Cross-Cultural Sources of Measurement Error in Substance Use Surveys

Abstract

We present an overview of the cross-cultural quality of survey reports of substance use behaviors in the United States. Empirical data from 36 published studies (1977-2003) are examined to evaluate the reliability and validity of substance use reports across cultural groups. In these studies, race/ethnicity are used as proxy indicators of respondent culture. In general, the available research suggests that, with a few exceptions, the quality of survey data on racial and ethnic disparities in substance use is often limited by differential measurement error. A conceptual paradigm is presented to consider a wide range of potential causes for these differences in measurement error that includes two dimensions: emphasis on negative vs. positive behavior patterns, and emphasis on internal vs. external causal factors. These two dimensions yield four potential models that are useful in understanding variations in substance use measurement error: the cultural deficit model, the cultural conflict model, the mainstream conformity model, and the cultural distrust model. Future research should focus on the ability of each of these alternative models to account for cultural variability in the quality of substance use reporting.

Race/Ethnicity Differences in the Validity of Self-Reported Drug Use: Results from a Household Survey

“…17 This model posits that race/ethnicity differences in underreporting reflect a divergence in values between African American and White respondents, especially with regard to privacy 18 and collectivism. 19 In support of this perspective, Johnson and Bowman 6 point out that research has suggested that African Americans have expressed more concerns about privacy and confidentiality in research than others. This model suggests that if race/ethnicity differences in substance abuse reporting validity exist, they may diminish in statistical models that control for respondent perceptions of privacy and question threat.…”

“…15,16 Regarding the former, Johnson and Bowman 6 note that "the basic notion is that low income African-Americans are locked in a cycle of poverty primarily because they have a poverty of culture." Misleading survey reports regarding substance abuse behavior (i.e., substance use underreporting on surveys) are viewed as an outgrowth of culturally divergent values and norms.…”

“…In their review, multiple studies suggested that African American and Hispanic respondents were characterized by less valid (or reliable) survey reports than White respondents. Accordingly, Johnson and Bowman 6 offered four theoretical explanations for this phenomenon which are briefly summarized below.…”

“…Thus far, the evidence for race/ethnicity variation in substance abuse reporting validity has been suggestive but not conclusive. 6,7 Theoretical Perspectives on Validity Differences in Surveys Johnson and Bowman 6 provided a comprehensive review of the literature that suggested consistent "cultural differences" in substance use reporting validity. In their review, multiple studies suggested that African American and Hispanic respondents were characterized by less valid (or reliable) survey reports than White respondents.…”

Neighborhood-level Determinants of Delayed HIV Diagnosis and Survival among HIV-positive Latinos, Florida 2000-2011

“…Although the proportion of Latinos with history of IDU (10%) was consistent with previous studies , IDU might be underreported among Latinos (Johnson and Bowman, 2003), particularly among those who are foreign-born and undocumented. Second, we were limited to using ZIP codes to define neighborhoods.…”