Evidências de validade e fidedignidade da escala de gratidão (G-20) na população de adultos brasileiros

Abstract

This work was divided into two papers, one theoretical and one empirical. Regarding the first, it was a systematic review that aimed to verify the instruments available in the world to measure gratitude and its psychometric properties. In addition, verifying the predominant population and in which phenomena the instruments found assess the gratitude. This study had its search for papers in four databases (PsycINFO, Scopus, Web of Science and PubMed) resulting in a total of 36 selected papers. 16 original scales were found, the remainder being indicated as adaptations and validations. The scale with the highest number of adaptations and validations was GQ-6, with Ecuador, China and Brazil with the best evidence. Most of the samples were composed of women and the phenomenon of gratitude in a dispositional way was the most used. GRAT together with two scales developed in Japan demonstrated the highest reliability indices of the original scales. The second study of this dissertation, the empirical, aimed to provide evidence of validity of the Gratitude Scale (G-20) for the Brazilian adult population. The specific goals were to verify the structure of the questionnaire through confirmatory factor analysis, to analyse the internal consistency of the scale, as well as the validity of concurrent and convergent criteria. This is a quantitative, instrumental, descriptive and cross-sectional study. Individuals from the adult population, between 18 and 59 years old, who agreed to participate in the study were included. The exclusion criteria were the participants who answered all the items with the same answer. Data collection was performed online, after accepting the Informed Consent Form, which the project was approved by Ethics Committee with number CAAE:26530819.7.0000.5336. The convenience sample consisted of 348 participants from the five Brazilian regions. The ages of the participants ranged from 18 to 59 years, being 77% female. At the time of the study, 74% (n=264) of the participants had no children and 87% declared their race to be white. In addition, 44% (n=159) of the sample considers following the Catholic religion. Regarding education level, 28% (n=100) had complete higher education, and incomplete higher education are 27% of the sample (n=97). The instruments used were sociodemographic data questionnaire, G20, GQ-6, DASS-21, PANAS and Religiosity Scale P-DURELL. The total scale demonstrated satisfactory reliability (α = 0.92 and Ω = 0.71) and in all dimensions of the instrument, pointing to a high internal consistency. In addition, the analysis of the data adjustment for the four-factor structure showed an acceptable value for a good adequacy of the model (RMSEA = 0.068, CFI = 0.956 and TLI = 0.949) and the temporal stability showed significant results. Regarding the correlations, it can be seen that the majority of the correlations were significant, the highest regarding the intersection between Gratitude and Religiosity. Thus, it is concluded that the G-20 is a scale with psychometric properties suitable for use in adults in Brazil.