Adolescent Depression: Children's Depression Inventory Norms, Suicidal Ideation, and (Weak) Gender Effects
Chartier, George M., Lassen, Maureen K., Adolescence
Although the presence of depressive symptomatology in children and adolescents has been amply demonstrated, much less is known about the phenomenon in these age groups than in adults. Further, the inconsistency of findings of the empirical research is cause for serious concern (Kazdin & Petti, 1982). In order to help clarify a variety of conceptual issues, there is a pressing need to learn more about normative patterns of depression among normal as well as clinical populations (Achenbach, 1982), and to examine the possible contributions of developmental level, gender, and other relevant variables to variations in normative parameters (Kazdin & Petti, 1982; Smucker, Craighead, Craighead, & Green, 1986; Worchel, Nolan, & Wilson, 1987).
One of the more popular measures of depression in younger age groups, the Children's Depression Inventory (CDI; Kovacs & Beck, 1977) has generated some normative data in several studies of nonreferred adolescent samples. However, this data base is still relatively small, and the findings have been inconsistent with respect to age and gender differences among 12-18-year-olds. Specifically, Finch, Saylor, and Edwards (1985) reported higher levels of depression among the older adolescents in their sample, while others have reported either no age differences (Kovacs, 1980/81; Saylor, Finch, Baskin, Furey, & Kelly, 1984; Worchel et al., 1987) or patterns that were unclear (Doerfler, Felner, Rowlinson, Raley, & Evans, 1988; Smucker et al., 1986). With regard to gender, female adolescents have evidenced both significantly more depression than have males (Doerfler et al., 1988; Worchel et al., 1987) and less than males (Finch et al. 1985), or have shown no or unclear gender differences (Kovacs, 1980/81; Saylor et al., 1984; Smucker et al., 1986). The inconsistency of these findings on the CDI is at odds with the widely held view that adolescent girls are much more likely than boys to experience depressive symptoms (McGrath, Keita, Strickland, & Russo, 1990; Nolen-Hoeksema, 1990).
The primary purpose of the present study was to provide additional normative data for the CDI on 7th- to 12th-grade subjects with particular attention to age and gender differences. In addition, the study allowed us to examine responses to the CDI item concerning suicidal ideation, and to compare the results with those obtained from adolescents on similar items on other self-report inventories. The potential importance of both aspects of this study is highlighted by statistics that show suicide to be the third most common cause of death among adolescents (Achenbach, 1982).
The initial subject pool consisted of 226 students in a junior high school (grades 7-9) and 705 students in two senior high schools (grades 10-12) in 1980 in a moderate-size Midwestern city. Approximately 89% of the parents and their children consented to participate in the study. Of the 825 consenting students, 33 were either absent on the day of testing or failed to complete portions of their materials. Thus, the total number of subjects actually included was 792. According to their own reports, more than 80% of the students were living in two-parent households. For the junior high sample, approximately two-thirds of the fathers and one-half of the mothers had at least some college education. For the senior high students, the corresponding estimates were 44% and 29%, respectively. The subjects were predominantly Caucasian.
Materials and Procedures
The CDI (Kovacs, 1980/81), a downward extension of the Beck Depression Inventory (BDI), is a self-report measure designed to assess the severity of various symptoms of depression in adolescents and children as young as age seven. Each of the 27 items consists of three sentences; the subject is asked to choose the one that best describes him or her over the past two weeks. The CDI appears to possess good psychometric properties, including an adequate ability to discriminate between referred and nonclinical samples (Carey, Faulstich, Gresham, Ruggiero, & Enyart, 1987; Kovacs, 1985; Saylor et al. …