Association of fake-good MMPI-2 profiles with low Beck Depression Inventory scores.

Scafidi, Frank A. ; Field, Tiffany ; Prodromidis, Margarita 等

Women are especially prone to depression following childbirth (Brockington & Kumar, 1982; Gotlib, Whiffen, Wallace, & Mount, 1991; Hopkins, Marcus, & Campbell, 1984). The prevalence of postpartum depresssion has been found to range from 7% to 33% (Gotlib, Whiffen, Mount, Milne, & Cody, 1989; O'Hara, Neunaber, & Zekoski, 1984). Several investigators have suggested that maternal depression is associated with early mother-infant problems and with emotional and behavioral problems in children (Caplan, Coghill, Alexandra, Robson, Katz, & Kumar, 1989). Depressed mothers are less active, less playful, and less responsive during face-to-face interactions (Field, Sandberg, Garcia, Vega-Lahr, Goldstein, & Guy, 1985; Cohn, Campbell, Matias, & Hopkins, 1990). At a later age, children of depressed mothers perform less well on the Bayley mental scales, exhibit more negative emotions, and have more emotional and behavioral problems than do children of nondepressed mothers (see Zuckerman & Beardslee, 1987, and Field, 1995, for reviews).

Since postpartum depression seems to have a negative impact on mother-infant interactions and on developmental outcome, it is important to identify mothers who exhibit depressive symptoms. The Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961) is the most frequently used self-report instrument for identifying postpartum depression (Field et al., 1985; Gotlib et al., 1991; O'Hara et al., 1982; Pfost, Stevens, & Lum, 1990; Steer, Scholl, & Beck, 1990). The BDI has high convergent validity with psychiatric ratings of depression severity (Beck et al., 1961; Bumberry, Oliver, & McCLure, 1978). Although it has high sensitivity and specificity for detecting clinical depression (Barrera & Garrison-Jones, 1988; Oliver & Simmons, 1984), there is a paucity of research examining extremely low BDI scores.

Mothers with extremely low scores on the BDI (total score = 0, 1, or 2) exhibit more depressed behavior in face-to-face interactions with their infants than do mothers with high scores (an indicator of depression) (Field, Morrow, Healy, Foster, Adelstein, & Goldstein, 1992; see also Lyons-Ruth, Zoll, Connell, & Grunebaum, 1986, for results using the Center for Epidemiological Studies Depression Scale). Several possible explanations for these unexpected findings have been offered, including denial of symptoms, defensiveness, or a need to look good to others (Field et al., 1992). The purpose of the present study was to investigate this phenomenon. Thus, the validity scales of the Minnesota Multiphasic Personality Inventory 2 (MMPI-2) were administered to determine whether low-BDI mothers were "faking good."

METHOD

Sample

The sample consisted of 79 mothers from 14 to 21 years of age (mean = 18.1). based on their BDI total scores, they were divided into three groups: low BDI (scores = 0, 1, 2), nondepressed (scores = 3-9), and depressed (scores [greater than or equal to] 13). The mothers were primarily single (74%), of varying ethnicity (37% African American, 35% Hispanic, and 28% Caucasian), from a low socioeconomic background, and had an average of 10 years of education.

Procedure

Within 24 hours after delivery, each mother was administered a social history questionnaire, the BDI, and the validity scales of the MMPI-2. All were presented in an interview format to control for differences in reading levels.

Measures

Demographic data. Information on age, marital status, ethnicity, and total number of pregnancies was gathered. Socioeconomic status was determined using the Hollingshead Two Factor Index of Social Status.

Beck Depression Inventory. The revised Beck Depression Inventory (Beck, Rush, Shaw, & Emery, 1979; Beck & Steer, 1987) assesses a wide range of symptoms associated with depression. Responses to the 21 items of this self-report inventory are made on a 4-point scale, ranging from 0 to 3 (total scores can range from 0 to 63). The BDI is applicable for use with individuals 13 years of age or older (Steer & Beck, 1988).

