Gender differences in social desirability and social approval bias in dietary self-report

Social desirability (the tendency to respond in such a way as to avoid criticism) and social approval (the tendency to seek praise) are two prominent response set biases evident in answers on structured questionnaires. These biases were tested by comparing nutrient intakes as estimated from a single 24-hour diet recall interview (24 HR) and a 7-day dietary recall (7DDR). Data were collected as part of the Worcester Area Trial for Counseling in Hyperlipidemia, a randomized, physician-delivered nutrition intervention trial for hypercholesterolemic patients conducted in Worcester, Massachusetts, from 1991 to 1995. Of the 1,278 total study subjects, 759 had complete data for analysis. Men overestimated their fat and energy intakes on the 7DDR as compared with the 24HR according to social approval: One unit increase in the social approval score was associated with an overestimate of 21.5 kcal/day in total energy intake and 1.2 g/day in total fat intake. Women, however, underestimated their dietary intakes on the 7DDR relative to the 24HR according to social desirability: One unit increase in the social desirability score was associated with an underestimate of 19.2 kcal/day in energy intake and 0.8 g/day in total fat. The results from the present study indicate that social desirability and social approval biases appear to vary by gender. Such biases may lead to misclassification of dietary exposure estimates resulting in a distortion in the perceived relation between health-related outcomes and exposure to specific foods or nutrients. Because these biases may vary according to the perceived demands of research subjects, it is important that they be assessed in a variety of potential research study populations.     

Self-appraisal, personal desirability, and perceived social desirability of personality traits.

200 college sophomores took the MMPI 3 times. Analysis showed the following results: (1) Variation of order or time of testing in the personal or social condition seemed to make no difference. (2) Intercorrelations between the 3 forms were low enough to show that the 3 tasks were different. (3) Variance on all 3 forms was lower than in the standardization of the MMPI. (4) The mean profiles were highly similar in all 3 groups. (5) The clinical scales were in general highest on self, as were the Si, A, and obvious scales. The L, K, and subtle scales were in general lowest on self. (6) Results based on analysis of item discrepancies rather than scale comparisons showed that the discrepancies tended to be very sizable. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Hopelessness, social desirability, and construct validity.

M. M. Linehan and S. L. Nielsen (see record 1982-01443-001) questioned the construct validity of the Hopelessness Scale on the basis of its covariation with the Edwards Social Desirability Scale. The present author suggests that social desirability (SD) should not be invoked as a potential confound unless the obtained covariation is theoretically inconsistent or is so overlapping as to make the respective scales redundant with respect to factorial content. It is argued that the "sacred cow" status of SD as a potential confound to the interpretability of psychological tests should be reexamined. (9 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The frequency of self-reference statements as a function of generalized reinforcement.

Skinnerian-oriented research has demonstrated that interpersonal verbal behavior may be influenced by conditions of operant conditioning. Within this framework, the present research tests the effect of an interviewer's attention and assent on the frequency of self-reference statements made by an interviewee. These interviewer variables were observed to function as a reinforcing state of affairs without the interviewee being aware of the conditions of the experiment. Several explanations are offered for the results. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Relationship between probability of endorsement and social desirability scale value for a set of 2,824 personality statements.

A group of 47 male and 48 female judges rated 2824 personality statements for social desirability using a 9-point rating scale. Another group of 110 male and 111 female Ss described themselves in terms of the same set of 2824 statements by answering each "true" or "false." The correlation between probability of a "true" response and social desirability scale value for the combined sex groups was .892. The distribution of the social desirability scale values of the 2824 statements was distinctly bimodal. These results are in accord with another large-scale study in which 1647 personality statements were investigated. In view of the large number of personality statements involved in these 2 studies, it is suggested that a correlation of .90 between probability of endorsement and social desirability scale value and a bimodal distribution of the scale values of personality statements may be characteristic of the population. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Effect of social desirability instructions on several MMPI measures of social desirability.

Wiggins (1959) evaluated several MMPI measures of the social desirability variable in a role-playing experiment, and concluded that his Sd scale was the preferred measure. The effectiveness of several social desirability measures in detecting role-playing subjects was cross validated in a sample of psychiatric patients by the contrasted groups method. Contrary to Wiggins' conclusion, the results indicated that the Sd, the Cof, the SD, and the MMPI L and K scales were of comparable efficiency in discriminating between the control and experimental conditions in the role-playing experiment, even though Sd and Cof were specifically designed for this purpose. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Construct validity and social desirability.

