Psychometric evaluation of the Intersectional Discrimination Index for use in Brazil

Cad Saude Publica. 2024 Sep 23;40(9):e00009724. doi: 10.1590/0102-311XEN009724. eCollection 2024.
[Article in English, Portuguese]

Abstract

This cross-sectional study evaluated the configural and metric structures of the Intersectional Discrimination Index (InDI), an instrument that measures anticipated (InDI-A), dat-to-day (InDI-D), and major (InDI-M) discrimination. Data from a broader study, focused on the impacts of discrimination on the mental health of women living in Brazil, were used. Approximately 1,000 women, selected according to a convenience sampling scheme, answered the InDI and questions about sociodemographic characteristics in an electronic form that was administered in 2021. Exploratory factor analyses and exploratory structural equation modeling were applied to the first half of the sample; for the second, confirmatory factor analysis was conducted. Taken together, the findings suggest that each of the three measures is one-dimensional. However, unlike the study that originally proposed the InDI for use in Canada and the United States, we observed the presence of residual correlations in the three subscales evaluated, all of which were suggestive of content redundancy between specific pairs of items. The three measures showed moderate to strong factor loadings and acceptable fit to the data. InDI exhibited reasonable internal validity, potentially becoming a valuable instrument for investigating the health effects of intersectional discrimination in Brazil. Future studies should evaluate the consistency of these findings, examine the scalar structure of the instrument, and analyze its invariance among different marginalized groups.

Este estudio transversal evaluó las estructuras configural y métrica del Intersectional Discrimination Index (InDI), un instrumento que mide la discriminación anticipada (InDI-A), diaria (InDI-D) y mayor (InDI-M). Se utilizaron datos de una encuesta más amplia, centrada en los impactos de la discriminación en la salud mental de las mujeres que viven en Brasil. Aproximadamente 1.000 mujeres, seleccionadas por conveniencia, respondieron el InDI y preguntas sobre características sociodemográficas en formulario electrónico, aplicado en el 2021. Mientras que en la primera mitad de la muestra se realizaron análisis factoriales exploratorios y se realizó un modelado por ecuaciones estructurales exploratorias, en la segunda se realizó un análisis factorial confirmatorio. En conjunto, los hallazgos sugieren que cada una de las tres medidas es unidimensional. Sin embargo, a diferencia del estudio que originalmente propuso el InDI para su uso en Canadá y Estados Unidos, observamos la presencia de correlaciones residuales en las tres subescalas evaluadas, todas ellas sugestivas de redundancia de contenido entre pares específicos de ítems. Las tres medidas presentaron cargas factoriales de moderadas a fuertes e índices de ajuste aceptables. El InDI exhibió indicadores de validez interna razonables, convirtiéndose potencialmente en un instrumento valioso para investigar los efectos de la discriminación interseccional para la salud en Brasil. Los estudios futuros deben evaluar la consistencia de estos hallazgos, examinar la estructura escalar del instrumento y analizar su invariancia entre diferentes grupos marginados.

MeSH terms

  • Adolescent
  • Adult
  • Brazil
  • Cross-Sectional Studies
  • Factor Analysis, Statistical
  • Female
  • Humans
  • Middle Aged
  • Psychometrics*
  • Reproducibility of Results
  • Social Discrimination
  • Socioeconomic Factors*
  • Surveys and Questionnaires / standards
  • Young Adult