Propriedades psicométricas da Escala Subjetiva De Carga Mental de Trabalho (ESCAM) versão para o Brasil

• Ano de defesa: 2023
• Autor(a) principal: Lima, Mauren Pimentel
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• dARK ID: ark:/26339/0013000019v07
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Enfermagem; UFSM; Programa de Pós-Graduação em Enfermagem; Centro de Ciências da Saúde
• Palavras-chave em Português: Psicometria; ESCAM; Validade; Confiabilidade; Modelagem por equações estruturais; Psychometrics; Validity; Reliability; Structural equation modeling; CNPQ::CIENCIAS DA SAUDE::ENFERMAGEM
• Link de acesso: http://repositorio.ufsm.br/handle/1/32979

This study aimed to evaluate the psychometric properties of the Subjective Mental Workload Scale (ESCAM), version for Brazil. ESCAM is a Spanish tool used to evaluate the global mental load perceived by the worker. It was validated in terms of content for Brazil by Cerdótes (2020). It has 20 items, with five dimensions on a five-point Likert scale. Psychometric and cross-sectional methodological study that comprised the measurement, structural and predictive relevance stages. Structural invariance, internal consistency, convergent validity, discriminant validity, collinearity, as well as model fit and prediction indices, were analyzed. The study population included administrative technicians in education at the Federal University of Santa Maria. The sample consisted of 450 participants. Data collection was carried out using an online questionnaire with sociodemographic, labor and health profile data, ESCAM Brazilian version, Job Stress Scale and Resilience Scale. The data were generated in an Office Excel spreadsheet, version 365. Data analysis was carried out by means of Statistical Package for the Social Sciences (SPSS, version 26), Statistical Analysis System (SAS, version 9.2) and SmartPLS® (version, 4.1.0.0) software, using descriptive and multivariate statistics through structural equation modeling with partial least squares (PLS-SEM). To this end, theoretical models formed by hypotheses were established. Ethical precepts were respected. Among the participants, 65.30% were women, 32.40% were aged between 30 and 40 years, 33.10% had master’s degrees and 44.70% had a length of service between 6 and 15 years. The scale showed satisfactory internal consistency, with reliability indicators between 0.70 and 0.95. It also presented convergent validity due to the average variance extracted, with values greater than 0.5. The scale achieved discriminant validity according to the Fornell-Larcker criterion, with the highest correlation being lower than the highest and HTMT<0.9. Multigroup confirmatory factor analysis demonstrated invariance (equivalence) of ESCAM in the gender groups, with an explanation coefficient (R²=0.256). The model fit indices were appropriate (VIF<0.50), indicating that there was no collinearity. The theoretical models had most of the hypotheses confirmed by the structural coefficients (H1: b¹ 0; tcal. >1.96; p<0.05). The evaluation of predictive relevance indicated good accuracy of the models (Q2>0). The ESCAM Brazilian version presented adequate psychometric properties for the analyzed population.