Comments on the article by R. Hogan and R. A. Nicholson (see record 1989-00097-001) on the construct validity of personality tests and argues that Hogan and Nicholson attributed to A. L. Edwards (1953) conclusions that Edwards never made, concerning the use of the Minnesota Multiphasic Personality Inventory (MMPI) and the influence of social desirability. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Repression-sensitization and social desirability responding.

3 groups of male and female Ss were given Byrne's R-S scale and the Marlowe-Crowne Social Desirability (SD) scale. The scores on these 2 scales were correlated and significant negative correlations were obtained for all 3 groups. The magnitude and consistency of these correlations appear to reflect sufficient independence between these dimensions to justify considering them as separate variables. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Social-learning theory, social desirability, and the MMPI.

Recent factor analytic interpretations of the MMPI have emphasized response sets as sources of variance, and the tendency to answer items in socially desirable or undesirable ways has been interpreted as the primary factor. It is contended that the primary factor is better interpreted as degree of psychopathology and both social-learning theory and empirical data are presented to support this view. In brief, socially desirable behaviors are typically those which conform to group standards, and the deviant behaviors which define psychopathology are usually socially undesirable. Accordingly, the tendency for MMPI scales to load inversely to social desirability on the primary factor is a consequent of the actual inverse relationship between deviancy and social acceptability of behavior. (61 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Person perception, marriage adjustment, and social desirability.

Extended B. Murstein's (see record 1971-26678-001) stimulus-value-role theory to account for the relationship of person perception scores to marital adjustment. The hypotheses predicted that (a) similarity, self-acceptance, accuracy of predicting other's responses, and role compatibility would be correlated with marital adjustment; (b) when the perceptual target was the man, the correlation with marital adjustment would be higher than when it was a woman; (c) women would show a higher self-acceptance-marital-adjustment correlation than men; and (d) intraperceptions would be more highly correlated with marital adjustment than interperceptions. 60 young married middle-class couples took a 20 adjective bipolar checklist under 8 different "sets" (self, ideal self, spouse, ideal spouse, and predictions of these 4 sets for the partner). Results generally support the hypotheses. (22 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Comment on social desirability.

Comments on R. Hogan and R. A. Nicholson's (see record 1989-00097-001) article on construct validation in personality measurement and argues that social desirability (SD) is not simply an early theory of item response. SD explains item endorsement in terms of social normative processes and accounts for the lack of differential validity often observed among personality scales. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Locus of control and social desirability.

Hypothesized that there would be a negative correlation between the Edward's Social Desirability Scale and Rotter's Internal-External Control Scale. 102 army mental health clinic outpatients, 218 army stockade prisoners, 99 alcoholic inpatients, and 48 new careers participants were administered both tests. All correlations were negative, ranging from -.7 to -.29, and were significant at the .025 level or greater. It is suggested that since internalizers feel that they have some control over reinforcements, they may try to influence the dispensers of these reinforcers by behaving in socially acceptable ways, while externalizers, who feel they have no control over reinforcements, may not be so motivated. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Individual viewpoints of social desirability.

The attempt to control for social desirability in personality inventories by means of a forced-choice technique tacitly assumes the unidimensionality of the construct. Utilizing a methodology which allows multidimensionality to emerge both within individuals and within stimuli, it is demonstrated that at least 6 individual viewpoints exist with respect to social desirability judgments of selected MMPI items. The perceptual spaces of the 6 "idealized individuals" differed from one another and from the perceptual space of the "average" individual. It is concluded that controlling social desirability by the pairing of items on the basis of "average" social desirability scale values, as in the forced-choice technique, is an inadequate control at the individual level. (43 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A redefinition of social desirability.

A redefinition of an item cluster previously subsumed under the category "social desirability" to "desirability for well-being." (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Mastery motivation and the need for approval in older children and their relationship to social desirability response tendencies.

