Matching Items (6)
Filtering by
- All Subjects: Clinical Psychology
- All Subjects: Psychology--Statistical methods.
- Genre: Doctoral Dissertation
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
Research has consistently shown that gay/lesbian/bisexual (GLB) or sexual minority youth are at an increased risk for adverse outcomes resulting from the stress caused by continual exposure to negative events (e.g., victimization, discrimination). The present study used a nationally representative sample of adolescents to test mechanisms that may be responsible for the differences in offending behaviors among sexual minority and heterosexual adolescents. Specifically, this study tested whether bisexual adolescents received less maternal support than did heterosexual adolescents because of their sexual orientation, thus increasing the likelihood that they run away from home. This study then examined whether the greater likelihood that bisexual adolescents running away would lead to them committing a significantly higher variety of income-based offenses, but not a significantly higher variety of aggression-based offenses. This study tested the hypothesized mediation model using two separate indicators of sexual orientation measured at two different time points, modeled outcomes in two ways, as well as estimated the models separately for boys and girls. Structural equation modeling was used to test the hypothesized direct and indirect relations. Results showed support for maternal support and running away mediating the relations between sexual orientation and offending behaviors for the model predicting the likelihood of committing either an aggressive or an income offense, but only for girls who identified as bisexual in early adulthood. Results did not support these relations for the other models, suggesting that bisexual females have unique needs when it comes to prevention and intervention. Results also highlight the need for a greater understanding of sexual orientation measurement methodology.
ContributorsMansion, Andre (Author) / Chassin, Laurie (Thesis advisor) / Barrera, Manuel (Committee member) / Grimm, Kevin J. (Committee member) / Toomey, Russell B (Committee member) / Arizona State University (Publisher)
Created2018
Description
Researchers who conduct longitudinal studies are inherently interested in studying individual and population changes over time (e.g., mathematics achievement, subjective well-being). To answer such research questions, models of change (e.g., growth models) make the assumption of longitudinal measurement invariance. In many applied situations, key constructs are measured by a collection of ordered-categorical indicators (e.g., Likert scale items). To evaluate longitudinal measurement invariance with ordered-categorical indicators, a set of hierarchical models can be sequentially tested and compared. If the statistical tests of measurement invariance fail to be supported for one of the models, it is useful to have a method with which to gauge the practical significance of the differences in measurement model parameters over time. Drawing on studies of latent growth models and second-order latent growth models with continuous indicators (e.g., Kim & Willson, 2014a; 2014b; Leite, 2007; Wirth, 2008), this study examined the performance of a potential sensitivity analysis to gauge the practical significance of violations of longitudinal measurement invariance for ordered-categorical indicators using second-order latent growth models. The change in the estimate of the second-order growth parameters following the addition of an incorrect level of measurement invariance constraints at the first-order level was used as an effect size for measurement non-invariance. This study investigated how sensitive the proposed sensitivity analysis was to different locations of non-invariance (i.e., non-invariance in the factor loadings, the thresholds, and the unique factor variances) given a sufficient sample size. This study also examined whether the sensitivity of the proposed sensitivity analysis depended on a number of other factors including the magnitude of non-invariance, the number of non-invariant indicators, the number of non-invariant occasions, and the number of response categories in the indicators.
ContributorsLiu, Yu, Ph.D (Author) / West, Stephen G. (Thesis advisor) / Tein, Jenn-Yun (Thesis advisor) / Green, Samuel (Committee member) / Grimm, Kevin J. (Committee member) / Arizona State University (Publisher)
Created2016
Description
The comparison of between- versus within-person relations addresses a central issue in psychological research regarding whether group-level relations among variables generalize to individual group members. Between- and within-person effects may differ in magnitude as well as direction, and contextual multilevel models can accommodate this difference. Contextual multilevel models have been explicated mostly for cross-sectional data, but they can also be applied to longitudinal data where level-1 effects represent within-person relations and level-2 effects represent between-person relations. With longitudinal data, estimating the contextual effect allows direct evaluation of whether between-person and within-person effects differ. Furthermore, these models, unlike single-level models, permit individual differences by allowing within-person slopes to vary across individuals. This study examined the statistical performance of the contextual model with a random slope for longitudinal within-person fluctuation data.
