Latent variable and clustering methods in intersectionality research: systematic review of methods applications.

Area of Science:

• Health Equity Research
• Quantitative Social Sciences
• Intersectionality Studies

Background:

• The intersectionality framework is increasingly used in quantitative health equity research to address social power dynamics.
• Person-centered methods, particularly clustering techniques, are identified as suitable for capturing intersectional complexities.
• A systematic review is needed to evaluate the application of these methods in conjunction with intersectionality theory.

Purpose of the Study:

• To systematically review the use of person-centered clustering methods in quantitative intersectionality research.
• To assess how these methods align with intersectionality theory in practice.
• To identify advantageous applications and missed opportunities for these methods within intersectional frameworks.

Main Methods:

• Conducted a multidisciplinary systematic review of quantitative studies explicitly using an intersectional approach and clustering methods.
• Extracted study characteristics, including the application of intersectionality and chosen clustering techniques.
• Included English-language studies, identifying 16 eligible papers from an initial pool of 782.

Main Results:

• Sixteen studies met eligibility criteria, primarily employing latent class analysis, latent profile analysis, or other clustering methods.
• Studies predominantly used cross-sectional data, were led by U.S. authors, and published in psychology, social sciences, and health journals.
• While most papers defined intersectionality and cited key authors, engagement with methodological literature was limited; clustering variables focused on social identities, dimensions, or processes, commonly resulting in four distinct classes/clusters used in further analyses.

Conclusions:

• Latent variable and clustering methods demonstrate diverse applications within intersectional quantitative research, reflecting varied theoretical-methodological alignment.
• The review highlights specific contexts where these methods are beneficial and suggests potential for broader utilization.
• Further research can explore refined integration of these statistical approaches with intersectionality theory to enhance health equity investigations.