Healthcare Training Institute - Quality Education since 1979
CE for Psychologist, Social Worker, Counselor, & MFT!!
Because of concerns about the inadequacy of a symptom approach to depression, as well as recognition of the advantages of considering the continuities between life experiences in normal psychological states and many forms of psychopathology, especially depression, colleagues and I (Blatt, D'Afffitti, & Quinlan, 1976, 1979) constructed a 66-item questionnaire on which individuals rate themselves on a wide range of life experiences frequently associated with depression but not directly considered symptoms of depression. These items were gleaned from classic case reports of patients with affective disturbances (e.g., Bibring, 1953; M. B. Cohen, Baker, Cohen, Fromm-Reichman, & Weigert, 1954; S. Freud, 1917/1957e) that described the life experiences of depressed patients (e.g., how they relate to people, how they feel about themselves, the ways they conduct their lives). Approximately 150 statements were constructed on the basis of a review of this clinical literature on depression, and from this list 66 items that represented a relatively broad range of life experiences associated with depression were selected. Items tapped issues such as a distorted or depreciated sense of self and others, dependency, helplessness, egocentricity, fear of loss, ambivalence, difficulty in dealing with anger, self-blame, guilt, loss of autonomy, and distortions in family relations. These 66 items are presented to individuals with the request that they rate themselves on 7-point Likert-type scales that range from strongly disagree (1) to strongly agree (7). The scale includes items that are presented in both negative and positive directions. This instrument, which we labeled the Depressive Experiences Questionnaire (DEQ), was initially group-administered to 500 female and 160 male undergraduates enrolled in a local state college.
Because of significant differences between
men and women on many DEQ items, the data for men and women were analyzed
separately. Principal-components factor analyses (PCAs) with varimax rotations
identified three orthogonal factors that appeared in both male and female college
The Dependency factor consists of items that are primarily externally directed; refer to interpersonal relations; and contain themes of abandonment, loneliness, and helplessness and the desire to be close to, related to, and dependent on others. High-loading items reflect concerns about being rejected, hurting or offending people, and having difficulty in managing anger and aggression for fear of losing the gratification someone could provide. The Self-Criticism factor consists of items that are more internally directed and reflect feelings of guilt, emptiness, hopelessness, dissatisfaction, and insecurity. Items with a high loading on this factor reflect concerns about a failure to meet expectations and standards, an inability to assume responsibility, feeling threatened by change, ambivalence about self and others, and a tendency to assume blame and feel critical toward oneself.
The third DEQ factor, Efficacy, involves items indicating a sense of confidence about one's resources and capacities. The items with high loadings on this factor contain themes of high standards and personal goals, a sense of responsibility, inner strength, feelings of independence, and a sense of pride and satisfaction in one's accomplishments. Individuals who score high on this factor are characterized by goal-oriented strivings and feelings of accomplishment, but not by extreme competitiveness.
Exhibit 3.1 The Five Items With the Greatest Loading for Each of the Three Factors of the Depressive Experiences Questionnaire______________________________________________________________________________________
Factor II: Self-Criticism
Factor III: Efficacy
It is important to stress that the DEQ was not developed to assess anaclitic and introjective dimensions of depression; rather, it was developed to assess the everyday life experiences of individuals with depression, and the factors of Dependency (relatedness) and Self-Criticism (self-definition) emerged, confirming the independent clinical observations and theoretical formulations of two independent dimensions (i.e., interpersonal relatedness and self-definition) as primary sources of depression (Blatt, 1974). It is also important to note that the DEQ yields factor scores: Each item, depending on its factor loading, contributes to the score on each of the three factors. Thus, a computer program is necessary to score the DEQ. The DEQ and the DEQ-A, as well as the scoring programs in the SPSS and SAS, can be obtained from S. J. Blatt, Yale University, Department of Psychiatry, 25 Park Street, New Haven, CT 06519 ([email protected]) and from D. C. Zuroff, McGill University, Department of Psychology, 1205 Dr. Penfield Avenue, Montreal, Quebec, H5A 1B1, Canada ([email protected]).
