High school student responses to different question formats assessing race/ethnicity

Abstract

Purpose: In 2005, the format for assessing race/ethnicity on the national Youth Risk Behavior Survey (YRBS) was changed from one to two questions. The 2005 Chicago YRBS included the single-question and two-question formats, providing an opportunity to identify how the change affects reporting of race/ethnicity.

Methods: Students in grades 9-12 (n = 808) were asked at the beginning of a 91-item questionnaire, "How do you describe yourself?" with "Hispanic or Latino" as one of several response options. At the end of the questionnaire, students were asked, "Are you Hispanic or Latino?" and then "What is your race?"

Results: Using standard algorithms to categorize students, 10.6% were classified as White, 46.2% as Black, 33.2% as Hispanic/Latino, 9.2% as other, and .9% as missing based on the single-question format. The two-question format yielded a similar distribution: 10.3% White, 41.1% Black, 34.8% Hispanic/Latino, 8.2% other, and 5.7% missing. The difference between these distributions was explained by the disproportionate number of Black students who left multiple questions blank at the end of the questionnaire. Regardless of whether the single-question or two-question format was used, 92.0% of students were classified the same (kappa = .85) when comparing distributions of the four-category race/ethnicity variable.

Conclusions: These results suggest self-reported race/ethnicity among high school students is similar regardless of which question format is used, and the changed format will not affect the ability of YRBS data users to conduct trend analysis by race/ethnicity.