Spurious Relationships: Modeling the Effect of Steady Dating and Marriage on Deviance among Taiwanese Youth

• In Event: International Perspectives on Crime, Violence, and Victimization Over the Life-Course

Abstract

Using marginal structural models for counterfactual inference and a novel application of Bayesian Item Response Theory (IRT) models to the measurement of crime, this paper investigates the effect of intent to marry on deviance and of marriage on smoking for Taiwanese youth between the ages of 20 and 26. The research draws on the Taiwan Youth Project (TYP), a longitudinal survey that presents challenges for analysis because the items measuring deviance change every year. The problem is solved by using vertical scaling techniques from the field of educational assessment to place latent IRT measures of deviance onto a common scale. Fixed and random effects models indicate that the mere intention to get married (either in general or to one’s dating partner) has a spurious association with general deviance. Due to data limitations, analyses of marriage itself are restricted to its effect on smoking: while fixed effects models show that Taiwanese youth who are married tend to not smoke, inverse probability weighting models that adjust for selection into marriage and survey attrition show that getting married does not actually cause youth to quit. Together, the results suggest that steady dating and early marriage do little to deter Taiwanese youth from deviance.

Authors

• Richard Callahan, University of Washington

• Wen-Hsu Lin, Academia Sinica