is prevented
from withdrawing by special valves in the penile veins which can
be closed by reflex action. Thus, in essence, this process con-
stitutes a hydraulic system which utilizes blood as its fluid.
The small flacid penis is thereby enlarged and distended to the
limits of tough fascial sheath, and, concurrently, is rendered
hard and firm, and thus adapted for vaginal penetration.

The change in volumetric pressure results in a circumference change of

the shaft. Obviously, an erection is only one facet of the male sexual

33

response, but it is a necessary step for consummatory behavior. Penile
erections to varied stimuli are, then a valid indication of potential
for sexual activity involving that stimulus.

0f the three modalities of measuring sexual arousal, the strain
gauge is the most valid (Barlow, 1970). Self-reports were consistently
found to be unreliable (Walker, 1973; Abel et al., 1977) and other
physiological measures were found to be non-specific to sexual arousal
(Zucherman, 1971).

The self-administered strain gauge used was in accordance with one
described by Barlow et a1. (1970). The ring surrounds the penis but
causes no constriction or discomfort.

A diagrammatic sketch of the apparatus is presented in Figure 2.
Elgiloy stock, a corrosion-resistant surgical spring material 0.003 in.
(0.85 mm) thick, purchaseable at the Elgin Watch Company, Elgin, Illinois,
was chosen as the ring and support material, because it shows no expan-
sion with temperature. A double thickness at the gauge platform, 0.006
in. (1.7 mm) was used to prevent breaking of the gauges. Two Budd C6-141
strain gauges, with a gauge factor of 1.02, were used, purchasable at the
Budd Company, Phoenixville, Pennsylvania. It is important to use the
recommended size gauges, because smaller gauges have produced erratic
results (Barlow, et al., 1970). The two 120.0-phm gauges were wired in
a series, the common point (6) used as the output terminal and the others
(2 and 3) attached to the source voltage, as supplied by the polygraph
or batteries. Thus, as the resistance of the strain gauge was altered
by expansion of the ring, the voltage at pin 6 was increased (in milli-
volts) relative to pin 4. It was this difference that the techniques

polygraph recorded. The gauges were installed using BLll SC-4 waterproof

34

  
 

 

.cor mu m
3mm
“uh“. nun 0
and "IO
M ORCHID“. 3’00.
% I mu “1C!!!"
006! ”NM
D11: 8 HT!!! a 1'!” “III: "a

cunt:

V.
3 . ® 111: mm mm cannon
I. A m... etc-111. tux-ox:

 

77'! “4‘! 1??
CW1! nuns

“fls'M

M I
L i (D m was mm
\ ' v.0. sax us, momm'nu, n.
C cam

“—'@ © MLMl-um-WLTOI can.

 

 

 

.N)’ "1:. gmgcu IWSCS ‘ IIST‘IWITAT‘IO‘
MEI-UV
9 . UALM, MSSAD‘L’SIm ’
.. I
| .

 

Figure 1

Schematic Diagram Used for Strain Gauge

 

Figure 2

Xerox of Photograph of Mounted Strain Gauge

35

cement, purchasable at Baldwin, Lima, Hamilton Corp., Walham, Massachu-
setts. The completed assembly was tygon-dipped to make it less sensitive
to handling and to provide a smooth coating to prevent skin irritation.
The instrument costs $25.00 (1980) to build and weighs approximately
0.125 pounds (Figure 2).

In the present research the strain gauge was wired into a Pre-
amplifers Techniques TRBBBG Recorder with a Techniques TSC 820 strain
transducer, for convenient write-out of circumference change utilizing
the 20 k ohm d.c. input. Two matched variable resisters were used as
the balance of the bridge and were installed in the plug to the pre-
amplifier. The leads from the strain gauge were connected to the ter-
minals of the preamplifier which, in turn, was wired to the recorder.

A suppression correction was used to monitor the absolute value of
gauge output, therefore allowing for later calculation and transforma-
tion to full range scores. The least amount of gauge output represented
the baseline flaccid score and the greatest amount of gauge output
represented the upper limit erection score. Response to the rape scenes
was calculated as a percent value between subjects' individual baseline
erection and flaccid score.

Independent Variables--Victim
Affect State

 

 

The victim response to a sexual assault was simulated by presenting
subjects with an audiotaped description of a rape scene that was read
by a female. Each vignette portrayed a particular victim's behavioral

and affective response to a rape and lasted approximately two minutes.

36

The scripts were as much as possible, controlled for confounding
variables by standardizing time, place, and weapon characteristics while
manipulating the victim's emotional state. The five vignettes follow.

Vignette--Anger-Fight

As you walk up the driveway you spot her in the back yard. She's
got a nice ass--big, round. You want to see her naked. It's
almost dark. Her husband will probably be home soon. She turns,
and sees you. You fondle the knife in your pocket. It's there.
Just in case. She looks surprised. You stare at her breasts.
They're big too. You really want to suck on them. She tells
you to get out of her yard. She says, "You won't get near me--

I won't let you." There's a couch in the back of the garage.

You show her the knife, and you tell her you're going to fuck

her on that couch. She's really fighting now. You twist her

arm tight while you pull your cock out. She spits on you. You
slap her face and then tug at her pants. You can see her cunt.
Sweet bushy hair. You're really hot now. You throw her on the
couch. Her legs are locked tight together. She covers her pussy
with her hand, "You dirty bastard!" You slap her again and you
get on top of her. She screams. There's nothing she can do.
You've got her pinned. You pull her legs apart and shove your
cock right in her deep. It feels good 'cause you're in her so
deep. She's still fighting you. Tossing and twisting. But it
feels good--the way she's moving her hips. It's really turning
you on the way she's squirming. You tell her to keep doing that
'cause you're going to come. It feels good. And you want to
come deep inside her. She's crying now and you just keep coming.
Your cock feels so swollen. It's throbbing. You feel so good.

Vignette--Fear

As you walk up the driveway you spot her in the back yard. She's
got a nice ass-~big, round. You want to see her naked. It's
almost dark. Her husband will probably be home soon. She turns,
and sees you. You fondle the knife in your pocket. Its' there.
Just in case. She looks startled. You stare at her breasts.
They're big too. You really want to suck on them. ”Leave me
alone. Go away." You grab her arm and lead her towards the

gargage. Your cock is hard now. You can't wait to fuck her.

She starts to cry. She's trembling. "Oh God, leave me alone,
please. My kids . . ." There's a couch in the back of the garage.
You tell her you're going to fuck her on that couch. You hold
her arm while you pull your cock out. She begs you not to touch
her. You tug at her pants. You can see her cunt now. Sweet
bushy hair. You push her backwards on the couch. She covers

her pussy with her hand. You knock it away. Her legs are stiff,
heavy, locked tight together. You pull them apart and stick
your cock in. "No, no." You're pushing hard, so you can get in
her deep. She's whimpering now. There's nothing she can do.

37

She's so afraid of you. She tosses her head back and forth and
when she does her hips move. You tell her to keep doing that
cause you're going to come. It feels good. And you want to come
deep inside her. She's crying hard now and you just keep coming.
Your cock feels so swollen. It's throbbing. You feel good.

Vignette--Ca1m-Reasoning

As you walk up the driveway you spot her in the back yard. She's got
a nice ass--big, round. You want to see her naked. It's almost
dark. Her husband will probably be home soon. She turns, and sees
you. You fondle the knife in your pocket. It's there. Just in
case. She looks curious. You stare at her breasts. They're big
too. You really want to suck on them. She asks you who you are.
You rub your cock. "Look, whatever it is you want let's talk about
it." You grab her by the arm and start dragging her towards the
garage. Your cock is hard now. You can hardly wait to fuck her.
"Look, let's talk, you don't need to do this. If you have a pro-
blem, we can talk about it." There's a couch in the back of the
garage. You tell her you're going to fuck her on that couch. "I
don't want to fuck with you. You don't really want to force me.
You're a good looking guy. You can get lot's of girls, you don't
need to do this. Let's talk, and if you still want to, then we can
fuck. It's just not good this way." You hold her arm while you
pull your cock out. "I have a husband and kids. That's why I don't
want to. If you're lonely, we can talk about it. Did something bad
happen to you today?" You tug at her pants. You can see her cunt
now. Sweet bushy hair. You push her backwards on the couch. "Wait,
please, I told you I don't want to do this, we can figure out some-
thing for you. You can find a better way to get what you want."

She covers her pussy with her hand. You pull it away and spread her
legs apart. You shove your cock right in her. "Stop, I don't want
this. I don't know you. We should at least talk first." You push
harder so you can get in deep. It feels good 'cause you're in her
so deep. She goes limp. There's nothing she can do. She barely
moves. You put your hands under her ass and make her rock up and
down. It feels good the way her hips move. You tell her to keep
doing that 'cause you're going to come. It's really turning you on.
"Keep moving! Keep moving!" you say. It feels so good. And you
want to come deep inside her. She's crying now, and you just keep
coming. Your cock feels so swollen. It's throbbing. You feel

so good.

Vignette--Submission

As you walk up the driveway you spot her in the back yard. She's
got a nice ass--big, round. You want to see her naked. It's
almost dark. Her husband will probably be home soon. She turns,
and sees you. You fondle the knife in your pocket. It's there.
Just in case. She's startled. You stare at her breasts. They're
big too. You really want to suck on them. "Who are you--what are
you doing in my yard?" You rub your cock. 'You grab her arm and
lead her toward the garage. "I'll do whatever you want, just don't
hurt me." Your cock is hard now. You can't wait to fuck her.

She starts to tremble. There's a couch in the back of the garage.

38

You show her the knife and you tell her you're going to fuck her on
that couch. She starts to cry. Her arms go limp. She stands there
while you pull your cock out. You point to her pants. She slips
them off and you can see her cunt. Sweet busy hair. You're really
hot now. You push her onto the couch. She looks scared. You pull
her hand away from her pussy and you spread her legs apart. She
just lays there whimpering with her face turned away from you. You
shove your cock right in her. You're pushing hard so you can get
in her deep. It feels good 'cause you're in her so deep. She's

so stiff. She barely moves. You put your hands under her ass and
make her rock up and down. I feels good the way her hips move.

You tell her to keep doing that 'cause you're going to come. It's
really turning you on. "Keep moving, keep moving," you say. It
feels so good. And you want to come deep inside her. She's crying
now and you just keep coming. Your cock feels so swollen. It's
throbbing. You feel so good.

Vignette--Lust

As you walk up the driveway you spot her in the back yard. She's
got a nice ass--big round. You want to see her naked. You fondle
the knife in your pocket. It's there. Just in case. You stare at
her breasts. They're big too. You really want to suck on them.

She watches you rub your cock. She smiles. You take her arm and
lead her towards the garage. There's a couch in the back of the
garage. You point to it and tell her you want to fuck her there.
She just smiles again. She watches as you pull your cock out. She
really wants it. You tug at her pants. She slips them down and
you can see her cunt. Sweet busy hair. She opens her blouse and
you nuzzle her tits. She laughs a little. Then she falls back on
the couch. She spreads her legs wide apart. You shove your cock
in. She cries out. You're pushing hard 'cause you want to get
deep inside her. You can feel her muscles tighten around your cock.
It's so warm in there. And wet. In a low voice she says, "Oh fuck
me, fuck me." You startlidVing in her, and she moans, "More, more."
She wraps her legs around you and she thrust her hips forward.

