Adaptação e validação de uma escala para mensurar a autoeficácia e eficácia coletiva em ações de aconselhamento em alimentação saudável para professores

Abstract

This dissertation aims to evaluate the psychometric properties of a scale to measure self efficacy and collective effectiveness in counseling actions on healthy eating, according to the recommendations of the Food Guide for the Brazilian Population (GAB2) adapted for teachers. This is a methodological study that was developed with teachers from municipal schools in the urban and rural areas of Montes Claros, MG. To assess the psychometric properties of the scale, four steps were performed: scale adaptation, content, construct and reliability validation (factorial, convergent, discriminant, internal consistency and reproducibility) and face validation. The research was submitted and approved by the Ethics and Research Committee of the State University of Montes Claros under opinion nº 3,586,107. Data were collected from December 2019 to June 2021 and were electronically categorized and processed using the Statistical Package Software for Social Science (SPSS), version 22.0 and Amos. The scale was adapted to the school context and subsequently content validation was carried out, with the participation of 11 experts, including 10 nutritionists and 1 pedagogue. The Content Validity Index (CVI) found was equal to 0.98, confirming the approval of the scale. In the third stage, a questionnaire structured into seven sessions was used: identification, socioeconomic conditions, training and professional performance, approach to content on food and nutrition in the classroom, anthropometric conditions, eating conditions and, finally, the scale to be validated ( GAB2 for Teachers). A total of 595 teachers from the city participated in the study, 92.4% of whom were female. Construct validity was observed, with adequate goodness-of-fit indices: χ²/gl=4.55, RMSEA=0.077; CFI=0.11; TLI=0.90 and GFI=0.90. Factor loadings were positive, ranging from 0.56 to 0.83, and the correlation between domains was adequate and statistically significant (r=0.70; p=0.001). The scale also presented convergent and discriminant validity and proved to be a reliable instrument, with values of Average Variance Extracted (AVE) equal to 0.50 and 0.57, in the two domains of self-efficacy (AE) and collective effectiveness (AEC) , respectively, adequate Composite Reliability (CC), AE=0.917 and AEC=0.940, Cronbach's Alpha equal to 0.95 and Positive Intra-Class Correlation Coefficient (ICC), AE=0.85 and AEC=0.80. Apparent validity was confirmed with 80% approval by potential users of the scale, showing that the language and understanding of the items were satisfactory. Therefore, the GAB2 version, 10 adapted for teachers, showed satisfactory psychometric properties to measure self-efficacy and collective effectiveness in counseling actions on the Brazilian food guide. Educational materials were produced that have already been implemented in the educational context of the municipality, as a form of feedback to the studied public.