Minnesota Multiphasic Personality Inventory 2. The Minnesota Multiphasic Personality Inventory 2 consists of three validity scales and ten clinical scales. T-scores of 65 or higher are considered to be out of the normal range. For this investigation, only the validity scales were used: L, F, and K. The L (Lie) scale consists of 15 items on minor flaws and weaknesses that most people would admit. T-scores of 65 and above indicate that respondents are deliberately trying to present themselves in an unrealistically favorable light. The F (Symptom) scale consists of 60 items representing a wide range of symptoms. Normal respondents usually endorse fewer than five items. This scale yields an index of the respondents' cooperativeness and ability to provide useful information. High scores indicate that respondents are answering randomly or are deliberately trying to look bad. The K (Defensiveness) scale consists of 30 items covering different areas in which a person may deny problems. These items are more subtle than are those on the L scale. Scores of 65 and above are indicative of a defensive response style and a reluctance to disclose personal information. T-scores for these three scales were plotted to identify fake-good profiles, typified by elevated scores on the L scale. A cutoff of five endorsed items was used to identify elevations on the L scale (see Graham, Watts, & Timbrook, 1991), in addition to fewer than five on the F scale.

RESULTS

Data analyses were performed using the SPSS-X computer program (SPSS, Inc., 1988). Analyses of variance suggested that the groups did not differ on parity or age, but the nondepressed group had a higher level of education than did the depressed and low-BDI groups, F(2, 76) = 4.42, p [less than] .05. Chi-square analyses indicated that the groups did not differ on marital or employment status. However, the groups differed on racial composition, [[Chi].sup.2](4) = 18.5, p [less than] .001; the low-BDI group included more Hispanics than did the two other groups, and the depressed group included more African Americans (see Table 1).

A multivariate analysis of variance revealed differences among the groups on the three validity (F, K, and L) scales, F(4, 152) = 9.80, p [less than] .001. Univariate analyses indicated that the groups differed significantly on the F scale, F(2, 76) = 13.8, p [less than] .001. Post hoc analyses (Tukey's HSD) indicated that, compared with both the nondepressed and low-BDI groups, the depressed group had higher F scores. In addition, the nondepressed group had a higher F score than did the low-BDI group. The groups also differed on the K scale, F(2, 76) = 4.26, p [less than] .05. Post hoc analyses revealed that the depressed group had lower K scores than did the two other groups. No significant group differences were noted for the L scale (see Table 2).

Table 1

Means (and Standard Deviations) and Percentages for the Demographic
Data

Measure Low BDI Nondepressed Depressed

Age 18.2 (2.0) 18.3 (2.1) 17.8 (2.0)

Education Level 9.6 (2.6) 11.2 (1.6) 10.1 (1.2)

Race
Caucasian 20% 32% 33%
African American 17% 44% 57%
Hispanic 63% 24% 10%

Marital Status
Married 27% 36% 14%
Single 73% 64% 86%

Pearson chi-square analysis revealed that a significantly higher percentage of mothers in the low-BDI group, as compared with the nondepressed and depressed groups, had fake-good profiles, [[Chi].sup.2](2) = 16.9, p [less than] .001 (see Table 3). Not all of the mothers with fake-good profiles were in the low-BDI group, however; several in the nondepressed group also had this profile.

Analyses of variance for the demographic data revealed no significant differences between mothers with fake-good and valid profiles on age, race, educational level, and marital status. based on scoring criteria, the mothers with fake-good profiles had higher L scale scores, F(1, 77) = 14.2, p [less than] .001, and lower F scale scores, F(1, 77) = 16.3, p [less than] .001. In addition, the mothers with fake-good profiles had higher K scale scores, F(1, 77) = 18.8, p [less than] .001.

