Development of the Credibility of Claim and Credibility of Victim Scales: Assessing Law Enforcement Officers' Determination of Rape Claim Credibility

Background and Purpose: Victims of rape must interact with law enforcement officers (LEOs) to obtain justice. LEOs are entrusted to make determinations about the credibility of a rape claim and whether the case should proceed to prosecution. LEOs are often accused of endorsing stereotypic and harmful beliefs about rape claims. Several scales exist to measure endorsement of overt rape myths, however these are not an accurate measure of the subtle myths LEOs may hold.

Research suggests that LEOs believe there is such thing as a “real” rape and an “ideal” victim. The credibility of a rape claim is linked to how much the claim adheres to these standards. The more a rape claim differs from the standard the less credible the claim is perceived to be. Currently, there is no measure to assess these constructs – credibility of the situation and credibility of the victim.

The purpose of this investigation was to develop and validate two measures- Credibility of Claim and Credibility of Situation. In practice, such measures can be used to design and assess the effectiveness of victim-focused trainings and interventions for LEOs aimed at reducing endorsement of these stereotypes.

Methods: Two scales were developed to measure subtle rape claim stereotypes, the Credibility of Victim Scale (CVS) and the Credibility of Situation Scale (CSS). Each consisted of 6 subscales, scored on a 10-point semantic-differential. For the CVS, respondents were asked to indicate how credible a “real rape” claim would be coming from victims who varied on their ideal victim characteristics. For the CSS, respondents were asked to indicate how credible a rape claim coming from a “ideal victim” would be as it varied on “real rape” characteristics. After the item pool was revised based on expert feedback, the survey was administered to law enforcement officers and criminology majors at a major university.  A total of 101 usable surveys were collected using a three-tiered purposive sampling method.

Results: Subscales on the CVS demonstrated acceptable reliability (ranging from .89 to .98), however a global stratified alpha was unacceptably low (.61). Reliability alphas for the CSS subscales were all acceptable (ranging from .90 to .98) and the global stratified alpha was acceptable (.76).  Using a confirmatory factor analysis, the CVS failed to reach acceptable levels on four out of six fit indices examined, even after model respecification. The factor structure of the CSS was more successful, as it reached acceptable levels on three out of the six fit indices. Due to poor reliabilities and factor structure results, tests of construct validity were not performed.

Conclusions and Implications: This investigation provided initial evidence of two constructs pertaining to rape myths – Credibility of Victim and Credibility of Situation. While not completely successful, the results did not suggest a complete failure. More work can be done with the item pools and factor structure to yield results that are statistically and conceptually acceptable. The results suggest that these two constructs may be a right step toward the goal of measuring an elusive and complicated concept.