Your hands are all over her. You're really ot. It's really turn-
ing you on the way she's squirming. You tell her to keep doing
that 'cause you're going to come. It feels good, and you want to
come deep inside. Her eyes are closed now, and she's still moaning.
She's tossing her head back and forth. She's clutching your
shoulders. She's going to come too. You can feel her whole body
quiver. She's squealing now and you keep coming. Your cock feels
so swollen. It's throbbing. You feel so good.

Four pretests and one pilot study were conducted to ensure equal
arousal levels in the victim response tapes and to validate the labels
given to their content by the experimenter. The experiment began with
four vignettes of similar levels of eroticism to ensure that, in the

experimental condition, differences in rapist arousal, could be

39

attributed to their differential responses to the four types of victim
states. The fifth tape, the lust response, was an exception, because,
by its nature, it is, and is designed to be, more arousing. In the
pilot study, all five tapes were played for 17 felons who were incar-
cerated for various violent offenses and were currently involved in
group therapy at the prison. The felsons were asked to listen to the
tapes and rate them on a written survey (Appendix B) for their levels
of sexual arousal. They were also asked to identify the victim's
affective state. The results of the pilot study are presented here in
a percentile distribution format (Table 3). The arousal levels of
Tapes One through Four were very similar given either 1 or 2 rating of
not sexy at all or a little sexy. Tape Five, as expected, was rated as
almost twice as arousing by this group of non-sex offenders. For all
tapes, the pilot subjects selected the same label as the experimenters
more than half the time, and for most tapes, there was still higher
agreement. Pretest Tapes One through Four were considered to be proto-
types, and were used for feedback purposes to the actress. The labels
"Afraid" and "Gave In" seemed the most confusing for the pretest groups.
The confusion was reflected in the labeling of tape "Acquiescing" where
the scores on the pretest were distributed between "Afraid” (35%) and
"Gave In" (63%) labels. The labeling schema was retained, because the
accurate label "Gave In" was still used twice as many times as the
inaccurate label "Afraid." The rest of the labels were scored in what
might be termed a ”purer" manner. An example of the purer labeling is
contained in the scoring of the lust tape which received 94% of the

scores for "Turned On."

40

TABLE 3

Vignette Pre-Test Results Response Distribution

 

 

 

 

 

 

 

 

 

 

Tape 1 Tape 2 Tape 3 Tape 4 Tape 5
Angry, Terrified Acquiesing Calm Lust,
Physically Pleading Quickly Reasoning Want Sex
Resisting
(score)
Not Sexy (1) 13 13 14 8 3
At All
75
3 A Little (2) 3 2 2 8 6
-' Sexy
7;
§Sexy (3) l 2 1 .3
$-
‘F Quite (4) 1 1
Sexy
Very (5) 4
Sexy
Angry 10
i=3
g Afraid 4 l6 6 l
(D
‘5, Gave In 2 1 ll 1 1
<
'3‘, Tried to
Q Talk Him
-‘ Out of It
Turned 0n 1 l 16
Average 1.3 1.7 1.3 1.6 2.6
Arousal

Score

 

41

Descriptive Categories

The descriptive categories used for purposes of this study were from
the character types described in Chapter I (Rada, 1978): (l) the psycho-
tic rapist; (2) the situational stress rapist; (3) the masculine identity
conflict rapist; (4) the sadistic rapist; and (5) the sociopathic rapist.
The placement of rapists in each category was performed as follows.

Prior to a felon being sentenced to the Department of Corrections,

a Pre-sentence Investigation Report is drafted to assist the judge in
making an enlightened judgment concerning the felon's sentence. That
Pre-sentence Investigation Report is forwarded with the felon at the
time of incarceration, and becomes a part of the felon's permanent
institutional record.

The content of the Pre-sentence Investigation Report includes an
arresting officer's description of the offense, an offender's description
of the offense, and the victim's description of the offense. A home
visit is made and opinions of the immediate family or spouses are
enlisted. Finally, a social-behavioral workup is constructed about the
offender, which includes sections on family history, criminal history,
educational development, career development, military activity, substance
abuse problems, religious orientation, and mental-health history. The
Pre-sentence Investigation Report was the document from which the rapists
were estimated on the descriptive categories

Table 4 is a list of the descriptive categories with an item
description that identifies them. The existence of a descriptive cate-
gory can be verified by its manifest behavior. It is assumed that the
descriptive categories are heterogeneous; inclusion of one category is

not assumed to be at the exclusion of other descriptive categories,

42

TABLE 4

Descriptive Categories With Item Description

 

Category

Rating Sheet Source*

 

Tendency toward victim violence

Tendency toward weapon violence

Tendency toward victim
degradation

Tendency toward homosexuality

Tendency toward mental illness

Tendency toward passive or
aggressive personality

Environmental press

Single continuous item with five
levels: Range of verbal threat
to excess use of violence with
death. (Item #20 Appendix C)

Single continuous item with five
levels: Range of unseen threat
of weapon to use of weapon with
death. (Item #21 Appendix C)

Five dichotomous items exploring
range of sexual—sadistic behaviors.
(Item #1, 2, 3, 4, Appendix C)

Three dichotomous items to assess
three levels of homosexual inter-
action. (Item #11, 12, 13 Appen-
dix C)

Five dichotomous items to assess
past history of professional con-
tacts and level of problem. (Item
#6, 7, 8, 9 10 Appendix C)

Single continuous item: Calls for
subjective summary assessment by
rater. (Item #22 Appendix C)

Six dichotomous items to assess
situational personal-economic
stress areas of functioning.
(Item #14, 15, 16, 17, 18, 19
Appendix C)

 

*See Appendixll for complete rating sheet.

43

since the quantum of a descriptive category exists on a continuum from
marginal existence in the personality to dominance of the personality.
The categories were transposed to an item evaluation sheet (Appen-
dix C) for easy use by the raters. A random sample of 62 subjects was
taken from the list of 700 sex offenders incarcerated at Jackson and

was classified on all descriptive categories by the raters.

W
The gathering of data for the present research was conducted in two
separate settings. The first was the classification of the rapist popula-
tion on the descriptive categories. The second was the experimental
situation in which the rapists were exposed to the independent variable

victim affect state.

Population Classification

 

The ten raters were all mental health professionals working within
the Department of Corrections for at least one year. The age range was
from 24 to 50 years old, and included both male and female, single and
married individuals. The staff was composed of M.A.s or beyond in
counseling, clinical psychology or clinical social work. Initially, the
raters were contacted by a face-to-face interview with the researcher.
The contact was followed up with a reminder letter (Appendix E) a few
days prior to the classifying of the population on the descriptive cate-
gories. After the population classification, a thank-you letter was
sent to the raters (Appendix F).

Using departmental clinicians had numerous advantages. First, they
understood the confidentiality of the documents with which they were
working. Second, they had experience at recognizing the source document

in a folder which can accumulate a voluninous amount of documents. Last,

44

as departmental clinicians, they were already approved by the Michigan
Department of Corrections and, hence, did not cause any concern on the
department's part in terms of confidentiality and liability.

Prior to assessing the rapists' Pre-sentence Investigation Reports,
a training session was conducted. The Population Rating Sheet (Appen-
dix C) and a sample of the Population Rating Instruction Sheet (Appen-
dix D) were distributed and explained. An example Pre-sentence Investi-
gation Report was assessed using all items of the Population Rating
Sheet. A question and answer period followed. All raters then met
together in a large office at the Prison Complex and reviewed the source
documents for the same six hour period on a Saturday in May. The re-
searcher also remained available throughout the six hour work period to
further clarify any questions.

During the six hour work period all participants "blind" rated the
same seven subjects to obtain a reliability estimate of the raters. The
interrater reliability was assessed by a percent agreement matrix among

all raters for each item on the population rating sheet.

Experimental Procedure

 

Prior to the beginning of the experimental data collection, the
necessary arrangements for rooms were made with the hospital administra-
tion, and sponsoring physician Dr. Bruce Cole. The research required
the use of two separate rooms. One room was used as a debriefing room.
In the carpeted experimental room, subject booths were created by the use
of moveable hospital screens. Each booth was equipped with a cloth-
upholstered arm chair, headphones, and strain gauge. Individual subjects
were seated in individual subject booths that faced a common movie

screen, thus maintaining individual privacy. Full-light illumination

45

Movie Screen

 

 

 

Movable
Hospital
Screens
S.| S2 S3 54
Research Table
Tape Recorder PreAmp
Audio Mixer Polygraph

 

 

Each subject booth equipped with chair, headphones, strain gauge.

Figure 3

Experiment Room Layout

46

was used during entry and departure from the experiment. During the
experiment, the lights were off and the windows were covered with both
shades and dark curtains, thus creating a dark room.

Behind the subjects was the research table which held the pro-
jector, preamplifier, the polygrath recorder and a tape recorder. A
mixer was used to allow the experimental tape to be played to all sub-
jects simultaneously. Throughout the experiment, the researcher remained
at the research table, out of sight of the subjects.

The second room used as the debriefing room was an adjacent office.
The hallway between the two offices was used as a waiting room, and the
security personnel remained seated there.

The subjects' consent to participate was a two-step process. The
research was conducted on a weekend to minimize upsetting the institu-
tional routines. In Step One, the subjects received an introductory
letter (Appendix A) on Wednesday of the research week, announcing the
availability of a research project for their participation. Specifics
of the present research were purposely omitted from Step One consent.
Indicating that the present research was concerned with sexual deviance
could have skewed the sample. A possible group of subjects could have
eliminated themselves based on what they might have fantasized as
negative ramifications of acknowledging their sexual deviance. On the
weekend mornings of the research, all subjects who received letters were
placed "on call" in groups of feur at sequential times. By choosing not
to show up for their "call" at the hospital at the designated time, the
subject exercised his option not to participate.

Upon arriving at the scene, subjects who chose to participate were

seated in the waiting area. The security personnel had been directed

47

not to engage the subjects in conversation, particularly as related to

the research. In order to create a non-personal, professional atmosphere,
the experimenter wore a white coat and addressed the subjects in a
friendly, but firm, manner. At the beginning of each rotation, the
researcher entered the waiting area and invited the subjects into the
experimental room by stating the following:

Transcript
Pre-Experimental Instructions

Gentlemen, just have a seat and place on a set of headphones. All
instructions, information, and your right to choose not to parti-
cipate are all prerecorded for uniformity.

(subjects seated)

Gentlemen, you are being asked to participate in a research project
on human arousal or turn on. The experiment involves your watching
sexy movies and listening to sexy sounds, while your response is
being measured. Your response will be recorded by a pen-graph
recorder and a penis circumference gauge. Mr. Andy Marcin, a
Physician's assistant, will help you put on the circumference

gauge. The gauge works a lot like a blood pressure instrument.

Mr. Marcin will make sure that the gauge is put on right. Once

it is in place don't touch it or yourselves. The total experiment
will last approximately one hour with a debriefing session imme-
diately afterwords. All responses are being tabulated and presented
in group form so nobody will know how any one person responded.

Your participation is greatly appreciated as it will allow for com-
pletion of my doctorate in psychology at Michigan State University.
Additionally, for participating I will place $5.00 in your institu-
tional account on Tuesday morning, August 13, 1980 for participating.
Before beginning it is now necessary for each of you individually to
decide if you wish to participate. If you do not want to partici-
pate simply take off your headphones and replace them on the hook.
Leave the room immediately and the officer will sign your pass.

