Abstract

The present study attempted to detect the faking-good response set effectively on the MMPI in social situations such as applying for a job and counseling, At intervals of one week, 105 male and 104 female college students took the MMPI under the standard instruction(ST), and again under the faking good instruction(FG), College male MMPI records were compared with 129 male MMPI records(J) which were gathered from the public official job entrance examination. In the present study, ten indices were used to detect the faking good response set on the MMPI. These were : L raw score ; K raw score ; F-K raw score ; L+K raw score ; Wiener's(1948) obvious items(WO) ; Wiener's(1948) subtle items (WS) ; Wiener's(1998) O-S score (WO-WS) ; Park, Yi, Ahn, and Kim's(1991) obvious items (MO) ; Park, et al.'s(1991) subtle items (MS) ; Park, et al.'s(1991) O-S score (MO-MS), The MMPI profiles of J and FG were obviously different from that of ST, the profile of J was similar to that of FG. In the MMPI profile of FG, L ways higher than 70, F was lower than 40, K was higher than 60, and scale 5(Mf) and scale 9(Ma) were higher than any other scale, scale 0(Si) was lower than any other scale. In discriminating ST and FG for each faking good detection index, the optimum cutting scores which had the highest hit rates were acquired. As a result, except for WS and MS, all detection indices were expected to detect effectively the faking good response set on the MMPI. Compared to Wiener's subtle-obvious scale for the effectiveness of the faking good detection, Park, et al.'s(1991) subtle-obvious scale showed little difference. Contrary to WS and MS, WO and MO were considered to be valuable as the faking good detection indices.