Abstract
Historically, instruments normed on White people have been used with other populations implying the concepts measured are acultural. Similarly, instruments developed using both women and men have assumed there aren’t differences between the groups that may be salient. However, nuances exist and pervade each assessment. This article examines the construct validity of three commonly used health behavior assessments: Five‐Facet Mindfulness Questionnaire, Perceived Stress Scale, and the John Henryism Active Coping Scale. With a sample of 223 Black/African American women living in the United States, the analyses used the Rasch’s partial credit model as a measure of unidimensionality. The findings reflect the complexity of measuring subpopulations using instruments normed on one group or in aggregate. Many of the items misfit due to lack of predictability or the suppression of multiple dimensions. The items that did not fit the model described activities such as “control over irritations” and having “distressing thoughts” that are experienced differently due to the intersections of discrimination that disproportionately affect Black women in comparison to their White female and Black male counterparts. Subsequently, this article advocates for the development of tailored measures of health behavior when the construct or phenomenon is experienced differently.