-—v—_
-—-‘
v.-

——————

REMOTE
TORAGE

$
' [son

'-

Date

" IIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIII

1293 00646 4097

 

LIBRARY
Michigan State
University

 

 

 

This is to certify that the

thesis entitled

CONFLICT RESOLUTION BEHAVIOR
IN DATING RELATIONSHIPS: _
TYPOLOGIES AND PREDICTORS

presented by

w. John Curtis

has been accepted towards fulfillment
of the requirements for

M.A. degree in Psychology

  

/," ‘ { Major {)rofessor
/ Ga y Sto‘liak

November 4 , 1987

 

MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

REMOTE STORAGE

PLACE IN RETURN BOX to remove this checkout from your record.

TO AVOID FINES return on or before date due.

 

I DATE DUE

DATE DUE

DATE DUE

 

AMI; 341 12038

 

 

 

 

 

 

 

 

 

 

 

 

 

i“
2/17 20:: Blue FORMS/DateDueForms_2017.indd - 99.5

 

 

‘Ct‘

fl“

(II
\o

CONFLICT RESOLUTION BEHAVIOR
IN DATING RELATIONSHIPS:
TYPOLOGIES AND PREDICTORS

By
w. John Curtis

A THESIS

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

MASTER OF ARTS

Department of Psychology

1 987

ABSTRACT

CONFLICT RESOLUTION BEHAVIOR
IN DATING RELATIONSHIPS:
TYPOLOGIES AND PREDICTORS

By

W. John Curtis

A sample of 671 undergraduates was surveyed with the Conflict
Resolution and Abuse Questionnaire (CRAQ) concerning the types of conflict
resolution behaviors they utilized in their current dating relationship.
Respondents were also surveyed concerning the characteristics of their dating
relationship, the types of conflict resolution behaviors used by their partner,
and the history of violence in their family of origin.

A factor analysis of the CRAQ items yielded seven factors representing
seven types of conflict resolution behavior. A cluster analysis of the subjects
resulted in a typology of conflict resolution behavior.

Length of and depth of committment in the relationship. amount of
violence received from the dating partner, and violence in the family of origin
were used to predict conflict resolution behavior.

The strongest finding of the study indicated that violence in dming
relationships is a reciporical phenommeneon.

ACKNOWLEDGMENTS

Throughout this process of writing a Master's Thesis, many people have
been helpful, supportive, and loving in their own way.

Gary Stollak, to me you have been a friend, mentor, teacher, and
philosopher. Your words and your courage to apply philosophy and theory to
your everyday life have more often than not been an inspiration to me. In my
struggle to begin to comprehend human behavior and interaction, you have
acted as a guide to the at once unknown and often discouragingly
unknowable realm of human existence. Your belief in the ultimate and
attainable goal of knowledge has motivated and encouraged me when I felt
unable to go any further. You have invited me into your home and l have felt at
home there. We have talked about Asimov and Clarke. For all of this I am
eternally grateful, and wherever and whenever, I will often and fondly
remember.

Larry Messe, you helped and helped, unselfishly and patiently giving
time, knowledge, and advice. Especially when it counted and was needed.
Thank you.

Susan Frank, thank you for the very helpful comments and insights on
the analyses. They put me on the right track.

Gene Maguin. A friend and colleague. One who has helped me in
immeasurable ways to begin to understand and tame the computer and
statistics beasts. Also, one who always listened to whatever complaints I
might have had about anything, and often provided words of wisdom
concerning life and this violence issue we have gotten ourselves into. The
innumerable conversations and ideas were both inspirational and

entertaining, to say the least. And for all the glasses of water. Thank you.
Amy, thank you for love, understanding, and comfort. You inspired in
ways you never knew.
Melvin and Nina, two wonderful people from Virginia, without whom I
would not have begun this journey, thank you.

TABLE OF CONTENTS

Page
CONFLICT RESOLUTION BEHAVIOR IN
DATING RELATIONSHIPS. TYPOLOGIES
AND PREDICTORS . . . . . 1
INTRODUCTION . . . . . . 1
Measurement of Conflict
Resolution Behavior . . . . . 4
The CRAO. . . . . 8
Reciprocity and Sex Differences in
Conflict Resolution Behavior . . . . 10
Violence in the Family of Origin . . . 13
Conflict Resolution and Duration
and Type of Dating Relationship . . . 17
HYPOTHESES . . . . . . 19
METHOD . . . . . . 20
Subjects . . . . . . 20
Instrument . . . . . . 23
Derivation of Scales . . . . . 26
Procedure . . . . . . 27
RESULTS . . . . . . 28
Factor Analysis . . . . . 28
Cluster Analysis . . . . 31
Multivariate Analysis of Variance . . . 35
Individual Tests of Predictors . . . 35
Summary of Cluster Differences . . . 37
Discriminant Function Analysis . . . 37
Discriminating Variables . . . . 40

Classification Results . . . . 43

DISCUSSION

APPENDICES

A:

FF???) [9999?

Conflict Resolution and Abuse
Questionnaire .

Conflict Tactics Scale (Form A)

Conflict Tactics Scale (Form N)

Index of Spouse Abuse .

Frequency of Responses to CRAO

Items 1-67 . .

Scree Plot of Eigenvalues

Items and Item Loadings of CRAO Factors

Plot of Group Centroids on Two
Discriminant Functions

REFERENCES

vi

8

74
75
76
77
87
92

93

Table

10.

11.

12.

LIST OF TABLES

Length of respondents' current or most
recent dating relationship

Depth of comittment of respondents'
current or most recent dating relationship

Frequency of endorsement of parent-child
violence items

Frequency of endorsement of parent-parent
violence items .

Coefficient alphas of subscales
Alpha coefficients for CRAO factors

Pearson correlations between factors
of the CRAO

Cluster populations

Cluster means on seven CRAO factors

Cluster standard deviations on seven
CRAO factors

Means and F-ratios of dependent
variables for each group

Characteristics of discriminant
functions derived

vii

Page

21

22

23
27

30

31

34

13.

14.

15.

16.

17.

18.

19.

21.

24.

Mean discriminant functions evaluated
at group means (group centroids)

Standardized canonical discriminant
function coefficients .

Correlations between discriminant
functions and discriminating variables

Discriminant function classification
results . .

Frequency of responses to CRAO
items 1-67 . .

Appropriate/Generic factor of the CRAO
Moderate violence factor of the CRAO
Avoid factor of the CRAO

Severe violence factor of the CRAO
Distraction factor of‘the CRAO .

Verbal Attack/Coercion factor of the CRAO

Symbolic/Indirect violence factor
of the CRAO .

viii

39

41

42

77
87
88
89
89
90
90

91

LIST OF FIGURES

Figure Page

1. Scree plot of eigenvalues . . . . 86

2. Plot of four group centroids on two
discriminant functions . . . . 92

INTRODUCTION

A large body of work has focused on the the content and dynamics of
conflict resolution between intimates (e.g., Gottman, 1979; Peterson, 1983).
This work has largely involved assessment of marital conflict and coding of
couples discussing one or more areas of conflict.

As expected, research (e.g., Straus, Gelles, & Steinmetz, 1980; Yllo &
Straus, 1981; Makepeace, 1981; Sigelman, Sony, 8: Vlfiles, 1984) has
demonstrated that both married and dating couples utilize physically
coercive/violent behavior(s) in the resolution of conflict within their
relationships. Straus, et al. (1980), in their survey of 2143 American families,
found that approximately 17% reported that they had experienced at least
one act of spousal violence in the last year. Similarly, in his ground breaking
survey of violence among college dating couples, Makepeace (1981) found
that 23% of those surveyed reported they had experienced at least one act of
violence in a dating relationship. This literature has, until recently, pn'marily
been aimed at establishing that such violence occurs. Only recently has work
begun to establish some of the basic correlates of violent behavior in intimate
relationships (e.g., Arias & O'Leary, 1984; Lane & Gwartney~Gibbs, 1985).

To date, no research has attempted to integrate the findings from the
marital/premarital violence literatures and the more general area of marital
conflict assessment. It is reasonable to assume, given the prevalence of
marital violence, that some couples who have participated in studies
assessing marital conflict (e.g., Gottman, 1979) also experience violence in
their relationships. They simply have not been asked, and will not likely be
demonstrating this behavior in front of an observer or video camera. However,

1

2
it seems probable that violent conflict resolution behavior must certainly be
exerting a powerful onfluence on the styles of conflict resolution exhibited in
such studies, and in fact has been demonstrated to have a dramatic impact on
the functioning and structure of the marital relationship (Walker, 1979).

The present study assessed the kind and frequency of conflict resolution
behaviors occurring in dating relationships and of some of the correlates of
these behaviors. The study assessed the frequency of a broad range of
conflict resolution behaviors, including physically violent and verbally and
emotionally abusive behaviors, as well as positive, constructive behaviors.
These behaviors were assessed by the Conflict Resolution and Abuse
Questionnaire (CRAO) (see Appendix A), an instrument developed to
determine how often a wide range of specific behaviors are exhibited and
experienced during the subjects' current or most recent (if not currently
involved) dating relationship. Specifically, each subject was asked to report
on the frequency with which slhe exhibited such behaviors, and also how
often slhe was the recipient of these same behaviors.

The first phase of the study involved a factor analysis of the conflict
resolution items from the CRAO. The items from the CRAO represent a broad
range of conflict resolution behaviors, and were found to be grouped into
factors, each of which represents one specific style or class of conflict
resolution behavior. The development of such factors provided an opportunity
to investigate relationships among several different classes of violent and
non-violent conflict resolution behaviors, and the association of various
' conflict resolution behaviors with a number of hypothesized predictors of
violent conflict resolution behavior.

The first area investigated was the respondents' reported conflict
resolution behavior. Specifically, what types, if any, of non-violent methods of
conflict resolution do subjects use who also report using violence to resolve
conflicts ? Do these who exhibit violent behaviors to resolve conflict also
exhibit fewer non-violent, or prosocial conflict resolution behaviors? Also, are
the. non-violent behaviors exhibited by respondents who also report abuse

3
ciifferent from those methods of non-violent conflict resolution reported by
overall non-violent, or less violent subjects ?

Secondly. conflict resolution behaviors were related to the reported
level of violence in the family of origin. The first question investigated in this
context concerned the relationship between being the recipient of violence as
a child (perpetrated by parent(s)) and the reported conflict resolution behavior
in the respondents' current or most recent dating relationship. Does being the
recipient of violence as a child relate to being more or less likely to be the
perpetrator and/or recipient of maladaptive, or even violent conflict resolution
behavior in a dating relationship ?

Further, information concerning physical violence witnessed between the
respondents' parents was compared with the reported conflict resolution
behaviors in the respondents' dating relationship. Is witnessing violence
between parents associated with violent conflict resolution behavior in dating
relationships ? If so, is one more likely to be a recipient and/or a perpetrator of
maladaptive or violent conflict resolution behavior ?

A third area investigated was the relationship between the duration and
level of committment of the respondent's relationship and the type of conflict
resolution behavior reported. Do those involved in a more committed type of
relationship (i.e., dating exclusively or discussing marriage) report different
conflict resolution behaviors than those in uncommitted relationships 7
Further. does the amount of time one has spent in the relationship relate to the
type of conflict resolution behavior utilized in the relationship ? This area of
inquiry may begin to shed some light on the development of conflict resolution
behaviors (within the temporal context of a dating relationship).

Finally, a fourth area investigated was the extent of reciprocity of conflict
resolution behaviors. The questions addressed here are: Is being the recipient
of one type or types of conflict resolution behavior(s) asssociated with using
those same types of behaviors, or are there subjects who appear to utilize a
set of conflict resolution behaviors regardless of the conflict resolution
behavior experienced by them.

4
Each of these four areas were examined as predictors of different types
or categories of conflict resolution behaviors. Also, sex differences within each
of these predictors were investigated.

II III: II'IB ll' El'

Previous investigators have detailed the relationship between different
intensities and frequencies of reported interspousal violence (e.g., Straus,
1979), and the relationship of this violence with other kinds of conflict
resolution. However, this work has been based largely upon a 14 or19 item
measure of conflict resolution strategies developed by Straus (1979) called
the Conflict Tactics Scales (CTS) (See Appendix B ). This instnrment was
developed by Straus as a technique to measure the ways in which family
members resolve conflicts. The conflict resolution tactics to be measured by
the CTS were chosen initially as a means to test the hypothesis that greater
use of intellectual andreasoning tactics for resolving conflict will result in less
use of verbally or physically aggressive methods of conflict resolution
between family members (Straus, 1974).

The items which comprise the original CTS (as well as the revised form)
were designed to fit into three categories of conflict resolution: 1) the use of
rational discussion and reasoning (called the Reasoning scale by
Straus,1979). 2) the use of verbal and nonverbal acts which symbolically hurt
the other, or the use of threats to harm the other (called the Verbal Aggression
scale), and 3) the use of physical force against another person as a means of
conflict resolution (called the Violence scale). The original CTS (Form A)
consists of a 14 item list of actions which might occur in conflicts. The items
start with those low in coerciveness and become gradually more coercive and
destructive towards the end of the list. The respondent is asked to indicate the
number of times each action occurred during the past year, ranging from
"Never" to "More than 20 times". The CTS is designed to measure conflict
resolution tactics used by all nuclear family members (e.g., between siblings,

5
parent and child, etc.), as well as between spouses.

A revised version of the CTS (Form N) (see Appendix C) was developed
for a national interview survey (Straus et al., 1980). This scale has 19 rather
than 14 items. Minor revisions include a greater focus on the Verbal
Aggression and Wolence scales (via the dropping of one Reasoning item and
adding two Verbal Aggression and three Vrolence items), changes in the
wording of some of the items, a place to record if the act ever happened was
added, and the reponse category range was increased from 0 to 5 to O to 6.

Straus' (1979) factor analysis of the CTS confirmed the three major
conflict tactics included in the CTS. The factor analyses were computed for
interspousal conflict resolution using the sample described in Straus (1974)
with N-385, and replicated with the sample described in Straus et al. (1980)
with N-2143. The first sample, using Form A of the CTS produced three
factors, each of which identifies with one of the three conceptual Conflict
Tactic (CT) Scales. With the second sample, using a different form (Form N),
Straus obtained four factors. The extra factor corresponds to two new items on
this form not included in Form A, specifically, the use of a knife or a gun. Thus,
this additional factor represents a separate dimension, life threatening
violence. The factor stnrcture for husband to wife violence is very similar to the
structure obtained for wife to husband violence.

Analyses of the national sample's responses found that of subjects who
reported engaging in little or no verbal aggression, less than 0.5% also
reported physical violence. Of those couples who reported a low to moderate
level of verbal aggression, 5% also reported physical violence. Nineteen
percent of couples in the upper middle range of verbal aggression reported
violence, and 56% of these couples who were in the top 25% in verbal
aggression reported physical violence.

These results seem to indicate that the use of verbal aggression in
conflict resolution is associated with the use of physical violence. However,
given the limitations of Straus' CTS (there are only three items which could be.
considered verbally aggressive behaviors), it is impossible to determine what

6
other, if any, nonviolent strategies violent couples employ when attempting to
resolve conflict. Further, Straus (1979) used an orthogonal rotation for his
factor analysis, and thus sought to minimize the possibility of interfactor
correlations. Straus (1979) does not report which, if any, of the Reasoning
scale items were used by violent versus nonviolent subjects, and only
mentions that the reasoning items form one of the four factors derived by the
factor analysis.

Straus (1979) also reports on the internal consistency reliability of the
CTS. For Form A, item-total correlations were computed for each of the three
subscales. For the Reasoning scale, the mean item-total correlation (r) for the
husbands was .74 and .70 for the wives. On the Verbal Aggression scale,
mean item-total correlations for husbands and wives were .73 and .70,
respectively. For the Violence scale, the mean correlations for husbands were
.87 and .88 for wives. '

For Form N, Straus (1979) computed‘the Alpha coefficient of reliability for
the three subscales. For husband to wife, Coefficient alphas for the
Reasoning, Verbal Aggression and Violence sacles were .50, .80, and .83
respectively. For wife to husband, coefficient alphas for the three scales were
.51, .88, and .88. These coefficients are reasonably high for the Verbal
Aggression and Violence scales. However, the low coefficients for the
Reasoning scale are a function of the small number of items (three) which
make up the Reasoning scale (Straus, 1979).

The CTS Form A items were included in a list of 23 conflict areas used by
Jorgensen (1977) in a study investigating the amount of marital distress in
marriages where the partners are of differing social class background.
Jorgensen does not specify what the conflict areas were. The questionnaire
consisted of asking the respondents how often (from never to always) they
and their spouses engaged in each behavior listed during a marital conflict. In
a factor analysis performed on the items of this scale, Jorgensen (1977) found
three factors, corresponding to three levels of intensity in marital conflict: 1) a
high intensity factor, which consists entirely of items involving physical assault

7
and force, 2) a medium intensity factor, which includes acts of verbal or
symbolic aggression (eg., insulting the other or breaking something), and 3) a
low intensity factor, which consists of items involving reasoning, empathy, and
providing information.

Hudson and McIntosh (1981 ), in an attempt to expand upon Straus' CTS
developed the Index of Spouse Abuse (ISA) (see Appendix D). This is a
30-item scale designed to measure the severity of physical (ISA-P) and
nonphysical (ISA-NP) abuse that is inflicted upon a woman by her spouse or
partner. The investigators derived a complex scoring system utilizing item
weights to account for the fact that some of the items represent very serious
forms of abuse and others do not. Their scoring formula produces scores that
range between 0 and 100. Hudson and McIntosh used the ISA to study 398
graduate and undergraduate female college students who were either
married or involved in an ongoing dating relationship with a male partner. The
purpose of this study was to confirm the presence of two dimensions of abuse
which were purportedly measured by the ISA; physical and nonphysical.
Thus, the authors performed a factor analysis of the responses of this sample.
The 25 items of the Index of Marital Satisfaction (IMS) (which was also
administered to this sample) were also included in the factor analysis as a
check on the factorial validity of the ISA items. Thus, it was expected that three
factors would emerge: one for the IMS, and two for the ISA, and if the IMS and
the ISA were conceptually distinct, that the IMS items should lead highest on
the first factor, and the ISA items should load highest on the second and third
factors. This procedure was intended as a check on whether or not the ISA is
merely an alternate form of measuring marital discord. Three clear factors
emerged from the factor analysis, which confirmed that the ISA is a measure
distinct from marital discord as measured by the IMS. Specifically, all of the
IMS items loaded highest on the first factor, and none of the ISA items had
their highest loading on this factor. The physical abuse items from the ISA
loaded on the second factor at .50 or better, and none had loadings of this
magnitude on the other two factors. Finally, nearly all of the nonphysical

8
abuse items loaded at .40 or better on factor three.

Reliabilities of the two ISA subscales were examined utilizing coefficient
alpha as an estimator. The ISA-P and ISA-NP subscale reliabilities averaged
across two independent samples were approximately .92 and .93,
respectively. These coefficients are high, and support the claim of Hudson and
McIntosh (1981) that the ISA subscales are unidimensional measures.

MEAD

The CRAO is an instrument designed to be more comprehensive than
either the CTS or the ISA in its assessment of a much wider range of conflict
resolution and interpersonal (including sexual) behaviors. Specifically, the
CTS is limited to the assessment of a narrow range of possible behaviors
within conflict resolution. It lists a total of 13 violent and 5 nonviolent conflict
resolution behaviors. The CRAO includes 67 items which are descriptions of a
wide range of behaviors that might occur when a couple is confronted with
conflict.

