Composite abuse scale : validação psicométrica preliminar de uma escala de violência por parceiro íntimo contra as mulheres
Resumo
Resumo: Segundo a Organização Mundial da Saúde (OMS) uma em cada três mulheres sofre algum tipo de violência durante sua vida, o que equivale a cerca de 736 milhões de mulheres no mundo. Contudo, apesar da extensa amplitude dessa epidemia, no Brasil, não há instrumentos adequados para mensuração, tipificação ou para mostrar a frequência da violência contra a mulher brasileira. Por esta razão, este trabalho tem o objetivo de realizar a validação preliminar em português brasileiro da Composite Abuse Scale (CAS), por meio de uma avaliação das propriedades configurais e métricas com ênfase na validade interna. A CAS é um instrumento validado a nível mundial em diversos idiomas, que, além de tipificar, mede e fornece dados como a frequência e a severidade da violência por parceiro íntimo (VPI). Para isto foram utilizados dados secundários da pesquisa "Criança em Casa", coletados entre julho e setembro do ano de 2020, de forma online através do Google Forms, no qual as participantes foram convidadas a responder de forma voluntaria a um survey com questionários, entre eles a CAS. Com as duas coletas se obteve o total de 869 participantes que responderam a CAS pelo menos uma vez e 121 que responderam duas vezes. A fim de se checar a propriedade métrica da CAS, analisamos a confiabilidade do instrumento com o calculado do Alfa de Cronbach, que investiga o constructo interno da escala, e o coeficiente de Kappa, que explora a equivalência da CAS. Já para a analisarmos a propriedade configural, foi realizada a Análise Fatorial. Ademais, ao final foi aplicado um teste preliminar da validade interna escalar a partir do Teste de Hipótese. Com a população de 869 participantes calculou-se o Alfa de Cronbach, a Análise Fatorial e os testes de hipóteses e com as 121 participantes calcularam-se o coeficiente de Kappa. Para organizar os dados foi utilizado o Excel e para a análise dos dados, o software SPSS. Foi constatado que a escala possui consistência interna, já que todos os Alfas de Cronbach foram acima de 0,90 e abaixo de 0,95, não apresentando redundância. O coeficiente de Kappa não atingiu o esperado (> 0,4) pois ocorreu o paradoxo do Kappa, entretanto, a concordância foi maior que 82,6% demonstrando que a CAS possui equivalência. A análise fatorial resultou em 7 fatores e a análise de verificação de médias das respostas das participantes mostrou que não houve viés de aquiescência na amostra. Já o teste de hipótese é um teste piloto, feito a partir da regressão logística, foi realizado em duas etapas, nas quais foi testado o quanto as variáveis independentes influenciavam as variáveis dependentes, muitas delas se mostraram significativas [p>0,005], enquanto outras não. Podemos concluir que a CAS é uma escala complexa e válida para estudar a VPI no contexto brasileiro. Abstract: According to the World Health Organization (WHO), one in three women suffers some type of violence during their lifetime, which is equivalent to around 736 million women in the world. However, despite the extensive scope of this epidemic, in Brazil, there are no adequate instruments for measuring, typifying or showing the frequency of violence against Brazilian women. For this reason, this work aims to carry out the preliminary validation in Brazilian Portuguese of the Composite Abuse Scale (CAS), through an evaluation of the configural and metric properties with an emphasis on internal validity. The CAS is an instrument validated worldwide in several languages, which, in addition to typifying, measures and provides data such as the frequency and severity of intimate partner violence (IPV). For this, secondary data from the "Criança em Casa" survey was used, collected between July and September 2020, online through Google Forms, in which participants were invited to voluntarily respond to a survey with questionnaires, among them to CAS. With the two collections, a total of 869 participants were obtained who responded to the CAS at least once and 121 who responded twice. In order to check the metric property of the CAS, we analyzed the reliability of the instrument using Cronbach's Alpha calculation, which investigates the internal construct of the scale, and the Kappa coefficient, which explores the CAS equivalence. To analyze the configural property, Factor Analysis was carried out. Furthermore, at the end a preliminary test of scalar internal validity was applied based on the Hypothesis Test. With the population of 869 participants, Cronbach's Alpha, Factor Analysis and hypothesis tests were calculated and with the 121 participants the Kappa coefficient was calculated. Excel was used to organize the data and SPSS software was used for data analysis. It was found that the scale has internal consistency, as all Cronbach's Alphas were above 0.90 and below 0.95, showing no redundancy. The Kappa coefficient did not reach what was expected (> 0.4) because the Kappa paradox occurred, however, the agreement was greater than 82.6%, demonstrating that the CAS is equivalent. The factor analysis resulted in 7 factors and the analysis of checking the means of the participants' responses showed that there was no acquiescence bias in the sample. The hypothesis test is a pilot test, carried out using logistic regression, it was carried out in two stages, in which it was tested how much the independent variables influenced the dependent variables, many of them proved to be significant [p>0.005], while others no. We can conclude that the CAS is a complex and valid scale for studying IPV in the Brazilian context.
Collections
- Dissertações [116]