Social anxiety profiles and academic self-attributions in secondary school students. What are we really talking about?: Theoretical, methodological, and statistical preciseness

• Published in: European journal of education and psychology Vol. 15; no. 2; pp. 43 - 56
• Main Authors: Inglés Saura, Cándido J, Fernández Sogorb, Aitana, García Fernández, José Manuel, Urbán, Dori J. A
• Format: Journal Article
• Language: English
• Published: 2022
• ISSN: 1989-2209; 1888-8992; 1989-2209
• DOI: 10.32457/ejep.v15i2.1970

The aim of this study was to identify the relationship between academic self-attributions and subclinical social anxiety in a sample of Spanish adolescents and examine statistically significant differences in academic self-attributions among subgroups of socially anxious youth. Random cluster sampling was conducted. The final sample consisted of 717 Spanish students enrolled in Secondary Education (51% girls) aged 14 to 17 years (M = 15.68, SD = 1.16). The Social Anxiety Scale for Adolescents (SAS-A) and the Sydney Attribution Scale (SAS) were administered. Cluster analysis identified three clusters: (1) students with high social anxiety (n = 102) and high scores on fear of negative evaluation (FNE), anxiety toward strangers or new social situations (SAD-N), and anxiety in social situations in general (SAD-G); (2) students with moderate social anxiety (n = 290) and moderate scores on FNE, SAD-N, and SAD-G; and (3) students with low social anxiety (n = 325) and low scores on FNE, SAD-N, and SAD-G. Multivariate analyses (MANOVA) examined differences in the academic self-attributions across the three clusters of subclinical social anxiety, finding statistically significant differences in the six types of academic self-attributions (Wilks Lambda = .89, F(12,714) = 7.11, p < .001, η2 = .06), including success attributed to ability, success attributed to effort, success attributed to external causes, failure attributed to ability, failure attributed to effort, and failure attributed to external causes. The implications of these findings for Psychology and Education professionals are discussed.