versión On-line ISSN 1668-7027
COSENTINO, Alejandro C. y CASTRO SOLANO, Alejandro. Argentinian adaptation and validation of the Marlowe-Crowne Social Desirability Scale. Interdisciplinaria [online]. 2008, vol.25, n.2, pp. 197-216. ISSN 1668-7027.
Crowne and Marlowe (1960) developed the Marlowe-Crowne Social Desirability Scale (MCSDS) that is a widely used scale to measure social desirability defined as "the need of subjects to obtain approval by responding in a culturally appropriate and acceptable manner" (p. 353). Several studies from psychology and medicine areas have included the MCSDS. The MCSDS is used both to asses and to control for response bias in a socially desirable manner as well as to measure psychological concepts, such as defensiveness or need for approval. Moreover, the MCSDS is the standard measurement to categorize the four coping styles from Weinberger, Schwartz, and Davidson model (1979). Since the MCSDS first appeared, investigators have proposed several changes such as, short forms, computerized administrations, translations and adaptations to diverse cultures. In this study of development of the Escala de Deseabilidad Social de Crowne y Marlowe (EDSCM) is described. The EDSCM Scale is an adaptation to the Argentinian culture of the original MCSDS full scale with a paper-and-pencil administration form. A pilot study with a sample of Psychology student volunteers was conducted to adapt the MCSDS to Argentinian culture. One hundred nineteen psychology college students (93 women and 26 men (M = 28 years; SD = 9.6) from two universities of the Ciudad Autónoma de Buenos Aires (República Argentina) volunteered to participate in reliability and validity studies of the EDSCM Scale. The student sample participants completed the EDSCM, the Beck Depression Inventory-Second Edition (BDI-II), and the Eysenck Personality Questionnaire (EPQ). For this student sample, the EDSCM scores ranged from 0 to 27, with a mean of 14.13 (SD = 4.99). EDSCM scores were normally distributed with no outlier values. The internal scale reliability of the EDSCM measured with Cronbach's alpha was .76. No statistically significant mean difference between sexes was found for the EDSCM Scale. A statistically significant association between the L Scale of EPQ and the EDSCM, with a large correlation effect size, provided convergent validity evidence for the EDSCM Scale. With regard to the divergent validity of the EDSCM Scale, a small to medium correlation effect size between the BDI-II and the EDSCM provided evidence that the EDSCM is a social desirability scale independent from psychopathology. A sub-sample of students (n = 52) who previously completed the EDSCM Scale, completed the EDSCM again with instructions to imagine being in a hiring process. Comparing the normal (M = 15.94; SD = 4.81) and the differential instruction condition (M = 28.67; SD = 4.67) a statistically significant increase in the EDSCM scores was detected. To provide known-groups validity, the EDSCM scores derived from an adult sample were compared with the EDSCM scores derived from an applicant sample. The adult sample included 120 participants (73 women and 47 men; M = 35.6 years; SD = 15.9) and the applicant sample included 134 people (61 women and 73 men; M = 28.9 years, SD = 3.4). Both for the adult sample and the applicant sample, there was not statistically significant sex difference for the EDSCM Scale. The applicant condition (M = 23.66; SD = 5.37; a = .75) showed statistically significant higher EDSCM scores than the normal condition (M = 16.09; SD = 5.18; a = .84). In this article, the validity and reliability of the EDSCM Scale have been analyzed, as indicated by the study of convergent and divergent validities, differential instructions technique, and known-groups technique, as well as internal consistency analyses. All consdired, the findings presented appropriate psychometric properties of the Argentinian culture adaptation of the MCSDS. The use of EDSCM Scale is suggested for researches related to psychology and medicine areas in Argentinian samples.
Palabras llave : Marlowe-Crowne Social Desirability Scale; Adaptation; Argentina; Psychometric properties; Foreign language translation; Test validity; Test reliability.