All of the items from Straus' CTS are Included within the 67 conflict
resolution items of the CRAO. However, unlike on the CTS, each item
represents only one behavior. For example, CTS (Form N) item n. reads
“Kicked, bit, or hit with a fist“. On the CRAO, each of these behaviors, kicking,
biting, and hitting with a fist, is a separate item. On the CTS, if a person
endorses item n., there is no way to know which of the violent behaviors they
are referring to. The distinction between these behaviors may not be
necessary, as each is simply a violent, assaultive act that could produce tissue
damage. However, there is no evidence to indicate that a person who hits is
also likely to bite and or kick when resolving conflict. Also, one who bites may
not hit with their fist during a conflict. The use of the CTS does not allow the
distinction between which of the act or acts the respondent is referring to. With
the CRAO, since each item represents only one behavior, there is no

9
uncertainty about the behavior of interest.

Another advantage of having an expanded number of conflict resolution
items is that it enables the measurement of a wider variety of both negative
and positive conflict resolution tactics. Straus (1979) includes three general
forms of conflict resolution on the CTS: reasoning, verbal and nonverbal acts
which symbolically hurt the other, and physical force. However, these three
categories of conflict tactics do not represent a broad range of human
behavior during conflict, and are certainly not exhaustive. For example, a
person may also attempt to avoid conflict (e.g., put off talking about a
problem), deny conflict (e.g., not accept that there is a problem), bring in a
third party to help resolve the conflict, or attempt to distract the other from the
issue, to mention a few alternatives.

Also, even within Straus' three categories distinctions could be made
between various forms of reasoning (e.g., discussing the issue calmly vs.
suggesting possible solutions), and different types of verbal assault (e.g., an .
insult vs. yelling), and although Straus distinguishes between extreme
violence (using a knife or a gun) and other forms of physical assault, there is
no sharp distinction between different violent behaviors, for example, varying
intensities of violence (e.g., squeezing someone's arm, pulling hair, or
slapping a lower body part vs. slapping the face, hitting someone with an
object vs. choking the person, etc.).

Most importantly, Straus does not (and given the brevity and generality of
the CTS is not able to) delineate the relationship(s) between the different
reported conflict tactics. For example, if a respondent indicates on the CTS
that he or she is the perpetrator of a violent act, such as hitting with a fist,
Straus generally does not report what other violent or nonviolent acts he or
she uses.

Hudson and McIntosh (1981) overcome some of the shortcomings of the
CTS with their ISA instrument. However, the ISA items are not presented in
the context of conflict resolution. The respondent is simply instructed to report
the degree of abuse experienced in her (the ISA is only intended for female

1 0
respondents) relationship with her partner. Thus, a direct comparison with the
CTS is not possible, as the two questionnaires elicit different information. The
ISA is a list of physically and nonphysically (psychologically and sexually)
abusive acts which presents a wider range of behaviors than the CTS.
However, there is no way to determine in what context the reported acts took
place (i.e., during a conflict, or with no apparent reason, etc.). Also, there are
no positive or nonabusive acts included as items on the ISA. Therefore, the
factor analysis could only reveal two factors, physical and nonphysical abuse,
which provide little information beyond the validation of scale items.

8' 'I IS D'II i8 III
8 II' Bl'

An important issue to be addressed in any inquiry concerning behavior
ocurring between two people Is the extent of reciprocity of that behavior. For
the most part, research in the area of violence between spouses has focused
upon the one-way nature of such interaction (i.e., male assaulting female)
(e.g., Pagelow, 1981; Roy, 1977; Gelles, 1974; Walker, 1979). However,
recent surveys of married and dating couples have indicated that violent
interactions between intimate heterosexual couples is indeed reciprocal (e.g.,
Straus et al.,1980; Steinmetz, 1980; Sigelman at al.,1984; Lane 8:
Gwartney-Gibbs, 1985). In fact, there seem to be no significant sex differences
in the reported use of violence in both married and dating couples. However,
in order to determine the reciprocity of abuse, one must actually specify that
the interaction is indeed reciprocal within each couple, rather than simply
indicate that there are no sex differences in the behavior across subjects.

In their survey of 2143 families using the CTS, Straus, et al. (1980) found
no statistical difference between the rates of violence reported by men and
women. The only differences revealed indicated that twice as many wives as
husbands threw something at their spouse. The rate for kicking, biting, and

1 1
hitting with an object is also higher for wives than for husbands. Husbands
had higher rates for pushing, shoving, slapping, beefing up, and using a knife
or a gun. Thus, even though overall violence rates are similar for husbands
and wives, it appears from these data that husbands are overall more likely to
use the most severe and potentially life threatening conflict resolution tactics
than wives.

Straus et al. (1980) report that although most of the violence towards
husbands by women in this sample seems to be committed in self-defense,
nearly 30 percent of the women committed violent acts that could not be
classified as such. Data from several other studies of violent marriages further
reflect this phenommenon of reciprocity in abuse (Steinmetz, 1977; Steinmetz,
1980). However, Steinmetz notes that when one focuses on more severe
violence, differences between husbands and wives do emerge. Steinmetz
(1977) provides data based on police records and a random sample of
families which estimates that 7 percent of the wives and 0.6 percent of the
husbands would be victims of physical abuse resulting in serious tissue
damage or even death by their spouses.

In a survey study of violence in dating relationships, Sigelman et al.
(1984) found no overall difference between sexes in the commission of at
least one form of physical aggression. They found that women, in fact, show a
greater tendency to engage in some of the more serious forms of violence
(e.g., using a weapon). Also, the men in their sample reported having been
the recipient of at least one such serious act more frequently than women.
Sigelman et al. (1984) also found that physical violence in dating
relationships was usually mutual as well. They report that for both men and
women, having physically abused a partner in some way was strongly related
to having been physically abused in some way; 47.3 percent of the men in
their sample and 40.2 percent of the women reported both abusing partners
and being abused themselves. On the other hand, 34.5 percent of the men
and 40.5 percent of the women reported being neither perpetrators nor
recipients of violence.

1 2

From a sample of 371 college undergraduates surveyed concerning
violence in their relationships. Laner and Thompson (1982) and Laner (1983)
also reported no sex differences in violence between dating couples.
Although not reported, on the basis of the frequency data broken down by sex,
there appears to be no large differences between rates of violence reported
by men and women. Further, Laner (1983) found that 67 percent of the men
and 48 percent of the women stated that violent events in the relationship
were mutual. '

Lane and Gwartney-Gibbs (1985) using data from a survey of 325
college undergraduates also confirm the finding that violence between dating
couples is inflicted and experienced equally by men and women. However,
more males reported having perpetrated more serious forms of assault
against women (e.g., using a weapon or causing severe tissue damage).
These authors further found that males were more likely to report having
experienced violence on multiple occasions with multiple partners, wheras
most of the women in the sample reported having expereineced violence only
once with one person. A survey of 461 college students by Bernard and
Bemard (1983) also supports the finding of mutuality of violence.

On the other hand, Makepeace (1983) found, in his survey of 244 college
undergraduates, that men were more likely than women to report having used
some type of violence in the context of a dating relationship. Further, the rate
of using severe violence against a partner was found to be 2.5 times higher for
males than for females. However, when examining the rate of reported
receptive violence, there were no significant sex differences.

Stats 8. Pirog-Good (1987), in a survey of 505 college undergraduates,
found that women are more likely than men to receive violence in a dating
relationship. Howevr, women in their sample reported using the same amount
of violence as the men. The scoring system used by these investigators did
not permit analysis of sex cfifferences across various levels of severity of
violence.

It is apparent that verbal and physical assault between married or

1 3
unmarried couples is equally likely to be perpetrated and/or received by men
and women, although there is some evidence indicating that this may not be
true for all forms of violence, and there may be sex differences in using versus
receiving violence. All of the studies cited above used the CTS or a
modification thereof in the gathering of their data Thus, the subjects involved
were reporting on a narrow range of primarily violent behaviors used to
resolve conflict. However, given the limitations of the CTS, it is impossible to
determine the mutuality of other nonviolent or maladaptive forms of conflict
resolution, and how these other forms of conflict resolution might be related to
violent conflict resolution and related sex differences.

For quite some time investigators have demonstrated a link between
violence and aggression in the home and violence perpetrated by children
(e.g., McCord, McCord, and Howard, 1961; Bandura, Ross and Ross, 1961;
McCord, McCord, and Howard,1963; Bandura and Walters, 1963;
Bandura,1973; Patterson, 1976). Early investigators established that children
who are raised in homes where they are consistently exposed to aggressive
and violent acts by their parents are more likely than children in
nonaggressive homes to be aggressive and violent towards peers. Bandura
(1973) hypothesized that this transmission of aggressive behavior from
parents to children takes place through social learning; children learn through
the observation of their parents' aggressive behaviors and the reinforcements
their parents' receive as a result of such behavior.

Early investigators in the area of spouse abuse (e.g., Gelles,1974;
Owens and Straus, 1975; Carroll, 1977) observed that both the victims and
perpetrators of conjugal violence were more likely to have experienced
violence in their families' of origin, both as witnesses and victims. In a

1 4
nationwide survey conducted by the President's Commission on the Causes
and Prevention of violence, Owens and Straus (1975) found that adults'
approval of violence was as highly correlated with being a victim of violence
as a child as was committing a violent act as an adult. Carroll (1977) found a
clear relationship between the severity of parental punishment and reports of
family violence in his sample of 375 families.

Straus et al. (1980) also gathered extensive data on the respondents'
history of violence in their family of origin. Men who had seen their parents
physically attack each other were almost three times more likely to have been
violent toward their wives during the past year; 35% of the sons of violent
parents versus 10.7% of the sons of non-violent parents reported using
violence against their wives. The statistics for women parallelled those of the
men; 26.7% of the women who had observed their parents engage in violence
against each other reported using violence against their husbands, as
compared to 8.9% of the daughters of non-violent parents.

Straus et al. (1980) also reported that, in general, the more punishment
received as a child, the greater the rate of violence in marriage 15 or more
years later. Specifically, people whose parents did not hit them as teen-agers
had the lowest rate of violent marriages. Approximately one out of four of
those who reported receiving physical punishment as a child also reported
being violent toward his or her spouse during the survey year. Of the men who
were not physically punished as teenagers, only two out of a hundred
reported violence toward their spouse. Straus, et al. (1980) report that this
same low rate also applies to assault of husbands by women who had not
been hit as teen-agers. They also hypothesized that subjects who both
witnessed their parents being violent toward each other and experienced
physical punishment themselves would report the most conjugal violence,
more than either group by itself. This hypothesis was also confirmed. The
group having experienced both types of violence during childhood had the
highest rate of violence in their own marriages (25% reported conjugal
violence).

1 5

In a survey of violence in dating relationships Bernard and Bernard
(1983) found that in their sample 73% of the abusive males had either
experienced or observed abuse in their families of origin, while only 32% of
the nonabusive males had either experienced or observed abuse in their
families of origin. Of the abusive females in the sample, 50% had either
experienced or observed violence in their families of origin, while 23% of the
nonabusive females had observed or witnessed abuse in their families of
origin. They further reported that there was no relationship between the type of
exposure a child had to abuse in the family of origin and whether or not they
became abusive themselves.

Laner and Thompson (1982) reported similar findings concerning
childhood exposure to violence and violence in a dating relationship. It was
found that 60% of the women and 67% of the men who had experienced
childhood violence were also the recipients of abuse in courtship. However,
for both sexes, those who had not experienced childhood violence still “had
about an even chance” (p.236) of experiencing violence in a dating
relationship. Also, it was found that of those subjects who had experienced
violence during childhood, approximately 50% reported inflicting violent
behaviors in dating relationships. However, 33% of those who did not report
childhood violence also reported inflicting violence in a dating relationship.
However, Laner and Thompson (1982) did not collect data concerning the
extent of violence between parents witnessed by subjects. The authors found
that being the recipient of violence as a child did increase the Iikliehood of
receiving or perpetrating violence in a dating relationship, although a
significant proportion of those who did not receive violence in childhood also
reported violent experiences in dating. However, they note that this
association may have been stronger had data concerning witnessing of
violence been gathered. They provided no evidence other than frequency
statistics to support any of these claims.

In a survey of 505 college undergraduates, Stats 8: Pirog-Good (1987)
utilized a probit model to estimate both using and receiving violence in a

1 6
dating relationship for males and females. These investigators found that
experiencing violence in childhood was a significant determinant of receiving
violence for males in a dating relationship. However, this variable did not
significantly increase the Iikliehood of either males or females using violence,
or of females receiving violence in a dating relationship. Further, witnessing
violence as a child did not have a significant effect on using or receiving
violence for men or women. Stats & Pirog-Good concluded that these
historical factors may be more likely to explain violence in more serious
relationships.

In separate discriminant analyses for college males and females,
Sigelman et al. (1984) linked involvement in abuse to having been abused as
a child, but only found a significant relationship for the women in their sample.
Further, there was no significant relationship between having witnessed
parental violence and dating violence for either males or females in their
sample. Thus, with data subjected to a somewhat more sophisticated
multivariate statistical analysis (i.e., a discriminant analysis), there is not as
clear a relationship between childhood history of violence and courtship
violence as was previously reported in other studies. The reason for this
ciscrepancy with other studies is not clear. Sigelman et al. (1984) suggest that
their correlational and discrrninant analyses offered a ”methodological
waming' about research on this sensitive topic: the possibility that socially
desirable responding, especially among women, leads to underreporting of
experience with abusive behavior. However, these authors do not make it
clear why socially desirable responding would result in low and/or
nonsignificant correlations in such data, other than implying that
underreporting could lead to these statistical problems.

Further, exposure to violence as a child might not be as important a
factor in violence in dating relationships for men as it is for women due to sex
differences in socialization. Boys tend to be socialized in a more aggressive
fashion than girls (e.g., being exposed to more physical play by their fathers,
being given toy guns, (3. l. Joe and Rambo dolls to play with, etc.), regardless

1 7
of whether or not they are physically abused by their parents or they observe
their parents being violent toward each other. Evidence indicates that overall
boys are more aggressive than girls (e.g., Hartup,1977; Maccoby and Jacklin,
1980). Thus, boys may not require the direct exposure to childhood abuse or
parental abuse to become involved in dating violence as a young adult. On
the other hand, women, who are not usually socialized In an aggressive way,
may not be prone to courtship violence unless there is a history of violence in
their childhood.

It is apparent that early studies of marital and dating violence found
strong evidence for a link between family violence in chioldhood and the use
of violence in a current intimate relationship. However, more recent studies
utilizing multivariate statistical techniques have demonstrated that this
relationships is not as strong as previously believed, and in one study (Stats &
Pirog-Good, 1987), history of violence was not associated with the use of
violence for either sex.

I: II'IB II' III I’
II IDI' BII' l'

Three studies investigating dating violence have examined the
relationship between the duration and level of committment of relationships
and the type of conflict resolution behavior in the relationship. Laner and
Thompson (1982) found that regardless of the type of violent conflict
resolution behavior, relationships defined as serious and meaningful (exact
definition not provided by the authors) showed higher rates of abusive and
aggressive conflict resolution behaviors than those defined as casual.
Sigelman at al. (1984) found that 74% of the relationships with abusive
conflict resolution behaviors were characterized as beyond the ”casual dating"
stage and into either the “serious dating” stage or beyond (i.e., engagement or
cohabitation) when the violence began.

Stats 8 Pirog-Good (1987) found that length of relationship was a

1 8
significant determinant for using and receiving violence in a dating
relationship for males, but was not associated with using or receiving violence
for females. Measuring degree of seriousness by the combination of frequent
dating and lengthy relationships, Stats & Pirog-Good (1987) also found that
being involved in a more serious relationship was associated with using
violence for males. However, they did not find this direct relationship between
length and depth of relationship and using violence for females.

Other studies of dating violence have not included reports on the level of
involvement or duration of the relationships where maladapive conflict
resolution behaviors are reported. Studies of violent conflict resolution
behavior between marital partners by definition only include couples at one
level of stated relationship commitment. Further, these studies generally do
not ask respondents about violence that might have occurred in this or
previous relationships before the marriage. Thus, very little information exists
concaming the evolution of conflict resolution behavior in heterosexual
relationships, and the different behavioral characteristics of conflict resolution

at varying levels of commitment to and duration of the relationship.

Hypotheses

. Reports of being the recipient of violence as a child will be
associated with reports of current or recent coercive conflict
resolution behavior.

. Reports of witnessing violence between parents as a child will be
associated with reports of current or recent coercive conflict
resolution behavior.

. Those who report a longer duration relationship will report using
more coercive conflict resolution behaviors.

. Those who report a more committed relationship will report using
more coercive conflict resolution behaviors.

. Those who report being the recipient of current or recent

coercive conflict resolution behaviors will also report using
coercive conflict resolution behaviors.

19

Method

Subjects. A sample of 671 undergraduates completed the CRAO. The
respondents were students in introductory psychology classes at Michigan State
University who participated in the study in exchange for extra credit points. Of
these,10 indicated that they were married, and were dropped from the final
samme, which was 661. The sample was 54% female and 46% male. The
sample ranged in age from 16 to 44, with a mean age of 19.6 years. Racially, the
sample was 90% white, 8% black, and approximately 2% Hispanic, Asian, and
others. Fifty nine percent of the respondents reported that they were currently in
a relationship; the 41% of the sample who reported they were not currently in a
relationship completed the CRAO based on their most recent relationship.
Differences between these two groups on the hypothesized variables were
tested via Chi Square, and none were significant.

Seventy three and nine tenths percent of the respondents reported living
with their biological mother and father during high school, 8.1% reported living
with their biological mother and a step-father, 2.9% reported living with their
biological father and step-mother, 12.4% reported living in single mother homes,
0.9% reported living in single father homes, and 1.8% reported living with
adoptive parents.

Twenty seven and seven tenths percent of the respondents reported that
their fathers had graduated from college and 28.9% reported that their fathers
had gone to graduate or professional school. Sixteen and seven tenths percent
of the fathers had reportedly graduated from high school only, and 5.0% had
only completed some high school. For mothers, 28.3% of the respondents
reported that their mothers had graduated from college and 14% reported that
their mothers had gone to graduate or professional school. Twenty eight and
three tenths percent of the mothers had graduated from high school only, and
3.8% had only completed some high school.

Seventy six and nine tenths percent of the respondents reported that their
fathers held white collar or professional jobs, while 15.7% reported their fathers

2O

21
holding blue collar jobs (e.g., factory workers). Fifty seven and four tenths
percent of the respondents reported that their mothers held white collar or
professional jobs,10.3% reported their mothers with blue collar jobs, and 28.4%
of the respondents reported that their mother was a homemaker only.
The proportions of respondents in relationships of different lengths and levels
of committment are shown in Tables 1 and 2. Tables 3 and 4 show the
frequency of reported child abuse and abuse witnessed between the
respondents' parents.

