121.4
Cumulative Risk/CR and Developmental Outcomes Among Youths in High-Risk Families: Results from a Nationally Representative Taiwan Sample
CR was created by 8 variables measuring different aspects of family risks. OLS regression models with gender, ethnicity, and Squared CR as independent variables were applied for each of two outcome variables (i.e. academic performance for delinquent outcomes). The findings support a linear model (Squared CR coef.=0.01, p=0.07 for academy model, coef.=0.001, p=0.92 for delinquency model) indicating those who experience more risks are more likely to develop negative outcomes. With one exception, a single risk as well as CR predicted both outcomes. The risk factor, losing someone during the great earthquake, only predicts academic performance but not delinquency. Moreover, with one exception, the effect of CR is similar across ethnic groups. The effect of CR on academic performance is lower among aboriginal Taiwanese comparing to the reference group, Minnan (coef.=0.11, p=0.02).
Although more research is needed to better understand the difference between aboriginal Taiwanese and Minnan, the research findings confirm the linear CR model, highlight the influence of CR on youths’ academic and behavior outcomes and suggest potential differences among ethnic groups. Findings also support the comprehensive prevention for youths in high-risk families, particularly those experiencing multiple risks.