Paper
Saturday, July 24, 2004
This presentation is part of : Method Testing
A Psychometric Examination of English and Spanish Versions of The Revised Conflict Tactics Scales
Cynthia D. Connelly, PhD, 1. Child and Adolescent Services Research Center (CASRC), San Diego, CA; 2. University of San Diego Hahn School of Nursing, San Diego, CA, USA, Rae R. Newton, PhD, 1. California State University Fullerton - Department of Sociology; 2. Child and Adolescent Services Research Center, San Diego, CA, USA, and Gregory A. Aarons, PhD, Child and Adolescent Services Research Center, San Diego, CA, USA.
Learning Objective #1: Identify alternative interpretations regarding the validity of the five-factor model and the generalizability of the CTS2 across language groupings
Learning Objective #2: Understand that simply combining results across language groupings may obscure important differences in rates of endorsement and patterns of responses reflecting theoretically important cultural, educational, and economic differences
Purpose: The Conflict Tactics Scale-Revised (CTS2) was developed in response to the critique and recommendations to improve and enlarge the Conflict Tactics Scales (CTS) measure of intimate partner violence (IPV). Translations have been developed in several languages; however, psychometric findings have only been reported for the English version. For research on IPV to progress with scientific rigor, the reliability and validity of existing measures must be empirically supported for use with ethnically diverse samples and with non-English speaking populations. The present study examined the psychometric properties of the CTS2, based upon the responses of English (n=211) and Spanish (n = 194) speaking Latina women.

Method: Secondary data analysis was conducted on data provided by 395 community based, self-identified Latinas, (mean age of 34.9 years, SD = 11.37), living in San Diego, California. Inclusion criteria: completion of the CTS2 and missing no more than two CTS2 items.

Results: Confirmatory factor analytic results supported a-priori, a five-factor model, specifying factors of negotiation, minor psychological aggression, severe psychological aggression, minor physical assault, and severe physical assault. In unconstrained two group models, loadings were of similar magnitude across language of administration, with the exception of the physical assault scales. Unconstrained and constrained model comparisons showed scale structure varied significantly by language group for physical assault. Internal consistency of total scale scores was satisfactory (Cronbach’s alpha .70-.84), however, subscale alpha’s ranged from .46 - .80.

Conclusions: While results of our study show overall comparability for English and Spanish speaking Latinas, we note that simply combining the results across language groups may obscure important differences in rates of endorsement and patterns of responses that may reflect theoretically important cultural, educational, and economic differences.

Research supported by NIDA K01-DA15145, NIMH MH01695, MH50313, MH55282, and California Wellness Foundation, Stuart Foundation, and Department for Social Services, State of California.

Back to Method Testing
Back to 15th International Nursing Research Congress
Sigma Theta Tau International
July 22-24, 2004