The relative strength of mastery motivation and need for approval was tested among 2 groups of 11-yr-olds, 40 high and 40 low scorers on the Children's Social Desirability Scale. Mastery motivation was inferred from the amount of time that Ss spent in 1 of 2 discrimination tasks, a challenging (unsolvable) or a solvable problem. To measure need for approval, Ss were tested in either a social-reinforcement or an E-absent condition. As predicted, low scorers spent more time in the unsolvable than the solvable task, with negligible condition differences. High scorers spent more time in the social-reinforcement than in the E-absent condition, with negligible task differences. On the solvable task, high Ss demonstrated poorer learning than low Ss. Sex differences revealed greater mastery motivation for boys and greater need for approval for girls. The need to study the developmental course of these motives and to investigate the antecedents of the group differences obtained is emphasized. (29 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Agreeing response set and social desirability.

This is a rejoiner to 2 papers (see 36: 3HF73E, 3HF80E) which suggest that "social desirability set" had been confounded with an "agreeing response set"; Couch and Keniston attempted to confirm or reject this assumption. Factor analysis of responses to MMPI scales, Edwards' Social Desirability scale (SD), and Couch and Keniston's scale measuring agreement response set (OAS), indicated that SD and OAS responses sorted out into separate, orthogonally related factors. Moreover, a 3rd factor was discerned, suggesting that SD is measuring the admission or denial of anxiety rather than conformity to conventional values. Further research was called for to confirm the findings. From Psyc Abstracts 36:01:3HF75C. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Attitudes toward instructors, social desirability, and behavioral intentions.

Examined whether student attitude toward instructors was related to subsequent behavior, such as the behavioral intention to take further courses with that instructor, and whether social desirability affected this relationship. 158 undergraduates completed instructor ratings, and a short form of the Marlowe-Crowne Social Desirability Scale. Principal components analysis and varimax rotation isolated 6 instructor rating factors. Ss also participated in a mock preregistration procedure assessing their intention to reregister for the same or different instructors. Results indicate that the 1st instructor rating factor, Instructor Skill, correlated .72 with intention to take further courses with that instructor. No other relationships or interactions were significant. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Will the real social desirability please stand up? Hopelessness, depression, social desirability, and the prediction of suicidal behavior.

Articles by the 2nd author and S. Nielsen (see PA, Vols 67:1443 and 69:12939) and J. Nevid (see record 1983-11675-001) have examined and debated the clinical significance of the negative relationship between the Hopelessness Scale and the social desirability (SD) response style. K. Petrie and K. Chamberlain (see record 1983-32939-001) found that the Marlowe-Crowne Social Desirability Scale had no influence on the prediction of suicidal behavior. The present authors argue that this study did not address the question of SD response set in self-reported hopelessness, as the Marlowe-Crowne scale is factorially distinct from existing SD scales. The clinical utility of including SD assessment in the risk assessment and prediction of suicidal behavior was analyzed using both a general population sample (n?=?197) of Ss in a previous study by the 2nd author and Nielsen (1981) and a psychiatric sample (n?=?96) who participated in a study by the 2nd author et al (1983). Results indicate that prediction accuracy was enhanced by including SD assessment, particularly among psychiatric patients and when historical reports of suicidal behavior were involved. Nevertheless, using only hopelessness, depression, and SD assessments in suicide risk prediction still produced a dangerously high level of false negative cases in both general population and psychiatric samples. (21 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A children's social desirability questionnaire.

A Children's Social Desirability (CSD) questionnaire was constructed and administered to 956 Ss in Grades 3, 4, 5, 6, 8, 10, and 12. A direct question ("yes-no") form of the scale was presented to children in the 3rd, 4th, and 5th grades in individual testing sessions, and a true-false form was given older Ss in group sessions. For both forms of the questionnaire, split-half reliabilities and test-retest (1-month interval) reliabilities were high. Socially desirable responses were more frequently given by younger children than by older children, by dull Ss than by bright Ss, by girls than by boys, and by Negro children than by white children. Social class, size of family, and ordinal position were not associated with CSD responses. Questions were raised pertaining to the generality of this response and pertaining to the premises on which this measure of social desirability, as well as that of Crowne and Marlowe, is based. (PsycINFO Database Record (c) 2010 APA, all rights reserved)