If you wish to continue, simply sit back and listen. I will con-
tinue with more instructions in a moment. Decide now. (music 1.5
m1n.

At this time, subjects had observed a penis circumference gauge,
as each subject booth contained a gauge, hanging next to the chair. 0f
the 62 subjects who were screened to participate, 30 indicated a willing-
ness to participate by showing up for their appointment at the hospital

unit. Of these 30, only one subject exercised his right not to

48

participate at this step of the informed-consent procedure. The remain-
ing 29 subjects chose to participate by remaining in the room, and were

exposed to the following procedure.
Instructions Transcript

Thank you for your decision to participate in this experiment.
Before beginning it will be necessary for you to sign an informed
consent, which is on the clipboard below your seat. Pick it up
now. The first piece of paper reads as follows:

(Appendix G)
Please sign your legal signature on the first blank line, then
write your institutional number on the second line, I have already
signed my name on the third line. After signing your name, place
the board back under the seat. Sign now. If during the experiment
you change your mind or need to leave raise your hand and Mr. Marcin
will remove the gauge and direct you out. Now Mr. Marcin will come
to each of you individually and put on the gauge. Please assist

by pulling your pants and undershorts down below your balls, then
sit back down.

(experimental procedure)

When the penis circumference gauge had been properly placed and
inspected by the on-site physician, the experimental stimuli were pre-

sented in a repeated measures format (Table 5).

TABLE 5

Experimental Procedure: Stimulus Presentation Fonnat

 

 

Type of Erotic Relaxing Tape R T R T R T R T Erotic
Stimulus Film Music 1 M 2 M 3 M 4 M 5 Film
Duration 8 5 3 4 3 4 3 4 3 4 3 8

(in minutes)

 

The victim audio tapes were then played for the subjects. As the sub-
jects arrived at the experimental site, they were assigned an order of

tape presentation, determined by a Latin Square formula (Winer, 1971).

49

The first group of subjects received order one and the second group
received order two, etc. (see Table 6). Order two tapes were only used
once because of tape breakage. Hence only two subjects received exposure
to order two. A film approximately eight minutes long and of sufficient
erotic content to ensure arousal in the experimental situation was used
to obtain a maximum erection score. One of the films showed a black
heterosexual couple, and the other a white heterosexual couple. Two
films were used in keeping with the fact that most rapes are intra-
racial (Amir, 1971) and many of the subjects were black. After the
stimuli were presented all subjects listened to the following instruc-

tions:

Post Experimental Instructions

Thank you gentlemen. The experiment is complete. Mr. Marcin will
now help you in removing the gauges. Please remain seated and

do not touch yourself or the gauge. Mr. Marcin will now come to
each of you individually. After the gauge is removed pull your
pants up and sit back down.

(music)

Pick up the clipboard under your seat. The second page reads as
follows:

(Appendix H)
Sign your name using your legal signature on the first blank line.
Then put your prison number on the second line and my signature is
on the third line. The last line says: I am tenninating my parti-
cipation and would like to consult with a physician. If you want
to go to the emergency room mark that line. Place the clipboard

back under your chair. Once again thank you and now take off your
headphones and hang them back on the hooks.

After signing the experimental release, the subjects were taken to
the debriefing room. The debriefing was done by a Ph.D. Clinical

Psychologist employed by the Department of Corrections at the time of

the experiment. The debriefing room was a physician's office. Additional

50

Table 6

Latin Square: Order of Tape Presentation

 

Number of Subjects

 

Order Receiving This Order
1 T1 12 T3 T4 TS 7
2 12 13 T4 15 T1 2
3 T3 T4 15 11 T2 7
4 T4 15 T1 12 T3 7
5 15 T1 T2 T3 15 __6
N = 29

 

Note: T1 = Anger; T2 = Fear; 13 = Acquiescance; T4 = Reasoning;
15 = Lust.
cushioned arm chairs were positioned in the room to create a circle.
The debriefing was an unstructured group intervention in which the
primary aim was to assess any ill effects resulting from the experiment.
Had there been difficulties that required further attention, a subject
would have been referred to the Psychological Services outpatient unit
in the same complex. Since the experiment was conducted in the infinnary,
a seriously decompensating subject could have been immediately referred
to the inpatient Clinical Services Psychiatric Unit. During the debrief-
ing, the subject received an Experimental Explanation Sheet (Appendix I)
detailing their involvement. Of the 29 subjects participating, none
requested follow-up, nor were any judged by the debriefing clinician as

needing referral or follow-up.

51

design

This study was an ex post facto experimental design which analyzed
two different relationships. The first of these relationships was that
of the subject's sexual arousal to the victim's emotional state, as
represented by audio tape simulations of rape. The second analysis was
that of the relationship between the subject's sexual arousal to the
victim's emotional state and his own descriptive classification.

The design used to answer the first question, as diagrammed, in
Table 7, was a repeated measures order by subjects by tape presentation
design. The analysis of the second relationship was based on the
assumption that the subjects would be different on scores of arousal and
descriptive categories. Hence, the design answers the question of there
being any identifiable corresponding uniform change on the two sets of
variables. A correlational analysis was considered particularly appro-
priate, as it defined the magnitude and direction of any corresponding

change.

Experimental Hypothesis

I. Rapists will differ on sexual arousal as measured by penis
circumference while listening to audio vignettes of rape
scenes. It is hypothesized that there will be significant
differences on sexual arousal scores.

II. Rapists are different in descriptive categories and presumed
to be different in sexual arousal. It is hypothesized that the
correlation between arousal scores and descriptive category

scores will be significant.

52

TABLE 7

Repeated Measures Design

 

Tape 1

Tape 2

Tape 3

Tape 4

Tape 5

 

Order 1

Order 2

Order 3

Order 4

Order 5

(D mmmmmmm

U1

mmmmmmm

MMMMMMMMMMMU’U’U’

 

53

Analysis

The size of the sample, representative adequacy of the sample, and
the use of internal data, suggested the applicability of parametric
statistics (see Table 2, p. 31). Scores for sexual arousal were obtained
by a linear transformation of the subject's microamp scores to percent of
full arousal scores. Variations in penis size and circumference gauge
output resulted in different microamp baseline readings for each subject.
To make inter-subject comparisons it was necessary to standardize their
arousal scores. Percentile scores were chosen for the linear transfor-
mation because of their use in previous related studies (Abel et al.,
1975, 1977; Barlow, 1970), and their use would not change the experi-
mental outcome. The assumptions of the multivariate analysis of variance
called for normality of sample independence of scores and homogeneity of
variance. The linear transformation to percentage scores relabeled the
terms without distorting the distribution. A recheck of the assumptions
reveals no changes with the score transformation. The assumption of the
independence of scores remains violated. Without being robust this
assumption can only be violated by the use of a repeated measures format
which this research employs. The homogeneity of variance is preserved
as the sample size remains constant. The assumption of normality which
was violated by the original distribution remains robust as the sample
size remained at 27 scores for each cell of the data analysis. Hence,
inferences about the size of a type I error were not affected by the
linear transformation of scores.

Hypothesis I was tested using a two-way multivariate analysis of

variance with repeated measures format. The F-test of significance was

64

performed for within-subject tape effects and for order of presentation
as a confounding variable.

Hypothesis II was tested using the Pearson Product-Moment Correla-
tion Coefficient. Scores on descriptive category items were linearly

transformed from item scores to additive and cumulative trait scores.

Summary

Of the 62 subjects randomly selected to participate, 29 volunteered
to have their sexual arousal monitored while listening and viewing
sexually arousing stimuli. Sexual arousal was measured with the use of
a penis circumference gauge. Exposure to sexually arousing stimuli con-
sisted of listening to simulated rape scenes and viewing erotic films.
Audio tapes and visual films were pretested to validate their arousal
level. A multivariate repeated measures design was used to test for
stimuli effect and order of presentation effects.

Subjects' descriptive category scores were obtained by evaluation
on social and behavioral items. Trained raters assessed historical court
documents to obtain subjects' item scores. Rater reliability was esti—
mated by having all raters "blindly'I rate the same seven subjects. A
correlational design was used to estimate the uniform change between

arousal and descriptive categories.

CHAPTER IV

ANALYSIS

In this chapter, the results of the study with the appropriate
statistical analyses are presented. The first presentation is a descrip-
tion of the tabulation and transformation of arousal scores. The
research hypotheses and their respective analyses are then discussed.
Finally, an examination of extraneous effects, some of which influenced
the hypotheses, are presented.

One data transformation was performed. The data transformation
converted erection responses from graph tracings to percent erection
scores. Some erection responses exceeded the recording limits of the
pen chart, therefore, a suppression correction gauge was used to monitor
the absolute values of the gauge output. The conversion formula used to
obtain the subject arousal score is presented in Figure 4. For the
entire experimental condition, the gauge resistance value (GR) was left
at .01 microvolts. All of the subjects were monitored for penis strain
resistance value (PSR) prior to the beginning of the experimental con-
dition. These two scores were added together to give the beginning
resistance value of the pen graph in microvolts or (BR). The divergent
tracing scores at each of the five experimental tapes were then added

to the beginning resistance value. The summation yielded the test

55

56

GR+PSR=BR 1_(§y
BR + TCR = |TCR| BR Beginning Resistance Value

PSR Penis Resistance Value

GR Gauge Resistance Value

TCR Test Condition Resistance Value
Figure 4

Subject Arousal Score Tabulation Formula

condition resistance (TCR) absolute value at each experimental condi-
tion (Figure 4).

In Figure 5 conversion of the test condition resistance value (TCR)
to a percentile value of erection at Test Condition (PVTC) is represented.
The TCR-flaccid was subtracted from the TCR-erection, yielding the full
range of resistance values for each subject. To individually define the
percent erection value at each of the experimental conditions, the test
condition value was obtained, using the fonnula described in Figure 4.
The TCR-Flaccid value was subtracted from the test condition resistance
value yielding the divergence from the flaccid resistance value (DFR).
The DFR was then divided by the full range for each subject, which
resulted in the percent value of the erection at each of the experimental
conditions (PVSTC). These percentile values were used in the multi-
variate analysis of variance as well as in the correlation analysis.

The second data transformation was that of changing the item scores
on the population rating sheet (Appendix C) to cluster scores as defined
by Table 4 (p. 42). The clusters of items were reduced to total scores
by computing the proportion score for each cluster. The total scores
were computed accounting for missing data; that is, if a person had only
three data points out of five, then the total score was divided by

three. These items were retained in their clusters because the total

57

|TCR-EI - |TCR-F| = RR Iggy
|TCR| - |TCR-F| = |DFR| TCR Test Condition Resistance Value
E Erection
F Flaccid
.053 = PVTC X Experimental Stimuli
RR RR Range of Resistance Values

DFR Dispersion from Flaccid
Resistance Value
PVTC Percent Value at Test Condition
Figure 5

Percentile Transformation Formula

score still suggested an increased intensity of the variable. For
example, each of the items included in the tendency toward victim violence
was considered to add to the overall severity of the crime. This resulted
in each subject receiving a cluster score on all of the descriptive cate-
gories. These cluster scores were used in the Product Moment Correlation
Coefficient to analyze the relationship between arousal and the descrip-
tive categories.

