Advertisement
Published in:
01-09-2020
Simpson’s Paradox in LGBTQ+ Policy: a Case Study
Published in: Sexuality Research and Social Policy | Issue 3/2021
Log inActivate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Abstract
Introduction
Simpson’s paradox occurs when trends found in the underlying data disappear or are reversed when groups are aggregated. Because reported data are used to guide policymaking, understanding and being able to identify instances of Simpson’s paradox is crucial to LGBTQ+ policy.
Method
The article offers a theoretical introduction to Simpson’s paradox before looking to a recent LGBT poverty study as a case study of the paradox and of its dangers in the context of LGBTQ+ policy. It then offers suggestions for minimizing the impact of the paradox for researchers and data users.
Results
The Williams Institute’s LGBT Poverty Study reported a much higher percentage of cisgender bisexual women living in poverty compared to other cisgender LGB groups. The rate of poverty was comparable to that of transgender respondents. However, the pattern was absent from age-disaggregated data and arose from the proportionately younger age of cisgender bisexual women in the study.
Conclusion
Aggregate statistics and how they are reported may misrepresent the causal relationship between group belonging and outcome.
Policy implications
Researchers should seek to minimize the risk of readers falling prey to the paradox by deliberately discussing the impact of confounding variables in the abstract or executive summary of their study, and by varying how they elect to report, represent, and disseminate the information. Policymakers and other data users should be cognizant of and attentive to Simpson’s paradox when interpreting studies.