Table 1
Length of Respondents‘ Current or Most
Recent Dating Relationship

 

 

Length Proportion of Respondents
Less than 1 month 11.6%
1 to 2 months 13.4%
3 to 6 months 20.4%
7 to 12 months 17.1%
1 to 2 years 22.7%
3 to 5 years 10.1%
6 or more years 1.2%

 

Note. N-648

 

22

Table 2
Depth of Committment of Respondents' Current or
Most Recent Dating Relationship

 

 

 

Depth Proportion of Respondents
Dating nonexclusively 28.8%

Dating exclusively 44.4%

Discussing marriage 17.6%

Engaged 3.3%

Living together 1.8%

Note. N-653

Table 3

Frequency of Endorsement of Parent-Child Violence Items

 

 

Item Percent Reporting
Threaten to hit you or throw something at you 38.5
Throw something at you 15.8
Push, grab, or shove you 36.7
Spank you 78.1
Slap you 48.0
Kick, bite, or hit you 9.7
Beat you up 2.1
Threaten you with a gun or knife 0.9
Use a gun or knife on you 0.1

 

N016. N2647

23'

Table 4
Frequency of Endorsement of Parent-Parent Violence Items

 

 

 

Item Percent Reporting
Threaten to hit or throw something at each other 23.0
Throw, hit,kick, or smash something 31.1
Throw something at each other 11.0
Push, grab, or shove each other ' 13.7
Slap each other 7.0
Kick, bite, or hit the other 4.9
Beat up the other 2.4
Threaten each other with a gun or knife 3.0
Use a gun or knife on each other 0.4
Note. N-646

mm Subjects were administered the CRAO. This instrument was
designed to assess the frequency and intensity of both appropriate and
coercive behaviors in the domains of conflict resolution, verbal/emotional
interactions, and sexual interaction. Development of the CRAO was undertaken
by Curtis, Maguin & Stollak (1987) to bring together and expand upon previous
studies of marital violence which have used stnrctured clinical interviews (e.g.,
Walker, 1979) and/or brief questionnaires covering only selected and narrow
areas of behavior (e.g., Straus, 1979; Hudson & McIntosh, 1981) to assess the
level of violence and/or coercion between couples. The CRAO integrated these
three areas of interaction into one asessament device utilizing a common
methodological format.

2 4

The first section of the CRAO consisted often questions concaming
demographic characteristics of the respondent. Another set of demographic
questions on the CRAO included items that pertain to the marital, educational,
and occupational status of the respondents' parents.

The remainder of the first section included several scales designed to
assess the dating and sexual histories of the respondent. In addition,
information concaming characteristics of the current or most recent relationship
was gathered. This included length of involvement, depth of commitment, and
degree of sexual intimacy. Also, items asked about the frequency of alcohol
and/or drug use in the context of the current or most recent relationship.

The final section of the first part of the questionnaire included two scales to
measure the degree of violence in the respondent's childhood. The items
comprising these scales were taken from the CTS (Straus, 1979). The first
sale asked the respondent to indicate which of nine violent behaviors he or
she ever saw his/her parents engage in. The second scale asked the
respondent which of nine violent behaviors his or her parents ever exhibited in
their encounters with him or her.

The first set of items in the main body of the questionnaire (nos. 1-67)
asked the respondent to report on the frequency of a variety of conflict
resolution behaviors employed between him/herself and his/her current or most
recent boy/girlfriend. Each item is a behavior that could be employed to resolve
interpersonal conflict, and the respondent was asked to rate approximately how
many time he/she used this method of resolving conflict with his/her current or
most recent boy/girlfriend by way of a frequency scale,1 to 5, where 1-never,
2. 2 to 4 times, 3- 5 to 10 times, 4-11 to 20 times, and 5. 21 or more times
during the course of the relationship. Each subject was then asked to report on
the methods used by his or her boy/girlfriend using the same rating scale.

Items comprising this scale were taken from a number of different sources
in an attempt to represent the widest possible range of conflict resolution
behavior. Many items representative of violent behavior were taken from
Straus' (1979) CTS. Items representing positive problem solving were derived

2 5
from Meichenbaum's (1977) problem solving steps. Other items were derived
from clinical reports of battered women and their violent husbands. Some Items
were the result of brainstorming possible conflict resolution behaviors, with the
general categories of positive, avoiding, negative (nonviolent), and violent
behaviors guiding development.

A second set of items (nos. 68-93) measured the frequency of varying
styles of emotionally and/or psychologically abusive behavior that might take
place between dating couples. A third set of items (nos. 94-123) were designed
to assess the frequency of various sexual behaviors between dating couples.
The purpose of these items was to determine the frequency of mutually
pleasurable and acceptable sexual behavior between dating couples as well
as the frequency of coerced, unpleasant sexual behavior experienced and/or
perpetrated by the respondents.

The items from these two other sections of the CRAO were not utilized in
the present study. These items are not posed in the context of conflict
resolution. Rather, they measure whether or not any of these behaviors have
ever occurred in the relationship, regardless of context. However, the first 67
items refer specifically to behaviors occurring during the resolution of a conflict
situation. Hence, these first 67 items were the focus of the present study.

Also included as part of the questionnaire was a 15 item validity scale,
which is a modification of the Short-Form Marital Conventionalization (MC)
Scale (Edmonds,1967; Edmonds, Withers, 8 Dibatista,1972). The original scale
as developed by Edmonds (1967) was designed to be administered to married
couples. Hence, the wording of some of the items was modified slightly so that
the scale would be applicable to unmarried dating couples. The scale consists
of 15 statements which the respondent marked either true or false as they
applied to the relationship reported on. This validity scale was included as a
check for random responding by subjects, as well as to explore the relationship
between respondents portraying their relationship in the direction of social
desirability and reports of physical, emotional and/or sexual abuse Maguin,
Curtis, 8 Stollak, 1987). Questionnaires with extreme scores were not included '

2 6
in the analysis.

W. The length of the current or most recent relationship
was measured on a seven-point scale, with one being the shortest length (less
than a month) and seven being the longest (six or more years). This was the
length of relationship scale.

The level of commitment in the current or most recent relationship was
measured on a six-point scale from dating nonexclusively to marriage (1 to 6).
This was termed the depth of committment scale.

History of child abuse was measured by the respondent indicating which
of nine abusive behaviors he or she ever experienced at the hands of his/her
parents. The items were taken from Straus' (1979) Conflict Tactics Scale. One
of the nine items, “spanked“, was marked by approximately 80% of the
respondents. It was felt that since this was appareme such a normative
behavior In this sample that it should not be considered for the purposes of the
study as an abusive behavior, and was thus not Included in the analysis. Each
item was scored as zero or one, for not checked or checked. Each respondents'
score on this scale was the resulting sum of the items. This scale was called the
child violence scale.

The violence between parents witnessed by the respondents was
measured by a similar nine item scale taken from Straus (1979). This scale was
termed the parental violence scale. Again, respondents were asked to indicate
which of the behaviors they ever witnessed their parents engage in or display.
A score was derived for this scale by summing the responses based on the
zero-one scoring.

A dating violence received scale was derived by scoring respondents for
the level of violence received in their current or most recent relationship based
on reports of 17 possible violent conflict resolution behaviors received from
their partner. These 17 items were selected from the second half of the CRAO
which has items that are identical to those in the first half, but where the
question asked is which of these have you experienced as opposed to inflicted.
These 17 items were taken from the moderate and severe violence factors

2 7
derived from the factor analysis. A recipient score was derived by summing the
responses to these seventeen items. Thus, the lowest possible score is zero,
and the maximum is 68.
The coefficient alphas of each of the violence scales were calculated
(T able 5) and were found to be of sufficient magnitude that the scales could be
considered unidimensional measures of the reported events (Nunnally, 1978).

 

 

 

Table 5

Coefficient Alphas of Subscales

Scale Coefficient Alpha
Violence Received in Relationship .906
Receiving Childhood Abuse .702
Witnessing Parental Violence .774
Note. N-630

mum. Respondents signed up for a psychology experiment entitled
“Behavior in Dating Relationships.“ Each respondent was instructed on the
sign-up sheet as to the date, time, and location of the ”experiment." Groups of
approximately 40-50 undergraduates were recruited for each administration of
the CRAO. These group administrations took place over a period of three
months in large classrooms located on campus.

Before each session, subjects were read a brief set of instructions
describing the CRAO and how to fill it out. Subjects were told that the purpose
of the study was to investigate conflict resolution behaviors of those involved in
dating relationships. No reference was made to any specific behavior (e.g.,
violence) as one of interest. The CRAO took most respondents approximately
45 minutes to complete.

Results

The frequencies of endorsement for the first 67 items of the CRAO are
presented in Table 17 In Appendix E. As expected, the typical, non-coercive
conflict resolution behaviors were reported as rather common, wheras the
more coercive and violent behaviors were reported less frequently.

W. The first step in the formal analysis was a factor analysis
of the conflict resolution behaviors endorsed (C RAO items 1-67). A principal
components procedure was employed, with an oblique (oblimin) rotation of
the initial factor solution. For the oblimin rotation, Delta was set at 0, creating a
fairly oblique (correlated) solution (Nie, Hull, Jenkins, Steinbrenner 8 Bent,
1975). Seven factors were retained for final rotation. This decision was based
on a SCREE test of the eigenvalues (see Figure 1, Appendix F) (Cattall,
1978). An oblique rotation was employed rather than an orthogonal rotation
on the conceptual grounds that various conflict resolution behaviors are in fact
related, and thus correlated. It is unreasonable to assume that an individual's
use of one behavior (e.g., arguing) is not also sometimes accompanied by
another different class or classes of behavior (e.g., yelling, hitting an object, or
even before the yelling began, calm discussion). Further, an oblique rotation
allows correlational relationships between factors (i.e., different classes of
conflict resolution behaviors) to be examined.

The seven factors derived included 59 of the scale's original 67 items.
The item inclusion criterion was a factor loading of at least .30. In addition,
three omitted Items, #4 “Used alcohol or drugs to settle the conflict”, #7 Prayed
with him/her or alone“, and #45 ”Called police” were excluded for both
conceptual and quantitative reasons. Conceptually, these items are not
reflective of commonly employed conflict resolution strategies. They are more
specialized or unusual strategies which were rarely endorsed. In fact, these
non-violent items were endorsed less often than many of the violent items on

28

 

2 9
the questionnaire. Further, the exclusion of these three items significantly
improved internal reliabilities (as calculated via coefficient alpha) of the factors
initially derived from the factor analysis.

Two items loaded on two of the factors. Item # 39, 'Soothed or calmed
him/her“, loaded on the Appropriate/Generic factor as well as the Distraction
factor. This is not surprising, as this conflict resolution behavior could refer to
both a generally positive behavior and/or a way of directing attention away from
conflict. Item if 47, 'Yelled or screamed at him/her” loaded on the
Appropriate/General factor and the Verbal Attack/Coercion factor. This item
loaded moderately on both factors, thus it was felt that it should be retained.
This item clearly fits in the Verbal Attack/Coercion factor. However, it is not clear
why it loaded on the Appropriate/Generic factor. Perhaps this Item was
interpreted by some respondents as a normative, and not overly negative
behavior. Thus, it is correlating with other, more clearly positive behaviors.

With a listwise deletion of missing data (analysis performed using only
cases with nonmissing values on the 64 items used), and the exclusion of
subjects who indicated they were married, the final N for the factor analysis was
588. The items in each factor and their factor loadings are shown in Tables
18-24 in Appendix G.

The first factor contains items that seem to reflect generally appropriate
and/or common conflict resolution behaviors. Not all of the items are positive or
facilitative. However, none of the items is violent, and none of them involves
behavior that is overtly coercive. The second factor consists of items that are
violent. These are all items that involve the use of some type of physically
coercive behavior that is utilized to force the will of the actor upon the receiver
in the resolution of a conflict. However, these are the moderate violence items
on the CRAO. The items comprising the third factor generally reflect behaviors
employed to avoid dealing directly with conflict. Factor four consists of the most
violent items on the questionnaire. The fifth factor contains items which reflect
attempts to avoid conflict by distracting the other person. Factor six represents
generally negative verbal attack behaviors, that are not physically coercive.

3 0
These behaviors perhaps serve to amplify conflict, and are reflective of
inappropriate, non-constructive conflict resolution behaviors. The seventh and
final factor extracted is one which contains the items of violence directed at
objects during conflict.

Composite scale scores were computed based on the sum of the
responses to the items In the seven factors using the 0-4 frequency scale of the
CRAO. The internal consitency (coefficient alpha) of the items of each
composite are presented in Table 6. The coefficient alphas are of sufficient
magnitude such that each of the composite scales for the extracted factors
appears to meet the conventional standards for satisfactory internal consistency
(Nunnally, 1978).

Table 6
Alpha Coefficients for CRAO Factors

 

 

 

Factor Coefficient Alpha
Appropriate/Generic .937
Moderate Violence .886
Avoid .854
Severe Wolenca .674
Distraction .842
Verbal Attack/Coercion .936
Symbolic/Indirect .725
Note. N-588

The Pearson interfactor correlations were calculated for the raw subscale
scores. These are presented in Table 7. As can be seen, these correlations
were for the most part moderate to high, indicating a great deal of
interrelatedness among the subscales. The only two nonsignificant correlations

3 1
were between the Severe violence subscale and the Appropriate/Generic and
Distract subscales. The Severe and Moderate violence scales were moderately
correlated, while there was a much higher correlation between the Moderate
and Symbolic violence subscales. There was a somewhat low correlation
between the Severe violence subscale and the Symbolic violence subscales.
The Verbal Attack/Coercion, Appropriate, and Avoid subscales were also highly
correlated. The lowest significant correlations were between the Severe
violence subscale and the Verbal Attack, Symbolic violence and Avoid
subscales. There was also a low but significant correlation between the
Moderate violence subscale and the Distract subscale. Otherwise, the factors
generally appear to be moderately to highly intercorrelated.

Table 7
Pearson Correlations Between Factors of the CRAO

 

Factor 1 2 3 4 5 6 7
AppropGen Mod.Viol. Avoid SevereVioI. Distract Verbl.Attack Symblc.

 

1 1 .00

2 .35‘" 1 .00

3 .57‘“ .33“‘ 1 .00

4 .03 .27'“ .1 4*“ 1 .00

5 .64‘“ .21 ”' .38'“ -.01 1 .00

6 .75'" .75'“ .65’” .1 4*“ .49'” 1 .00

7 .54‘” .54‘” .52’” .1 7"" .32’" .63”* 1 .00

 

Notes. Correlations based on N-591

‘“ p< .001

W. The second step of the analysis was to develop profiles
of respondents through a cluster analysis of cases based on the scores on the
composite subscales. These scores formed the data set for the cluster analysis.
A k-means clustering technique (Dixon, 1981) was employed whereby the

3 2
cases were partitioned into clusters based on these raw score-based
composite scales. For clustering, the data were standardized univariately using
the standard deviation of the complete sample. The final number of cases
utilized for the cluster analysis (after deletions due to missing data) was 635.
After experimentation with solutions of varying numbers of clusters, a five
cluster solution was viewed as the most parsimonious without losing important
detail, given the number of subjects. Cluster populations, means, and standard
deviations are presented in Tables 8-10.

In general, these clusters seemed to represent distinctive and logical
conceptual groupings of cases. The first cluster contained only three cases (this
cluster remained even in a four cluster solution). This cluster clearly included
the most violent of the respondents, with scores on the Severe and Moderate
violence subscales elevated well above the mean. These respondents were
deemed Life Threateners. However, it is important to note that any conclusions
concaming these respondents should be tempered by the large standard
deviations of the subscale means. For this reason, and because of the small
number of respondents in the cluster, it was decided to exclude this cluster from
any subsequent analyses.

The second cluster included those cases that are close to themean on all
the subscales except for Severe violence, on which they were below the mean
(in fact, they have the lowest mean on this composite). They were slightly below
the mean on the moderate violence subscale, but were still within one standard
deviation. These respondents were termed Positive/Average Conflict
Confronters. The third cluster seemed to be comprised of those respondents
who were below the mean on all seven subscales. This cluster may represent a
group of people who generally respond passively to conflict, or perhaps are in
short term relationships and thus have not experienced a great deal of conflict
in the relationship. This group of people might best be called lnexperienced
Nonresponders. The fourth cluster consisted of the fewest people aside from
cluster one. These respondents reported a great deal of violent behavior, but
did not report perpetrating the most severe acts of violence. They were well

 

 

3 3
above the mean on the Moderate violence scale, and are elevated on the
Severe violence scale, although not by one standard deviation or more. In
addition, they were very high on all of the other scales. This group was termed
the Coercive/Active Conflict Resolvers. The fifth cluster contains people who
were above the mean on all of the scales except the two violence scales, on
which they fell near the mean. These respondents were called the
Belligerent/Active Conflict Resolvers.

 

 

Table 8

Cluster Populations
Cluster N
Life Threateners 3
Positive/Average 255
lnexperienced 203
Coercive/Active 39
Belligerent/Active 135

 

Total 635

 

Table 9

34

Cluster Means on Seven CRAO Subscales

 

 

 

 

 

 

Approp' Moderae Severe Verbd Syn'bolic

Ctrster Genera Violence Avoid Violence Distract Attack Violence
Lie Threateners 29.67 12.67 23.67 7.33 1 1.00 31 .33 6.00
Positive/Average 31.05 1 .08 10.17 0.03 14.40 13.40 1.20
lnexperienced 16.70 0.19 5.73 0.03 7.53 5.40 0.39
Coercive/Active 46.08 13.84 18.26 0.18 16.26 35.44 6.38
Belligerent/Active 42.89 2.08 17.78 0.03 16.40 27.43 3.72
Grand Mean 29.66 1.81 10.94 0.08 12.72 15.18 1.82
Table 10
Cluster Standard Deviations on Seven CRAO Factors

Amopi Modeiae Severe Verbd Synbolic

Chalet General Violence Avoid Violence Distract Attack Violence
Lle Threateners 5.51 16.74 1 1 .02 1 .53 2.65 15.57 4.58
Positive/Average 7.94 1 .78 4.55 0.27 3.82 6.28 1 .30
lnexperienced 7.25 0.57 3.41 0.30 3.57 4.26 0.79
Coercive/Active 6.07 6.32 6.24 0.51 3.90 9.19 2.83

Belligerent/Active 7.28 2.1 1 5.77 0.28 3.48 8.77 2.12

 

35

W. The five dependent variables were

subjected to a 4 (group) x 2 (sex) multivariate analysis of variance. The
multivariate F for group was significant (p<.001). However, the multivariate F of
the interaction of group x sex was not significant (p>.10). Since there was no
multivariate effect for the interaction of group and sex, univariate F's were
examined with males and females combined for each of the dependent
variables by groups.

W. The relationships between each of the five
hypothesized predictors of coercive behavior In dating relationships and the
groups resulting from the cluster analysis were tested by one-way ANOVAs,
with group membership as the independent variable.

The length of the current relationship was a significant predictor of group
membership. The mean of the lnexperienced group (group three) was
significantly lower (p<.05) than the means of all other groups according to a
Tukey HSD procedure. Also, the means of the Coercive/Active and the
Belligerent/Active groups (groups four and five) were significantly higher than
the mean of the Positive/Average group (group two) (p<.05) (see Table 11).
There was no difference between the lnexperienced and Positive/Average
groups, as well as no difference between the Belligerent/Active and
Coercive/Active groups.

36

Table 11
Means and F Ratios of Dependent Variables For Each Group

 

 

Varlmle Pcsiive/ lnexperienced Coercive/Active Beigerert/ F(3,619)
Mace m

Length of

Relationship 3.783 2.73), 4.596 423c 43.42“-

Depth of

Relationship 2.063 1.76:, 2.41 a 2,19a 1050‘m

Child

“demo 1 .47 1 .44 1 .97 1 .82 2.69'

Parent

W 0.92a 0.77a.b 1.58c 1 .27c 4.65“

Receiving

Violence In

Curran

“Mbmmp 1.368 0.373 1 1 .44b.c 2'89b,d 80.20."