Research Hypothesis I

I. There will be significant differences on sexual arousal scores
as measured by penis circumference.

In Table 8, degrees of freedom (DF), mean square (MS), F-test ratio
(F), and significant level (P), are given. Both the research hypothesis
(main effects for tape) and the confounding effects hypothesis (tape and
order of presentation interaction) were tested. 0f the 29 subjects
participating, 100% showed variations in their arousal scores. Eighty-
six percent (N = 25) of the subjects' arousal scores were influenced by
order of presentation. Both the main (p<.001) and interaction effect
(p>.001) were found to be statistically significant. Therefore, the
research hypothesis is supported but confounded by the significant

interaction effect of order of presentation.

58

TABLE 8
Repeated Measures MANOVA Summary

 

 

Source DF MS F P
Hypothesis I 1.22 1737870.37 170.50 .0001
(Tapes)
Hypothesis II 4.22 58.68 5.02 .0001

(Tapes & Order)

 

Research Hypothesis II

II. The correlation between arousal scores and descriptive

categories scores will be significant.

Table 9 is the matrix of correlations scores for the descriptive
category scores and their respective victim arousal score using the
product moment correlation coefficient. Occupational stress category
and angry victims yielded the strongest positive direct correlation
(r = .50). Victim degradation category and submissive victim produced
the strongest negative or inverse correlationship (r = -.25). One corre-
lation, “angry" victim style and rapist category "need for weapon vio-
lence," yielded no relationship (r = .00).

Figure 6 charts the correlation matrix (Table 9) between arousal
scores to types of victim and descriptive category scores. Plotted
across the descriptive categories are the correlation scores of the five
victim affect tapes.

Only two of the descriptive categories were recognized as signifi-
cantly correlating with arousal to rape tapes. These were occupational
stress and aggress1ve aescr1pt1ve trait. Occupational stress correlated

significant (r > .35) with arousal response to victim emotions of

59

TABLE 9

Correlation Values for Descriptive Categories
by Arousal Tapes

 

Descriptive Categories

 

 

 

: 73

O C O)

OF 0 >

H 0!- Q) Q) 1'1-

6 U) r— 4—1 U U 00')

EU v—m cum rum E: :: >111

'0-0‘6 ('50) 44m (101 ~1-QJ 00) PG)

+35- “: w-OJ 30.) 4d!— D-f— U13-

0:» :l— LS- US- 00 I130 mm

'1-Q) 0)!— GH UH ’1'". w-:— (601

>0 2H EU) 04/) >> 3> O—<

Submission -.25 -.O9 .04 15 - O8 - O7 47
U)

§Anger .02 -.07 .23 .50 -.08 .00 .36
|—

7; Reasoning .02 .10 .07 41 .10 - ll .26
V)

g Fear .13 -.20 .13 13 .05 -.11 .20
<

Lust .01 -.06 .10 .46 17 -.20 .18

anger (r = .50), lust (r = .46), and reasoning (r = .41). Aggressive

descriptive correlated significantly (r > .35) with arousal response to
victim emotions of submission (r = .47) and anger (r = .36). Other
variables tended to cluster about the a-axis (r = .00) and did not indi-
cate positive or negative relationships of significant magnitude. The
other variables correlated less than .35 with arousal response to rape
vignettes. Validity of variables as predictors of a behavior are not
used below .35 correlations (Borg & Gall, 1963). Statistical signifi-
cance tables for correlation coefficients at the p<.05 indicate using a
r 36.5 (Borg & Gall, 1963). The more conservative .35 value was chosen
because the variables were constructed with these predictive values in

mind. Hence, all but one correlation are not equal to zero, but are of

Magnitude & Direction of Correlation

6O

DESCRIPTIVE CATEGORIES

 

0)
m >
C m o'—
0 cu m
.,.. s. m
+-‘ 44 a) 0.1 CD
to m m m u u s.
a 0 :3 a = r
S- C S- f; r— 02)- 2
an r— +1 s: o o
1% E: V’ .9 E; E: '
l— 4-’ Q)
E u- to to E c >
.F- G *J Q 0'- o 'F'
4—1 +3 w- 3 +4 D. U1
0 c s. u u as m
'I- Q) (U U -r- 0.1 co
> Z Z O > 3 a.
.60
(Submission)
.40
(Anger)
/
/ (Reasoning)
.20 (Fear)
/ (Lust)
I
/
I
(+)
o 00 K
.20
.40

Figure 6

Graphic Representation of Correlation Values for Descriptive
Categories Ryzérousal Tapes

61

insufficient magnitude to warrant further use of the descriptive cate-

gory instrument.

Factors Influencing the Hypothesis

A factor which was tested statistically was recognized as affecting
each of the hypotheses. Hypothesis I, which tested for arousal scores
across tapes, may have been particularly influenced by order of presen-
tation. Because of the repeated measures format, arousal fatigue may
have confounded the results.

Two operations were conducted which attempted to reduce and estimate
the influence of the confounding variable of order and fatigue. The
first of these was in the presentation of the stimuli. The audio tape
vignettes were presented in a counter-balance Latin Squares design
(Campbell & Stanley, 1963). Second, a main effect for order was tested
in the repeated measures multivariate analysis. The test was opera-
tionalized by crossing subjects with tapes and nesting them within order
of presentation (Table 7, p. 53). Table 8, (p. 58) contains the obtained
means squares and the respective F-tests for the interaction of tapes
and order. A significant interaction effect was noted. In Figure 7 the
interaction effect can be observed in the plotted cell means. Differ-
ences in arousal were observed, however, they are significantly influ-
enced by order of presentation.

The influence of order of presentation was observed on individual
scores. In 86% (N = 27) of the subjects, the first type of victim
stimuli presented elicited the greatest arousal. Likewise the last
victim stimuli elicited the least arousal scores. In 14% (N = 14) of
the subjects this effect was not found. It seems logical that

susceptibility to fatigue has a range of response. Hence, 14% of the

Level of Arousal in Percent

 

62

80
75 +6
7 - ‘4" :11 ’ c (Fear) - ' -
0 4 - -
B if a ”05° / o. A (Anger) .....

0 (Reasoning) xxxxx

8 (Submission)

 

5
0
Order
I II III IV V
Figure 7

Graph of Cell Means to Demonstrate Interact'
1on
Effect of Order of Presentation and
Tape Stimulus in MANOVA

63

subjects not having fatigue effect suggests that individual variation
does exist.

The testing of Hypothesis II yielded its own source of confounding
influences. This source was the reliability of the descriptive cate-
gories. If they could not be reliably scored, discussions concerning
their validity were superfluous to this project. Two operations were
conducted to reduce and assess the impact of the reliability of scoring
of the personality trait instrument. First, the expert raters were given
a training session prior to their evaluation of the rapists on the social-
behavioral items. Second, the reliability of the raters was estimated by
having raters "blind“ rate seven files that each of the other raters also
rated. The interrater reliability was assessed by a percent agreement
matrix among all judges for each item on the population rating sheet
(Appendix J). Table 10 is the summary of the interrater reliability data
which was shown in the form of average percent agreement for each item
by all ten raters. The range of average agreement was 36.30% through
100%. The total mean percent agreement across all judges on all items
was 78.4%.

The least agreed upon items were "Passive-Aggressive" personality
continum (36.30%), "Occupational Job Decline" (42.22%), and "Occupation
Dismissal" (48.89%). The "Passive-Aggressive" item was the least con-
crete item and was a summary intuitive judgment of the raters. Hence,
it was expected this item would be agreed upon least because of raters'
'individual biases. Also occupational histories were minimum in the
source document resulting in a greater vulnerability to rater biases.

In contrast the most agreed upon items were concrete behaviors or

diagnoses and were consequently least vulnerable to rater biases. The

64

TABLE 10

Average Percent Agreement Among Raters (N = 10)
on Items in Descriptive Categories

 

 

Item Number Item Label Percent Agreement
1 Intercourse 93.33
2 Fellatio 100.0
3 Sodomy 75.56
4 Mutilation 93.33
5 Ritualistic Degradation 100.0
6 Outpatient 64.44
7 Inpatient 81.48
8 Diagnosed Neurotic 100.00
9 Diagnosed Character Disorder 63.70

10 Diagnosed Schizophrenia 72.59
11 Passive Homosexual 93.33
12 Predatory Hmosexual 93.33
13 Cooperative Homosexual 93.33
14 Relationship Separation 62.96
15 Relationship Fighting 70.37
16 Relationship Divorce 86.67
17 Occupation Dismissal 48.89
18 Occupation Quit 54.07
19 Occupation Job Decline 42.22
20 Victim Violence 57.78
21 Weapon Violence 74.81
22 Passive-Aggressive 36.30

Average Percent Agreement = 78.4%

 

65

most agreed upon items were "Fellatio" (100%), "Ritualistic Degradation"
(100%), and I'Diagnosed Neurotic" (100%).

The traditional level of percent agreement for direct observation
of behavior is 70% (Bruning & Kintz, 1968). Because the raters were
limited in this study by having to rate second hand data (i.e., court
reports instead of observations), the reliability was expected to be
somewhat lower. Observing second hand data may explain the range of
64% variation in agreement on some items. More importantly, the mean
percent agreement of all judges on all items exceeded by 8.4% the level
one would expect if the judges were directly observing behavior (70%).

Hence, sufficient reliability of judges was obtained.

Summary

The results of the data analysis were presented in this chapter.
Data transformations which were performed to complete the analyses were
presented first. The first of these transformations consisted of con-
verting microvolt resistance values to percentile erection scores. The
second included the converting of item responses by the expert raters
into variable cluster scores. Next, the analyses of the hypotheses
were described. In testing Hypothesis 1, a repeated measures multi-
variate analysis indicated that there were main effects of tapes. How-
ever, there was also an interaction effect for tapes and order of pre-
sentation. Hence, the null hypothesis was rejected, however, the
significance of the results were confounded by the interaction factor.
A correlational matrix used to test Hypothesis 11 indicated that the
correlation between arousal to rape vignettes and scores on descriptive
categories were other than zero. However, only two of the seven cate-

gories were noted to have correlation coefficients of significant

66

predictive value. Hence, further use of the instrument in classifying
rapists was not recommended. Finally, a presentation of the factors
influencing each of the hypotheses was described. Two of these were
noted, and operations were described that were conducted to assess the

impact of the confounding variables.

CHAPTER V

SUMMARY, CONCLUSION, DISCUSSION, AND IMPLICATIONS

Summary

Rape is the fastest growing reported crime of violence in the
United States. Two identified problems were addressed in the present
research. The first of these was the~advocacy of advising potential
victims of rape to use physical defense in defending themselves. Inher-
ent in this advice is the assumption that all rapists are alike and will
respond in kind. The second issue of this research was to test the
validity of a classifications system which had the potential for usage
in a correctional setting.

Numerous classifications systems that are based on motivation
(Cohen, 1971), object choice (Brancle, 1952), Id impulses (Williams,
1965), and others were reviewed. The common weakness of all the classi-
fications systems reviewed was the lack of validating research.

The present study used the classificatory framework suggested by
Rada (1978) as the basis for the descriptive category system used in
this research. He proposed five different symptomatic or characterlogi-
cal types of rapists based on descriptive categories. These included:
(1) the psychotic rapist; (2) the situation stress rapist; (3) the

masculine identity conflict rapist; (4) the sadistic rapist; and (5) the

67

68

sociopathic rapist. From these characterological types descriptive
categories were derived.

