Autoconceito e vergonha externa em adolescentes: validação de instrumentos para o Brasil

Abstract

Self-concept can be considered a set of beliefs that individuals hold about themselves. Shame, on the other hand, is a profound, complex, and self-conscious emotion that emerges around the age of two, requiring self-awareness and an understanding of interpersonal relationships. This study's main objective is to validate the Piers-Harris Children’s Self-Concept Scale (PHCSCSV1-6) and Other as Shamer Scale (OAS) instruments for Brazilian Portuguese and test their psychometric properties. A methodological study was developed in two stages, with the first being the transcultural adaptation of the questionnaires. In the second stage, the psychometric properties of the Brazilian versions (BR-PHCSCS and B-OAS) were tested on a sample of 325 school adolescents aged 12 to 18 from two cities in Northeast Brazil. Psychometric properties of the BR-PHCSCS and B-OAS instruments were analyzed, including internal consistency (Cronbach's Alpha and McDonald's Omega), reproducibility (instrument reapplication after seven days - intraclass correlation coefficient [ICC]), convergent validity (correlation with the Depression, Anxiety, and Stress Scale [DASS-21]), discriminant validity (association with sociodemographic variables), predictive validity (association with dental caries and occlusal aspects), and factorial validity (exploratory [EFA] and confirmatory factor analysis [CFA]). The study was approved by the Research Ethics Committee with Human Beings of the State University of Paraíba (CAAE: 60413722.0.0000.5187). Data were analyzed descriptively and using Mann-Whitney, Kruskal Wallis, Spearman's Correlation, EFA, and CFA tests (α=5). The BR-PHCSCS and B-OAS instruments showed satisfactory internal consistency (Cronbach's Alpha = .79 and .91 and McDonald's Omega = .78 and .91) and reproducibility (ICC = 0.97 and 0.96), as well as convergent validity (rs= -0.575, p<0.001; rs=0.663 p<0.001). There was a statistically significant association between BR-PHCSCS scores and the type of school (p=0.033), number of children (p=0.027), caregiver's education level (p=0.008), and monthly income (p=0.018) (discriminant validity). B-OAS scores were significantly higher in female adolescents (p=0.003) and those with a monthly income of up to one minimum wage (p=0.025) (discriminant validity). There was no significant association between instrument scores and dental caries, occlusal aspects, and the need for orthodontic treatment. EFA suggested a unifactorial structure for B-OAS (41.6% of explained variance), confirmed by CFA (X2= 345.504, df=135, CFI= 0.096, TLI= 0.893, RMSEA= 0.069, and SRMR= 0.048). BR-PHCSCS showed a multidimensional factorial structure with six factors explaining 45.3% of the variance. The instruments demonstrated satisfactory psychometric properties for the group of Brazilian adolescents studied, and despite not being associated with the analyzed oral conditions, they can be reliably used for other situations related to external shame and self-concept.