Gender differences in the expression of depressive symptoms in middle adolescents: an extension of earlier findings

Adolescence, Winter, 1993 by Pierre Baron, T. Leanne Campbell

Epidemiological studies in Western culture consistently report a higher rate of unipolar depression for females than males (e.g., Baron & Perron, 1986; Nolen-Hoeksema, 1987). Based on research demonstrating that expression of depressive symptomatology elicits interpersonal rejection (e.g., Coyne, 1976), especially for males (Hammen & Peters, 1977), and that individuals view depression as more stereotypically feminine (Landrine, 1988), researchers have argued that depression may be more socially unacceptable for males than it is for females (Warren, 1983). Consequently, individuals may learn to express their psychological distress in gender-appropriate ways.

Baron and Joly (1988) identified gender differences in the expression of depressive symptoms in a sample of adolescents (aged 12 through 17 years) scoring 15 or more on the Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961). Discriminant function analyses revealed that male adolescents exhibited symptoms of irritability, work inhibition, social withdrawal, and sleep disturbance, whereas females manifested symptoms characterized by body image distortion, loss of appetite, weight loss, sadness, and dissatisfaction. Similarly, gender-role appropriate symptom expression was found in a sample of clinically depressed adult clients (Vredenburg, Krames, & Flett, 1986), and in depressed subsamples of college students (e.g., Hammen & Padesky, 1977; Padesky & Hammen, 1981). In contrast, Campbell, Byrne, and Baron (1992) found that females had higher mean scores on all of the BDI and the Reynolds Adolescent Depression Scale (RADS; Reynolds, 1986) items that significantly discriminated between male and female nonclinical early adolescents.

The present study contributes to the body of research aimed at examining the expression of depressive symptoms in adult and adolescent clinical and nonclinical samples. The purpose of the study was to examine whether, consistent with the findings based on data obtained at the first phase of the longitudinal study, females continue to have higher mean scores on all discriminating items as mid-adolescence is approached. Unlike previous research, the present study investigates gender differences in the expression of depressive symptoms among the same nonclinical sample over time. This design permits a description of normative patterns in the expression of depressive symptoms, and facilitates the understanding and interpretation of depression scale scores for males and females in other clinical and nonclinical samples.

METHOD

Sample and Procedure

Data were derived from a larger longitudinal study aimed at examining the prevalence of depressive symptoms in francophone adolescents (Baron & Groulx, 1990; Baron, Regimbald, & Groulx, 1991). Participants were assessed at two time points, approximately one year apart. Data for this study were obtained from a total of 153 subjects (98 females, 55 males) aged 14 through 16 years (M = 14.9 years). Informed consent was obtained, and subjects were ensured of confidentiality. Testing took place in the context of intact classes using translated versions of the BDI (Baron & LaPlante, 1984) and the RADS (Baron & DeChamplain, 1990) as two of several instruments.

Instrumentation

Beck Depression Inventory. The BDI is a 21-item self-report scale presented in multiple-choice format. Developed to measure depth of clinical depression, it assesses symptoms related to cognitive, affective, behavioral, and somatic components of depression (Beck et al., 1961). Each item on the inventory corresponds with one of these symptoms, and numerical values ranging from zero to three are assigned to each of four statements to indicate degree of severity. Subjects are asked to describe the way they have been feeling for the past week. Total scores range from 0 to 63, with higher scores indicative of greater severity of self-reported depression.

Coefficient alpha estimates of .79, .87, and .80 have been reported for a clinical sample of adolescents (Strober, Green, & Carlson, 1981), a nonclinical sample of adolescents (Teri, 1982), and a nonclinical sample of francophone adolescents using Bourque and Beaudette's (1982) adapted version of the BDI (Baron & Laplante, 1984), respectively. Strober et al. (1981) obtained a test-retest reliability (5-day period) of .69 for the English version, and Baron and Laplante (1984) reported a test-retest reliability (8-week period) of .74 for the French version of the BDI.

Reynolds Adolescent Depression Scale. The RADS is a 30-item self-report questionnaire designed to assess the severity of depressive symptomatology in adolescents aged 13 through 18 years. Although no particular time frame is specified, items are worded in the present tense to elicit current symptom status. The instrument utilizes a four-point Likert-type format and requires the adolescent to endorse whether the symptom-related item has occurred: "almost never," "hardly ever," "sometimes," or "most of the time." Total score responses can range from 30 to 120, the higher scores indicating greater levels of depressive symptoms. A cut-off score of 77 delineates a level of symptom endorsement consistent with clinical depression (Reynolds, 1986).