Twenty-nine randomly selected incarcerated male rapists out of 60
who were invited to participate in the present research study on sexual
arousal did so. Prior to their participation, their institutional files
were evaluated by raters on sociobehavioral items which were later
clustered into descriptive category scores. The rapists themselves
observed visual and auditory sexually arousing stimuli while their
sexual arousal was monitored using a penis circumference gauge. The
sexually arousing stimuli were pretested to validate them for adequacy
of arousal and accuracy of labeling the victim emotional state during
the vignettes. Graph tracings on the pen recorder were transformed into
percent erection scores. Each subject listened to five rape audio
vignettes and observed two erotic movies, yielding a baseline erection
score, a baseline flaccid score, and five experimental scores all pre-
sented in repeated measures format.

The analysis of data consisted of (a) data transformation; (b) the
tests of the hypotheses; and (c) assessing the influence of confounding
factors.

Two hypotheses were tested. The first was that rapists are differ-
ent in their arousal score to audio tape vignettes. The second was that
rapists would obtain correlational scores on sexual arousal and descrip-
tive categories not equal to zero.

The first hypothesis was tested using a multivariate analysis of
variance. Tests were constructed for main effects of tapes and order of
presentation. Main effects were found for both tapes and order of

presentation (fatigue). Order of presentation significantly influenced

69

the results. Hence, the research hypothesis of differences on tape
vignettes was supported, but the influence of the confounding effect of
fatigue limits the implication, i.e., differences of rapists on tapes
could have been the stimuli or fatigue or both.

The second hypothesis test was constructed by a decision rule:
non-zero (r = .00) correlation coefficients indicated rejection of the
null hypothesis. Correlation coefficients -.35 5 r_§ .35 were of signi-
ficant magnitude to warrant further use of the personality trait system.
A correlation matrix was developed using scores on the descriptive cate-
gories and sexual arousal scores. Individual correlation scores were
derived using the Product-Moment Correlation Coefficient. Only two of
the descriptive categories were recognized as significantly (r = +/-.35)
correlating with arousal to rape tapes. Occupational stress correlated
significantly with anger, lust, and reasoning victim vignettes. The
aggressive descriptive category correlated significantly with submission
and anger victim vignettes. The other descriptive categories (86%),
although not having zero correlation, were found to have less than .35
correlation and were judged to be of insignificant predictive value.
Further use of the descriptive category instrument was not recommended.

Order of presentation of stimuli was found to significantly influ-
ence the data. Adequate rater agreement was obtained so rater reliability
was probably not a confounding factor. The raters were trained prior to
their rating of institutional information. Additionally, reliability of
the raters was estimated by having all raters "blind“ rate seven files.
The interrater reliability was assessed by a percent agreement matrix
among all judges for each item on the rating sheet. The range of average

agreement was 42% through 100% and the total mean percent agreement

70

across all judges on all items was 78.4%. The traditional level of per-
cent agreement for direct observation of behavior is 70% (Bruning & Kintz,
1978). Hence, sufficient reliability of judges was obtained, therefore,

this influence was found not to influence significantly the results.

Conclusions

 

l. A significant difference exists for subjects' arousal scores
when presented with vignettes of rape scenes. A significant
effect was found for the confounding variable of order of
presentation, thereby confounding the results of the initial
hypothesis.

2. Other than zero correlation coefficients were found for scores
on descriptive categories and sexual arousal to vignettes.
Only two of the descriptive categories were found to signifi-
cantly correlate (greater than r = .35) with arousal scores

(Borg & Gall, 1971).

Discussion

 

It seems clear on the basis of the findings reported above that
fatigue effect and the low predictive validity of the descriptive cate-
gory system significantly overshadowed the research hypothesis. Penis
circumference as a measure of sexual arousal was found to be a reliable
and easily obtained measure. Findings were consistent with those of
Bancroft et a1. (1966), i.e., changes in penis circumference equal changes
in microamps of the strain gauge. Second, movement artifacts were easily
identifiable as sharp deflections on the graph rather than the gradual
sloping deflections of the experimental condition. No evidence of
suppression effect was found, as all 29 subjects that were tested

registered erections of some intensity. Responses were not found to be

71

stable across design; that is, order of presentation of stimuli was
identified as influencing arousal scores.

The order effect might be reasonably attributed to fatigue. The
erection scores were successively lower with each presentation. The
results showing a fatigue effect, although logically predictable, are
surprising, since current research, similar in design and duration
(Abel, et al., 1977), made no mention of a fatigue factor. This is an
interesting finding, reported here for the first time, and ought to be
taken into consideration for future research. An alternative explanation
is that for certain persons a stimulus is not arousing unless preceeded
by certain other stimuli which serve as a kind of readiness experience.
Similarly stimuli ordinarily exciting to a person might be perceived as
bland if preceeded by other stimuli.

Two explanations for the unexpected appearance of an order effect
are offered. First, unreported order effects may have existed in other
studies. That is, the existence of an order effect may not have been
considered and therefore not investigated. Second, a difference in the
design of this study may have made fatigue a more likely outcome than
in the previous research.

A review of the studies (Abel, et al., 1977) used as models for the
design of this study revealed no mention of a fatigue effect. No analy-
ses to determine the effect of the order of stimulus presentation were
reported. Two factors may account for the fact that fatigue was not
predicted or detected in these studies. First, each subject was pre-
sented the same stimulus tape more than once. Second, the highest score
for the two trials was used in the analyses. In such a design, an order
effect would be less likely to be considered. Consequently, it is diffi-

cult to know the source of fatigue from the literature.

72

Several possible methodological factors may have produced the order
effect in the present study. Abel et a1. (1977) refocused their subjects
between each stimulus tape. Each subject was given instructions to
"attempt to see themselves 'right there' in the scenes and to try to
see the experience exactly as it was described" (Abel et al., 1977,

p. 901). These instructions may have had the effect of reminding the
subjects to approach each tape separately and freshly. This refocusing
procedure was not part of the present study. If the monotonic decreases
in scores found in this study were the result of a decrease in interest
or attention, rather than physical fatigue, adoption of the procedure of
Abel et a1. (1977) might have alleviated possible effects of fatigue.
The alleviation might also have been accomplished by dividing the pro-
cedure into two sessions, with the subjects receiving two presentations
of tapes in each session. Such a correction could, however, create new
problems, such as greater variability in the environmental factors of
the experimental condition and variation in the subjects' attitudes from
one session to the next.

The predictive invalidity of the descriptive category system was a
second weakness identified in this study. Descriptive categories are
likely to have more predictive validity if there is direct manipulation
by the subjects of the test instrument. Observations of secondary-
source material, while reliable, are too distant from the subject to be

of sufficient predictive validity.

Subjective Observations
The subjective observations of the experimenter fall into two broad

categories: remarks about the relationship of the present research to

73

rules of thumb for counseling potential victims, and use of the re-
searcher's descriptive category classifications system to assist in
corectional decision-making.

Although the main research hypothesis was confounded and, hence, its
significance questioned, certainly it deserves attention. That is, main
effects were found for differences in responses to the tapes. While
this study is inconclusive concerning whether rapists respond in the
same manner to all victim vignettes, it certainly raises doubt as to
whether that is a valid assumption. Individual subjects were found to
respond differently to different victim affects, and some were parti-
cularly aroused by an angry fighting victim. To make a statement that
victims should not fight their assailant is, at this time, as ill-advised
as counseling victims to fight their assailant. As frustrating as it
is, the best possible advice, at this time, would appear to be for
victims to deal with each rapist individually, taking into account the
size of the rapist and victim, the exhibited use of force by the rapist,
the threat of exhibition of a weapon, and the confidence of the victim
to assess the impact of her strategy, and to adjust accordingly through
the progression of the rape.

The second question of the advocacy of the present research's
descriptive category system to assist in decision making is more clearly
answered. Only two of the traits, Job Related Stress and Aggressive
Personality Trait, were found to significantly correlate with sexual
arousal to deviant stimuli. One would hope that a classification system
would be of adequate predictive value in identifying deviant actors.
Multiple correlations, when used as individual predictors, were found

to increase the accuracy of the prediction when they correlated .35 or

74

more with the criterion. Within this research study, five of the seven
personality traits did not attain that level of correlation. Hence,
future use of this study's descriptive category system for classifica-

tion in a correctional setting is not recommended.

Implications for Future Research

 

One need for future research is the need to confirm or reject the
question of variance of sexual arousal within levels of deviance. A
longitudinal presentation of the stimuli, allowing for adequate recovery,
or a larger sample size could counter fatigue effects which were identi-
fied in this study.

A second question concerning the action of victims and its potential
effect upon the rapist remains unanswered. This remains a significant
question, as its answer fonns the basis for the advising of potential
victims.

Third, using existing psychological tests in which the rapists them-
selves manipulate the instrument could yield greater sensitivity in dis-
criminating types of rapists. Using existing tests has the advantage
of prior validity testing and general acceptance of an instrument.

Finally, if the above three questions could be answered, an area
of research to be considered would be the development of a predictive
study using a multiple regression equation to predict the potential for
rape. When taking into account actions of the victims, personality
traits of the rapist, and the outcome criterion of rape, predictive
instruments could be developed which might aid in the early identifica-

tion and treatment of potential offenders.

APPENDICES

APPENDIX A

LETTER TO SAMPLE

APPENDIX A
LETTER TO SAMPLE

Dear

Your name has been generated from a Department of Corrections com-
puter list to participate in a research project. The research is a
perception experiment. This is pp; a medication experiment. The
results of the research will be analyzed and presented in group form,
thereby protecting the confidentiality of all individuals.

The experiment will take approximately one hour and you will be
given $5.00 for your participation. It is not expected that you will
experience any aftereffects from your experience in this research.

You will be placed on call either Saturday, August 9 or Sunday,
August 10. Watch your block call board for Specific items. The research
is being conducted on the second floor of the Infirmary Central Complex.
After having the procedure explained, you may choose to participate or
withdraw.

I am conducting this research for completion of my requirements for
a Ph.D. in counseling psychology. This project will complete many years
of formal education. I am greatly appreciative of your participation.

Thank you,

Michael Kalisewicz

75

APPENDIX B

PRE-TEST EVALUATION SHEETS

APPENDIX B
PRE-TEST EVALUATION SHEETS

WRITTEN SURVEY

 

TAPE 1
THE WOMAN WAS:
afraid angry gave in calmly tried turned on
to talk him
out of it
THE TAPE WAS:
not sexy a little sexy quite sexy very sexy
at all
TAPE 2
THE WOMAN WAS:
afraid angry gave in calmly tried turned on
to talk him
out of it
THE TAPE WAS:
not sexy a little sexy quite sexy very sexy
at all
TAPE 3
THE WOMAN WAS:
afraid angry gave in calmly tried turned on
to talk him
out of it
THE TAPE WAS:
not sexy a little sexy quite sexy very sexy

at all

76

77

TAPE 4
THE WOMAN WAS:
afraid angry gave in calmly tried turned on
to talk him
out of it
THIS TAPE WAS:
not sexy a little sexy quite sexy very sexy
at all
TAPE 5
THE WOMAN WAS:
afraid angry gave in calmly tried turned on
to talk him
out of it

THIS TAPE WAS:
not sexy a little sexy quite sexy very sexy
at all

APPENDIX C

POPULATION RATING SHEET

APPENDIX C
POPULATION RATING SHEET

 

 

Inmate number Rater
Female victim ;:16 years old Inmate presently é=50 years old
yes no yes no

Instructions: Answer all questions by circling the appropriate answer.

