Psychological Reports, 1990, 66, 1003-1009. O Psychological Reports 1990
PARENTAL DIVORCE DURING CHILDHOOD AND USE O F VIOLENCE I N DATING RELATIONSHIPS ' ROBERT E. BILLINGHAM AND KATHLEEN R. GILBERT Department of Applied Health Science, Indiana University Summary.-355 women and 173 men university students between the ages of 17 and 23 yr. completed a questionnaire which included Straus's Conflict Tactic Scales, and from which whether the respondents had experienced the divorce of their parents could be assessed. Two hypotheses were tested. First, individuals from divorced families would be more likely to report that violence existed in their current relationship than would individuals from intact families. Second, in those relationships which included violence, individuals from divorced families would report higher levels of violence than would individuals From intact families. Analyses indicated that individuals from divorced families were no more likely to report themselves as experiencing violence in their current relationship than were their cohorts from intact families. However, when the conflict tactic scores for men from dvorced, men from intact, women from divorced, and women from intact families were compared, differences were found. Men from divorced families reported levels of violence for themselves that approached sigdicance. Also, men from divorced families reported that their partners' verbal aggression approached sigdicance and that their partners' use of violent behavior reached significance when compared to those of the other groups. These results indicate that having experienced a divorce in one's family may have a greater effect on later relationships for men than for women.
Research on long-term effects of divorce on children has only recently become possible, but that fact is not consistent with interest in the topic. This lack of research is more a reflection of the historically small numbers of children who have experienced parental divorce; however, about one-third of all children born in the 1970s will experience a parental divorce (Bane, 1979). The youngest of these children are just now entering their late teens and early twenties, so the effects that divorce will have on these children's own relationships can only now be investigated. The prognoses for these relationships do not appear to be particularly satisfactory. Kelly (1981) found that, even those subjects who had adjusted well to their parents' divorce expressed fear of involvement and had relationships that were short lived. The least well adjusted had rushed into heterosexual activity and seemed "incapable of malung judgments about the adequacy of their relationships . . ." (p. 137). These effects were found five years after the divorce had occurred. Wallerstein (1987) in a 10-yr. follow-up of children who had been younger (six to eight years old) than the Kelly subjects (13 to 18 years old) at the time of their parents' divorce found these 'Address correspondence to Robert E. Billingham, Department of Applied Health Science, Indiana University, Bloomington, IN 47405.
R. E. BILLINGHAM & K. R. GILBERT
subjects to be profoundly unhappy in their current relationships and concerned about future relationships. McLanahan and Bumpass (1988), using data from the National Survey of Family Growth (NSFG), reported that women from one-parent families were more likely to give birth before marriage, marry and bear children early and have their own marriages break up. Finally, Billingham, Sauer, and Pillion (1989) found significant differences in the sexual attitudes and behaviors of college students, with those from divorced famdies being significantly more permissive and sexually active than their peers from intact families. Interestingly, one of the areas of heterosexual dating for which the impact of divorce has not been investigated is dating violence. This is surprising in light of the fact that an estimated 25 to 30% of the currently dating couples experience some form of violence in their relationships (Billingham, 1987; Laner, 1983; Makepeace, 1981). Further, recent research (Raymond & Brusch, 1989) has suggested that psychological abuse in dating relationships is also a common occurrence. Therefore, the present study was designed to evaluate the effect of divorce in one's family on the use of violence in current dating relationships. To that end, two questions guided this study: first, are indviduals who have experienced parental divorce more likely to be violent in their current relationships than are individuals whose parents are not divorced? Second, in those relationships which do include violence, does coming from a divorced family produce a difference in the pattern in the use of violence?
Respondents The respondents were gleaned from a larger study designed to investigate various sexual attitudes and behaviors evidenced by university students. I n this study respondents were limited to those who were white and between the ages of 17 and 23 yr. This group was selected because the investigators believe they would be most likely to have the cognitive, intellectual, and financial resources available to minimize the effects of parental divorce.
Procedure One thousand questionnaires (500 for men and 500 for women) were distributed in classes offered at a large university in the rnidwest. These classes were selected because they were general elective courses which attracted students from a variety of schools and majors. They were Human Sexuality, Marriage and Family Interaction, Drug Use in American Society, Life Span Human Development and Personal Health. The principal investigator introduced the study to the students attending the classes, oversaw the distribution of questionnaires, and instructed the students on how to return their responses.
