Prioritizing and mapping the distribution of high-risk audience segments: Rural teens and seat belt behavior

Background.  Young adults are disproportionately involved in fatal car crashes. In fact, motor vehicle accidents are the leading cause of death among teenagers. Furthermore, young people demonstrate lower seat belt use than their older counterparts. Data from the National Survey on Drug Use and Health was used to (1) identify particularly high-risk segments of teens and young adults and (2) locate these segments geographically. In cases where it is impractical or cost-prohibitive to conduct topic-specific surveys (e.g., for hard-to-reach populations), publicly available data via geographic information systems (GIS) software can be an effective option in market segmentation efforts with a strong geographic component. 

Methods.  Chi-Squared Automatic Interaction Detection (CHAID) was used to model seat belt behavior based on several demographic variables. Logistic regression techniques were used to validate the CHAID model. Finally, GIS software was used to determine target geographic areas. 

Results.  Population density emerged as a key behavioral predictor, with risky behavior increasing with residence rurality. Two rural segments were identified as at risk for both driving and riding in the front seat without a seat belt:  non-Hispanic white males (particularly low income) and non-Hispanic black males. Rural, non-Hispanic black females were identified as at risk only for risky passenger behavior. Based on these segments, maps were generated to illustrate the relative efficiency of media spending in particular markets. 

Conclusions. Rural teens and young adults demonstrate heightened levels of risky seat belt behavior. By estimating how many people within various media markets fall into distinct segments, messaging can be efficiently tailored and disseminated.