 

 

 

 

 

 

 

Victim Degradation Homosexual History
1 intercourse yes no 11 passive yes no
2 fellatio yes no 12 predatory yes no
3 sodomy yes no 13 cooperative yes no
4 mutilation yes no
5 ritualistic Situational Stress
degradation yes no
Relationship
Mental History 14 separation yes no
15 fighting yes no
6 outpatient yes no 16 divorce yes no
7 inpatient yes no
8 neurosis yes no Occupation
9 character 17 dismissal yes no
disorder yes no 18 quit yes no
10 schizophrenia 19 job decline yes no
psychosis

Instructions: Place an X over the most severe level of behavior reported.

 

20 Victim violence

 

 

verbal minor physical more than excess use excess use

threat contact to force necessary physical of contact of violence

submission contact to force less death with death
submission

78

79

21 Weapon violence

 

no unseen threat seen threat use of use of
weapon of weapon of weapon weapon weapon
threat less death with death

22 Instructions: Place an X over subject's routine level of functioning.

 

2 l 0 1 2
Passive Aggressive

APPENDIX D

POPULATION RATING INSTRUCTION SHEET

APPENDIX D
POPULATION RATING INSTRUCTION SHEET

Victim Degradation: Sexual acts or there attempts are intended.
Mutilation refers to cutting, biting or other forms of sadistic
abuse. Ritualistic degradation infers that the victim in
addition to being sexually abused is expected to perform unusual
acts (i.e., acts w/animals, ingestion of feces).

 

Mental History: Refers to any and all mental health contacts either
during incarceration or while free.

Homosexual History: Refers to any and all homosexual activity either
during incarceration or while free.

Situational Stress: Prior to sex offense subject is being evaluated
on. Relationship does not have to be marriage. Some with
occupation: prior to sex offense subject is being evaluated on.

Victim Violence: Your impression taking into account reports from
police, victim, subject and doctor report.

Weapon Violence: Your impression taking into account report from
poliCe, victim, and subject.

PassiveaAggressive: Your impression of how subject would routinely
be viewed by others (i.e., a normal subject whose crime is his
only explosive act would be classed passive 2).

If a subject denies guilt, disregard using the victim's and police
reports to make your judgments.

80

APPENDIX E

REMINDER LETTER TO RATERS

APPENDIX E
REMINDER LETTER TO RATERS

Reception and Guidance Center
S P S M
(date)

Dear (name):

This is to remind you of your commitment to assist me with my
dissertation research. The place to meet will be in the central complex
records office on (day, date, time). Please be prompt. A brief

instruction-question period will be conducted first. Coffee and dough-
nuts will be available.

It is expected the project will take some afternoon time. We will
break for lunch. Lunch will be provided at the Quality Inn on 1-94
and Elm Road Exit.

See you (day, time).

Thank you,

Michael Kalisewicz

81

 

APPENDIX F

POPULATION RATERS THANK YOU

APPENDIX F
POPULATION RATERS THANK YOU

Reception and Guidance Center
S P S M
(date)
Dear (name)

I would like to thank you for your generosity in assisting me with
my dissertation research. I have been working toward my degree for
five years. Your efforts have greatly assisted me in the completion of
my dissertation.

If I can ever assist you in similar enterprises, please do not
hesitate to call upon me.

Thank you,

Michael Kalisewicz

82

APPENDIX G

INFORMED CONSENT--PERMISSION TO PARTICIPATE

APPENDIX G
INFORMED CONSENT--PERMISSION TO PARTICIPATE

On (day, date), I have had explained to me the experimental pro-
cedures involved in the Experiment in Perception being run by Mr.
Michael Kalisewicz.

I am knowledgeable of the fact that the present research is for
Mr. Kalisewicz's dissertation and that he does not represent the
Michigan Department of Corrections in his endeavors.

I am knowledgeable of the fact that the data in this research will
be analyzed and presented in group form, thereby protecting all indi-
viduals to their right to privacy.

Being knowledgeable of all of the above, I have decided freely and
without pressure to participate in the research on perception.

 

Resident‘s name

 

Resident's number

 

Researcher's name

83

 

 

 

APPENDIX H

RELEASE FROM THE EXPERIMENT

APPENDIX H
RELEASE FROM THE EXPERIMENT

On Saturday, August 9, 1980, or Sunday, August 10, 1980, I have
participated in an experiment on sexual perception. After participation
in the experiment, I was involved in a debriefing session where any
questions and concerns I may have had have been resolved.

I am terminating my participation without any ill effects known to
me. I have been advised whom to contact if I should need a followup
interview.

 

Resident's Name

 

Resident's Number

 

Researcher's Name

I am terminating my participation and would like to consult with
a physician. Check here

84

APPENDIX I

EXPERIMENTAL EXPLANATION SHEET

APPENDIX I
EXPERIMENTAL EXPLANATION SHEET

You have just been part of a study on human sexuality. The equip-
ment was measuring your arousal to two films and five different taped
scenes. The films and tapes were designed to turn you on, and the
music in between was to help bring you down. The penis circumference
gauge measured your response. I was particularly interested in the
reSponse of men who have been imprisoned in correctional institutions
for different offenses. As a result of this study, I will receive my
doctoral degree in psychology; so I am especially appreciative. The
five dollars will be deposited in your account at SPSM on Tuesday,

August 19, 1980.

Thank you.

Michael Kalisewicz

85

APPENDIX J

ITEM RELIABILITY DATA

Interrater Reliability:

Percent Agreement

 

 

 

 

Item 1: Intercourse
Average = 93.33
Jl J2 J3 J4 J5 J6 J7 J8 J9
J2 100.0
J3 100.0 100.0
J4 66.7 66.7 66.7
J5 100.0 100.0 100.0 66.7
J6 100.0 100.0 100.0 66.7 100.0
J7 100.0 100.0 100.0 66.7 100.0 100.0
J8 100.0 100.0 100.0 66.7 100.0 100.0 100.0
J9 100.0 100.0 100.0 66.7 100.0 100.0 100.0 100.0
J10 100.0 100.0 100.0 66.7 100.0 100.0 100.0 100.0 100.0
J = Judge.

86

88

Interrater Reliability: Percent Agreement

 

Item 3: Sodomy
Average = 75.56

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 100.0
J3 100.0 100.0
J4 33.3 33.3 33.3
J5 66.7 66.7 66.7 66.7
J6 66.7 66.7 66.7 66.7 100.0
J7 100.0 100.0 100.0 33.3 66.7 66.7
J8 100.0 100.0 100.0 33.3 66.7 66.7 100.0
J9 100.0 100.0 100.0 33.3 66.7 66.7 100.0 100.0
J10 66.7 66.7 66.7 66.7 100.0 100.0 66.7 66.7 66.7

 

J = Judge.

89

Interrater Reliability: Percent Agreement

 

Item 4: Mutilation
Average = 93.33

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9

J2 100.0

J3 100.0 100.0

J4 100.0 100.0 100.0

J5 66.7 66.7 66.7 66.7

J6 100.0 100.0 100.0 100.0 66.7

J7 100.0 100.0 100.0 100.0 66.7 100.0

J8 100.0 100.0 100.0 100.0 66.7 100.0 100.0

J9 100.0 100.0 100.0 100.0 66.7 100.0 100.0 100.0
J10 100.0 100.0 100.0 100.0 66.7 100.0 100.0 100.0 100.0

 

J = Judge.

9O

Interrater Reliability: Percent Agreement

 

Item 5: Ritualistic Degradation

Average = 100.0

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9

J2 100.0

J3 100.0 100.0

J4 100.0 100.0 100.0

J5 100.0 100.0 100.0 100.0

J6 100.0 100.0 100.0 100.0 100.0

J7 100.0 100.0 100.0 100.0 100.0 100.0

J8 100.0 100.0 100.0 100.0 100.0 100.0 100.0

J9 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
J10 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

 

J = Judge.

91

Interrater Reliability: Percent Agreement

 

Item 6: Outpatient
Average = 64.44

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 33.3
J3 66.7 66.7
J4 100.0 33.3 66.7
J5 100.0 33.3 66.7 100.0
J6 100.0 33.3 66.7 100.0 100.0
J7 33.3 66.7 66.7 33.3 33.3 33.3
J8 66.7 66.7 100.0 66.7 66.7 66.7 66.7
J9 66.7 100.0 100.0 33.0 100.0 100.0 66.7 100.0
J10 33.3 68.7 66.7 66.7 66.7 66.7 100.0 66.7 66.7

 

J = Judge.

92

Interrater Reliability: Percent Agreement

 

Item 7: Inpatient
Average = 81.48

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9

J2 66.7

J3 66.7 100.0

J4 100.0 66.7 66.7

J5 100.0 66.7 66.7 100.0

J6 66.7 100.0 100.0 66.7 66.7

J7 100.0 66.7 66.7 100.0 100.0 66.7

J8 66.7 100.0 100.0 66.7 66.7 100.0 66.7

J9 66.7 100.0 100.0 66.7 66.7 100.0 66.7 100.0
J10 100.0 66.7 66.7 100.0 000.0 66.7 100.0 66.7 66.7

 

J = Judge.

93

Interrater Reliability:

Percent Agreement

 

Item 8:

Diagnosed Neurotic

Average = 100.0

 

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9

J2 100.0

03 100.0 100.0

J4 100.0 100.0 100.0

J5 100.0 100.0 100.0 100.0

J6 100.0 100.0 100.0 100.0 100.0

J7 100.0 100.0 100.0 100.0 100.0 100.0

J8 100.0 100.0 100.0 100.0 100.0 100.0 100.0

J9 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
J10 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

J = Judge.

94

Interrater Reliability: Percent Agreement

 

 

 

Item 9: Diagnosed Character Disorder
Average = 63.70
J1 J2 J3 J4 J5 J6 J7 J8 J9

J2 66.7

J3 33.3 66.7

J4 66.7 33.3 66.7

J5 100.0 66.7 33.3 66.7

J6 33.3 66.7 100.0 66.7 33.3

J7 66.7 100.0 66.7 33.3 66.7 66.7

J8 100.0 66.7 33.3 66.7 100.0 33.3 66.7

J9 66.7 33.3 66.7 100.0 66.7 66.7 33.3 66.7
J10 66.7 100.0 66.7 33.3 66.7 66.7 100.0 66.7 33.3

 

Judge.

 

95

Interrater Reliability: Percent Agreement

 

Item 10: Diagnosed Schizophrenia

Average = 72.59

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 66.7
J3 66.7 100.0
J4 66.7 33.3 33.3
J5 66.7 100.0 100.0 33.3
J6 66.7 100.0 100.0 33.0 100.0
J7 33.3 66.7 66.7 66.7 66.7 66.7
J8 66.7 100.0 100.0 33.3 100.0 100.0 66.7
J9 66.7 100.0 100.0 33.3 100.0 100.0 66.7 100.0
J10 33.3 66.7 66.7 66.7 66.7 66.7 100.0 66.7 66.7

 

J = Judge.

