In contrast to the two previous applications that focused on CFA first-order models, this application examines a CFA model that comprises a second-order factor. In this chapter, we test for the validity of scores on a hierarchically ordered Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961) as they relate to nonclinical adolescents. The example is taken from a study by Byrne, Baron, and Campbell (1993) and represents one of a series of studies that have tested for the validity of second-order BDI structure for high school adolescents in Canada (Byrne & Baron, 1993,1994; and Byrne, Baron, & Campbell, 1993,1994), Sweden (Byrne, Baron, Larsson, & Melin, 1995, 1996), and Bulgaria (Byrne, Baron, & Balev, 1996, 1998). Although the purpose of the Byrne et al. (1993) study was to test for the equivalence of BDI structure across gender for Canadian high school students, this chapter focuses on factorial validity as it relates only to the female sample (n = 321). (For details regarding the sample, analyses, and results, see the original article.)
The BDI is a 21 -item scale that measures symptoms related to cognitive, behavioral, affective, and somatic components of depression. Although originally designed for
use by trained interviewers, it is now typically used as a self-report measure. For each item, respondents are presented with four statements rated from 0 to 3 in terms of intensity and asked to select the one that most accurately describes their own feelings; higher scores represent a more severe level of reported depression. For the study from which the current application is taken, scores were recoded such that their scale points ranged from 1 through 4 rather than from 0 through 3. As discussed in chapter 4, the CFA of a measuring instrument is most appropri ately conducted with fully developed assessment measures that have demonstrated satisfactory factorial validity. Justification for CFA procedures in this instance is based on evidence provided by Tanaka and Huba (1984) and replicated studies by Byrne and associates (cited previously) that have shown BDI score data to be most adequately represented by a hierarchical factorial structure. As such, it is argued that the three first-order factors are explained by some higher order factor that, in the case of the BDI, represents a single second-order factor of general depression.