 

Note. Within each row, means with different subscripts differ at the .05
level of significance according to a Tukey HSD procedure.

Depth of involvement in the current or most recent relationship was also a
significant predictor of group membership overall. The means of the
Coercive/Active, Positive/ Average, and Belligerent/Active groups were higher
(more deeply committed) than the mean of the lnexperienced group at the .05
level of significance.

Experiencing of abuse by parents was a significant predictor of group
membership overall, although there were no significant between group
differences. Also, witnessing violence between parents was a significant

3 7
predictor. According to the Tukey procedure, the means of the
Belligerent/Active and Coercive/Active groups were significantly higher (more
parental violence witnessed) (p<.05) than the mean of the lnexperienced
group.

Finally, being the recipient of abusive behavior in the dating relationship
was the strongest predictor of group membership. The mean of the
Coercive/Active group was significantly higher at the .05 level than all other
groups. Also, the mean of the Belligerent/Active group was significantly higher
than the means of the lnexperienced and Positive/Average groups at the .05
lavel.

W. In the overall ANOVA, the respondents in
the lnexperienced cluster had shorter relationships than respondents in all
other clusters. The respondents in the Coercive/Active and Belligerent/Active
clusters had longer duration relationships than respondents in the
Positive/Average group, although length of relationship did not differ between
the Coercive/Active and Belligerent/Active groups. Respondents in the
lnexperienced cluster also reported relationships that were less committed than
respondents in all other groups.

These respondents in the Belligerent/Active and Coercive/Active clusters
reported witnessing more parent violence than those in the lnexperienced
cluster. Finally, those in the Coercive/Active cluster reported receiving more
violence in the current or most recent relationship than those in all other
clusters. Also, those in the Belligerent/Active cluster reported more violence in
the current or most recent relationship than those in the Positive/Average and
the lnexperienced clusters.

W. Given that the variables selected as
hypothesized predictors of group membership were significant predictors of
group membership, a discriminant function analysis was employed to examine
which variables had the most power to discriminate between the four groups.
As with the oncway analysis of variance, since the group x sex interaction in
the multivariate analysis of variance was not significant, an overall analysis of

38

males and females combined was performed.

Of a maximum of three possible functions (in discriminant function

analysis, maximum number of functions-ngroups-1), two were significant, with
a combined x2 (15, N3592)-322.18, p<.001. After removal of the first function,
x2 (8, N-592) was 68.14 (p<.001). However, after removal of the second

function, x2 (3, N-592) was only 2.13, p>.50.

Table 12
Characteristics of Discriminant Functions Derived

 

 

 

Function I Function lI Function Ill

Eigenvalue .542 .1 19 .004
Canonical Correlation .593 .326 .060
Percent of Variance 81.53% 17.92% 0.55%
Wilks Lambda (After Removal) .577 .890 .996
Notes. N-592

Number of independent variables-5

Number of groups-4

The canonical correlations (RC) presented in Table 12 are correlations

between the entire set of predictors and the criteria of group classifications. The

eigenvalues (related to the canonical correlations by the formula 91 =- chi) and

their associated canonical correlations taken together reflect the relative ability

of each function to separate the groups. As can be seen, the canonical

39

correlations of the first two functions are moderately high, while the correlation
for function three is negligible, thus confirming the significance of the first two
functions and the relative nonlmportance of the third. Consequently, only the
first two functions (I and II) were employed for the remainder of the discriminant
analysis computations.

Table 13 presents the mean discriminant scores for each group on the
respective functions. The group centroids summarize the group locations in the
space defined by the discriminant functions. It appears that the first discriminant
function is discriminanting the Coercive/Active group from the other three. The
second discriminant function separates the Coercive/Active group from the
BelligerentlActive group and the Positive/Average group, but the discriminating
ability of this function Is somewhat less than that of the first function. Overall, the
dscnmination of the three nonviolent groups from each other is not as clear as
that of the violent group from these three.

 

 

 

Table 13
Mean Discriminant Functions Evaluated at Group Means
(Group Centroids)
Cluster Function 1 Function 2
Positive/Average -0.092 0.207
lnexperienced -0.654 -0.351
Coercive/Active 2.322 -0.706
BelligerentlActive 0.481 0.371
Note. N-592

Figure 2 (see Appendix H), a plot of the four group centroids on the two
discriminant functions, illustrates the separation of the four groups by the

40

dscriminant functions. This again shows that the first function serves primarily
to distinguish between the Coercive/Active group and the lnexperienced
group. The other two groups fall in the middle, and are not as clearly
distinguished, although the first function does discriminate them somewhat
from the Coercive/Active group. The second function does not serve to
cfiscriminate among the four groups, although it slightly discriminates the
Coercive/Active group form the BelligerentlActive and Positive/Average
groups. However, these-two groups (BelligerentlActive and Positive/Average)
are very close together on the second function. As can be seen geometrically,
the magnitude of discrimination of the second function is not as great as the
discriminanting ability of the first function and it seems that function one is the
primary discriminator.

WW Table 14 presents the standardized canonical
discriminant function coefficients, which represent the relative contributions of
the variables to the two functions. Receiving violence in a relationship seems
to make the greatest contribution to the first function. All of the other variables
appear to be of minor importance on this function, with the exception of length,
which makes a moderate contribution. Oh function two, receiving violence has
a high negative coefficient. Length of relationship makes the greatest positive
contribution to the second function. The other discriminating variables appear
to make negligible contributions to this function.

41

Table 14
Standardized Canonical Discriminant Function Coefficients

 

 

Variable Function 1 Function 2

Receiving violence

in current relationship 0.813 -0.582
Receiving parental

violence 0.059 0.072
Witnessing parental

violence 0.191 0.103
Length of current or

most recent relationship 0.466 0.869
Depth of current or

most recent relationship 0.000 0.006

 

The relationship between the individual discriminating variables and the
functions can be seen in Table 15, which presents the Pearson
product-moment correlations between the functions and the five discriminating
variables. According to Klecka (1980), the vector of correlations of variables
with a discriminant function, and not the vector of the partial discriminant
functions (as in Table 14), provides the most descriptive summary of the
behavioral correlates of the discriminant function.

42

 

Table 15

Correlations Between Discriminant Functions and

Discriminating Variables

Variable Function 1 Function 2

 

Receiving violence

in current relationship 0.863‘ 0.489
Witnessing parental

violence 0.1 96* 0.1 04
Receiving parental

violence 0.144' 0.058
Length of current or

most recent relationship 0.541 0.804’
Depth of current or

most recent relationship 0.291 0.319’

 

Note. Correlations marked with an asterisk in the first
column correlate more highly with Function 1 than with
Function 2; the reverse is true in the second column.

It is evident from these correlations that receiving violence in the current
relationship was most strongly related to function one. Since this function
served to discriminate the most violent group from the least violent group, the
results suggest that receiving violence in the relationship is associated with
perpetrating violence In the relationship, and is a useful variable in
discriminating violent from nonviolent groups. Further, this variable was
moderately and negatively correlated with the second discriminant function,
indicating that not receiving violence in the current relationship was associated
with this function. Since the second function most clearly discriminated

43

between the Coercive/Active group and the Positive/Average and
Belfigerent/Active groups, this would seem to indicate that these latter two
groups were those who report not receiving violence. However, in referring
back to the ANOVA, it is evident that this was not the case. Generally, it is
unclear how well function two discriminantes among any of the groups, and
interpretations made on the basis of group locations on this function must
obviously be made with some reservation.

Having witnessed parental violence during childhood and having been
the victim of abuse during childhood were also more highly correlated with the
first function, although these correlations were low, and not much higher than
the correlations of these variables with the second function. Thus, these two
history of violence variables were somewhat associated with the
Coercive/Active group, but this relationship was not strong.

Length of relationship was highly correlated with the second function, but
also had a strong correlation with function one as well. This may suggest that
those in the longest term relationships are not in the most violent group, while
being in a relationship of moderate duration is associated with violence (i.e.,
the moderate correlation with function one). The correlations of depth of
involvement and the two discriminant functions, while lower than those
betweeen the functions and length of relationship, indicate the same pattern,
with highly committed relationships being associated with the second function,
and relationships characterized by moderate levels of committment correlated
with the first function. However, these differences in correlations are not as
pronounced as those involving length of relationship.

W. As shown in Table 16, together, the two functions
correctly classified 44.7% of the most violent subjects (Coercive/Active cluster),
68.9% of the Positive/Average subjects, 59.8% of the lnexperienced subjects,
and only 9.6% of the BelligerentlActive cluster of subjects. The two discriminant '
functions correctly classified 51.9% of all cases correctly, which represents
reasonable improvement over chance classification (25%) of subjects into the

44

four groups. The discriminant functions incorrectly classified most of the
subjects from the BelligerentlActive group, erroneously predicting them as
members of the Positive/Average group. Also, the two functions classified less
than half of the subjects in the most violent group correctly. However, this is not
surprising when consideration is made of the inabilty of the second function to
discriminate these two groups at all, and the minor discrimination of these two
groups by the first function. The discriminant functions more accurately
classified the subjects in the two nonviolent groups.

Table 16
Discriminant Function Classification Results

 

 

Predicted Cluster
Membership
Actual Cluster No. of Cases PosJAv. Inexp. C/A BIA
Positive/Average 235 68.9% 26.8% 2.1% 6.4%
162 63 5 8
lnexperienced 194 39.9% 59.8% 0.0% 0.5%
77 1 16 0 1
Coercive/Active 38 28.9% 7.9% 44.7% 1 8.4%
1 1 3 17 7
BelligerentlActive 125 67.2% 16.8% 6.4% 9.6%
84 21 8 1 2

 

Note. Numbers under percentages are actual number of cases classified.

Discussion

Factor analyses of previous instruments measuring conflict resolution
behavior have yielded either separate physical and nonphysical abuse factors
(Hudson 8 McIntosh, 1981) or factors distinguishing between reasoning, verbal
aggression, and physical violence (Straus, 1979). These categories seemed to
represent a rather limited view of the great variety of possible conflict resolution
behaviors. The administration of the CRAO in this study revealed a wealth of
information concaming conflict resolution behavior in college students' dating
relationships. Specifically, it revealed that college students utilize a wider
variety of conflict resolution behaviors than simply physical violence, verbal
aggression, and reasoning in their dating relationships.

The factor analysis of the CRAO in the present study yielded seven
correlated factors, each representing distinct categories of conflict resolution
behavior. The first factor contains items that represent generally appropriate
conflict resolution behaviors. These include behaviors that represent positive
problem solving strategies (e.g., suggesting possible solutions, acknowledging
feelings about the problem, etc.) as well as more generic sorts of conflict
resolution behavior (e.g., arguing or becoming angry). Even though some of the
behaviors represented by this factor do not fall under the rubric of positive
problem solving (e.g., as discussed by Meichenbaum (1977)), none of the items
represent behavior that is physically coercive (i.e., an attempt to impose a
problem solution by the use of physical force or threat thereof). This factor is
similar to Straus' Reasoning Scale, but contains more items that represent a
broader range of positive problem solving behavior beyond just reasoning.

The second nonviolent factor, the Avoidance factor, contains, as the name
implies, items which represent behaviors that are an attempt to avoid the
conflict. Ignoring the problem, changing the topic of discussion, or leaving the
room are clearly ways of avoiding a conflict situation. This class of conflict

45

4 6
resolution behaviors was not included by Straus (1979) on the CTS or Hudson
8 McIntosh (1981) on the ISA, thus not taking into account the posilble
relationship between avoiding a conflict and using violence or any other
behavior during a conflict.

The items that comprised the Distraction factor were similar to those of the
Avoid factor, and certainly represented behaviors that could be utilized to avoid
conflict. However, they were different in that, for the most part, they represented
attempts to distract the other person by doing something that would divert
attention away from the conflict. This distracting behavior is generally
something which is nonaggressive, moving toward the other type of behavior
(e.g., becoming affectionate or making sexual advances). However, one of the
items, ”Went for walk, jogged or worked out“ did not seem to clearly fit in this
factor, and would seem to more appropriately go into the Avoid factor. However,
this item did not load on that factor. Respondents may not have viewed this item
as a way of directly avoiding conflict, but rather as a more positive behavior
designed to placate or distract the other person or possibly the actor
himself/herself.

The Verbal Attack/Coercion factor of the CRAO was made up of items
representative of behaviors that are verbally coercive, such as talking down to
the person, lecturing the person, and nagging. Although not physically
coercive, these behaviors are attempts to make the other person submit to the
will of the actor in a conflict by belittling and/or verbal badgering, and represent
what would be deemed verbal coercion

The Symbolic/Indirect Violence factor was the first of the three violence
factors of the CRAO. Although it only comprised four items of moderate to low
loadings, it appeared to be comprised of items that were distinctive enough to
warrant this being a factor unto itself. These items clearly represent violent
behavior that is directed at objects rather than a person. It is possible that in
some instances kicking an object or throwing an object may be an attempt to
frighten, shock, or intimidate the person into submitting in a conflict, or
performed out of the fnrstration aroused by the continuing failure to resolve the

4 7
conflict.

These two factors (Verbal Attack/Coercion and Symbolic Violence)
roughly parallel Straus' (1979) Verbal Aggression scale of the CTS. However, it
is clear from the present results that there were at least two unique domains of
behavior represented by these two factors, and not one verbal aggression
factor as Straus proposes. They also were similar to Hudson 8 McIntosh's
(1981) nonphysical abuse factor of their ISA Again, however, it is felt that the
two separate factors derived from the CRAO explain this dimension of conflict
resolution in greater detail. Verbal aggression (e.g., yelling at a person) and
kicking a chair are two very different behaviors that could very well result in
different conseuencas for those involved.

The Moderate Violence factor of the CRAO contained items that are
representative of moderately violent conflict resolution behaviors. These are
behaviors that might be employed to physically force the actor's will on another
during conflict. These behaviors could result in accidental or minor injury to the
recipient, although one of them, 'Choked or strangles him/her" seems to be
more serious than the the other items, since this behavior could result in death.
However, most of the behaviors represented by the items in this factor would

generally not be life threatening.

The Severe Violence factor included items that represent serious, life
threatening violence. These include the use of a weapon (e.g., a knife or gun),
or other behaviors (e.g., strangulation) that would clearly endanger the life of
the recipient.

These two violence factors, moderate violence and severe violence,
roughly correspond to Straus' violence and life-threatening violence scales
from the CTS. They also are similar to the physical abuse scale of the ISA.

The factor analysis performed by Straus (1979) utilized an orthogonal
rotation, in which the factors are assumed to be independent and unrelated.
Thus, he does not present interfactor correlations. Hudson 8 McIntosh (1981)
also report an orthogonal rotation of the factors form the ISA. The present factor
analysis of the CRAO utilized an oblique rotation, thus allowing for correlations

4 8

between factors. The correlations obtained indicated that all of the factors were
for the most part moderately correlated at a high level of statistical significance.

Specifically, these correlations suggest that those respondents who
reported utilizing appropriate conflict resolution behaviors also utilize verbal
attack behaviors characterized by items in that factor, as well as using
dstracting, symbolically violent, and avoidance behaviors. The
Positive/Appropriate factor was not related to the use of severe violence, and
only moderately with the use of moderate violence.

The use of verbal attack, symbolic violence, and moderate violence are
also moderately intercorrelated. However, Avoid is also moderately correlated
with Symbolic and Verbal Attack, but not with Moderate violence. Finally,
Severe violence is moderately correlated with Moderate violence, as might be
expected, but has a low to nonsignificant correlation with the other factors.

The overall pattern of moderate intercorrelations seemed to suggest that
respondents are reporting the use of widely diverse conflict resolution
behaviors, and generally tend not to utilize one or a few classes of behavior(s)
to the exclusion of others. However, factors that defined behaviors that are the
most divergent from each other, such as ApprOpriate/Generic and Severe
violence are not correlated. Further, factors that are not as different, but which
do not represent conceptually similar classes of behavior, for example, Severe
violence and Verbal Attack, are correlated, but only at a modest level. Thus,
those factors which represent similar, nonviolent forms of conflict resolution
behavior are most highly correlated. However, It is interesting to note that the
highest interfactor correlation is between the Appropriate/Generic and the
Verbal Attack factors. Perhaps the behaviors represented by the items
comprising these two factors are considered normative modes of conflict
resolution, and respondents do not characterize the Verbal Attack items as
'bad", or different from what have been labelled the appropriate, positive
behaviors.

After the derivation of the seven factors from the CRAO items, the next step
in the analysis was to cluster respondents according to their scores on

4 9
composite subscales derived from each of the factors. This procedure yielded
five clusters, or types of respondents. A typology of conflict resolution was thus
developed based upon these seven groups of conflict resolution behavior. This
particular type of analysis has not previously been reported in either the spouse
abuse or dating violence literatures.

The Ufa Threatener cluster consisted of only three subjects. These were
respondents who reported a great deal of life threatening violence. These
subjects had very high responses on the two violence factors, as well as
somewhat elevated scores on the Avoid and Symbolic violence subscales.
Otherwise, respondents in this cluster had scores on all other subscales that
were generally close to the mean. They obtained a very high score on the
Severe Wolence subscales, but their score on the Moderate Violence factor
was less than one standard deviation above the mean of that factor (however,
the standard deviation of that subscale for this group was actually greater than
the mean on the subscale for this group). Thus it seemed that these
respondents, while using some moderately violent and nonviolent coercive
strategies were primarily using a great deal of life threatening violence.

It is unclear whether these respondents were responding In a way to
mislead the investigator (a check of the validity scale of these protocols did not
indicate random responding, nor did an examination of the pattern of reponses
to the CRAO conflict behavior items), or if they were accurately reporting a great
deal of violence in their current or most recent relationship. There was some
doubt about the validity of these respondents' protocols, since these
respondents were reporting implausibly high levels of violence in their dating
relationships (e.g., reporting having shot their partner with a gun four to ten
times). Due to these doubts about the validity of these protocols the small
number of respondents in this cluster, these respondents were excluded from
further analyses. However, it is important to note that these very violent subjects
may have been honestly reporting their behavior, as other studies in dating
violence (e.g., Sigelman et al, 1984) have consistently noted a small group of
such highly coercive respondents.

5 O

The second cluster, which consisted of 40% (N-255) of the respondents,
was designated the Positive/Average cluster. Respondents in this cluster were
near the mean on all of the subscales. Also, they had the lowest score (along
with the lnexperienced group) on the severe violence factor, and the second
lowest on the moderate violence factor.

The proportion of males and females In this cluster was similar to the
proportion in the entire sample. Respondents in this cluster report receiving
very little violence in the current relationship, current or most recent
relationships of average length and depth, and average amounts (low) of
childhood violence. The discriminant function analysis was only able to clearly
ciscriminate this cluster from the most coercive cluster. This cluster, the most
populous, is felt to represent the typical respondent, who is using an average
amount of a wide range of conflict resolution strategies, but who does not
typically exhibit violence in a conflict.

The third cluster, the second largest, contains 32% (N-203) of the
respondents. It also has nearly equal proportions of males and females. This
group of respondents were designated the lnexperienced conflict resolvers.
Across all subscales they have scores on the predictors that are well below the
mean. This group of respondents reported the shortest and least committed
relationships. Thus, the low scores are most likely not indicative of passivity in
conflict (although in some individual cases this may be true), but rather reflect
the low number and intensity of conflicts experienced in a short, uncommitted
relationship. They also report receiving the least amount of violence in the
current or most recent relationship, which seems to further support the notion of
a low number and intensity of conflicts.

