Utility of the General Validity Scale Model: Development of Validity Scales for the Co-parenting Behavior Questionnaire

Abstract

UTILITY OF THE GENERAL VALIDITY SCALE MODEL: DEVELOPMENT OF VALIDITY SCALES FOR THE CO-PARENTING BEHAVIOR QUESTIONNAIRE Kimberly M. Parker, B.A. A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science at Virginia Commonwealth University. Virginia Commonwealth University, 2010 Director: Dr. Arnold L. Stolberg Emeritus Professor of Psychology Validity scales for child-report measures are necessary tools in clinical and forensic settings in which major decisions affecting the child and family are in question. Currently there is no standard model for the development and testing of such validity scales. The present study focused on 1) creating the General Validity Scale (GVS) Model to serve as a guide in validity scale development and 2) applying this model in the development of validity scales for the Coparenting Behavior Questionnaire (CBQ), a child-report measure of parenting and co-parenting behaviors for children whose parents are divorced. Study 1 used the newly developed GVS Model to identify threats to CBQ validity and to develop procedures for detecting such threats. Four different validity scales were created to detect inaccurate responding due to 1) presenting mothering, fathering, and/or co-parenting in an overly negative light, 2) rating mothering and fathering in a highly discrepant manner, 3) inconsistent item responses, and 4) low reading level. Study 2 followed the GVS Model to test the newly developed scales by comparing CBQ responses produced under a standard instruction set to responses from contrived or randomly generated data. Support for the ability of each validity scale to accurately detect threats to validity was found.