The current study investigated the concurrent and reciprocal relations between heavy drinking and dating violence for men and women across the first three years of college. The frequency of heavy drinking and dating violence was found to be relatively stable across time for both genders. With the use of multi-group path analyses, we determined that the better fitting model was one that allowed the paths in the model to vary by gender, suggesting that the relation between alcohol use and dating violence over time differed for men and women.
Based on the alcohol myopia model, we hypothesized that alcohol would be positively associated with concurrent dating violence. This, however, was only supported for men and only during their freshman year of college. This finding may be due to the fact that aggressive behavior of heavy drinking individuals is disinhibited by their alcohol use (Curtin et al., 2001
; Giancola, 2000
; Steele & Josephs, 1990
), or it may be that these men spent more time in situations where alcohol and aggression often co-occur (e.g., bars, large parties) during their freshman year. Consistent with a coping model, we also expected that dating violence would predict future heavy drinking in order to cope with negative feelings such as guilt over the aggression. Lastly, based on the idea that drinking leads to aggressive behaviors, we hypothesized that heavy drinking would predict future dating violence. These latter two hypotheses were not supported as instead heavy drinking during sophomore year predicted dating violence for women in their junior year, but no significant longitudinal associations were found for men.
It is somewhat surprising that heavy drinking did not predict subsequent dating violence for men; however, research in this area has not been consistent. At least one other study of heavy drinking and dating violence by men likewise failed to find an association over time even though they were related cross-sectionally (Foshee et al., 2001
). Dating violence for men may be influenced more by the situational effects of heavy drinking, which would be consistent with a concurrent rather than a longitudinal association. Additionally, individuals in romantic relationships during college may not be as committed to their partners as married couples. Because heavy drinking has been associated with relationship discord or dissatisfaction (Testa, Livingston, & Leonard, 2003
), these relationships may end before a pattern of dating violence develops.
The results of the current study also did not support earlier work which concluded that dating violence victimization and perpetration predicted future heavy drinking for women. These previous findings, however, represented either indirect effects through other relationship factors, or associations over a six-year period which represent more long-term effects than were investigated in the current study (Martino et al., 2005
, Testa, Livingston, & Leonard, 2003
). Substance use may be a way to cope with the negative consequences of dating violence (e.g., depression, PTSD); however, these effects may not be immediate, but rather develop over time, or may be related to the occurrence of other factors such as relationship dissatisfaction (Anderson, 2002
; Kilpatrick et al., 1997
It is important to note that heavy drinking was predictive of subsequent dating violence for women, with heavy drinking in sophomore year predicting dating violence during their junior year. This longitudinal relation may be explained by a tendency for heavy drinking women to associate with other heavy drinking individuals (for a review see Borsari & Carey, 2001
). They may be more likely to form unhealthy relationships based on mutual abuse of alcohol (Leonard & Das Eiden, 1999
; Leonard & Mudar, 2003
) thereby increasing the likelihood of aggression occurring in the context of their dating relationships. Additionally, women who are drinking heavily may experience discord or dissatisfaction in their dating relationships. For those individuals who do not know healthy techniques to resolve conflicts in interpersonal relationships, a pattern of aggression consistent with common couple violence may begin (Johnson, 1995
). Specifically, these dating partners may engage in verbal aggression as a result of relationship conflict that over time may lead to physical aggression (Schumacher & Leonard, 2005
The current study improved upon the methodology from previous research by investigating the concurrent and reciprocal longitudinal associations between heavy drinking and dating violence across a three year period. Several limitations of the current study, however, should be noted. Heavier drinking individuals were more likely to fail to complete surveys or drop out of the study than individuals who were not heavy drinkers. This pattern, however, was not observed with respect to dating violence. Individuals who reported a greater frequency of dating violence were no more likely to have missing data or drop out of the study compared to those who did not report dating violence. Because heavy drinkers are more likely to engage in other aggressive behaviors including dating violence (e.g., Murray et al., 2008
), missing data from these individuals may have resulted in lower estimates of dating violence. Relationship status was assessed at each semester survey, but it was not known whether aggression occurred with the same or different dating partners within and across assessment periods. Relationships during college, especially those that involve aggression, may be relatively brief, and could therefore influence any longitudinal relations between alcohol and aggression. If an aggressive relationship ended and participants were not in another romantic relationship during subsequent assessment periods, they had no opportunity for dating violence. Moreover, because the nature of relationships may change throughout the assessment period, the relation between alcohol and aggression may also change due to partner characteristics or how the couple handles conflict under the influence of alcohol. Additionally, these data were retrospective for the past 3-months, which may have influenced accurate reporting of heavy drinking and dating violence. Studies of both heavy drinking and intimate partner violence have demonstrated accurate recall for similar reference periods (Fals-Stewart, Birchler, & Kelley, 2003
; Searles, Helzer, Walter, 2000
). Future studies on heavy drinking and dating violence might include assessments that allow participants to report on behaviors for the previous day using daily monitoring procedures (e.g., Neal & Fromme, 2007
) or interactive voice response (e.g., Parks, Hsich, Bradizza, & Romosz, 2008
Another limitation of the current investigation is that we did not collect separate assessments of dating violence victimization and perpetration. Thus we cannot determine the frequency with which participants were the victim or the perpetrator of their dating violence experiences. Nevertheless, most aggressive dating relationships involve mutual aggression in which individuals are both perpetrators and victims (Capaldi & Owen, 2001
; Johnson, 1995
; Vivian & Langhinrichsen-Rohling, 1994
; White & Chen, 2002
). Thus, the failure to distinguish these roles may be less important than identifying their mutual occurrence with heavy drinking. Additionally, a history of dating violence has been linked to current dating violence and drinking behaviors, but we did not assess dating violence prior to college. A more detailed history of dating violence would have facilitated our interpretation of the relations we observed, especially during the freshman year. It should also be noted that the findings presented in this investigation are specific to college samples and are not generalizable to young men and women more broadly. In addition, other forms of interpersonal aggression such as verbal aggression or sexual assault were not included in the current analyses; therefore, these results are only specific to the relation between heavy drinking and physical dating violence.
Limitations notwithstanding, the current study addressed a much-needed area for intimate partner violence by examining the longitudinal relation between heavy drinking and physical dating violence for both men and women. Although not found at every assessment period, dating violence and heavy alcohol consumption were concurrently associated for men, and heavy drinking predicted subsequent dating violence for women. Future studies should examine the contextual and relationship factors that may relate to the occurrence of dating violence. As Johnson (1995)
suggested, there may be a pattern for partner violence that begins with a verbal argument and escalates by one or both partners to mild physical aggression. Event-level analyses would allow for an examination of these potential patterns as well as partner characteristics and relationship factors, such as satisfaction and commitment, which may influence the occurrence of dating violence.
Results of the current investigation provide important implications for interventions that target dating violence and heavy drinking. Efforts should be made to identify men and women who are drinking heavily and address their high level of alcohol consumption in order to prevent dating violence from occurring. Given the concurrent association between heavy drinking and dating violence for freshman men, education and intervention programs should target both of these harmful behaviors preferably prior to college matriculation. Specifically, programs should instruct men of the possibility that heavy drinking and dating violence can be associated and teach strategies to reduce the likelihood that either behavior will occur. Heavy drinking should be identified among women and targeted prior to or during their sophomore year of college to reduce the negative effects associated with heavy drinking, including the possibility of subsequent dating violence. Men and women should also be provided with information about how to establish and maintain healthy relationships, including effective ways to address conflict in dating relationships, and skills and support for ending unhealthy relationships.