Individual Submission Summary
Share...

Direct link:

Poster #57 - Identification of separate anxiety and depression factors in referred children using the Preschool Socioaffective Profile

Thu, March 21, 4:00 to 5:15pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

An analysis of the Preschool Socioaffective Profile’s psychometric characteristics (PSP; LaFrenière, Dumas, Capuano, & Dubeau, 1992), in its French version (Dumas, LaFrenière, Capuano, & Durning, 1997), was performed using a sample of 466 preschoolers referred to a psychiatric clinic (30-74 months old, 120 girls). The PSP is specifically designed to assess patterns of affective expression, social competence, and adjustment difficulties in group settings with peers. It is completed by a daycare provider or preschool teacher. Although the original version was developed using a sample that included children from the clinical population (LaFrenière, Dubeau, Janosz, & Capuano, 1990), its norms and factorial structure were established using a convenient sample of 608 preschoolers and, for the French version, two samples of 979 (Quebec) and 800 (France) preschoolers (Dumas et al., 1997) that did not include clinical children. Yet, the factorial structure of the scale could be expected to differ as clinical children display a wider range of problematic behaviors both in terms of their nature and severity. The purpose of the present study is thus to verify whether the PSP’s subscales show adequate internal consistency and if the authors’ proposed factorial structure (Externalizing, Internalizing, Social Competence) can be replicated with a clinical sample.
Results showed that all Cronbach alphas were satisfactory (Cicchetti, 1994; Ponterotto & Ruckdeschel, 2007) with four subscales yielding alphas above .80 (irritable-tolerant, isolated-integrated, aggressive-controlled, resistant-cooperative), three above .70 (depressed-joyful, anxious-confident, egoistic-prosocial) and only one above .60 (dependant-autonomous). When analysed separately, similar results were obtained for boys and girls and for 30-47 and 48-74 months old. A confirmatory factorial analysis (Amos 24) also revealed the original orthogonal 3-factor model did not fit the data and was therefore rejected [χ2 (96, N = 466) = 847.95, p < .001] with many subscales loading on two different factors (Table 1). An exploratory factorial analysis and oblique rotation suggested instead a 4-factor solution with modest between-factor correlations (all r < .40). Those four factors (Table 2) explain 73% of the total variance. Two factors were the same as the original version (i.e., Social Competence and Externalizing Problems) while the items of the Internalizing Problems separated into two separate factors corresponding to anxiety (dependant, anxious) and withdrawal/depression (depressed, joyful-inversed-, isolated).
These results confirm the pertinence of using the PSP to assess social adaptation with clinical children. Moreover, given the importance of detecting early signs of depression now that there is prospective evidence for homotypic continuity from preschool to adolescence (Gaffrey, Tillman, Barch, & Luby, 2018), the capacity of the PSP to distinguish between different Internalizing Problems in a clinical setting makes it all the more useful. Indeed, the PSP could fill a significant gap in detecting early signs of depression since these signs have been traditionally very difficult to assess in young children (Gudmundsson et al., 2013). Future research is now needed to replicate this new factorial structure and to further assess its external validity. Meanwhile, this study highlights the importance of replicating models in different samples, especially when these vary in their composition.

Authors