96

Interrater Reliability: Percent Agreement

 

Item 11: Passive Homosexual

Average = 93.33

 

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9

J2 66.7

J3 100.0 66.7

J4 100.0 66.7 100.0

J5 100.0 66.7 100.0 100.0

J6 100.0 66.7 100.0 100.0 100.0

J7 100.0 66.7 100.0 100.0 100.0 100.0

J8 100.0 66.7 100.0 100.0 100.0 100.0 100.0

J9 100.0 66.7 100.0 100.0 100.0 100.0 100.0 100.0
JlO 100.0 66.7 100.0 100.0 100.0 100.0 100.0 100.0 100.0

J = Judge.

 

97

Interrater Reliability:

Percent Agreement

 

 

 

 

Item 12: Predatory Homosexual
Average =
J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 66.7
J3 100.0 66.7
J4 100.0 66.7 100.0
J5 100.0 66.7 100.0 100.0
J6 100.0 66.7 100.0 100.0 100.0
J7 100.0 66.7 100.0 100.0 100.0 100.0
J8 100.0 66.7 100.0 100.0 100.0 100.0 100.0
J9 100.0 66.7 100.0 100.0 100.0 100.0 100.0 100.0
J10 100.0 66.7 100.0 100.0 100.0 100.0 100.0 100.0 100.0
J = Judge.

98

Interrater Reliability:

Percent Agreement

 

 

 

Item 13: Cooperative Homosexual
Average = 93.33
J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 100.0
J3 100.0 66.7
J4 100.0 66.7 100.0
J5 100.0 66.7 100.0 100.0
J6 100.0 66.7 100.0 100.0 100.0
J7 100.0 66.7 100.0 100.0 100.0 100.0
J8 100.0 66.7 100.0 100.0 100.0 100.0 100.0
J9 100.0 66.7 100.0 100.0 100.0 100.0 100.0 100.0
J10 100.0 66.7 100.0 100.0 100.0 100.0 100.0 100.0 100.0

 

J = Judge.

99

Interrater Reliability: Percent Agreement

 

Iten 14: Relationship Separation

Average = 62.96

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 66.7
J3 33.3 33.

J4 66.7 33. 66.7

J5 66.7 33. 66.7 100.0

J7 66.7 33. 66.7 100.0 100.0 100.0

J8 33.3 33. 100.0 66.7 66.7 66.7 66.7

J9 66.7 33.

3

3

3

J6 66.7 33.3 66.7 100.0 100.0

3

3

3 33.3 66.7 66.7 66.7 66.7 33.3
7

J10 100.0 66. 33.3 66.7 66.7 66.7 66.7 33.3 66.7

 

J = Judge.

lOO

Interrater Reliability: Percent Agreement

 

 

 

 

Item 15: Relationship Fighting
Average = 70.37
J1 J2 J3 J4 J5 J6 J7 J8 J9

J2 66.7

J3 100.0 66.7

J4 100.0 66.7 100.0

J5 66.7 66.7 66.7 66.7

J6 100.0 66.7 100.0 100.0 66.7

J7 100.0 66.7 100.0 100.0 66.7 100.0

J8 66.7 33.3 66.7 66.7 33.3 66.7 66.7

J9 33.3 33.3 33.3 33.3 66.7 33.3 33.3 33.3

J10 100.0 66.7 100.0 100.0 66.7 100.0 100.0 66.7 33.3

J = Judge.

 

101

Interrater Reliability: Percent Agreement

 

 

 

 

Item 16: Relationship Divorce
Average = 86.67
J1 J2 J3 J4 J5 J6 J7 J8 J9

J2 66.7

J3 100.0 66.7

J4 100.0 66.7 100.0

J5 100.0 66.7 100.0 100.0

J6 100.0 66.7 100.0 100.0 100.0

J7 100.0 66.7 100.0 100.0 100.0 100.0

J8 100.0 66.7 100.0 100.0 100.0 100.0 100.0

J9 66.7 33.3 66.7 66.7 66.7 66.7 66.7 66.7
J10 100.0 66.7 100.0 100.0 100.0 100.0 100.0 100.0 66.7

J = Judge.

102

Interrater Reliability: Percent Agreement

 

Item 17: Occupation Dismissal

Average = 48.89

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 33.3
J3 0.0 33.3
J4 33.3 66.7 66.7
J5 33.3 33.3 66.7 33.3
J6 33.3 33.3 77.7 33.3 100.0
J7 33.3 66.7 66.7 100.0 33.3 33.3
J8 0.0 33.3 100.0 66.7 66.7 66.7 66.7
J9 33.3 33.3 33.3 66.7 0.0 0.0 66.7 33.3
J10 0.0 33.3 100.0 66.7 66.7 66.7 66.7 100.0 33.3

 

J = Judge.

103

Interrater Reliability: Percent Agreement

 

Item 18: Occupation Quit
Average = 54.07

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 33.3
J3 66.7 33.3
J4 33.3 0.0 66.7
J5 100.0 33.3 66.7 33.3
J6 66.7 66.7 33.3 0.0 66.7
J7 66.7 33.3 100.0 66.7 66.7 33.3
J8 66.7 33.3 100.0 66.7 66.7 33.3 100.0
J9 66.7 0.0 33.3 33.3 66.7 33.3 33.3 33.3
J10 100.0 33.3 66.7 33.3 100.0 66.7 66.7 66.7 66.7

 

J = Judge.

104

Interrater Reliability: Percent Agreement

 

Item 19: Occupation Job Decline

Average = 42.22

 

 

J1 J2 J3 J4 J5 06 J7 J8 J9
J2 33.3
J3 66.7 66.7
J4 33.3 33.3 0.0
J5 0.0 33.3 33.3 33.3
J6 66.7 0.0 33.3 66.7 33.3
J7 100.0 33.3 66.7 33.3 0.0 66.7
J8 100.0 33.3 66.7 33.3 0.0 66.7 100.0
J9 33.3 0.0 0.0 33.3 33.3 33.3 33.3 33.3
J10 66.7 0.0 33.3 66.7 33.3 100.0 66.7 66.7 33.3

 

J = Judge.

105

Interrater Reliability: Percent Agreement

 

Item 20: Victim Violence

Average = 57.78

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 66.7
J3 66.7 33.3
J4 66.7 100.0 33.3
J5 66.7 33.3 33.3 33.3
J6 100.0 66.7 66.7 66.7 66.7
J7 66.7 66.7 33.3 66.7 33.3 66.7
J8 66.7 33.3 33.3 33.3 100.0 66.7 33.3
J9 100.0 66.7 66.7 66.7 66.7 100.0 66.7 66.7
J10 33.3 66.7 0.0 66.7 66.7 33.3 33.3 66.7 33.3

 

J = Judge.

106

Interrater Reliability: Percent Agreement

 

Item 21: Weapon Violence

Average = 74.81

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 100.0
J3 66.7 66.7
J4 100.0 100.0 66.7
J5 66.7 66.7 100.0 66.7
J6 33.3 33.3 66.7 33.3 66.7
J7 100.0 100.0 66.7 100.0 66.7 33.3
J8 66.7 66.7 100.0 66.7 100.0 66.7 66.7
J9 100.0 100.0 66.7 100.0 66.7 33.3 100.0 66.7
J10 66.7 66.7 100.0 66.7 100.0 66.7 66.7 100.0 66.7

 

J = Judge.

win-A. L" -’;T n‘V'A

107

Interrater Reliability: Percent Agreement

 

Item 22: Passive-Aggressive

Average = 36.30

 

 

J1 J2 J3 J4 J5 J6 J7 J8 J9
J2 66.7
J3 33.3 33.3
J4 33.3 66.7 0.0
J5 33.3 33.3 66.7 0.0
J6 33.3 33.3 66.7 0.0 66.7
J7 0.0 0.0 0.0 33.3 0.0 33.3
J8 66.7 100.0 33.3 66.7 33.3 33.3 0.0
J9 33.3 66.7 0.0 100.0 0.0 0.0 33.3 66.7
J10 33.3 33.3 66.7 0.0 66.7 100.0 33.3 33.3 0.0

 

J = Judge.

REFERENCES

REFERENCES

Abel, G. G., D. H. Barlow, E. B. Blanchard, & et a1. "Measurement of
Sexual Arousal in Male Homosexuals: The Effects of Institutions
and Stimulus Modality." Archives of Sexual Behavior (1965), 4:
623-629.

_ “v"

 

Abel, G. G., D. Barlow, E. Blanchard, & 0. Guild. "The Components of
Rapists' Sexual Arousal." Archives of General Psychiatry
(August, 1977), Vol. 34 (8): 895-903.

 

v‘mi‘fifla. '.' ‘ ' s

Abel, 0. 0., E. B. Blanchard, 0. H. Barlow, & et a1. "Identifying
Specific Erotic Cues in Sexual Deviations by Audiotaped Descrip-
tions.” Journal of Applied Behavioral Analysis, 8 (1975), 247-260.

 

American Psychiatric Association. Diagnostic and Statistical Manual of
Mental Disorders. Washington, D.C., 2nd ed., APA (1968).

 

 

American Psychological Association. Ethical Standards of Psychologists.
Washington, D.C., APA, 1963.

 

American Psychological Association.. Ethical Principles in the Conduct
of Research with Human Participants. Washington, D.C., APA, 1973.

 

American Psychological Association. Publication Manual of the American
Psychological Association, 2nd ed. Washington, D.C., APA, 1974.

 

Amir, M. Patterns in Forcible Rape. Chicago: University of Chicago
Press, 1971.

 

Amir, M. "Patterns in Forcible Rape: With Special Reference to
Philadelphia, Pennsylvania.“ Dissertation Abstracts, 1965, 27,
A 1126.

 

"Forcible Rape." Federal Probation, 31 (1965), pp. 51-58.

Amir, M., & J. S. MacKellar. Rape: The Bait and the Trap. New York:
Crown Publishers, 1975.

 

Bancroft, J. H., H. G. Jones, & B. P. Pullan. "A Simple Transducer for
Measuring Penile Erections." Behavior Research and Therapy, 4
(1966). Pp. 239-241.

 

108

109

Barlow, D. H., R. Becker, H. Lutenberg, & et a1. "Mechanical Strain
Gauge Recording Penile Circumference Change." Journal of Applied
Behavioral Analysis, 3 (1970), 73-76.

 

Bessemer, S. "The Laws of Rape." Dissertation Abstracts, 1976, 37,
p. A6717.

 

Billings, V. "How to Decide You're Being Raped and What to Do About
It." Redbook, 1974, pp. 70- 71.

Bond, J. "Self-Defense Against Rape: The Joann Little Case." Black
Scholars (1975), 29-31.

Borg, W. & M. Gall. Educational Research an Introduction. New York:
David McKay Company Inc., 1971.

 

Brancale, R., A. Ellis, & R. Doorbar. "Psychiatric and Psychological
Investigations of Convicted Sex Offenders: A Summary Report."
American Journal of Psyphiatry (1952).

 

Brownmiller, S. "Antirape Techniques: Interview." Harper's Bazaar
(1976). Pp. 109-119.

Burgess, A. W. & L. Holstrom. "Coping Behavior of the Rape Victim."
American Journal of Psychiatry (April, 1976), Vol. 133, 413-418.

 

Byrne, D. & J. Sheffield. "Response to Sexually Arousing Stimuli as a
Function of Repressing and Sensitizing Defenses." Journal of
Abnormal Psychology (1965), Vol. 70, 114-118.

 

 

Calmas, W. E. "Fantasies of the Mother-Son Relationship of the Rapists
and the Pedophilia." Dissertation Abstracts, 1965, V26(05), p. 2875.