The fourth cluster, which contained 6% (N-39) of the respondents, was
named the Coercive/Active cluster. This was the smallest of the four clusters
retained for analysis. These respondents were above the mean on all the
subscales. They had the highest scores on the Appropriate, Verbal
Attack/Coercive, and Symbolic/Indirect Violence factors, as well as the highest
score (by several standard deviations) on the Moderate Violence subscale.

5 1
Also, of the four clusters retained for analysis, this cluster had the highest score
on the Severe Violence subscale. In general, these respondents seemed to be
utilizing every type of conflict resolution behavior with great frequency. Of
particular note, they were using a great deal of negative behaviors and
moderately violent behaviors but were also using many appropriate strategies,
and have not resorted to a great deal of severe as those in the Life Threatening
cluster.

This cluster consisted of 26 females and 14 males from the sample. This
dsproportionately high number of females in the Coercive/Active cluster was
an unexpected finding, to say the least. Other studies of dating violence (e.g.,
Laner 8 Thompson, 1982; Sigelman et al, 1984) have indicated that
approximately equal numbers of males and females perpetrate violent acts in
dating relationships. The high level of coercion reported by these women may
be acts of self-defense or in retaliation against male perpetrated coercion.
Matthews (1984) found that when violent acts occurred in a dating relationship,
39% of the males and 61% of the females indicated that they fought back in
response. This data Indicates that there may be some validity to the idea that
when females are assaulted, they tend to fight back the majority of the time.
However, the present data does not allow a determination of a causal
sequence, and one could just as easily say that these respondents receive
coercion because they perpetrate it first. The clinical case literature on spouse
abuse Indicates that in the majority of cases where women use violence it is in
self-defense or in retaliation against male initiated assaults (Walker, 1979;
1984). However, much of this literature is based on abused women reporting
on their husbands In therapy. Past findings in the area of dating violence have
clearly demonstrated that the perpetraticn of violent acts in dating relationships
is reciprocal (Sigelman at al, 1984; Lane 8 Gwartney-Gibbs, 1985). In the
present study this notion is further supported, with receiving violence in the
current relationship being the most powerful predictor of perpetrating violence.
However, the factors that might contribute to a dating couple becoming mutually
coercive are unclear due to the limitations inherent in survey research.

5 2

The data in the present study indicated that respondents in this cluster
reported being in somewhat longer and more committed relationships than
those in other clusters. This finding is consistent with findings by Laner 8
Thompson (1982) and Sigelman et al (1984) where subjects in longer and
more serious relationships reported more violence in the relationship. Being in
a more serious and ongoing relationship may bring more long standing,
serious conflict which could potetially lead to more extreme forms of conflict
resolution behaviors. Also, given the deeper level of commitment, there may be
a greater pull to maintain the relationship regardless of the amount of conflict
and resultant coercion.

Respondents in the Coercive/Active cluster reported having witnessed
more violence between parents than those in the two nonviolent clusters,
althogh there is no statistically significant difference in the amount of abuse
received as a child, although the mean for this group is higher than the means
of the other three on this predictor. A general association between violence in
the dating relationship and experiencing violence in the family of origin is
consistent with findings in almost all studies testing this hypothesis in both the
spouse abuse literature (e.g., Straus at al, 1980) and the dating violence
literature (e.g., Laner 8 Thompson, 1982; Bernard 8 Bernard, 1983).

The discriminant functions distinguished this cluster most clearly from the
other clusters. However, the discriminant functions correctly classified only 45%
of the respondents into the Coercive/Active cluster, while incorrectly classifying
29% of the respondents in this cluster into the Positive/Average cluster, and
18% of the respondents into the BelligerentlActive cluster. This may be
indicating that some of the respondents in this cluster are not as coercive, and
based on the discriminating variables they cannot be accurately classified into
this cluster and are placed into a less coercive or noncoercive cluster.

In general, the respondents in the Coercive/Active cluster reported utilizing
a great number of varying conflict resolution strategies. This may simply be the
result of a higher level of conflict in their dating relationships. Unfortunately, it
was beyond the scope of the present study to assess the level of conflict in the

5 3
respondents' current or most recent relationship.

The fifth cluster was designated the BelligerentlActive cluster.
Respondents in this cluster were well above the mean on all factors, except the
violence factors, on which they were near the mean. On the nonviolent factors,
they were very similar to the Coercive/Active group, except they scored slightly
lower on the Verbal Attack subscale than that group. This cluster consisted of
135 respondents (21% of the respondents), with nearly equal numbers of
females and males.

The respondents In this cluster reported receiving more violence than
members of the Positive/Average and lnexperienced clusters. Respondents in
this cluster reported slightly longer duration and slightly more committed
relationships than those respondents in the Positive/Average and
lnexperienced clusters as well. They generally did not report receiving more
violence in childhood, and although not statistically significant, they reported
witnesssing more violence in childhood. Thus, repondents in this cluster seem
to represent individuals who utilize many strategies to resolve conflict, as with
the Coercive/Active cluster, but not as many violent or negative strategies.

The cfiscriminant functions were not able to clearly distinguish this group
from the Coercive/Active cluster, and placed it somewhere between this group
and the other two noncoercive groups. Notably, the discriminant functions were
able to correctly classify only 10% of the respondents into this cluster. Sixty
seven percent of the respondents in this cluster were incorrectly classified into
the Positive/Average cluster, and 17% were misclassified into the
lnexperienced cluster. It seems that the chosen discriminating variables were
not able to distinguish these respondents from those in the Positive/Average
cluster. It may be that these variables were good at distinguishing extremes
(i.e., coercive from appropriate), but they may not be able to distinguish this
noncoercive, yet negative and aggressive group.

The discriminant function analysis attempted to utilize a set of variables to
distinguish between four of the clusters derived. The first discriminant function
was able to distinguish the Coercive/Active cluster from the other three. The

54
second discriminant function generally did not serve to clearly discriminate the
groups, although it did discriminate slightly the Coercive/Active group from the
BelligerentlActive and Positive/Average groups.

Receiving violence in the current or most recent relationship was most
highly correlated with function one. This correlation with the first function is
consistent with this variable as a very strong predictor of coercion as
demonstrated by the ANOVA, since function one is most clearly associated
with the Coercive/Active group.

Length of relationship and depth of relationship were correlated with the
first function, but these two variables had their highest correlation with the
second function. This, along with the ANOVA for length and depth may suggest
that the relationship between length and depth and coercion is not linear, but
rather an inverted 'U' relationship, where those in the shortest, least committed
and longest, most committed relationships do not report as much violence as
those in relationships of moderate length and depth. Violent dating
relationships may continue to a point, but with high levels of conflict and
violence may not last as long as relationships without violence. Thus,those
relationships of greatest length and duration are generally not violent.

Witnessing parental violence and receiving violence as a child were more
highly correlated with the first function than with the second. However, these
correlations are low, and do not differ greatly from the correlations with the
second function. However, the relationship between childhood violence and
the four groups as yielded by the discriminant analysis is somewhat ambiguous
at best. Sigelman et al (1984), in their discriminant analysis of abusive and
nonabusive college students, found that being abused as a child discriminated
violent females from nonviolent ones, but did not discriminate males. Their
results were also somewhat weak as in the present study. Also, Stats 8
Pirog-Good (1987) did not find strong support for the idea that violence is
intergenerational. However, in the present study these two variables were
predictive overall of group status according the the ANOVA, but did have the
least predictive power. It seems that history of childhood violence is somewhat

5 5
predictive of coercion in dating relationships. However, this historical variable
clearly does not have the predictive power of variables associated with the
current relationship. It may be that events in childhood simply may not have as
great an effect on current behavior, and there may be more variability in the
effect of these more temporally distal variables. Thus, being abused as a child
certainly makes one more likely to be an abusive dating partner, but as Straus
at al (1980) report, nearly half of the abusive respondents in their sample had
not experienced childhood violence.

Overall, the discriminant function analysis most strongly indicated that
receiving violence in the current or most recent relationship is the most highly
dscriminating variable of the four clusters. This seems to indicate, along with
the highly significant ANOVA results concaming this variable, that coercion in
dating relationships is reciprocal. When one partner is violent, the other tends
to strike back Researchers in marital violence (e.g., Steinmetz, 1980; Straus, et
al. 1980) have found clear indications that violence in marriages is reciprocal,
as have investigators in the dating violence field (e.g., Sigelman, at al., 1984;
Lane 8 Gwartney-Glbbs, 1985).

Although these studies, as well as the present study, demonstrate the
mutuality of such violence, there has been little but speculation concaming the
development and maintainence of mutually coercive behaviors in the context of
intimate, loving relationships. Due to the limitations of the survey method - the
predominant research methodology (present study included) - little is known
concaming the development and dynamics of violent behavior between
intimates. Initial investigations in the field of marital violence, primarily by
feminist oriented writers, tended to blame the male dominated social and
political hierarchy for male violence against women (e.g., Dobash 8 Dobash,
1979). However, it is now apparent that violence is a mutual event, albeit some
of the violence perpetrated by women Is certainly in self-defense against male
initiated acts (Straus et al, 1980).

Further, studies of dating violence to date have primarily gathered
demographic data as predictors of violence, and have not examined more

5 6
psychological and/or personality variables that may be related to conflict
resolution in dating relationships. Stats 8 Pirog-Good (1987) found that the
personality dimensions of instrumentality and expressiveness influence
receiving violence in dating relationships. However, these two dimensions did
not influence the use of violence in relationships. Future studies should
examine other personality characteristics of individuals in dating relationships
that may predispose them to both use and receive violence in resolving conflict.
Concepts such as mood, agoocontrol and resiliency, internal versus external
locus of control, and others need to be investigated as predictors of dating
violence so that more focused theoretical formulations concaming violence and
conflict resolution in general may be made.

Studies, to date, have only begun to provide the necessary information to
understand the specific causal events within a relationship. But we have even
less knowledge of events within the violent encounter itself that lead to violent
behavior. Clearly, violent behavior within a relationship is a symptom of a
dysfunctional system, where the man's experiencing might be similar to the
woman's (although his behavior might often be the more dangerous of the two).
Thus, couples, together, need to be a target of study, along with continued
surveys of individuals. Further, it is believed that the continued study of dating
couples rather than married couples would be beneficial. Dating couples are
just beginning to form relationships, and early patterns that may lead to a
lifetime of abuse, which might be obscured in couples who have been together
a long time, may be uncovered.

The present study determined that there are several different types of
behaviors utilized to resolve conflict, and that these behaviors have more
detailed texture than simply violent versus nonviolent. A more intensive study of
a smaller number of couples, perhaps selected on the basis of a mass
screening with an instrument such as the CRAO, is essential. Also, the data
from the present study clearly Indicated that coercive conflict resolution
behavior is reciprocal, and in large part dependent on the amount of time the
couple has been together.

5 7

In future research, young dating couples (ideally high school or college
couples) who have been together for varying lengths of time should be
interviewed concaming experiences in their families of origin, the history of
their dating relationships, and about the functioning of their current relationship.
One historical variable that may be of particular interest in the context of conflict
resolution behavior would be that of attachment. Hazan 8 Shaver (1987) have
demonstrated that individuals' perceptions of their relationship with a romantic
partner can be categorized into styles of attachment similar to those of
mother-infant attachment conceptualized by Ainsworth at al (1978). One might
hypothesize that dating couples who experience violent and other maladaptive
conflict resolution behaviors in their relationship may be insecurely attached in
their relationship. Also, gathering information about each partners own
attachment history by administering an attachment interview (George, Kaplan 8
Main, 1984) may reveal interesting differences between coercive and
noncoercive couples.

lnforrnation gathered about the current relationship should include some
measure of the amount of conflict in the relationship, as well as information
concaming what the couple does when they are together on a date (i.e., do
they go to movies, stay at home and play board games, how much do they drink
alcohol, etc.) and what they do when they are apart (e.g., how often and for how
long do they talk on the phone, do they have outside interests, etc.). Further
work might include having couples role play and resolve a conflict, real or
hypothetical, in a laboratory setting and coding the interaction using a
behavioral code such as developed by Gottman (1979), and examining
differences In behavior between coercive and noncoercive couples.

APPENDICES

APPENDIX A

Conflict Resolution and Abuse Questionnaire (CRAO)

APPENDIX A

Conflict Resolution and Abuse Questionnaire (CRAQ)

Please read this sheet before completing the questionnaire.

As stated In class this questionnaire deals with the topic of behavior
between persons who were or are dating, engaged, or married. The questions
which follow all address behaviors which have been reported or described by
men and women who are involved with each other. While each of the questions
refer to activities or actions which have occurred between couples, they may
not have occurred between you and your partner. Please answer each question
as honestly as possible. Many of the questions concern tepics that are
sensitive and personal and several of the questions concern actions that are
illegal. Because of the nature of the questions, it is critically important
that you make 59 identifying marks anywhere on this questionnaire. Doing
this gives you the anonymity which guarantees that you can never be linked
to your truthful responses to the items on this questionnaire. If, in the
course of completing this questionnaire, you come to a question which you
find to be objectionable for some reason. please feel free to skip that
question and go on to the next one. If you do not want to complete the
questionnaire. regardless of reason. Just return it as described below.

If you have questions about the items on this questionnaire. please feel
free to contact Dr. Gary Stollak, 353-8877. If, after completing this
questionnaire (or at any other time). you believe that you have committed
or been the victim of illegal behavior, it is important that you contact one
of the following persons or agencies listed below. If. after completing
this questionnaire, you wish to talk about your feelings, thoughts or
experiences, you may call any of the persons and/or organizations listed
below. There are no fees involved in these consultations and. in each

case. nobody will ask your name or any other identifying information.

Dr. Gary Stollak, Licensed Clinical Psychologist - 353-8877.
Council Against Domestic Assault - 372-5572.

Please return the questionnaire whether you have completed it or not.
according to the instructions on the attached sheet (page 1a).

He estimate that it will take about one hour to complete the questionnaire.

Reading these statements and completing the following questionnaire shall
constitute your informed consent.

59

Pleaee anewer the following queetione about youreelf.

ralationehipa.

your peat

and your current or aoet recent relationehip with

your boy/girl friend or. if earried. your huehend/wiie by aarking
your reeponeee on theee pagee.

Ieportant: So that your privacy can he protected pleeee do not write
your neae on thie queetionnaire.

1)

2)

3)

4)

5)

h)

7)

8)

9)

Uhat ie your age?

what ie your een (circle one)? u
what ie your ethnic origin (circle one)?

white Ilach fliepenic/Latino eeian other

How aany yeare of age were you when you had
your firet real date?

How aany people have you been out with
on eora than one date?

How aany relationehipe have you ever been involved
laeted (indicate nuaher next to each):

leee than a aonth
to 2 aenthe

to 6 aonthe

to 12 aonthe
to 2 yeare

to 3 yeare

or eora yeare

Oflt‘flfli‘

flow aany relationehipe have you ever been involved in
(indicate nueher next to each):

dated nonexclueively
dated exclueively
diecueaed earriage
were engaged

lived together

were aerried

in which

Check-each of the activitiee that you have ever engaged in:

holding hande
kieeing on eouth
petting ahove waiet
petting below waiet
eexuel intercouree

How aany partnere have you had eexual intercouree with?

(Ouaetione 10 to 15 are on the back of the page)

60

10) Are you currently involved in a relationahip (check one)?
Yea
lo

11) low long have you been in your current relationehip (or how long
were you in your eoat recent reletionehip) (check one)?

leee than a aonth -----

1 to 2 aonthe --..-
3 to 6 aonthe __..-
7 to 12 aonthe -..--
1 to 2 yeare ..---
3 to 3 yeare ....-
6 or aore yeare .....

12) which of the following phreeee beet deecribee your current or
aoet recent relationehip (check one)?

. dating nonexclueively

dating exclueively ---..
diecueeing earriege ..---
engage-ant -----
living together -----
aerriage -..--

13) Check each of the ectivitiae that you have engaged in in your
current or aoet recent reletionehip:

holding hande ..-..
kieeing on aouth -----
petting above waiet --..-
petting below waiet -__--
eeeual interc'ouree ...-..

14) How often hae your current or aoet recent boy/girl friend or.
if harried. huebendlwife been drunk or high on druga in your
preaence while you were eober?

lever

1 to 4 tieee

5 to 10 tiaee

11 to 20 tieee
21 or aore tiaee

15) How often were you drunk or high on druga in the preaence of
your current or eoet recent boy/girl friend or. if aerried.
hueband/wifa while helehe wee eober?

lever

1 to a tiaee

S to 10 tiaee

11 to 20 tiaee
21 or aore tinee

(Pleaee go on to next page)

a"

61

16) Iow often were you both drunk or high on druga together?

Never

1 to e tiaee

5 to 10 tieee

11 to 20 tieea
21 or eore tiaee

--- “

Pleaea enawer the following gueetione concerning the parent(e),
etep-perentte). or guerdianCe) that you lived with before coaing to
college or while you were a teenager.

17) The parent(a). etep-parentte). or guerdian(e) that you lived
with before coaing to college or while you were a
teenager wea(were) (check one):

your biological nether and biological father
your biological nether and etep father

your biological father and atep aother

your anther only

your father only

your relativee or footer/adoptive parente

18) Education of your nether/feeele parent (check one):

gueetion not applicable
leee than eigth grade
eoee high echool

high echool graduate
aeae college

college graduate
graduate or profeeeional

19) Occupation of aother/feaale parent (check one):

gueetion not applicable
hoaeaeker

blue collar

white collar
profeeeional

uneaployed _-----_

20) If your aother/feaale parent worke outeide the
hone. doee ehe work (check one):

gueetion not applicable
pert tine
full tine

21) Age of aotherlfaaale parent (if applicable):

(Oueetione 22 to 27 are on the back of eheet)

(z ‘

62

22) Education of your father/hale parent (check one):

gueetion not applicable
leee than eigth grade
acne high achool

high achool graduate
none college

college graduate
graduate or profeeeional

 

23) Occupation of fatherlnale parent (check one):

gueation not applicable
honenaker

blue collar

white collar
profeeeional

unenployed

 

 

2e) if your father/nala parent worke outeide the bone, doee he
work (check one):

gueetion not applicable __..__
part tine ......
full tine _...._

25) Age of father/nale parent (if applicable):

 

26) Did you ever nee your parent(e) (Check each itea that appliee).

Threaten to hit or throw eonething at each other.
Throw. hit. kick. or aneeh eonething.

Throw eonething at each other.

Pueh. grab. or above each other.

Slap each other.

hick. bite. or hit the other.

heat up the other. .

Threaten each other with a gun or knife.

Can a gun or knife on each other.

27) Did your parente ever do any of the following thinge
to you (check each itea that appliee):

Threaten to hit you or throw eonething at you.
Throw.aoaething at you.

Pueh. grab. or above you.

Spank you

Slap you.

lick. bite. or hit you.

beat you up.

Threaten you with a gun or knife.

“no a gun or knife on you.

