Evidências de validade da versão brasileira da Critical Thinking Disposition Scale

Abstract

Confl itos de interesse: nada a declarar. ResumoObjetivo: Realizar a tradução, adaptação cultural, análise de evidências de validade e confi abilidade da Critical Thinking Disposition Scale para o português do Brasil.Métodos: Estudo metodológico que envolveu tradução inicial, síntese das traduções, retrotradução, avaliação por comitê de especialistas (evidências de validade de conteúdo), pré-teste e envio aos desenvolvedores para avaliação. Após essa etapa foi ainda realizada a avaliação de evidências de validade baseadas na estrutura interna do instrumento e de confi abilidade.Resultados: A versão fi nal da escala obteve coefi ciente de validade de conteúdo com variação entre 0,88 e 0,96 para os itens. O tempo de aplicação variou entre 11 e 20 minutos. Análise fatorial exploratória indicou estrutura interna formada por um único fator com cargas fatoriais adequadas para todos os itens, bem como boa correlação item-total. O instrumento apresentou adequada evidência de confi abilidade (coefi ciente Ômega) igual a 0,79.Conclusão: O instrumento foi traduzido, adaptado, possui evidências de validade de conteúdo, de validade baseada na estrutura interna do instrumento e confi abilidade adequadas à cultura brasileira.

Investigating Psychometric Properties of the Turkish Version of Sosu Critical Thinking Disposition Scale: Evidence from Two Independent Samples

“…Previous literature indicates that the CTDS has been widely used by many researchers to determine individuals' CTD since the day it was developed. The CTDS has been translated into many languages such as Spanish (Bravo et al, 2020), Brazilian Portuguese (Luiz et al, 2021), Norwegian (Gerdts-Andresen et al, 2022, and Turkish (Akın et al, 2015). Besides, there are other studies investigating the psychometric properties of the original CTDS (Thomas & Hayes, 2021;Yockey, 2016).…”

“…Besides, there are other studies investigating the psychometric properties of the original CTDS (Thomas & Hayes, 2021;Yockey, 2016). After factor analytic procedures, some studies revealed that the original two-factor model proposed by Sosu (2013) showed the best model fit (Akın et al, 2015;Gerdts-Andresen et al, 2022;Thomas & Hayes, 2021) although some studies concluded that the onefactor model had better psychometric properties for the CTDS (Bravo et al, 2020;Luiz et al, 2021;Yockey, 2016).…”

“…Also, MI of the one-factor model was assessed across gender groups and the results revealed that one-factor model is invariant and it can be used across different genders. Luiz et al (2021) carried out a study with a sample group of 179 undergraduate students in Brazil to adapt the CTDS into Brazilian Portuguese and to investigate psychometric qualities of the Brazilian Portuguese version of the CTDS (BPV-CTDS). After translation procedure of the scale into Brazilian Portuguese, EFA was conducted to reveal the factor structure of the BPV-CTDS.…”

“…EFA indicated the presence of two factors with total variance of 44.88%. However, parallel analysis of the items and factors showed that there was inadequacy in this structure and Luiz et al (2021) conducted another EFA fixing the solution in a single factor and found out that 11 original items had good factor loadings onto the single factor (the lowest λ=0.49). Therefore, Luiz et al (2021) stated that BPV-CTDS is better with one-factor model.…”

“…However, parallel analysis of the items and factors showed that there was inadequacy in this structure and Luiz et al (2021) conducted another EFA fixing the solution in a single factor and found out that 11 original items had good factor loadings onto the single factor (the lowest λ=0.49). Therefore, Luiz et al (2021) stated that BPV-CTDS is better with one-factor model. As an evidence for the reliability of the BPV-CTDS, Omega coefficient was calculated and it was seen that the BPV-CTDS was reliable (Omega coefficient=0.79; 95%CI=0.74-0.83).…”