To test initially the stability of the factor structure, the female sample (n = 500) was randomly divided in half, and a factor analysis was conducted in each subsample (n = 250). Tucker's coefficients of congruence (Harmon, 1960) were computed among the three factors of the two subsamples. The factor structure had a high degree of stability; the coefficients of congruence of each factor with its split-half duplicate were all above .900. Whereas a moderate level of congruence was found between Dependency and Efficacy (in different samples, this correlation ranges from -.30 to -.49), the congruence of Self-Criticism with the other two factors was minimal. Coefficients of congruence were also computed between each factor in the total male (n = 160) and the total female samples (n = 500). The coefficients of congruence between each factor and its counterpart in the male and female samples were again high (>.80), and again, a moderate degree of congruence emerged between Dependency and Efficacy, but only minimal congruence was observed between Self-Criticism and the other two factors. Thus, the factor structure was highly similar and stable within split halves of the female sample and between the overall female and male samples, with some tendency for Dependency and Efficacy to share a moderate degree of congruence. The DEQ's factor structure has been replicated in a number of other nonclinical samples (e.g., Beutel et al., in press; R. C. Campos, 2002; Jae Im, 1996; Jerdonek, 1980; Luyten, 2002; Priel, Besser, & Shahar, 1998; Zuroff, Quinlan, & Blatt, 1990) in several countries, including Belgium, Canada, Germany, Israel, Korea, and Portugal.
The stability of the DEQ's factor structure was also demonstrated in data from a large sample of female and male college students from introductory psychology classes at a large rural state university (Zuroff, Quinlan, & Blatt, 1990). A total of 951 female students and 650 male students completed the DEQ. The scores were subjected to a PCA with three factors extracted by means of varimax rotation. The factors that emerged not only were highly congruent with the original female factor analysis, but they also emerged in the same order as in the original sample. Scales derived from these three factors had high internal consistency (M = .78) and substantial test-retest reliability (rs ranged from .72 to .89; Zuroff, Moskowitz, Wielgus, Powers, & Franko, 1983). These factors have been replicated in several other samples, including an extensive sample with a Dutch version of the DEQ (Luyten, 2002). Luyten (2002) and colleagues (Luyten, Fontaine, Meganck, et al., 2003), using PCA with varimax and Procrustean rotations of the DEQ responses of 928 students, 253 adults, and 136 psychiatric inpatients, found evidence for the factorial invariance of the Dependency and Self-Criticism factors (see also Blaney & Kutcher, 1991; Blatt & Zuroff, 1992; Riley & McCranie, 1990). The third DEQ factor, Efficacy, was replicated by PCA with only varimax rotation. Luyten, Fontaine, Soenens, and Corveleyn (2003) also conducted a formal content analysis of the DEQ by developing an explicit category system of nine categories of DEQ items: (a) personal standards, (b) self-evaluation, (c) relational styles and beliefs, (d) expectations from others, (e) self-esteem and variations in selfesteem, U) helplessness, (g) aggression/anger, (h) anxiety and fear, and (i) developmental history. On the basis of this content analysis, Luyten, Fontaine, Soenens, et al. (2003) differentiated between interpersonal and intrapsychic guilt. Interpersonal guilt (e.g., fear of hurting others) is associated with the Dependency factor, whereas intrapsychic guilt (e.g., not achieving or living up to moral standards) is associated with the Self-Criticism factor. In their content analyses, Luyten, Fontaine, Soenens, et al. (2003) also found that themes of helplessness, variations in self-esteem, and a fear of change appear on both the Dependency and Self-Criticism factors and thus appear to be central general issues in depression. Jae Im (1996), using PCA with varimax rotation, also replicated the DEQ factor structure in a mixed sample of Korean students, clients in a school counseling center, and psychiatric patients with a mood disorder and reported good levels of internal consistency and test-retest reliability.
