Attitudes & effectiveness of professional programs for primary prevention of violence

Abstract

The PREVENT (Preventing Violence through Education, Networking, and Technical Assistance) project trained violence practitioners across the USA in Primary Prevention of Violence techniques (PPV). The purpose of this study is to describe the development and psychometric properties of the subscales of the PREVENT PPV Self Assessment. Of 800 participants, 585 responded (73.1%). We analyzed the data using factor analyses, reliabilities and LISREL structural equation measurement models. The Perceived PPV Project Success subscale exhibited a one-factor structure (R-2 = .534, X-2/df = 1.28, p = .277, GFI = .99, NNFI = .99); PPV Self-Efficacy and Support showed a four-factor structure (R-2 = .583, X-2/df = 3.7, p = .00, GFI = .94, NNFI = .93); Perceived PPV Collaboration showed a three factor structure (R-2 = .544, X-2/df = 4.20, p = .00, GFI = .92, NNFI = .93); Anticipated Future PPV Work yielded a three-factor structure (R-2 = .656, X-2/df = 4.76, p = .00, GFI = .94, NNFI = .91); and PPV Confidence a marginally acceptable two-factor structure (R-2 = .759, X-2/df = 11.1, p = .00, GFI = .95, NNFI = .99). Perceived PPV Project Success demonstrated very strong predictive validity (chi(2)/df = 2.33, p = .0167, GFI = .984, RMSEA = .0585) of increased time devoted to PPV. These construct validated subscales represent a rich source of material for assessing professionals' attitudes and perseverance towards personal implementation of PPV. From these results, we recommend specific items from each subscale for further use by PPV trainers