Identifying visible minorities or racialized persons on surveys: can we just ask?

Abstract

Objectives Visible minorities are a group categorized in health research to identify and track inequalities, or to study the impact of racialization. We compared classifications obtained from a commonly used measure (Statistics Canada standard) with those obtained by two direct questions-whether one is a member of a visible minority group and whether one is perceived or treated as a person of colour. Methods A mixed-methods analysis was conducted using data from an English-language online survey (n = 311) and cognitive interviews with a maximum diversity subsample (n = 79). Participants were Canadian residents age 14 and older. Results Agreement between the single visible minority item and the standard was good (Cohen's Κ = 0.725; 95% CI = 0.629, 0.820). However, participants understood "visible minority" in different and often literal ways, sometimes including those living with visible disabilities or who were visibly transgender or poor. Agreement between the single person of colour item and the standard was very good (Κ = 0.830; 95% CI = 0.747, 0.913). "Person of colour" was more clearly understood to reflect ethnoracial background and may better capture the group likely to be targeted for racism than the Statistics Canada standard. When Indigenous participants who reported being persons of colour were reclassified to reflect the government definition of visible minority as non-Indigenous, this measure had strong agreement with the current federal standard measure (K = 0.851; 95% CI = 0.772, 0.930). Conclusion A single question on perception or treatment as a person of colour appears to well identify racialized persons and may alternately be recoded to approximate government classification of visible minorities.

Increases in Alcohol and Cannabis Use Associated with Deteriorating Mental Health among LGBTQ2+ Adults in the Context of COVID-19: A Repeated Cross-Sectional Study in Canada, 2020–2021

“…This item and our survey's partial conflation of sex/gender likely contributed to misclassification bias [64], especially for trans respondents, given current recommendations to use two-step measures for identification of trans survey respondents [65]. Self-report biases may similarly have undercounted racialized and/or Indigenous persons [66,67]. The pooled LGBT2Q+ identity measure also hindered our capacity to conduct sub-population analyses between LGBTQ2+ respondents (e.g., gay vs. lesbian vs. bisexual individuals)-this is an important limitation and an area for targeted research.…”

Access to health care among racialised immigrants to Canada in later life: a theoretical and empirical synthesis

“…This review article prefers ‘racialised’ over other terminologies such as ‘ethnocultural minority’, ‘ethnic minority’ and ‘visible minority’ to acknowledge the socially constructed nature of racialisation as a stratification process of unequal power relations between ethnic groups (Veenstra, 2009; Bauer et al , 2020). The study aims to synthesise both theoretical and empirical literature to develop and validate a theory-informed evidence-based framework that could explicate what (the substance) and how (the process and mechanism) socio-structural determinants lead to inequitable access to care experienced by racialised older immigrants in Canada.…”

Review of third-wave therapies for substance use disorders in people of color and collectivist cultures: Current evidence and future directions.

“…This is problematic due to the distinct needs of various cultures and does not acknowledge how participants experience their race, as studies typically rely on precategorized self-report data on race/ethnicity. Using surveys to identify POC may not fully reflect the experiences of participants with racialization (Bauer et al, 2020). Future research should also consider measures of acculturation or racial identity in addition to qualitative interviews regarding participant experiences of substance use treatment and the type of therapeutic elements they have found the most helpful (Greenfield et al, 2018).…”