(Pleaee go on to the next page)

63

MutWadchciflermtitisbmnmplidhym,murmtwmost
rant boy/girlfrinid. tr yor current a- mat recent relationship. If it is true as mpliad toyou,
mornirmatremtbw/girlfriaud.rmnrrmtrmostrmt relationship circlethe
kit: 1’. If it isialaenuipliaa toyw.yororrcitor mat mow/girlfrimdmrm
orrmtreutmutnlatWip.circktlwlattr F. liywnnotreportimonanrrent
“Wimmtlunltunsasmwmlmthavemplladdrimvnrnmtrantrelatiomhip.
oyou migct have felt while that relationship we still wing on.

i F l.MutimanwMy/girliriedmnmianMmet-Mpy. '

T F zmreldinmiphmtaprfctnm

T F 3.nyboylgirlirmhasalldtlugnlltinl havealweyswmta) lna boy/girlirlem
I F 4.linybo/Igrlirieldhuelyfnaltslntuawoedtlun.

I F 5. fly boy/girlfriuideidl Wash othw' coupletely.

T F QWeandlaqinMnmytwopn-minthiswrldmbe

1’ F 7.Iheve mmmnmwmmwmy alumna

T F 8.!va'ymihmglhevewwnybo/lwliriudhnplum

T F ulna-acetinnwhuilm miniag-Itmlddictimio-myboy/girliriad

T F iOJdlI't ihirtmecouldpuiblybehmpio'timmybw/girlfrinidaulwm
werewiihonemothu‘.

T F ILHYNHmiDMHNMDWUmitB.

T F IZIM'imlembmflhmwimmmMm
my/qirlfrwml.

I F l3. Whey/girlfriuumpldeiymadmpathiu with myeverymoni
T F lilhwemm'ettaimvrelatmeutmia‘amt

 

T F is.limp-minthewrlddtluoppnsitaaMbmavailablamdwllllmtnbe
MIWW|DWlmM|WU|u WMOWU'DU. "

‘IIIIllIIIlllllllllllllllllllllll

 

 

DO 3 is NOT WRITE IN THIS AREA

 

®O®®®©©O® 9
lo nutter how well a couple gate along. there are tinee OO©®®®OO®®
when they diaagree on eager or ninor nattere. get annoyed
about eonething the other pereon hae done. or Juet have OO®®®©®O®®
epata or fighte becauae they are in a bad need or are
tired. or for none other reaeen. Sena couplee have aany OO®®®®®O®®
different waye of eettling their differencen. while other
couplee have only a few waye of eettling their differencen. ®O®®®©QO®Q
we are intereeted in how frequently [9! have uaed each of
the following nethede te renolve cenflicta between 199m OOQ®®®®®®®
and your 9513.“! or [931 um boy/girl friend or. if
canied. your huaband/wife. Pleeee one the following ecale OOQGGQGOGQ
to indicate how often you have uaed each of the following
netbede of confronting conflicta with hie/her by blackening ®O®®O®®O®®
the apppropriate circle after each itea.

©0®®®©®®®O

if you find that a guention ie objectionable. leave it
blank and go on to the neat gueetien. ®®®©®©©O©®
Scale: OQCDGDGOOOCDG

o-never 2-dto10tinee 4-21orneretinee
1 I 1 to 3 tinee I I 11 to 20 tinee OOQGGQOOGQ
low often have you . . . ®O®®®OOO©©
1) Defanded youreelf verbally. . . . . . . OO©®®OOOOO
2) Ignored the problan. . . . . . . . . 0060900000
2) naked another pereen to help nettle thinge. . . 00609600636)
d) land alcohol or druga to nettle conflict. . . ©G®®®O®O®®
5) Gotten and at eonebody elee heceuae of conflict. ®®©®®©®O®®
a) Talked to a friend about the problen. . . , . . OO©®®©@O®O
1) Frayed with bin/her or alone. . . . . . . ®O©@®®@O®®
0) bone or aaid eonething to nake bin/her laugh. . . o©©®®®®0®®
9) lode annual advancee toward hie/her. . . . . OQ©®®®®®®®
10) heat for walk. Jagged or worked out. . . . . OO©©®©®O®®
11) Put off talking about the conflict. . . . . ooe®®®©®®®
12> ‘Talked down“ to hie/her. . . . . . . 0093930090
13) lagged at hie/her. . . . . . . . . 0090903000
14) Done eonething to hin/her that left a bruiae. 0900903330
15) Oenied that there wae a problen. . . . . . OOQOGOQ"OC
16) Slapped hinlher on face. . . . . . . . COOL, 130.3300
17) Deetreyed er danaged an object belonging to hie/her. 95:3: : 3 :33;-

‘Q -oa-oc CO .q 0. 9.. Q. 55‘ h-~b .l Oh. oh--.\

 

 

I
AA a. AAAAAA I
(‘in A .45.‘. 'A ..

 

 

65

 

J:-

II».

.- .--—- - - —A_.--A.. ~—
\

4 Javu vavv

ll .)

(l

i)

l
)
l
l
)
)

‘qA—‘§A

‘ ‘ ‘ g g

‘ ‘ ‘ A A

a v .1 v ‘d

‘ 4 cf ‘ J ‘1

in)

‘1’)

20)
21)

33)

" 20

as:
as)
27)
2e)
as)
an)
an

f2)
33

34)
35)
3‘)
37)

3.)
3’)

d1)
‘2)
O3)
44)

Q8)

 

 

Called hie/her bod nanote).

Stoopod foot.

lode or given an enouao for what happened.
Left the reenlhouae.

Infueed to talk about it.

Iorgainod or negotiated with bin/her.
Changed the topic.

lurt bin/her aeverely enough to require
attention.

Lectured bin/her.

Ilaoed hie/her for problen.

Apologinod to hie/her.

Sowoened hie/bar painfully e.g.. their wriet or are.

Slanned the deer.
Sulhod or peuted.
Shot bin/her with a gun.

) Threotoned to phynically horn bin/her.

Argued with hie/her.

Agreed that there won a problen.

lode fun of or ridiculed hie/her.
linicked or nechod bin/her.

Spohen or acted eorcaetically to hin/har.
Seothad or calned hie/her.

Stabbed or cut hie/her with a knife.
Igged hie/her on.

lecane affectionate towarda hie/her.
noted confuood 1....
won wrong.

lecone very logical
Coiled polioo.

‘91..“ ,0 we ‘0 a. ”on. p-.-)

nodica

acted like you didn‘t know wt

a.-umun-uuwuwuwuuuuuuuuuuuuuuuuuuuuuuuuuuuwun‘uwuu

 

DO j i NOT WRITE IN

THIS AREA

 

 

 

 

QGGOGOOOGC

'O” 3 OOGGGGGQGC

ea) lade a pernenal verbal attack agninet hie/her. QQQQGOQQCDG
d7) Telled or ncrenned at hie/her. . . . . . QQQQQQQQGC
66) Threatened bin/her with a knife. . . . . . @QQQQQQOGC
49) Suggeeted pooeible nolutionn . . . . . QQQQQQOOQC
50) naked bin/h. to help find nolutionn. . . . . QOQG QOOOOC
51) brought up hie/her poet nhortconingn or faulte. QQQGGQOOQC
$2) Forgiven hie/her. . . . . . . . . . QQQQQQQQQC
31) Acknowledged hie/her feelinga about the problen. QOQGCCOCOC
ed) Ordered hin/hnr around. . . . . . . . QQQQQQFQQC
SS) Choked or etraagled bin/her. . . . . . . QQQQQQQ COG;
as) licked an object (not hie/her). . . . . . QOQQQQQ QCE
S?) licked bin/her. . . . . . . . . . QQQQQQQ:3.:f
an) bitten bin/her. . . . . . . . . . 000.33.33.33:
SS) Pulled hie/her hair. . . . . . . . . QQQQQQQQQC?
60) Slapped hinlhnr on body. . . . . . . . OOGOQCQCQC‘:
61) lit bin/her with your fiat. . . . . . . QQQQQQQQQC
62) Thrown an object at hie/her. . . . . . . QQQQQQQQQQ
63) Thrown an object but not at hie/her. . . . . QQQQQQQQ QQ I
M) Thrown bin/her. . . . . . . . . . QQQQQQQQQQ
6S) lit hie/her with an ebyect. . . . . . . QQQQQQQ QQ :
66) lecane angry at bin/her. . . . . . . . OQQQQQQQQQ
67) Threatened bin/her with a gun. . . . . . OOCJOGCC QQQ
The next not of guentionn deal with behaviorn that night OCCCCC : C 3:?

occur between two people who are clean to one another. we
are intereeted in how frequently you have done each of the
following thinge to your snugly: or 5931 flflgfilz boy/girl
friend or. if narried. your huebnnd/wifa. Again. uao the
following acnla to indicate how often you have done each of
the following to bin/her by blackening the apppreprinte
circle.

Scale:
o-never 2-dte10tinee
loitoatinee 3-11te20tinee

a I 21 or nere tinee

(0went1one6lte’5areonthebnckoftheehaot)

. — A ‘ .- '

CAAA’
O \Jv‘vawvdv

 

,a‘,‘-\A«A.\A
CCV'V-rvvw-tv
AAA" «««««
oL‘p/Kd\;wa\rw/o‘
I
O fifl/‘AAAA‘,
l
V‘v4v\../Vv
AA ..........
‘VKVV ........
AAA‘A AAAAAA

 

 

 

 

-- —_..- --- .-

443“). an Aflmudlwd Action (Que! Ummllumly Irunhlmn

C)

C)

(J
C)
C

C)
C)

()
()

C)
)
C)
i)
(9 Q)
3
C)

(I

i)
C)
C)
C)
C)
C)
C)

C) C)
(D C) ‘

l)

C)

C)

(

C)

C)

C)
C)
i)
C)
C

C)
()

C)
)
(D
)

()
C)
k
C)
(

() (3

() C) C)

> ( (,

i) ()
C)
()

() C)

() C)

C)
C)

C) (J C) (D C)

i)

C)

(I)

i
i)
C

C)
()

l)
C)

)
l
)
l
t
)
l
)

v-r Va v v .v v v ‘1

l
3
)
l
)
l )
()

C) C) (3 C) C) C) C) (D (D
C) C) C) C) (D C) C) C) C) C)

()
C)

C)
C)
()

C)
C)
C) C)
C)

C) C) (J C)
l) C) C)

(j

l)

C)

C) C)

C) C)
‘ C)

C)
C) C)
()
C) (D C)

C) C)

l) C) l)
C)
(J C)

C) C)

l)
l)

i)
l)

(3
C)
C)

C)

(J
l)
C)

l )
l)

)
l
)
l
)
)
')
)

AA\\-\- \A

00000

C) C) C) (3 C) (3 C) C)

(3

(3
C
(J (
t) C
l) (,
C) C
() C)
- C)
C) (3 C)
C) C) (

C)

G)

(3

()

 

 

’1)

67_

lnbnrraeeod hie/her in public.

Spoken well of hin/hor in front of othern.

Told cruel or neon Jekee about hie/her in their
preaence.
Gotten upnot when holnhe went out with none eex

{time
Tried to tell bin/her who hie/her friendn nhould be.

Ootten upeet when he/ehe went out with oppoeitn nex
friende.

benandod that he/ahe account for hie/her eotivitioe anc
:tlzely acounod hie/her of having been out with another
:::::=; acounod bin/her having one with eenoena elee.
Conplained of hie/her flirting with othern.

lecueod hie/her of keeping nocrete.

Looked through hie/her pernonal belengingn.

Spied on bin/her.

bellied hinlher.

Onnondod hie/her obedience to your whine.

been )enleue or eunpicioun of hie/her friende.

Told bin/her that he/ehe can‘t get along without you.
Oenanded that hnlnhe nerve/wait on you.

Told bin/her that helnhe doenn't really underntand what
the world in like.

Oenanded that helehe not necinlixo with

friende.
Ordered hie/her around.

hie/her

Told hie/her that holnhe in attractive.

Told bin/her that he/abe in fun to be with.

Told bin/her that he/nha in ntupid.

Laughed at hie/her feelingn.

Told bin/her that you couldn't live without bin/her.
Oenplinonted hinlhor on loveneking abilitioo.

initiated dincueeien of annual feelinge and/or behavior

(Pl-one an on to the port none)

V.’

“M-A‘AMUMM“M‘A‘A‘A‘A‘A‘A‘A‘A‘A“‘A‘A“‘A“‘A‘A‘A‘AA“ A“““AA“‘A“A‘A““A‘AA“‘A“““A““‘AA‘AA‘A‘A‘“““A“- -- ‘

 

 

'Il

 

DO i - NOT WRITE

68

IN THIS AREA

 

101)
102)
103)
104)
ion)
106)
107)
ion)
ioe)
iio)
111)
112)
113)
116)
11S)
116)
111)
11a)
11’)
120)
12))
122)

123)

Page S

Told hie/her that hie/her lovenaking wen worao than
that of othern.

Told bin/her that you wanted to have aax with eenoena
elee.

Called bin/her 'loeao'. 'pronincoue“. or ”aany“. .

forced bin/her into annual activity in epite of hie/her
pretantn.

been unable to tell bin/her that annual activity in
undenired.

Touched or fondled hie/her genitale in epite of
hie/her pretantn.

Talked bin/her into annual intarceurao in epite of
hie/her pretantn.

Phynically forced hie/her to have annual interceurne.

conplinnotnd hie/her on appearance. . . . .

Told hie/her what in exciting or arouaing about
bin/her.
Told bin/her that annual feelinge or eotivitioe are
'dirty'.
Started a dincueeien about contraception with hie/her.

Threatened to leave hie/her if annual activity wan not
forthconing.

land annual devionn in epite of hie/her pretantn.

Told hie/her that halnha won a bad lover. . .

Threatened to phyeicnlly horn hie/her if annual
activity won not forthconing.

forced bin/her to act out behavior portrayed in
pornography.
Told hie/her that helahe wan ugly. . . . .

leld or cuddled bin/her when naked to do no. . .

lad intorcourno without contraception in epite of
hie/her pretantn.
Unfavorably coapnred hie/her to other non/wenon.

Told bin/her that holnhe in reepeneible for planning
you.

Oennnded'aexual activity no part of the relationehip.
Tried to aoduco othnrn in front of hie/her. . .
lecone eullen or withdrawn if annual activity wan not
forthconing.

Touched hie/her in public in epite of hie/her
pretantn. '

lode hie/her look at pornographic natarialn in epite
of hie/her pretantn.
Kieead hie/her in epite of hie/her pretantn. . .

(Plneaa go to the back of the page for the next part)

 

OOOOGOGOGC
OOQGGQGOGC
OO©®®OOO©C
OOQGGOCOOC
QOQQGGOOOC
OOQQGOOOOC
OOQGQOQOGC
GOGGQOOOOC
OOQGQCGOGC
OOGGGOQOQC
00096“ OCOOC

 

OOQOQOOOJC
OOOGGOOOCC
OOOGQOOOO:
OOQCGQGOOC

 

AAAAAA

......
v d'v" to v

 

‘75 U I; an Aflumdlwr Achon (Qua! Opportunity Inummun

.-I. Pill-m),
.5-

'-'.I'r' Ull'i I"

A, "I ll

 

\

C)

 

()
C)
C)
C)
C)
C)
()
()

a-

 

.- an---

')

e- —-a

C)

08
C) C)
C) C)

C)
C)

) C) C) C) C) C)
C) C) C) C) C) C)
C)

C)

C)
C)

C)
I)

O

C) C) C) C)
C) C) (D C)

()
C)

(3

U.

(DC
0000(

C) C)
C) C)
. (J
” ()

l

C)
C)
C)

C)
C)

C)
C)
C)
()
C) -
C)
C) C)

C) C)

()
C)
C) C)
C) C)

C)

C)
C)
C)
C) C)

C)

C)

C) C)
C)
C) C) C) -
C) C) C) (D C)

C) (D C) C) C)

C)
C) C) C) C) C)
(J C) C) C) (D C) C) C) C) C) C)

C)

C)

C)

- )

C)
C)
C)
Q
C)
C) C)

(D C) C) C) C) C) C)

\
( ) I
C) C) C) C) I) I
C) C CI C
(D C) C l) C
h N C) I
C) C
C) C
C)

(7
C)
C) C)
C)
‘ C)
C)
C)
C)

C)

C) C)
()
C)
C)
C)
' C)
‘ C)
C)
C)
C)

C)
C)

C)
C)
i)
C)
C)
C)
C)

K I
i
I"
I
l
(

‘a-‘a‘

\
C
C
C
(
C
C
C
C)
C)

C)
)
)
l
I
I
l
)

J
l

l
I
I
I

‘ - .1 - a. s

 

  

IDS!
huaband/wife haa

renolve conflictn between

the

each of the following nethede of confronting conflictn wit)
you by blackening the apppreprinte circle after each itea.
If you are not aura of hie/her behavior.
beet entinnte.

Scale:
O I never 2 I d to 10 tinee d I 21 or none tine.
1 I 1 to S tinee S I 11 to 20 tinee

I00
1)
2)
I)
6)
5)
6)
7)
I)
S)

10)

11)

12)
13)
1C)
15)
16)
17)
1.)

‘19)

21)

69

IN THIS - - 3""53 .-

ken“ . \ .53 J- b h brain";

Page 6
we are intereeted in how frequently your QQBBIII 03

[[93]: boy/girl friend or. if narried. you:
uaed each of the following nethede tI
llltfllBSILE Ind 199- PIIIII “I

following ncnle to indicate how often he/nhe hen unec

pleaae give you)

often hen he/ahe . . .

Oefended hie/her aelf verbally.

Ignored the problen.

baked another perncn to help nettle thinge.
land alcohol or druga to nettle conflict.
Gotten and at aonebody elne becauae of conflict.
Talked to a friend about the problen.
Frayed with you or alone.

been or aeid eonething to nake you laugh.
lode annual advancea toward you.

lent for wolk. Jogged or worked out.

Put of! talking about the conflict.

“Talked down” to you.

lagged at you.

been eonething to you that left a bruiao.
Denied that there won a problen.

Slapped you on face.

Ooetreyed or danaged an object belonging to you.
Called you had naneta).

Stopped feet.

lode or given an encuan for what happened.
Left the roen/houae.

(Pleaae go on to the next page)

 

 

DO i 6 NOT WRITE

70

IN THIS AREA

 

 

 

0000000366

"II 7 OGGOGGOOCDC

22) lafuaed to talk about it. . .. . . . . . OOCDOGOCDOGC
23) lergained or negotiated with you. . . . . . OOQOQQA‘COC
20 one.“ the topic. . . . . . . . . CDCDCDCDCDOCCOC
23) sort. you aeverely enough to require aedical attention. OOQOQCOOOC
2e) Lectured you. . . . . . . . . . . OOGOGGOOOC
27) Ileaed you for the problen. . . . . . . OOCDOGOCCOC
2n Apologired u. you. . . . . . . . . ©0®®©coooc§
2a) Squeeaed m painfully ..... your mu or are. ooooeocooc;
so) 91...... a. door. . . . . . . . . ooooocococg
31) Sulhed or pouted. . . . . . . . . OCCCC’CCC-CC;
32) Shot you with a gun. . . . . . . . . OOCOGCCCCCE
a) Threatened to phyeicnlly here you. . . . . OO©C=CCQ":C'
34) Argued with you. . . . . . . . . . COOS); ‘ :C 3:
as) Acuocleceee that there wee a problen. . . . OOGOCCSCQC
as) Iade fun of or ridiculed you. . . . . . 006000333:
37) Iiaiched or nocked you. . . . . . . . OOOOOCCCCC
3a) Spoken or acted earcaatically cocci-cc you. . . OOQOCCCCCC
39) soothed or calaed you. . . . . . . . 0060006300
40) Stabbed er out you with a knife. . . . . . OOOOGOOOOC
d1) lgged you on. . . . . . . . . . . OOQOCOOOGC
e2) Iecaae affectionate with you. . . . . . . OOOQCCC SOC
a) noted confuaed i.e.. acted like lie/aha didn‘t know what occeccccoc

waa 02”,:

M) Iecoae very logical. . . . . . . . . OVCQSCCCCC
es) Called police. . . . . . . . . . owe-3;: : : 3:
ea) Iadeperaonal verbal attachoayou. . . . . QQCC::C:C:
intellecci-ecreececezyou. . . . . . . bbc:‘;;::;i
ea) Threatened you with a knife. . . . . . . QCQC333 : j :;
49) Suggeated poaaible eolutiona . . . . . . “““““““ :

(OuaationaSOto'IOarethebackoftheaheet)

 

 

 

DO NOT

am 319......“ .. m. . .5. mxxmvw his) .. .f.. C. .-, C r:

C J

C)

"\
Id

C j)

C)

C)
C-)
C)

CC) 1)
C)

C)
C)

C)
C)

C)
C"
C)
C)

C)’()
)

C)
C
C)

C)
C)

C)
)

C)
C)
C)

C)
C)

C)

"IJJ‘.’

