Advancing survey science for intimate partner violence: The Partner Victimization Scale and other innovations.

Abstract

This reply addresses the key points raised by Straus as well as Jouriles and Kamata in their commentaries on Hamby (2016), including (a) that the Partner Victimization Scale (PVS) has already shown incremental validity because it has demonstrated a well-established form of validity, multimethod convergence, which some self-report measures cannot show, in addition to data on reliability and construct validity; (b) that it is not uncommon for new scientific technologies to lead to improvements in sensitivity as well as specificity, (c) that the PVS is a measure of intimate partner violence, not a measure of physical assault, which is why it also includes sexual violence (although gender parity is not found for the physical assault items); and (d) that the PVS does not refer to fear or any related terms. Additional data have replicated the PVS findings from Hamby's Study 4, and new findings from other researchers have also shown that changes in item wording can bring intimate partner violence (IPV) self-report in line with other indicators regarding gender patterns. Of importance, the items on at least 2 of these methodologies, the PVS and the new Youth Risk Behavior Survey, increase disclosure of victimization by females. The conceptual basis for understanding how improved scientific technology can increase sensitivity and specificity is presented. It is an exciting time in IPV measurement because several alternatives that address the decades-old controversy in multimethod divergence in gender patterns are now available. It is hoped that more scientific innovation will occur in the future. ©2016 American Psychological Association Language: en