 

Campbell, 0. T. & J. C. Stanly. Experimental and Qpasi-Experimental
Designs for Research. Chicago: Rand McNally Publishers, 1963.

Chuven, H. "The Effects of Depression Frustration.“ Dissertation
Abstracts, 1973, V34(05), p. 82297.

 

Clapperton, 0., Jr. "The Effects of Aggression Anxiety and Sex Differ-
ences on the Use of Repressing Defenses." Dissertation Abstracts,
1967, V 28(08), p. 83467.

Cohen, M. L., R. Garofalo, R. Boucher, & T. Sighorn. "The Psychology of
Rapists." Seminar Psychiatry, 3 (1971), 307-327.

 

Cox, D. R. Planning of Experiments. New York: Wiley and Sons, Inc.,
1958.

Daigitz, R. E. "The Propensity to Sanction Selected Forms of Deviant
Sexual Behavior." Dissertation Abstracts, 1976, V37(08).

 

110

DeRiver, J. P. Crime and the Sexual Psychopath. Springfield, 111.:
Charcles C. Thomas, 1958.

 

Drzazga, J. Sex Crimes and Their Legal Aspects. Springfield, 111.:
Charles C. Thomas, 1960.

 

Ellis, A. & R. Brancale. The Psychology of Sex Offenders. Springfield,
111.: Charles C. Thomas, 1956.

 

Evans, H. & N. Sperekas. "Sexual Assault Bibliography." Catalog of
Selected Documents in Psychology (November, 1976), Vol. 6.

 

Executive Office of the President. Office of Science and Technology.
Privacy and Behavioral Research. Washington, D.C.: United States
Government Printing Office, 1967.

:"E'

Fenichel, 0. The Psychoanalytic Theory of Neurosis. New York: Norton,
1945.

 

 

Finn, J. D. A General Model for Multivariate Analysis. New York: 1
Holt, Rinehart and Winston Inc.; 1974.

Freud, S. Three Essays on the Theory of Sexuality. In S. Freud, The
Complete Psychological Works of Sigmund Freud. London: Hogarth
Press, 1953.

Freud, S. A General Introduction to Psychoanalysis. London: Liver-
right,T1920.

Freud, S. Sexuality and the Psychology of Love. New York: Macmillan,
1963.

Freund, K., K. Knob, & F. Sidlarcek. "simple Transducers for Mechanical
Plethysmorgraphy of the Male Genital." Journal of the Experimental
Analysis of Behavior, 8 (1965), 169-170.

Gebhard, P. H., J. H. Gagnon, & W. B. Pomeroy. Sex Offenders. New York:
Harper and Row, 1965.

 

Glass, G. V. & J. C. Stanley. Statistical Methods in Education and
Psychology. Englewood Cliffs, N.J.: Prentice Hall, 1970.

Glueck, C. G. "Psychodynamic Patterns in the Sex Offender." Psychiatric
Quarterly, 28 (1954), 1-21.

Graves, W. J. "The Effect of Communication--Induced Sexual Arousal on
Instigation to ggression in a Free Choice Paradigm." Disserta-
tion Abstracts, 1976, V37(10), p. 85435.

 

Groth, N.'& A. Burgess. "Rape: A Sexual Deviation." Journal of
Orthopsychiatric, 3 (July, 1977), 400-406.

 

111

Guss, G. H. "The Woman's Role in the Victim-Offender Relationship for
Forcible Rape." Dissertation Abstracts, 1975, V36(11), p. 85761.

Guttmacher, M. & H. Weihofen. Psychiatry and the Law. New York:
Norton, 1952.

Haldeman, C. F. "Reactivity to Anxiety Provoking Stimuli in Criminal
Sexual Psychopaths." Dissertation Abstracts, 1973, V32(07),
p. 83498.

Hall, C. S. A Primer of Freudian Psychology. New York: World, 1954.

Hammer, E. "A Comparison of HTP's of Rapists and Pedophilias."
Journal of Projective Techniqpes, 18 (1954), 346-349.

Hammer, E. F. 8 B. C. Glueck. "Psychodynamic Patterns in Sex Offenders
a Four Factory Theory." Psychiatric QUarterl , 31 (1957), 325-345.

Henn, F., M. Herjanic, & R. Vanderpearl. "Forsenic Psychiatry: Pro-
files of Two Types of Sex Offenders." American Journal of Psy-
chiatry, 133 (June, 1976), 694-696.

Hess, A. G. "Rape a Psychosocial Study." Dissertation Abstracts,
1976, V133(6), 694-696.

 

Howell, L. "Clinical Research Impressions Regarding Murder and Sexually
Perverse Crimes." Psychotherapy and Psychosomatics, 21 (1954),
156-159.

"How to Protect Ourself From Rape." Good Housekeeping (1975), V. 181
(157). p. 39.

Hursch, C. J. The Trouble With Rape. Nelson Hall, Illinois, 1977.

Jaffe, Y. "Sex and Aggression: An Intimate Relationship." Unpub-
lished Doctoral Dissertation, University of California.

Kanin, E. J. "Male Sex Aggression." Dissertation Abstracts, 1964,

V35(9), p. 84709.

Kaplan, H. S. The New Sex Therapy. New York: Brunner Hazel, 1974.

Karacan, I. "Nocturnal Penile Tumescence and Sleep of Convicted Rapists
and Other Prisoners." Archives of Sexual Behavior, 3 (1) (January,
1974), 19-26.

 

Kardner, S. H. "Rape Fantasies." Journal of Religion and Health, 14
(January, 1975), 50-57.

 

Kercher, G. & E. Walker. "Reactions of Convicted Rapists to Sexually
Explicit Stimuli. Journal of Abnonnal Psychology (February, 1973),
46-50.

 

112
Kerlinger, F. N. a E. I. Pedhauzer. Multiple Regression in Behavioral
Research. New York: Holt, Rinehart, Winston, Inc., 1973.

Kernberg, O. Borderline Conditions and Pathological Narcissism. New
York: Jason-Aronson, 1975.

 

Knobb, S. B. "The Character Structure of Sex Offenders." American
Journal of Psychotherapy, 16 (1953), 64-70.

 

Koenig, R. "Rape: The Most Rapidly Increasing Crime." McCalls (July,
1973). 25-28.

Kole, J. "Rape and What to do About It." Harper's Bazaar (1976),

 

 

 

pp. 118-119. r
Kraft-Ebing, R. Psychopathia Sexualiss A Medico-Forensic Study. New ”
York: G. P. Putnam's Sons, 1965 (English Editiofi), 1886 (lst E
edition German). (
Lalcey, J. I. "The Evaluation of Autonomic Responses: Toward a General ,
Solution." Annual of the New York Academy of Science, 67 (1965), 1
123-164. L

 

Laws, 0. R. & H. B. Rubin. "Institutional Control of an Autonomic
Sexual Response." Journal of Applied Behavioral Analysis, 2 (1966),
93-99.

 

Lesko, C.-M. F. "Psycho-Socio-Legal Aspects of Rape." Dissertation
Abstracts, 1976, V37(1l), p. 85881.

 

Lopez, T. W. "Emotional Expression in the Adult Sex Offender." .Qis-
sertation Abstracts, 1970, V31(6), p. 83709.

 

Lowenstein, L. F. "The Psychology of Rape." New Law Journal (March,
1977), 201-224.

 

Mahler, M., F. Pine, & A. Bergman. The Psychological Birth of the
Human Infant. New York: Basic Books, 1975.

 

Mason, E. "How Can a Woman Avoid Rape." Intellect (1975), pp. 512-513.

Masters, W. & V. Johnson. Human Sexual Inadequacy. Boston: Little
Brown and Company, 1970.

Masterson, J., ed. New Perspectives on Psychotheraphy of the Borderline
Adult. New York: Brunner/M0281, 1978.

Mavissakalian, M., et a1. "Subjective and Erectile Responses to Com-
plex Erotic Stimuli in Homosexual and Heterosexual Males." British
Journal of Psychiatry, 126 (1975), 252-257.

 

113

Mehren, W. & I. Lehman. Measurement and Evaluation in Education and
Psychology. New York: Holt, Rinehart and Winston, 1969.

 

"Michigan Commission: New Light on Sex Offenders." Journal of Social
Hygiene, 38 (1952), p. 29.

Michigan Statistical Abstracts. M.S.U. Graduate School of Business
Administration, 1978.

 

Moser, R. J. "Effects of Repression-Sensitization, Arousal and Sex on
Physical Aggression." Dissertation Abstracts, 1972, V33(7),
p. 83317.

Offir, C. W. "Don't Take It Lying Down." Psychology Today (1975).

Pacht, A. R., S. L. Hallick, & J. C. Ehrmann. "Diagnosis and Treatment
of the Sexual Offender: A Nine Year Study." American Psychiatry,
118 (1962), 802-808.

Peters, J. J. "Children Who Are Victims of Sexual Assault and the
Psychology of Offenders.“ American Journal of Psychotherapy
(July, 1976), 398-421.

Rabin, A. I. Projective Technigues in Personality Assessment. New York:
Springer Publishing Company Inc., 1968.

Rada, R. T. "Alcoholism and Forcible Rape." American Journal of
Psychiatry, 132(4) (April, 1975), 444-446.

 

Rada, R. T. Clinical Aspects of the Rapist. New York: Grune and
Stratton, 1978.

Rafferty, W. "Pupillary Activity as a Measure of Sexual Interest in
Pedophilias, Rapists and Normals. Dissertation Abstracts, 1975,

V35(6-13), p. 3033.

"Rape Prevention Tactics." IM§_(July, 1974), pp. 114-115.

Reppuci, N. C. "Methodological Issues in Research with Correctional
Populations." Journal of Consulting and Clinical Psychology, 4
(AUQUSt, 1978), 727-727.

 

Rogel, M. J. "Biosocial Aspects of Rape." Dissertation Abstracts,
1976, V37(9), p. 84763.

 

Seghorn, T. K. "Adequacy of Ego Function in Rapists and Pedophiles.
D1ssertation Abstracts, 1970, V31(12), p. 87613.

Statistical Abstracts of the United States. Washington, D.C.: Publishers
United States Department of Commerce, Bureau of Census, 1978.

Vernon, L. J. "Fear Response of Rape Victims." Dissertation Abstracts,
1977, V38(7), p. 83421.

 

114
Vinsel, A. "Rape: A Review Essay." Personality and Social Psychology
Bulletin (Spring, 1977), 183-189.

Vuocolo, A. B. "The Administration of the New Jersey Sex Offender Pro-
gram." Dissertation Abstracts, 1967, V28(3), p. A1121.

Williams, A. H. "Rape-Murder-in-Slovenko R." Sexual Behavior and the
tap. Springfield, 111.: Charles C. Thomas, 1965.

 

Wile, I. S. "Sex Offenders and Sex Offenses: Classification and Treat-
ment." Journal Criminal Psychopathology, 3 (1941), 11-31.

Winer, B. J. Statistical Principals in Experimental Design. New York:
McGraw-Hill, 1962.

Winslow, R. W. & V. D. Winslow. Reality: Alternative World Views.
Boston: Allyn and Bacon, 1974.

Zuckennan, M. "Physiological Measures of Sexual Arousal in the Human."
Psychological Bulletin, 75 (1971), 297-329.

 

Ill 1111 ll 11111111111ill 11111111 H 1111 ll
3 1 293 1 04453547