~A~AA

A a -- .- A

-\ a An. A q

C)
C)

C
C)
C) C

n "(u/M," .')'I)’l’[l.lv ~I

(._)

C)
C)
C)

mu.- 4: l) u. (44,-”Hip,

n .- ‘4 v a'

-A‘.‘A§

‘ - \ a s
.

-‘an-s

. A s ‘ a

C)
C) C)

C;)
C)

C)

C)
C)

C) C)
C) C)

C) C)

C)

C)

C) C)

C)
U
C)

C)

C)

C)
Q)

‘00
C) C) C) C) C)

C)

C) C) C) C) C)

C)
J

C)

A

C)

4““

C)
C)

\_I

C)

C)
C)
C)

Q

C)
C)

C)
C) C) C)
C)

C)

C)

C) C)
C) C) '
C)
C) C) C)

)
C)
C)

4“

C) C)
C) C)

C)

C)
C)

C)

C)
C)

«R ‘

‘ s s A a

\.« e.‘q

C)
C)

 

71

WHITE IN

8
“I

THIS - -

 

X;
Page 0
SO) aaknd you to help find eolutiona.
51) brought up your pant ahortconingn or faulta.
S2) forgiven you.
53) acknowledged your fenlinga about the problen.
so) Ordered you around.
SS) Choked or atrangled you
Sh) licked an object (not you).
$1) licked you.
on) bitten you.
3’) 'ulled your hair.
00) Slapped on your body.
61) lit you with hie/her flat.
‘2) Thrown an object at you.
CS) Thrown an ob3ect but not at you.
04) Thrown you.
63) hit you with an object.
‘6) banana angry at you.
67) Threatened you with a gun.

The newt net of gunntione deal with behaviorn that nigh
occur between two people who are clone to one another. 0»
are intereeted in how frequently your GIBBIII or In!
[19:11 boy/girl friend or. if harried. your huaband/wif-
haa done each of the following thinge to 199. Again. ua-
thn following ncnle to indicate how often halnhe hen don.
each of the following to you by blackening the apppropriat
circle.

Scale:
0 I never 2 I d to 10 tinee e I 21 or nore tinee
1 I l to S tinee 3 I 11 to 20 tinee

low often han helahe . . .
ea) lnbarrannnd you in public.
69) Spoken well of you in front of othern.

70) Told cruel or nean Joken about you in your preaence.

l
J

‘72

D0 — — NOT WRITE

IN THIS AREA

 

’1)

’2)

3)
4)
5)
'6)
‘7)
b)
'9)
.0)
Cl)
)2)
3)
0)
-3)
«SD
.7)
A»
S)
‘0)
Cl)
‘2)
'3)
4)
S)
to)
17)

8)

'ege S
Gotten upaot when you went out with none new frienda.
Tried to tell you who your frieodn nhould be. .
Gotten upoet when you went out with. oppooite non
frienda.
Denanded that you account for your activitieo and tine.
Talaely acounod you of having been out with another
pornon.
Tolnely acounod you having non with aoneone elae.
Coaplained that you were flirting with othern. .
hocuned you of keeping eocrotn. . . . . .
Looked through your parnonal belonginga. . . .
Spied on you. . . . .' . . . . .
bullied you. . . . . . . . . . .
Doaandod your ......... to hie/her whine. . .
bean aeoleua or auapicioua of your frienda. . .
Told you that helabe can’t get along without you.
Dooebdod that you borvoiwoit on bin/bur. . . .

Told you that you don't roally underntand what the

world in llko.

Donanded that you not aocialiae with your frienda.
Ordarad you around. . . . . . . . .
Told you that you are attractive. . . . . .
Told you that you are fun to be with. . . . .
Told you that you are ntupid. . . . . . .
Laughed at your feelingn. . . . . . . .
Told you that you couldn't live without bin/her.-
coaplinented you on your lovonaking abilitiee.
lnitiatad diaouaaion of annual feelinga and/or behavior
with you.

Told you that'your lovenaking woe worne than that of

othern.
Told you that he/nhe wanted to have non with ncaoone

elae.
Called you ”looee'. 'proaiecoun'. or ”aany“. . .

(Mention. Q. fan 1” .9. III- ft- he-h'nr ov- .. -.

 

0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
0000000000
00000 000{
0000000 '
0000000
0000000000
0000000000
0000000000
0000000000
0000000000

 

I

f"
\UC

00
000

 

MSU): on Affamuhve Aaron/(qua! 0000! “may Inuvlulmn

~

‘I l)

 

(D C)

C) C)
C)

C)
C) C)
(.1) (J
C_) C) .
C) C)

C)

C.)

C)
C)

C) ()
C)
C)
C)

C) C) C) C)
C) C) (D C)
- C) ? C)
C) C)

C) C)
C) C)

C) (D C)

C)
C)
C)

_)

C) C)
C)
C)

C)
C)
C)

C) C)

(D

C)
C) C)

C)
C)

C.)
C)

C)
C)
C)
C)
C)

C)

C) C) C)
()
C)
C)

Z C) “
C)

C) C) C)
C)
C)

C)

C)

C)
(D
(J
C)

C

C)
C)
C)
C)
C)
C) C)
C) C) C)
'“ C) C)
C) (D C) C)
C)

C) C)
C)
C)
C) C)
C)
C) C)
C)

C)

C)
C)

-‘RQAAAA

 

C) C
C) C) C) C)

C) )
)
I
I
)
)
C J
(J
C)

C) C)
C) C)

C)
(D C) C) (3

C)

C) C) C) C)
C) C)

C)
C)
C)
C) C)
C)
C)
C)
Q
C)
C)

C) (3

C)

C)

C)

C)
C)
C)
C)
C)
C)
C)
C)
C)

C)
C)
C)

C)

(p)
C)
C)
u()

.wDCD

C)
C)

I
‘)
C)
P
C)
C)
C)
C)

 

 

  

” 100)

101)
10B)
100)
100)
100)
10‘)
107)
100)
100)
110)
111)
112)
113)
110)
115)
110)
117)
110)
11’)
120)
121)
122)

‘12,)

73

\\\\\\\V\\\\\ “

“AA““AA “M “

 

feroea you late aeuual aotlvlty la aplte or you
preteeta.

beau unable to tell you tbat aeuual aotlvlty la
uodealree.
Touobeelteeelee your gealtala lo aplte or your
erotaata.
Talked you late aeuual lataroouraa la aolte or your
mua
’byaleally loroee you to baye aeuual loteroourae.

Coeellaeateo you on your aeoeeraaoe.

Told you ubat la euoltlo. or aroualo. about you.

Told you tbat aeuual teella'a or aotlvltlee are
.M'- e

Itartei a dlaouaalea about eoatreoeotloo ultb you.
fbreeteoeo to leave you 1! aeuual aotlvlty uaa not
lertbooelag.

Deed aeuual earloea lo aplte or your proteeta.

Told you tbet you uere a bee lever.

Tbreateoea ta abyaloally bare you ll aerual aetlvlty
uea not tortboealag.

'leroee you to act out bebavler portrayed ln

DO'IOOrODhY-
Told you tbat you uere ugly.

Ield or euodlea you ubeo aabea to do ea.

lad lateroouree ultbeut oeotraoeptleo lo aplta or your
ta.
lulayerably eeeperea you to otber uoaeo/aeo.

Told you tbet you are reeeeoalbla (or pleealoo
blelber.
Deeaaoea aerual aotlylty aa part or tba relatlooablp.

Trlee to eeouoe otbera lo {rout oi you.

leee-a aullea er ultberaue l: aeuual aotlvlty uaa not
fortbeoela'.

?ouebee you lo yubllo la aolta or your protaeta.

leee you look et poroegraeblo aaterlala lo epite or
your proteata. ‘

llaaeo you lo aplta oi your protaata.

Tbaak you for eoeoletlo. tbla aurvey

 
  

”WM-“Muvuuuuuu“uuuusouuuuuu‘cuuuuwu“uwuuuuuuuuuuuuuuuuuuuu

 

APPENDIX 8

Conflict Tactics Scale (CTS) (Form A)

APPENDIX B

Conflict Tactics Scale (CTS) (Form A)

Hutch-Mo!!hIapyeu“IhavadeaaotaayaubaiaooaflIflet~aamatwIIhyouthI‘u.\Vc
mum. ymtowmmmtmumwmm mound-ounctmihhiuhufiwd.
Housman-humour“ onudaflhutblaplluu‘bdevlanhovhouofloa bazyouddluhuyoot

0' New:

I - Once that you:

2 ' “Duo 0: than timcs

J ' Often. but Ia: than one: a month
4 - About one: a mouth

5 ' Men than once a month

"

 

I

a. IMIoMIhhuunhuvdyaI-Iyn ........................ 0 I 2 J 4 S
uwmmmuuuvayduy............. ................. o I 2,343
c.Gotinfotaudoatobuokup-yuduonhlap........ ................ 0 I 2 3 4 S
4. amount: nmuudutohdpmdulbluphtuuto)” ............... .. o I 2 3 4 S
o. Argued homily butMtofyuIIIa. ..... . ................... 0 I 2 3 4 S
I. YcIchu ocimuIIod .......... . ......................... 0 I 2 3 4 S
a. Sultcda ornfluudtoldkabouul. ................. 0 I 2 3 4 S
h. Stamped ouloflha mom . . ................................ 0 I 2 3 4 S
I. mewmmmwutnotutuyvifa)otuuuhodsoaown¢ ............. ,.. 0 I 2 3 4 S
tThruuoodtohItatbrwmatha... .................. 0 I 2 3 4 S
.TIIMWIIIIIIIO! my“: ....... ...................... 0 I 214 5
I Pushednabbcdot Ilet .......................... ....... 0 I 2 3 4 S
cu.Iliumuicdwhiflhuhutaotwflhuayomo; ...... ........... .. 0 I 2 J 4 S
a. III:(o:IrI¢dIohII)haruvIIhlo.-IhI-;hud...'. ..................... 0 I 2 J 4 S

 

74

 

APPENDIX C

Conflict Tactics Scale (CTS) (Form N)

 

APPENDIX C

Conflict Tactics Scale (CTS) (Form N)

No tuuttut boo vol a couple pt- quuo. than on than: when they elae.-c on Not decisions. pt annoyed
about touted“; hmeuMb-nmuotfldmmthoy‘nianbudmmulottimd otfo:
«out way: of trying to settle that Mm I'nt going to «and a list
of to. thing that you and you: (husbandlpattnor) might have done when you Ind a dispute. and wound first
IIbeyou to tell nefotoaeboaobovot’touyou ddttin the putyw.

I
2
I
3'
I
I
I

 

 

 

 

 

 

 

MWICIMA 0.18 0.79 0.8"
II. dent- Husbandll‘artnet- live:
In at Yen In I’utYeat Happened
i- I-
“ 3 3 2 ' a
a 23 s 2 - 2%. s .3: 8
§:;5* 5%:252 f
8 “avg:- .25 ”gt-32'3- .. 'c
gag-3.233 §5¢3¢:§o $353
token-“WWW '0I234S6X 0I234$6X 12x
b. Gottnromttoutobuekup ,
(rout/hisItidooIthiw 0 I 2 3 4 S 6 X 0 I 2 3 4 S 6 X I 2 X
c. Itoudttiaotttiodtobn'ngh
mtohflputththtn’ 0I234S6X onzztsox I2X
I. Insultodonmatthoothaoot 0 I 2 3 4 S 6 X 0 I 2 3 4 S 6 X I 2 X
e. Sulkod uao/o: Infused to talk
about“ 0I234S6X 0I23456X 12x
r. Stompdoutotthutoo-ot
boo-Howard) 0I234$6X 0I23456X sz
{Cried 0 I 2 3 4 S 6 X 0 I '2 3 4 S 6 X I 2 X
DIdouuidmtoqIII
Metham- ~0I234S6X 01234sox sz
I. Threatemdtohttetthluveooa-
thingnttheotbuouo o I 2 3 4 3 6 X 0 I 2 3 4 5 6 X I2 X
. Th anuhodothtteruekd .
Juana; 0I234S6X 0I234$6X I2X
bfhmsotnethlnuttboothuono 0 I 2 3 4 S 6 X 0 I 2 3 4 S 6 X I 2 X
I. Footed. Mamas.
cannon:n ‘ O I 2 3 4 3 6 X 0 I 2 3 4 5 6 X I 2 X
nt.SIupnedtheotherona 0 I 2 3 4 3 6 X 0 I 2 3 4 5 6 X I2 X
n. Kicked.btt.othttvlthuflu 0 I 2 3 4 3 6 X 0 I 2 3 4 5 6 X I 2 X
o. Hitottnedtohttwlthtontethln; 0 I 2 3 4 S 6 X 0 I 2 3 4 S 6 X I 2 X
p. IIeatuptheothatono 0 I 2 3 4 S 6 X 0 I 2 3 4 5 6 X I 2 X
o.1‘hmtenodvtthahnlfeotgun 0 I 2 3 4 3 6 X 0 I 2 3 4 5'6 X I 2 X
t. Usedaknifeotgun 0 I 2 3 4 5 6 X t 0 I 2 3 4 S 6 X I 2 X
tOtheHPROBE): 0I234S6X 0I234S6X I2X

 

 

79. And what about you! (hub-adiputtnet)? Tell me how often he (ITEM) In the past yen. -'
For each Itetn dtelod etthot “Never” or “Don't Known" for BOTH respondent and partnet, ash;
80. Old you or you: (husband/partner) ever (ITI-ZM)?

 

75

APPENDIX D

Index of Spouse Abuse (ISA)

APPENDIX D

Index of Spouse Abuse (ISA)

INDEX OF SI‘OUSE ABUSL (ISA)

This questionnaire is designed to measure the degree of abuse you has e esperienced in your relationship with your
partner. It is not a tee. so there are no right or wruny answers. Answer each item as carefully and accurately as you
can by placing a number beside each one as follows:

I Never

2 Rarely

J Occasionally .

4 Frequently

3 Very Frequently

Please begin.

. My partner belittles me. (I)

. My partner demands obedience to his whims. (I1)

My partner becomes surly and angry if I tell him he is drinking too much. (I5)
My partner makes me perform sea acts that I do not enjoy or like. (50)

My partner becomes very upset if dinner. housework or laundry is not done when he thinks it
should be. (4)

My partner is yeast and suspicious of my friends. (1)

My partner punches me with his fists. (75)

My partner tells me I am ugly and unattractive. (26)

My partner tells me I really couldn‘t manage or take care of myself without him. (I)
I0. My partner acts like I am his personal servant. (20)

II. My partner insults or shames me in front of Others. (4|)

l2. My partner becomes very angry if I disagree with his point of sicw. (IS)

I3. My partner threatens me with a weapon. (82)

I4. My partner is stingy in giving me enough money to run our home. (I2)

IS. My partner belittlcs ntc intellectually. (20)

I6. My partner demands that I stay home to take care of the children. (I4)

I7. My partner beats me so badly that I must seek medical help. (91)

II. My partner feels that I should not work or go to school. (2|)

I9. My partner is not a kind person. (I3)

20. My partner does not want me to socialize with my female friends. (I!)

2|. My partner demands sea whether I want it or not. (32)

22. My partner screams and yells at me. (31)

3. My partner slaps me around my face and head. (80)

23

26

9999 999-

. My partner becomes abusive when he drinks. (6S)
. My partner orders the around. (29)
. My partner has notr'espeet for my feelings. (39) °
27. My partner acts like a bully towards me. (44)
21. My partner frightens me. (33)
29. My partner treats me like a dunee. (29)
30. My partner acts like he would like to kill me. (80)

Copyright 0 Sally R. McIntosh and Walter W. Hudson. I978.
P: 3. 4. 1. I3. l7. 22-24. 27. 21. 30.
NP: I. 2. S. 6. 1-I2. I4-l6. I8-2I. 25. 26. 29.

