Psychological and Demographic Predictors of Attrition Among Court-Ordered Batterers

Fred Buttell
College of Social Work
University of South Carolina
Columbia, SC 29208
(803) 777-8495
FAX: (803) 777-3498
fred.buttell@sc.edu
 
Michelle Carney
College of Social Work
University of South Carolina
Columbia, SC 29208
(803) 777-2147
FAX: (803) 777-3498
michelle.carney@sc.edu
Recently, a growing body of literature has focused on characteristics of men who batter and on the intervention programs directed at changing their violent behaviors.  Reviews of the batterer intervention program literature have consistently indicated that 60% to 80% of male batterers who complete treatment are no longer physically abusive towards their partners (e.g., Holtzworth-Munroe, Bates, Smultzer, & Sandin, 1997).  However, as all of these reviews indicate, the positive, single-site, program evaluations are confounded by high attrition rates. Specifically, researchers have consistently demonstrated that 40% to 60% of men attending the first session of batterer treatment actually fail to complete treatment(Gondolf, 1997). Consequently, in an effort to enhance retention rates by understanding why abusive men drop-out of treatment, researchers have attempted to identify characteristics that distinguish treatment completers from drop-outs (e.g., Hamberger, Lohr, & Gottlieb, 2000).
 
The purpose of this study was to: (a) investigate differences in demographic variables (i.e., age, employment status, educational attainment, marital status, alcohol and drug use, and arrest history) and psychological variables (i.e., social desirability, assertiveness, control of partner, and abusiveness) between treatment completers and drop-outs among abusive men entering a court-mandated treatment program; and, (b) create a predictive model that would correctly identify men at greatest risk of dropping out of the program.  The study employed a secondary analysis of 237 men, 127 treatment completers and 110 drop-outs, randomly selected from a larger pool of 784 men court ordered into treatment for domestic violence offenses.  Analysis indicated that very few of the demographic and psychological variables differentiated between treatment completers and drop-outs.  However, a logistic regression model was developed that correctly predicted treatment completion for 75% of the sample, which represented a 33% improvement over chance.  Implications of the findings for improving retention rates among court-ordered batterers were explored and discussed.
 
References