Academic journal article Violence and Victims

Capture-Recapture Analysis of Batterer Reassaults: An Epidemiological Innovation for Batterer Program Evaluation

Academic journal article Violence and Victims

Capture-Recapture Analysis of Batterer Reassaults: An Epidemiological Innovation for Batterer Program Evaluation

Article excerpt

Capture-recapture analysis is an analytical tool being used in epidemiological research to estimate incidence rates for missing cases and unreliable reporting. Its contribution to batterer program evaluations is examined through a capture-recapture analysis of a multisite evaluation of batterer programs (n = 853). Capture-recapture analysis is applied to various subsamples of the multisite study and to different outcomes (i.e., any reassault, "severe" reassault) to explore the utility of capture-recapture. Finally, the capture-recapture estimates are compared to only the women's reports and to adjusted women's reports (women's reports supplemented with arrest records and men's reports), which are the basis of existing batterer program evaluation. The capture-recapture reassault estimate for the 15-month follow-up is equal to the adjusted women's reports (39%), but is 7% greater than the women reports. The subsample estimates for the individual sites appear to vary as a result of unreliable or unavailable arrest records.

The evaluation of batterer programs has dramatically increased in recent years, but has been compromised by several methodological shortcomings (Gondolf, 1997a). One of the central shortcomings is the difficulty in establishing the incidence rate for reassault among batterer program participants. Most of the previous program evaluations have relied exclusively on the women's reports of reassault since the men tend to deny or minimize their violence (Tolman & Bennett, 1990). Some recent evaluations have adjusted the women's reports with reports of reassaults from their batterers and from arrest records (Gondolf, 1997b; Saunders, 1996). Low response rates in follow-up samples of typically 30% - 50%, however, make it difficult to ascertain a reliable reassault rate for the full sample (DeMaris, 1989; Heckert & Gondolf, 1997; Szinovacz & Egley, 1995). Even with a respectable response rate of over 70%, the underreporting of reassault and no response from those more likely to reassault may artificially lower the reassault rates.

An alternative analytical technique has been developed in epidemiology to project an incidence rate for an entire sample-including missing cases. Capture-recapture assumes that any incidence or prevalence study will inevitably have missed cases. The number of missed cases needs to be addressed through statistical analyses that estimate a rate for the entire sample including missed cases. It is difficult to compare reassault rates across studies, sites, or time without these estimates. Some capture-recapture proponents insist that any research report or grant proposal addressing incidence or prevalence rates should not be accepted without a formal statistical plan to assess and control for ascertainment problems (International Working Group, 1995a; Sudman, Monroe, & Cowan, 1988).

Capture-recapture analysis was devised to project prevalence rates of diseases in populations where under- and over-reporting persists and the entire sample cannot be reached (International Working Group, 1995a, 1995b). It has recently been applied to projecting rates of injuries among adolescents, as well as the prevalence of diabetes in developing countries (Fienberg, 1992; McCarty, Tull, Moy, Kwoh, & LaPorte, 1993). Capture-recapture has not as yet been applied to batterer reassault, but would likely make a substantial contribution to ascertaining reassault rates in batterer program evaluations. It would estimate a reassault rate for the full sample by combining women's reports with any man's report (regardless if the man's partner were reached during follow-up), and with arrest records for the entire sample of batterers during follow-up.

To explore the contribution of capture-recapture analysis, we applied this analysis to reassault data collected in a multisite evaluation of batterer programs. Our objective was (1) to determine the projected reassault rates for the full sample at each site, (2) to assess the difference between the reassault rate based on adjusted women's responses and the estimated reassault rates based on capture-recapture analysis, and (3) to identify any methodological issues that emerge in the process. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.