Table 2

Means (and Standard Deviations) for the MMPI-2 T-Scores

Validity
Scale Low BDI Nondepressed Depressed

L 58.7 (13.8) 57.1 (11.3) 54.2 (9.2)
F 49.4 (8.5) 56.9 (11.8) 66.5 (15.4)
K 47.9 (8.4) 47.1 (8.2) 45.8 (7.4)
Table 3

Percentage of Fake-Good and Valid Profiles

Profile Low BDI Nondepressed Depressed

Fake Good 48% 20% 0%
Valid 52% 80% 100%

Correlation analyses were performed to determine which variables were significantly related to the fake-good profile. As expected, the L (r = -.68) and F (r = .35) scales were associated with the fake-good profile. In addition, K scale scores (r = -.32) and BDI scores (r = .43) were associated with this profile. The BDI and K scale scores were entered into a stepwise regression analysis to determine the amount of variance in the fake-good profile accounted for by these variables (although the L and F scales also were correlated with this profile, they were not entered because these scales were used to classify the mothers and by definition would account for the greatest proportion of the variance). The results indicated that high K scale scores accounted for 28% of the variance, with low BDI scores accounting for an additional 14%, for a total of 42%. A discriminant function analysis was then performed to determine classification accuracy. The K scale and BDI scores correctly classified 90% of the fake-good and 71% of the valid profiles, Wilks' lambda (2, 77) = .72, p [less than] .001.

DISCUSSION

The analyses revealed that low BDI scores may be indicative of a fake-good profile. Together with the high K scale scores (defensiveness), these findings suggest that the low-BDI mothers in the Field et al. (1992) study who looked more depressed than the high-BDI mothers were "faking good" (and being defensive) in their responses to the BDI items. Thus, in previous studies, subjects with low BDI scores may have been misclassified as nondepressed, potentially attenuating expected depressed/nondepressed group differences.

Despite the strong relationship between low BDI scores and fake-good profiles on the MMPI-2, researchers and clinicians must be cautious in their interpretations since only 48% of mothers with low BDI scores had fake-good profiles. It is also important to note that although Hispanics were more prevalent in the low-BDI group and African Americans were more prevalent in the depressed group, racial differences were not found for the fake-good and valid profiles. Other factors, such as prevalence and type of psychosocial stressors, family cohesion, and social support, may be related to the different racial distributions in the low-BDI and depressed groups.

The relationship between low BDI scores and high K scale scores is noteworthy. High K scale scores have been associated with defensiveness, denial of problems, and desire to maintain the appearance of control and effectiveness. Therefore, it is not surprising that elevated K scale scores and extremely low BDI scores were indicative of "faking good."

In summary, these findings suggest that individuals with extremely low scores on depression-screening instruments require further assessment. Additional factors (e.g., defensiveness, social desirability, and denial) must be considered when classifying these individuals as depressed or nondepressed. Therefore, researchers and clinicians should use supplemental assessment instruments, such as the L, F, and K scales of the MMPI-2, when low BDI scores (0, 1, 2) are obtained.

The authors thank Holly Wilcox, Cynthia Mueller, and Kathy Hanson for their assistance with this research, the staff of Jackson Memorial Hospital, and the adolescent mothers who participated in the study. This research was supported, in part, by NIDA (#DA06900) and NIMH (#MH40779) grants to Tiffany Field.

REFERENCES

Barrera, M., & Garrison-Jones, C. V. (1988). Properties of the Beck Depression Inventory as a screening instrument for adolescent depression. Journal of Abnormal Child Psychology, 16, 263-273.

Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Press.

Beck, A. T., & Steer, R. A. (1987). Manual for the revised Beck Depression Inventory. San Antonio, TX: Psychological Corporation.

Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561-571.

Brockington, I. F., & Kumar, R. (1982). Motherhood and mental illness. New York: Grune & Stratton.

Bumberry, W., Oliver, J. M., & McClure, J. N. (1978). Validation of the Beck Depression Inventory in a university population using psychiatric estimate as the criterion. Journal of Consulting and Clinical Psychology, 46, 150-155.

Caplan, H. L., Coghill, S. R., Alexandra, H., Robson, K. M., Katz, R., & Kumar, R. (1989). Maternal depression and the emotional development of the child. British Journal of Psychiatry, 154, 818-822.

