Date of Award
Doctor of Philosophy (PhD)
Virginia Consortium Program in Clinical Psychology
Richard W. Handel
Scott M. Debb
Jennifer M. Flaherty
Serina A. Neumann
When respondents to self-report measures fail to answer in accordance with item content it distorts the accuracy of obtained test scores, degrading the ability of clinicians to use results to make accurate diagnoses and recommendations. The Variable Response Inconsistency (VRIN) scale was created for the MMPI-2 and was later revised as the VRIN-r scale for the MMPI-2-RF and MMPI-A-RF in order to detect and invalidate protocols over saturated with random responding. Analyses conducted by Pitta (2016) revealed that the VRIN-r scale for the MMPI-A-RF was not ideally sensitive to the detection of random responding in protocols. This study explored adding items drawn from another validity scale used to detect over reporting on the MMPI-A-RF (the F-r scale) to the VRIN-r scale to ascertain if the resultant hybrid scale was more sensitive to random responding. Using the MMPI-A-RF normative sample and an over reporting sample from the MMPI-A, analyses were conducted to identify the ideal number of F-r items to add to the VRIN-r scale that maximized sensitivity to random responding while maintaining specificity from detecting over reporting. F-r items were also added to another validity scale, the Combined Response Inconsistency (CRIN) scale, to ascertain if this hybrid scale was reliably sensitive to random responding. Analyses revealed that adding six F-r items to both the VRIN-r and CRIN scale greatly improved their sensitivity to random responding while maintaining adequate specificity from the detection of over reporting. Implications for these findings, limitations to the research design, and areas of future research are discussed.
LaTulip, Stefan E..
"Improving the Sensitivity of the VRIN-r Scale on the MMPI-A-RF"
(2019). Doctor of Philosophy (PhD), Dissertation, Psychology, Old Dominion University, DOI: 10.25777/azvz-2597