Abstract
OBJECTIVES: To examine the impact of racial/ethnic coding strategies on the estimated prevalence of risk behaviors among American Indian/Alaska Native (AI/AN) high school students. METHODS: Data from the national Youth Risk Behavior Survey (2017 and 2019) were analyzed (N=28,422). Racial/ethnic data were coded to identify "Multiracial/ethnic AI/AN students" and "AI/AN alone students." The prevalence of persistent feelings of sadness or hopelessness, suicidality, and violence victimization were compared across the coding schemes and with non-Hispanic White students. RESULTS: Of students who selfidentified as AI/AN, one in six (18%) were AI/AN alone. The prevalence of many health risk behaviors was significantly higher among AI/AN students than non-Hispanic/Latino White students. The precision of the risk behavior prevalence estimates, however, varied considerably. CONCLUSION: How racial/ethnic data were coded affected the precision of calculations of risk behavior prevalence among AI/AN students, who are often multiracial and of Hispanic/Latino ethnicity.