Does the Life History Calendar Method Facilitate the Recall of Intimate Partner Violence? Comparison of Two Methods of Data Collection

Abstract

Violence perpetrated by intimate partners (intimate partner violence, IPV hereinafter) is prevalent, and often recurrent, in women's lives (Straus & Gelles, 1990; Tjaden & Thoennes, 2000). Gaining information about lifetime IPV is crucial to assess the cumulative effects of IPV on women's well-being over the life course. This study compared two methods of collecting data regarding lifetime experience of IPV: the Life History Calendar (LHC) method (Freedman, Thornton, Camburn, Alwin, & Young-DeMarco, 1988) and a commonly used structured interview using behavior-specific questions.The LHC method combines a visual calendar with a semistructured interview schedule to help the respondents gain better access to long-term memory. In most studies of IPV, women's self-reports are the sole data source because of limitations associated with other data sources. First, hospital and police records are incomplete as they are subject to the victims' and perpetrators' willingness to report the incident. Second, because IPV is often committed in privacy, verifying respondents' experiences of IPV is difficult unless couples are observed 24 hours a day. Such an approach not only poses enormous ethical and practical difficulties, but also would likely affect the occurrence of IPV. Third, studies consistently have found that individuals are less likely to report their own acts of abuse than those of their partners (Browning & Dutton, 1986; Saunders, 1991; Szinovacz, 1983; Szinovacz & Egley, 1995) and that the tendency to respond in a socially desirable manner is associated with lower likelihood of reporting a person's own use of violence (Arias & Beach, 1987; Dutton & Hemphill, 1992).Thus, the use of partners' reports is not suitable when researchers are interested in women's experience of partners' violence. In light of the consistent finding that social desirability responding is not associated with women's self-reports of experiencing partners' abusive acts (Arias & Beach; Dutton & Hamphill; Saunders, 1986), the use of women's self-report appears to be the best viable option for obtaining information on women's experiences, especially lifetime victimization experiences of IPV. LIMITATIONS OF COMMONLY USED STRUCTURED INTERVIEW METHODS IN ASSESSING LIFETIME EXPERIENCES OF IPV Recall difficulties can challenge the validity of self-report of lifetime IPV. Research on autobiographic memory and survey methodologies has consistently found that the longer the reporting period, the more likely underreporting and inaccuracy (Cannell, Fisher, & Bakker, 1965; Cash & Moss, 1972; Rubin & Wenzel, 1996; Thompson, Skowronski, Steen, Larsen, & Betz, 1996; Wagenaar, 1986). However, studies of lifetime IPV have not incorporated devices to facilitate respondents' recall of lifetime IPV. For example, researchers typically use an existing instrument in its original format, or modified versions, such as the Conflict Tactics Scales (CTS) (Straus, 1979) and CTS2 (Straus, Hamby, Boney-McCoy, & Sugarman, 1996), to examine lifetime IPV experience, even though these instruments are designed to assess respondents' experiences within a one-year or shorter time frame. As a result, these instruments lack the adjustments necessary to facilitate respondents' recall, and reports of lifetime experiences of IPV may have been incomplete in previous studies. To our knowledge, no studies have examined the validity of these existing scales as a function of the length of a recall period; however, preliminary evidence suggests that these instruments may not capture the full extent of lifetime IPV experiences. Several studies found higher lifetime prevalence of partners' physical violence among younger respondents than among their older counterparts who had been exposed to risk of such abuse for a longer period of time (Lloyd & Taluc, 1997; Smith, 1990). As Yoshihama and Gillespie (2002) showed, this seemingly contradictory finding may result in part from problems with recall, especially with older respondents. …