The performance of male prisoners on the Marlowe-Crowne social desirability scale: II. Differences as a function of race and crime

Abstract

PROBLEMThe Marlowe-Crowne Social Desirability Scale rS purportedly measures social desirabilit,y free from contamination by psychopathology. The items were chosen from a population of items "defined by behaviors which are culturally sanctioned and approved but which are improbable of occurrence"'2, p. 350). They were judged to be indicative of neither psychological health nor maladjustment, when answered in the socially undesirable direction.The 31-C has been interpreted as a measure of iieed for social approval, but studies have variously supported(*) and rejected this view(5). The hypothesis also has been advanced and supported(6) that the M-C measures a form of situational discrimination, i.e., people high in SD determine the appropriate response separately for each situation they face. Other studies have suggested that the IM-C is a measure of vulnerable self-esteem (') and of repressive ego-defensiveness ( l ) .This study investigated the relationship between type of criminal offense, race, intelligence, educational achievement level, age, and M-C scores, and in addition determined the correlations between 13 MMPI Scales arid the RI-C. METHODThe sample consisted of 492 White, 108 Mexican-American and 182 Negro male prisoners drawn from a population being processed for prison placement at the Southern California Reception-Guidance Center of the Department of Corrections. Ss were tested within four weeks of their arrival and all 8s were included who were able and willing to take the RIMPI. The hl-C was given as part of a larger test battery and was administered by inmates, some of whom were Negro and Mexican-American. Intelligence was assessed by the Army General Classification Test (AGCT). Educational achievement level was measured by the California Achievement Tests. Criminal offenses were categorized as follows: white collar, e.g., embezzlement, forgery, N = 127; assault, N = 208; burglary, N = 267; rape (excluding statutory rape), N = 8; sex perversion, N = 59; drugs (possession and sales of narcotics), N = 104; and involuntary manslaughter, N = 9. RESULTSAND DISCUSSION Age (M = 29.9, S D = 9.24) was not significantly correlated with the M-C ( r = .05). I& ( A 1 = 103.5, S D = 14.14) and educational achievement level (M = 8.50, S D = 1.48) were both significantly correlated with the M-C (r = -.23 and -.27, respectively; p < .001). An analysis of variance of M-C scores (A4 = 19.2, S D = 6.37) by the seven crime categories indicated that the groups did not differ significantly in their scores (F = 1.15, df = 6,775, NS). An analysis of variance of M-C scores by race indicated that the three race groups did differ significantly in their scores (F = 16.47, df 2,779, p < .001). The scores for the three race groups were: White, M = 18.2, SD = 6.45; Mexican, M = 21.5, SD = 5.30; Negro, M = 20.4, SD = 6.19. A series of t tests indicated that the Whites differed from both the Mexican and Negro groups ( t = 4.91 and 3.98, respectively; p < .001) but that the Mexicans and Negroes did not differ from each other ( t = 1...

The Impact of Ethnicity and Response Bias on the Self‐Report of Negative Affect1

“…This underreporting may be conscious or unconscious; however, the MCSDS does not make this level of discrimination. The MCSDS has shown very good internal consistency, with estimates of Cronbach's alpha ranging from .73 to .88 (Crowne & Marlowe, 1964;Fisher, 1967;Paulhus, 1984;Tanaka-Matsumi & Kameoka, 1986).…”

“…The few experiments that have examined ethnic differences in response bias have usually found that Blacks report higher levels of social desirability or defensiveness than Whites (Adams & Krasnoff, 1989;Fisher, 1967;Goldsmith, Stith, & White, 1987;Klassen et al, 1975;Vernon, Roberts, & Lee, 1982). However, at least two experiments did not find this response bias (Shapiro, Goldstein, & Jamner, 1996;Welte & Russell, 1993).…”

Clinical Relevance of the Marlowe-Crowne Scale: A Review and Recommendations

“…This section summarizes results currently available. Fisher and Parsons (1962) and Fisher (1967Fisher ( , 1969 have reported relatively high MCS means (16.73, 19.10 and 19.20) for various inmate samples. Combined n of these studies was 1017, with average age of subjects being about 29.…”

“…Combined n of these studies was 1017, with average age of subjects being about 29. Consistent with racial differences reported in the general population by Klassen et al (1975), Fisher (1967 found blacks to score higher than whites. In other research (Pettijohn, 1977) prisoners scored quite low on the MCS (M= 12.63, n = 38), although these subjects, unlike Fisher's groups, were enrolled in college coursework.…”

Intellectual Ability Correlates of the Marlowe-Crowne Social Desirability Scale

“…The negative correlation with intelligence (Quick Test) is, for example, consistent with results from samples of children (cf. Crowne, 1979) and adult prisoners (Fisher, 1967) where different intelligence measures were used. Interestingly, the only similar study of college students (Dickstein & MacEvitt, 1971) failed to find such MCS related differences, a fact perhaps attributable to the substantially high ability levels present in that sample of students (mean WAIS Verbal 1Q= 135.03).…”