76

li||||||||lllllllllilllllli llll

 

‘Al'-'. “-2

APPENDIX E

Table 17: Frequency of Responses to CRAO Items 1-67

 

APPENDIX E

Frequency of Responses to CRAQ Items 1-67

Table 17
Frequency of Responses to CRAQ Items 1-67

 

Frecpency
Iem Never 1t03Ttmes 4to10Ttmes 11t020Ttmes 21orMore AtLeastOnce

 

#1 4.2 18.2 25.3 20.3 31.9 95.7
Detended

yourself
verbally

#2 20.5 28.6 27.4 15.3 8.3 79.5
Ignored
the problem

43 45.3 32.0 15.2 4.7 2.9 54.8
Asked
another
to help

#4 79.6 12.6 4.1 2.1 1.7 20.5
Used
alcohol or

drugs

#5 28.3 38.7 19.9 7.8 5.3 71.7
Gotten mad

at somebody
else

#8 5.7 21.7 28.7 20.8 23.2 94.4
Talked to
a friend

 

#7 62.6 14.2 8.8 5.3 9.1 37.4
Frayed

#8 3.3 10.4 13.2 19.4 53.7 96.7
Done or

said some-

thing to

make the

other laugh

 

77

78

Table 17 (Continued)
Frequency of Responses to CRAQ Items 1-67

 

Frquency
Item Never 1t03firnes 4to10Ttmes 11t020Ttmes 21orMore AtLeastOnce

 

#9 19.8 19.6 16.7 11.8 32.1 80.2

sexual
advances

#10 10.7 24.7 23.6 18.1 22.9 89.3
Well tor

wak' pggedi

OfWOl‘kOd out

#11 19.3 35.1 25.6 13.1 6.9 60.7

taldngabout
tlteoomliot

#12 43.9 28.3 16.1 8.0 3.8 56.2
Talked down
to him/her

#13 30.9 31.4 19.2 11.7 6.6 69.1

Nagged
athim/her

#14 83.2 13.1 2.7 0.8 0.3 16.9

something that
lettabrulse

#15 45.7 37.4 9.9 4.4 2.6 54.3
Denied

that there

was a

 

79

Table 17 (Continued)
Frequency of Responses to CRAO Items 1-67

 

Frecpency
Item Never 11031‘tmes 4t010Ttmes 11to201'tmes 21 orMore AtLeastOnce

 

#18 88.1 9.7 1.2 0.9 0.2 12.0
Slapped

him/her

on lace

#17 87.5 11.0 1.2 0.2 0.2 12.6
Destroyed

or damaged an

object

belonging to

hIther

#18 42.0 29.3 15.0 7.6 6.2 58.1
Called

htmlher

bad names

#19 62.5 22.4 10.0 3.0 2.1 37.5

Stamped
feet

#20 11.2 38.1 29.5 13.5 7.7 88.8
Made or

given an

excuse

#21 29.2 39.0 16.8 10.6 4.5 70.9
Left the
IOOfl'l/I'DUSO

#22 43.1 32.5 13.7 6.9 3.8 56.9
Retused to
13k

#23 23.0 33.2 25.8 10.7 7.3 77.0
Bargained
or negotiated

 

Table 17 (Continued)

Frequency of Responses to CRAO Items 1-67

80

 

Mm

Never

1t03Ttmes 4to10Ttmes 1110201'tmes 21 orMore AtLeastOnce

Fremency

 

#24

the topic

Hurt
severely
enough to
require

thedoor

#31
Sulsed

orpouted

Shot them

23.3

98.9

26.4

26.3

3.8

79.8

46.2

21.6

37.5

0.5

36.0

14.9

30.3

0.2

0.3

20.5

21.0

25.2

3.3

14.5

20.0

0.0

10.4

0.2

10.9

10.6

24.5

1.5

5.8

12.6

0.2

5.4

0.2

8.7

6.2

23.0

0.5

3.2

12.1

0.0

76.7

1.2

73.6

73.6

96.3

20.2

78.4

0.4

 

 

re“- “

81

Table 17 (Continued)
Frequency of Responses to CRAO Items 1-67

 

quiency
lam Never 1t03Times 4t0101'tmes 1110201'tmes 21orMore AtLeastOnce

 

#33 90.9 6.5 1.8 0.3 0.5 9.1
Threatened

to Wisely

hmn

#84 10.9 23.9 20.0 16.6 28.6 89.1
Argued

#35 7.3 32.2 25.6 19.1 15.9 92.8
Agreed

there was

a problem

#38 42.7 31 .3 15.4 6.3 4.2 57.2
Made fun

of them

#37 39.2 31.1 18.1 6.0 5.6 60.8
Mimicked

or mocked

#38 17.7 32.4 22.5 14.4 13.0 82.3
Spoken F‘-
#39 6.5 19.6 23.6 23.1 27.2 93.5

orcaltrted
them

#40 98.9 0.5 0.5 0.2 0.0 1.2

 

(PVT-’4 .._-. - 1”

Of cut
#41 48.6 25.3 14.1 7.9 4.1 51.4

himlher on

 

82

Table 17 (Continued)
Frequency of Responses to CRAO Items 1-67

 

 

Frecpertcy

lam Never 1to3Ttmes 4to10T1mes 11t020'l'tmes 21 orMore AtLeastOnce
#42 5.7 17.1 14.7 18.8 43.7 94.3
Became
wectbnate
#43 27.6 33.8 20.7 13.3 4.5 72.3
Acted
confused
#44 8.6 24.5 26.6 21.8 18.5 91.4
Became
VOW
beleal
#45 98.9 0.6 0.0 0.3 0.2 1.1
Caled
police
#46 46.1 28.5 14.2 6.2 5.1 54.0
Made a
personal
vetbal attack
#47 35.8 30.1 14.7 9.3 10.1 64.2
Yelled
or screamed
#48 98.6 0.8 0.2 0.5 0.0 1.5
Threatened
wih a
knife
#49 4.5 23.8 27.7 23.9 20.2 95.6
Suggested
possible

 

 

83

Table 17 (Continued)
Frequency of Responses to CRAQ Items 1-67

 

Fremertcy
Item Never 1t03T1mes 4t010Ttmes 11t020Times 21orMore AtLeastOnce

 

#50 7.7 26.8 26.8 20.2 16.6 92.4

sokttiens

#51 31.6 31.5 18.2 11.4 7.2 68.3
Brought

UP 938!
laults

#52 3.6 20.2 23.7 22.9 29.6 96.4
Forgiven

#53 2.6 17.5 24.1 28.0 27.8 97.4
Acknowledged

their

teellnos

#54 50.7 27.4 12.0 6.3 3.6 49.3

them
around

#55 98.0 1.7 0.2 0.2 0.0 2.1
Choked

orstrangled

#56 72.9 16.4 6.8 3.0 0.9 27.1
Kicked

an object
#57 95.5 3.2 0.9 0.5 0.0 4.6

 

them

 

84

Table 17 (Continued)
Frequency 01 Responses to CRAO Items 1-67

 

Fremency
Item Never 1t03'l'tmes 4to10Ttmes 11t020'l'1mes 21orMore AtLeastOnce

 

#58 94.3 3.8 0.9 0.5 0.6 5.8
Bitten
them

#59 91.9 5.4 1.2 1.4 0.2 8.2
PUIled
their hair

#60 79.0 13.4 4.8 1.4 1.4 21.0
Slapped

them

onbcdy

#61 89.0 7.2 2.1 1.5 0.2 11.0
Hi them
with

yourtlst
#62 91.0 5.1 3.2 0.6 0.2 9.1

anobiect
atthem

#63 74.5 17.6 5.7 1.4 0.8 25.5

anobiect
(notatthem)

#64 97.4 2.4 0.2 0.0 0.0 2.6
them
#65 92.0 4.5 2.6 0.3 0.6 8.0

Hit them
withan

 

9"”-

 

85

Table 17 (Continued)
Frequency of Responses to CRAO Items 1-67

 

Frecpency
Rem Never 1t031'1mes 4to10Ttmes 11to20‘I'tmes 21orMore AtLeastOnce

 

#66 16.6 24.1 17.9 20.0 21.4 83.4
Became

am
am

#67 99.4 0.5 0.0 0.2 0.0 0.7
Threatened
them with

agun

 

Note. N-671

 

fl

APPENDIX F

Figure 1: Some Plot of Eigenvalues

 

the nvel ues

APPENDIX F

Scree Plot of Eigenvalues

N
O

040)“)

35040-0)

‘00

«wmouodm

.»
.1.
..
i

Number of factors

Figure I. Scree plot of eigenvalues.

4r

 

APPENDIX G

Tables 18—24: Items and Item Loadings of CRAQ Factors

 

APPENDIX G

Items and Item Loadings of CRAQ Factors

Table 18
Appropriate/Generic Factor of the CRAQ

 

Item
number Item Factor loading

 

53 Acknowledged his/her feelings about

the problem .815

49 Suggested possible solutions .808
50 Asked him/her to help find solutions .784
52 Forgiven him/her .756
35 Agreed that there was a problem .755
28 Apologized to him/her .645
34 Argued with him/her .592
39 Soothed or calmed him/her .516
1 Detended yourself verbally .506
66 Became angry at him/her .453
47 Yelled or screamed at him/her .426
44 Became very logical .423
31 Sulked or pouted .399
23 Bargained or negotiated with him/her .384

 

 

87

 

 

88

Table 19
Moderate Violence Factor of the CRAQ

 

 

Item

number Item Factor loading
59 Pulled his/her hair .834
57 Kicked him/her .801
61 Hit him/her with your fist .788
60 Slapped him/her on body .752
58 Bitten him/her .673
62 Thrown an object at him/her .669
16 Slapped him/her on face .634
65 Hit him/her with an object .617
33 Threatened to physically harm him/her .606
29 Squeezed him/her painfully e.g.. their

wrist or arm .399

55 Choked or strangled him/her .353

14 Done something to him/her that left
a bruise .312

 

 

89

 

 

 

Table 20
Avoid Factor of the CRAQ
Item
number Item Factor loading
2 Ignored the problem .775
11 Put off talking about the conflict .767
22 Refused to talk about it .694
15 Denied that there was a problem .684
24 Changed the topic . .659
43 Acted confused i.e., acted like you didn't know
what was going on .476
5 Gotten mad at somebody else because
of conflict .422
3 Asked another person to help settle things .392
21 Left the room/house .360
Table 21

Severe Violence Factor of the CRAO

 

 

Item
number Item Factor loading
48 Threatened him/her with a knife .809
25 Hurt him/her severely enough to
require medical attention .720
40 Stabbed or cut him/her with a knife .642
32 Shot him/her with a gun .619
67 Threatened him/her with a gun .351

 

 

 

 

 

9 0
Table 22
Distraction Factor of the CRAQ
Item
number Item Factor loading
8 Done or said something to make him/her laugh .815
9 Made sexual advances toward him/her .780
42 Became affectionate towards him/her .777
10 Went for walk, jogged or worked out .577
39 Soothed or calmed him/her .475
Table 23

Verbal Attack/Coercion Factor of the CRAO

 

 

 

 

Item
number Item Factor loading
36 Made fun of or ridiculed him/her -.74e '
54 Ordered him/her around -.665
37 Mimicked or mocked him/her -.648
12 Talked down to him/her -.643
46 Made a personal verbal attack against him/her -.613
38 Spoken or acted sarcastically to him/her -.595 In:
51 Brought up his/her past shortcomings or faults -.577
41 Egged him/her on -.553
27 Blamed him/her for the problem -.551
13 Nagged at him/her -.520
26 Lectured him/her -.495
47 Yelled or screamed at him/her -.424

18 Called him/her bad name(s) -.412

 

 

91

Table 24
Symbolic/Indirect Violence Factor of the CRAQ

 

Item
number Item

Factor loading

 

56 Kicked an object (not him/her)
63 Thrown an object but not at him/her
30 Slammed the door

17 - Destroyed or damaged an object belonging
to him/her

-.671
-.645
-.489

-.344

 

 

#
IQ."

APPENDIX H

Figure 2: Plot of Group Centroids on Two Discriminant Functions

1;“

 

 

Second discriminant function

APPENDIX H

Plot of Group Centroids on Two Discriminant Functions

 

m
N
d . . Belligerent / Active
Poettive/Average
a
o
e
lnexperienced °
'1' Coercive/ Active
(\l
I
m
I

 

 

 

-3 -2 ~1 0 1 2 3
First discriminant function
Figure 2. Plot of four group centroids on two

discriminant functions derived from hupothesized
predictors of dating coercion.

92

 

 

LIST OF REFERENCES

 

 

 

LIST OF REFERENCES

Ainsworth, M. D. S. Blehar M. C. Waters, E. &Wall, S. (1978). 33mm

WWW Hillsdale. NJ:
Erlbaum.

Arias. l- 8- O'Leary. K- D- (1984). WWW

' ’ ' r ' . Paper presented at the 18th annual
convention of the Association for the Advancement of Behavior Therapy,
Philadelphia.

Bandura. A. (1973). Aggression; AsssisIJsaminnsnaLxsis Englewood Cliffs.
N. J.: Prentice-Hall.

Bandura, A. Ross, D. 81 Ross. S. A. (1961). Transmission of aggression

through imitation of aggressive models. WW3)
Esxsanscx. 63. 575-582

Bandura. A.. & Walters. Ft. (1963). SecaUsaminnandnstssnalibt
W. New York: Holt, Rinehart, & Winston.

Bernard, M. L, & Bernard. J. L (1983). Violent intimacy: The family as a model
for love relationships. W, 32, 283-286.

Carroll, J. C. (1977). The intergenerational transmission of family violence: The
long-term effects of aggressive behavior. W,1 289—299.

Cane" R (1978) IbsssisnflflsusssttastoLanaIstanbsnaxicraLandms
W. New York. Plenum Press.

Curtis. W. J. Maguin, E. &Stollak, G. (1987). W

.Paper presented at the annual
convention of the Midwestern Psychological Association, Chicago.

Dixon. W. J. (1981). W. Bekeley, CA.: University of
Califomia Press.

Dobash. R. E. 8: Dobash. R. P. (1979). MislsnssasainsmixsssAsasescains:
W. New York: Free Press.

93

 

 

ll."

94

Edmonds, V. H. (1967). Maritalconventionalization: Definition and

measurament MW 23 531533

Edmonds, V. H. W1thers, G., 8. Dibatista, B. (1972). Adjustment, conservatism,

and marital conventionalization. WM 34,,
96-1 03.

Gelles, R. J. (1974).
W. Beverly Hills, Calif.: Sage Publications.

George. 0.. Kaplan. M. 8. Main, M. (1984)- AnacnmsnLintsniisinLtnLadults
Unpublished manuscript, University of California, Berkeley.

Gottman. J. M. (1979). Mantaumsractinmfixnsnmsntaljmstinatinns New
York: Academic Press.

Hartup, W. W. (1977). Peerinteraction and the behavioral development of the
individual child. In E. Mavis Hetherington and Ross D. Parke (Eds)

Qnmsmnnranusadinssinsnildssxsbslssx (pp 345- -354) New York:
McGraw-Hill Book Co.

Hazan, C. 81 Shaver, P. (1987). Romantic love conceptualized as an attachment
process W 62. 511-524

. Hudson, W. W. 6Mclntosh, S. R. (1981). The assessment of spouse abuse:

Two quantifiable dimensions JnumsLsLMacnanaandJnsEamilx. .43.
873-885.

Jorgensen, S. R. (1977). Social class heterogamy, status striving, and
perceptions of marital conflict: A partial replication and revision of Pearlin' 5

contingency hypothesis JournaLsLMsmsnsandeLEamih: 69. 653-661
Klecka, W. R. (1980).)2isgfim1n30Lanalysjs. Beverly Hills: Sage Publications.

Lane, K. E., 6Gwartney-Gibbs, P. A. (1985). Violence in the context of dating
and sex JsumalsLEamilLlssuss. 6. 45-59

Laner, M. R. (1983). Courtship abuse and aggression: Contextual aspects.
SosisbcisaLSnsstnim 3. 69-83

Laner, R. L., &Thompson, J. (1982). Abuse and aggression in courting couples.
DsitianLBsnaxiscAnJmsLdissinlinanulsumaI 3. 229-244

Maccoby. E., & Jacklin, C. (1974). W333. Stanford,
Calif.: Stanford University Press.

 

95

Maccoby, E., 8 Jacklin, C. (1980). Sex differences in aggression: A rejoinder
and reprise. Qnildnsxsbnmsm. 51. 964-980.

Maguin, E., Curtis, W. J. 8 Stollak, G. (1987). WWW
goggles. Paper presented at the annual convention of the American
Psychological Association, New York

Makepeace, J. M. (1983). Life events stress and courtship violence. Eamily
leatisns. 62. 101 -109.

Matthews, W. (1984). Violence in college couples. WW3), 1.8.
150-158.

McCord, W., McCord, J. 8 Howard, A. (1961). Familial correlates of aggression

In nondslinqusnt male children JnumsLnLAbnnnnaLandfissiaLEsysnnlscx
32, 79-93.

McCord, J. McCord, W. 8 Howard, A. (1963). Family interaction as antecedent

to the direction of male aggressiveness. WWW
Esxsbclncx. 66. 239-242

Meichenbaum. D. (1977). WWW
3123:3333. New York: Plenum Press.

Nie, N. Hull, C. H., Jenkins, J. Steinbrenner, K. 8Bent, D. (1975). 5131(3))33)
nachacslcunasnsiaLssisnssslrsv sd--) New York: McGraw-Hill

NunnaIIy, J. C. (1978). W (rev. ed.). New York: McGraw-Hill
Book Co.

 

Owens, D. M., 8 Straus, M. A. (1975). The social structure of violence in PE

childhood and approval of violence as an adult. W. 1.
193-211.

Pagelow, M. D. (1981) WWW
Beverly Hills: Sage Publications.

 

W

Patterson, G. R. (1976). The aggressive child: Victim and architect ofa coercive
system. In E. J. Mash, LA. Hamerlynck, 8L. C. Handy (Eds.,) M31191

W New York: Brunner/Mazel.

Peterson, D. (1983). Conflict. In H. H. Kelley, E. Berscheid, A. Christenesen, J.
Harvey, T. Huston, G. Levinger, E. McCIintock, L. A. Peplau, 8 D. Peterson

(Eds). Insnsxstinlsoxsisbsuslstisnsnips (pp 360-396). San Francisco:

Freeman.

95

Maccoby, E., 8 Jacklin, C. (1980). Sex differences in aggression: A rejoinder
and reprise. W, 51, 964-980.

M39Uin. E., Curtis, W. J. 8 Stollak, G. (1987). WWW

muglgs. Paper presented at the annual convention of the American
Psychological Association, New York.

Makepeace, J. M. (1983). Life events stress and courtship violence. Eamily
56161193232. 101- 109

Matthews, W. (1984). Violence in college couples. W, 18,
150-158.

McCord, W. McCord, J, 8 Howard, A. (1961). Famtltal correlates of aggressmn

in nondelinquent male children. Wm,
52, 79- 93.

McCord, J, McCord, W., 8 Howard, A. (1963). Family interaction as antecedent

to the direction of male aggressiveness. WW
Balancing! 6.6. 239 242

Meichenbaum. D. (1977)- WWW
333323311. New York: Plenum Press.

Nie, N. Hull, C. H., Jenkins, J., Steinbrenner, K. 8Bent, D. (1975). 313151133)
naskansinntnssoctaLssisnssslrsv 9d--) New York McGraw-Hill

Nunnally, J. C. (1978). W (rev. ed.). New York: McGraw-Hill
Book Co.

Owens, D. M., 8 Straus, M. A. (1975). The social structure of violence in
childhood and approval of violence as an adult. WW, 1,
193-211.

PageIOW. M. D. (1981) WWW
Beverly Hills: Sage Publications.

Patterson, G. R. (1976). The aggressive child. Victim and architect ofaooercive
system. In E. J. Mash, L. A. Hamerlynck, 8 L. C. Handy (Eds.,) 53113119)

W. New York: Brunner/Mazel.

Peterson, D. (1983). Conflict. In H. H. Kelley, E. Berscheid, A. Christenesen, J.
Harvey, T. Huston, G. Levinger, E. McClintock, L. A. Peplau, 8 D. Peterson

(Eds). Insnsxsnniosumlnssrslaflsnsnins (pp 360-396). San Francisco:

Freeman.

 

 

96

Roy, M. (1977). A current survey of 150 cases. In M. Roy (Ed.),B_31131311mmg_n.
(pp. 25-44). New York: Van Nostrand Reinhold.

Sigelman, C. K., Berry,C. J., 8W1les, K. A. (1984). Violence in college students'
dating relationships WW. 6 530-548

Steinmetz, S. K (1977). Wife-beating, husband-beating- a comparison of the
use of physical violence between spouses to resolve marital fights. In M. Roy
(Ed.), W311 (pp. 63-71). New York: Van Nostrand Reinhold.

Steinmetz, S. K. (1980). Women and violence: Victims and perpetrators.
AmsdsaMnumaLnLEsxsbstnsLacx. 3.4. 334-350

Stets, J. 8 Pirog-Good, M. (1987). Wolence in dating relationships. 53313)
EsxsnnlscLQuansle. 60. 237-246.

Straus, M. A. (1974) Leveltng, CIVIIIty and violence in the family mm
W. 36. 13-29.

Straus, M. A. (1979). Measuring intrafamily conflict and violence: The conflict
tactics (CT) scales JoumaLnLManiamndesEamilx. 41. 75-88

Straus, M. A., Gelles, R. J., 8Steinmetz, S. K. (198mm
W. New York: Doubleday/Anchor.

Walker, L. E. (1979). W. New York: Harper and Row.

Walker, L E. (1984). W. New York: Springer.

Yllo, K. 8 Straus, M. A. (1981). Interpersonal violence among cohabitating and
married couples- Eamilxfislatisns. 36. 339-348.