Cohn, J. F., Campbell, S. B., Matias, R., & Hopkins, J. (1990). Face-to-face interactions of postpartum depressed and nondepressed mother-infant pairs at two months. Developmental Psychology, 26, 15-23.

Field, T. (1984). Early interactions between infants and their postpartum depressed mothers. Infant Behavior and Development, 7, 527-532.

Field, T., Morrow, C., Healy, B., Foster, T., Adelstein, D., & Goldstein, S. (1992). Mothers with zero Beck depression scores act more depressed with their infants. Development and Psychopathology, 3, 253-262.

Field, T., Sandberg, D., Garcia, R., Vega-Lahr, N., Goldstein, S., & Guy, L. (1985). Pregnancy problems, postpartum depression and early mother-infant interactions. Developmental Psychology, 21, 1152-1156.

Gotlib, I. H., Whiffen, V. E., Mount, J. H., Milne, K., & Cordy, N. I. (1989). Prevalence rates and demographic characteristics associated with depression in pregnancy and postpartum. Journal of Consulting and Clinical Psychology, 57, 269-274.

Gotlib, I. H., Whiffen, V. E., Wallace, P. M., & Mount, J. H. (1991). Prospective investigation of postpartum depression: Factors involved in onset and recovery. Journal of Abnormal Psychology, 100, 122-132.

Graham, J. R., Watts, D., & Timbrook, R. E. (1991). Detecting fake-good and fake-bad MMPI-2 profiles. Journal of Personality Assessment, 57, 264-277.

Hopkins, J., Marcus, M., & Campbell, S. B. (1984). Postpartum depression: A critical review. Psychological Bulletin, 95, 498-515.

Lyons-Ruth, K., Zoll, D., Connell, D., & Grunebaum, H. E. (1986). The depressed mother and her one-year-old infant: Environment, interaction, attachment and infant development. In E. Tronick & T. Field (Eds.), Maternal depression and infant disturbance (pp. 61-82). San Francisco: Jossey-Bass.

O'Hara, M. W., Neunaber, D. J., & Zekoski, E. M. (1984). Prospective study of postpartum depression: Prevalence, course, and predictive factors. Journal of Abnormal Psychology, 93, 158-171.

O'Hara, M. W., Rehm, L. P., & Campbell, S. B. (1982). Predicting depressive symptomatology: Cognitive-behavioral models of postpartum depression. Journal of Abnormal Psychology, 91, 457-461.

Oliver, J. M., & Simmons, M. E. (1984). Depression as measured by the DSM-III and the Beck Depression Inventory in an unselected adult population. Journal of Consulting and Clinical Psychology, 52, 892-898.

Pfost, K. S., Stevens, M. J., & Lum, C. U. (1990). The relationship of demographic variables, antepartum depression, and stress to postpartum depression. Journal of Clinical Psychology, 46, 588-590.

SPSS, Inc. (1988), SPSS-X User's Guide (3rd ed.). Chicago: Author.

Steer, R. A., & Beck, A. T. (1988). Use of the Beck Depression Inventory, Hopelessness Scale, Scale for Suicide Ideation, and Suicidal Intent Scale with adolescents. In A. R. Stiffman & R. A. Feldman (Eds.), Advances in adolescent mental health (pp. 219-231). Greewich, CT: JAI Press.

Steer, R. A., Scholl, T. O., & Beck, A. (1990). Revised Beck Depression Inventory scores of inner-city adolescents: Pre- and postpartum. Psychological Reports, 66, 315-320.

Zuckerman, B. S., & Beardslee, W. R. (1987). Maternal depression: A concern for pediatricians. Pediatrics, 79, 110-117.

Tiffany Field, Ph.D., Director, Touch Research Institute, and Professor of Pediatrics, Psychology and Psychiatry, University of Miami.

Margarita Prodromidis, Ph.D., Research Associate, Touch Research Institute, University of Miami.

Sonya M. Abrams, M.S., graduate student, Touch Research Institute, University of Miami.