NCJ Number
200183
Journal
Domestic Violence Report Volume: 8 Issue: 4 Dated: April/May 2003 Pages: 49-50,61,62
Date Published
April 2003
Length
4 pages
Annotation
This paper identifies methodological problems and presents other findings that weaken the conclusion of the Broward Experiment that there is "no benefit from batterer counseling" in cases of domestic violence.
Abstract
The Broward study is increasingly being used to oppose batterer programs in general and reinforce objections to funding them. There are several methodological issues and shortcomings, however, that would undercut the rigor and science of the study. There are measurement shortcomings that potentially challenge the claim of a "valid and rigorous test." The "no effect" regarding batterer counseling in the Broward Experiment could result from problems with the attitude tests, victim response rates, and the meaning of rearrests. The New York City and San Diego Navy experiments with batterer counseling have their limitations. There are also problems with the interpretation of the study findings. If in fact there was "no effect," this result could be caused by weak reinforcement by the court and probation, rather than by the counseling itself. The Duluth Model, which was tested in the Broward Experiment, involves gender-based cognitive-behavioral counseling within a highly integrated system of monitoring and sanctions for the men, as well as advocacy and support services for the woman victim. This system is not likely to have been fully implemented or maintained in the Broward Experiment. Experimental designs, such as the one used in the Broward Experiment, have difficulty in implementation and in accounting for program context and complexity. Other useful approaches for evaluating batterer programs are realistic, action-oriented, and fourth-generation evaluations (e.g., Dobash and Dobash, 2000; Guba and Lincoln, 1989; and Pawson and Tilly, 1997). 19 references