Matching Items (3)
Filtering by
- All Subjects: Mexican American
- Creators: Tein, Jenn-Yun
- Member of: ASU Electronic Theses and Dissertations
Description
Random Forests is a statistical learning method which has been proposed for propensity score estimation models that involve complex interactions, nonlinear relationships, or both of the covariates. In this dissertation I conducted a simulation study to examine the effects of three Random Forests model specifications in propensity score analysis. The results suggested that, depending on the nature of data, optimal specification of (1) decision rules to select the covariate and its split value in a Classification Tree, (2) the number of covariates randomly sampled for selection, and (3) methods of estimating Random Forests propensity scores could potentially produce an unbiased average treatment effect estimate after propensity scores weighting by the odds adjustment. Compared to the logistic regression estimation model using the true propensity score model, Random Forests had an additional advantage in producing unbiased estimated standard error and correct statistical inference of the average treatment effect. The relationship between the balance on the covariates' means and the bias of average treatment effect estimate was examined both within and between conditions of the simulation. Within conditions, across repeated samples there was no noticeable correlation between the covariates' mean differences and the magnitude of bias of average treatment effect estimate for the covariates that were imbalanced before adjustment. Between conditions, small mean differences of covariates after propensity score adjustment were not sensitive enough to identify the optimal Random Forests model specification for propensity score analysis.
ContributorsCham, Hei Ning (Author) / Tein, Jenn-Yun (Thesis advisor) / Enders, Stephen G (Thesis advisor) / Enders, Craig K. (Committee member) / Mackinnon, David P (Committee member) / Arizona State University (Publisher)
Created2013
Description
Ethnic enclaves, or neighborhoods with high ethnic densities, have been linked with positive health outcomes and lower crime rates. Using data from the Pathways to Desistance project, this study tested whether neighborhood Latino concentration prospectively predicted re-offense rates among a sample of Mexican American juvenile offenders (n = 247). Further, I tested whether the effect of neighborhood Latino concentration on re-offense was moderated by ethnic identity, Mexican orientation, and generation status. Covariates included demographics and risk factors for offending. Results showed that neighborhood Latino concentration, ethnic identity, Mexican orientation, and generation status were not predictive of re-offense rates. Gender, risk for offending, and time spent supervised during the follow-up period predicted re-offense rates one year later. The results highlight the importance of risk assessment for this high risk group.
ContributorsBui, Leena (Author) / Chassin, Laurie (Thesis advisor) / Knight, George (Committee member) / Tein, Jenn-Yun (Committee member) / White, Rebecca (Committee member) / Arizona State University (Publisher)
Created2018
Description
Internalizing symptoms are prevalent among adolescents, especially among Latinos, and can have negative consequences on health and development. Understanding the risk and protective factors leading to internalizing difficulties among Latino youth is critical. The current study sought to assess the effects of family risk and peer social rejection in the seventh grade on internalizing symptoms in the tenth grade, and the potential buffering effects of social support from family and from friends, among a sample of 749 Mexican American youth. Structural equation modeling was used to examine pathways from seventh grade family risk and peer social rejection to internalizing symptoms in the tenth grade. Perceived social support from family and perceived social support from friends were tested as moderators of these relations. Gender differences in these pathways were also assessed. Results showed that family risk did not predict tenth grade internalizing symptoms, but that peer social rejection predicted increased internalizing symptoms for girls. Furthermore, buffering effects were not confirmed; rather social support from both friends and family had no effect on the relation between family risk and internalizing symptoms, and high levels of social support from both sources amplified the effect of peer social rejection on internalizing symptoms. Secondary analyses suggested that at low levels of social support from both sources, peer social rejection predicted decreased internalizing symptoms for males. Limitations and implications for prevention and future research are discussed.
ContributorsJenchura, Emily C (Author) / Gonzales, Nancy (Thesis advisor) / Tein, Jenn-Yun (Committee member) / Luecken, Linda (Committee member) / Arizona State University (Publisher)
Created2015