The Dependency and Self-Criticism factors of the DEQ correlate significantly with traditional self-report measures of depression (e.g., the BDI [A.T. Beck et al., 1961] and the Zung Depression Scale [ZDS; Zung, 1965]). The Dependency factor usually has lower correlations than the Self-Criticism factor with these measures of depression, suggesting that it assesses an often overlooked dimension of depression. Although the Dependency and Self-Criticism factors are relatively independent in nonclinical samples, several studies (e.g., J. D. Brown & Silberschatz, 1989; Franche & Dobson, 1992; Jerdonek, 1980) found high correlations between these two factors in clinical samples. This led some researchers (e.g., Viglione, Clemmey, & Camenzuli, 1990) to question the validity of the DEQ in clinical populations. In general, however, these studies with clinical samples lacked a sufficiently large sample to test adequately the factor structure of the DEQ. Beutel et al. (in press), studying 304 participants in a community sample and 404 psychiatric ("psychosomatic") patients in Germany, replicated the three DEQ factors in their nonclinical sample but not in their clinical sample. Frank, Van Egeren, et al. (1997), however, replicated the DEQ factor structure in a large sample of adolescent inpatients using the adolescent form of the DEQ, the DEQ-A, which is discussed later in this chapter. Luyten, Blatt, & Corveleyn (in press), Luyten, Corveleyn, & Blatt (in press), in an extensive exploration of the DEQ in Belgium, concluded that the DEQ assesses primarily stable personality characteristics rather than depressed mood or other state characteristics.
In clinical samples, patients have higher Dependency and Self-Criticism scores than do nonpatients (Bagby, Schuller, et a!., 1994; Blatt et a!., 1982; Fairbrother & Moretti, 1998; Franche & Dobson, 1992; D. N. Klein, Harding, Taylor, & Dickstein, 1988; Lehman et al., 1997; Luyten, 2002; Rosenfarb, Becker, Khan, & Mintz, 1998). Luyten (2002) found that depressed patients scored higher on Dependency, but not on Self-Criticism, than did general psychiatric inpatients. He suggested that self-criticism (or perfectionism) may be a more general dimension in a wide range of disorders, whereas dependency may be uniquely related to depression. Lehman and colleagues (1997) found that depressed inpatients scored higher on both Dependency and Self-Criticism than did depressed outpatients, and patients with early-onset dysthymic disorder scored higher on Self-Criticism (perfectionism), but not Dependency, than did patients with major depressive disorder.
have higher dependency scores than males, but no gender differences are
usually found for Self-Criticism or perfectionism (e.g., Blatt, Hart, Quinlan,
Leadbeater, & Auerbach, 1993; Blatt et al., 1982; Brewin & FirthCozens,
1997; Chevron et al., 1978; Fichman, Koestner, & Zuroff, 1994, 1996, 1997;
Sanfilipo, 1994; Sato & McCann, 1998; T. W. Smith, O'Keeffe, & Jenkins,
1988; R. E. Steele, 1978; Zuroff & Fitzpatrick, 1995; Zuroff, Stotland, Sweetman,
Craig, & Koestner, 1995). Depression is two to three times more prevalent
in women, possibly because they are as vulnerable as men to issues of selfcriticism
but are also much more vulnerable to feelings of loneliness, abandonment, and
rejection. This combined vulnerability to feelings of loss and abandonment as
well as self-criticism places women at greater risk for depression. Equally noteworthy
are suggestions (Chevron et al., 1978; Sanfilipo, 1994) that gender-role characteristics,
rather than gender itself, are important in the study of depression, as well as
indications (e.g., Fichman et al., 1994; Mongrain & Zuroff, 1994; T. W. Smith
et al., 1988) of the particular vulnerability to depression of gender-incongruent
individuals, especially dependent men but also self-critical women.