PARENTAL DIVORCEPIOLENCE IN DATING
The questionnaire was distributed in a preaddressed envelope for campus mail. The students were told to complete the questionnaires at their leisure and to return them via campus mail to their instructor at a later class meeting, or through whatever means they wished. Of the 500 questionnaires l s t r i b u t e d to the women 432 (86.4%) were returned. Of these, 355 (82.1%) met the criteria for inclusion. Of the 500 questionnaires distributed to the men 223 (44.6%) were returned. Of these, 173 (77.5%) met the criteria for inclusion. I t is possible that the low response rate on the part of the men was due to the fact that the study was introduced via an attached cover letter which stated "to help us understand how an individual's famiIy and personal attitudes and beliefs are translated into behaviors within their interpersonal relationships." I t is possible therefore that a large number of males either did not view themselves as being in a relationship or they did not want to report their attitudes and behaviors concerning their relationships.
Instrument To ascertain the extent to whlch violence was used in the relationship, Straus's (1979) Conflict Tactic Scales were used. While these scales were initially developed to study conflict resolution strategies used by married couples, they have been used in a variety of published papers which have investigated dating violence. The scales consisted of 14 questions which range from "Tried to discuss the issue calmly" to "Hit or tried to hit the other person with somethng hard." Each respondent rated the 14 items on a scale w h c h ranged from 0 (never happened) to 5 (happened more than once a month). Three subscores were obtained by summing the responses. These were reasoning (Items 1 through 4), verbal aggression (Items 5 through 9), and violence (Items 10 through 14). The possible range for each of the subscales is as follows: reasoning, 0-20; verbal aggression 0-25; violence, 0-25. Each respondent was asked to report the number of months they had been in their current relationship. The responses ranged from 1 mo. to 98 mo. (8.2 yr.). The mean number of months reported was 18.7 (SD= 18.6) or approximately 1% yr. I n addition, the median number of months reported was 12.0 (1 yr.). Finally, 71.9% of the respondents reported that they had been dating their current partner for six months or more. To answer the first guiding question, i.e., are individuals from divorced families more likely to be involved in relationships which involve violence, a series of 2 x 2 (parents divorced by violent status) chi-squared analyses was conducted. The results indicated that individuals from divorced families were no more likely to be involved in violent relationships than were individuals
R. E. BILLINGHAM & K. R. GILBERT
from intact families [X,2 (N = 520)= .31, p > .05]. This same lack of association was found for both men [X,2 (N= 170) = .58, p>.051 and women Lx,Z (N = 350) = .54, p > ,051. We conclude that individuals from divorced families are neither more nor less likely to be involved in violent relationships than are their cohorts from nondivorced families. The answer to the second guiding question, i.e., are there differences in the reporting of the use of Straus's Conflict Tactics Scales based on the presence or absence of divorce in one's family history, proved to be somewhat more complex. First, the respondents were divided into two groups. The first group reported that neither they nor their partners had used any form of violence in their present relationship (ns = 109 men and 212 women). The second group was comprised of those respondents who reported that violence in some form had occurred in their present relationship (ns = 6 1 men and 138 women). TABLE 1
MEANSAND STANDARD DEVIATIONS FOR CONFLICT TACTICSSUBSCALES BY SEX AND
Conflict Tactics Scales n
Respondents' Scores Reasoning M SD Verbal Aggression M SD Violence
DIVORCE STATUSFORNONVIOLENT GROUP
Divorce None Yes
Nondivorce Men Women
Divorce Men Women
10.72 4.79 4.98 4.26
Respondents' Pattners' Scores Reasoning M 10.23 SD 5.25 Verbal Aggression M 5.06 SD 4.29 Violence NA
For the nonviolent group a series of 2 x 2 (sex by parents divorced) analyses of variance were conducted for the respondents' report of their own reasoning and verbal aggression behaviors and for their reports of their partners' reasoning and verbal aggression behaviors. The results of these analyses indicate that no significant differences based on either sex or divorced status, and there was no significant interaction between sex and divorced status on any of the subscores; see Table 1. I n sum, for those relationships which
PARENTAL DIVORCE/VIOLENCE IN DATING
do not include violence, individuals from divorced families report use of conflict tactics patterns which are similar to those of their cohorts from intact families . For the violent group, a series of 2 x 2 (sex by parents divorced) analyses of variance were conducted for the respondents' report of their own reasoning, verbal aggression and violence conflict tactics, and for their reports of their partners' reasoning, verbal aggression and violent behaviors. Unlike the nonviolent group, significant differences and interactions were noted; see Table 2. TABLE 2 MEANS AND STANDARD DEVIATIONS FORCONF'LICT TACTICS SUBSCALESBY SEX AND DIVORCE STATUS FORVIOLENT GROW Conflict Tactics Scales
Nondivorce Men Women
Divorce Men Women
11.36 10.20 4.86 4.93
12.51 11.22 4.41 3.73
4.03 2.56' 5.34 4.00 * p < .05. tInteraction, p