A Monte Carlo simulation was used to generate data based on the contextual multilevel model, where sample size, effect size, and intraclass correlation (ICC) of the predictor variable were varied. The effects of simulation factors on parameter bias, parameter variability, and standard error accuracy were assessed. Parameter estimates were in general unbiased. Power to detect the slope variance and contextual effect was over 80% for most conditions, except some of the smaller sample size conditions. Type I error rates for the contextual effect were also high for some of the smaller sample size conditions. Conclusions and future directions are discussed.
A Monte Carlo simulation was used to generate data based on the contextual multilevel model, where sample size, effect size, and intraclass correlation (ICC) of the predictor variable were varied. The effects of simulation factors on parameter bias, parameter variability, and standard error accuracy were assessed. Parameter estimates were in general unbiased. Power to detect the slope variance and contextual effect was over 80% for most conditions, except some of the smaller sample size conditions. Type I error rates for the contextual effect were also high for some of the smaller sample size conditions. Conclusions and future directions are discussed.
ContributorsWurpts, Ingrid Carlson (Author) / Mackinnon, David P (Thesis advisor) / West, Stephen G. (Committee member) / Grimm, Kevin J. (Committee member) / Suk, Hye Won (Committee member) / Arizona State University (Publisher)
Created2016
Description
The current dissertation combines the results of three recently published papers focused on impulsive personality traits, alcohol and cannabis co-use, and alcohol misuse. The first study used alcohol administration to test whether different impulsivity facets conferred risk for heightened alcohol craving through subjective alcohol response patterns. The second study built on this study by incorporating alcohol and cannabis co-use into the acquired preparedness model, using longitudinal data. This study tested whether generalized impulsivity and sensation seeking during late adolescence predisposed individuals to co-use alcohol and cannabis, and whether co-use was indirectly associated with alcohol problems ten years later through positive expectancies and heavy drinking. The third study tested whether impulsive traits moderated the day-level effect of alcohol and cannabis co-use (vs. alcohol-only) on within-person deviations in drinking quantity on drinking days. Through the combination of these three studies, this dissertation seeks to further advance the literature on multiplicative ways in which personality enacts risk for heavy drinking, alcohol and cannabis co-use, and alcohol problems.
ContributorsWaddell, Jack T. (Author) / Corbin, William R. (Thesis advisor) / Chassin, Laurie (Committee member) / Grimm, Kevin J. (Committee member) / Meier, Madeline H. (Committee member) / Arizona State University (Publisher)
Created2024
Description
The current study assessed whether the interrater reliability and predictive validity of fidelity ratings differed significantly across the modalities of audio and video recordings. As empirically supported programs are moving to scale, attention to fidelity, the extent to which a program is delivered as intended, is essential because high fidelity is needed for positive program effects. Consequently, an important issue for prevention science is the development of feasible and acceptable methods for assessing fidelity. Currently, fidelity monitoring is rarely practiced, as the typical way of measuring fidelity, which uses video of sessions, is expensive, time-consuming, and intrusive. Audio recording has multiple advantages over video recording: 1) it is less intrusive; 2) equipment is less expensive; 3) recording procedures are simpler; 4) files are smaller so it takes less time to upload data and storage is less expensive; 5) recordings contain less identifying information; and 6) both clients and providers may be more willing to have sensitive interactions recorded with audio only. For these reasons, the use of audio recording may facilitate the monitoring of fidelity and increase the acceptability of both the intervention and implementation models, which may serve to broaden the scope of the families reached and improve the quality of the services provided. The current study compared the reliability and validity of fidelity ratings across audio and video rating modalities using 77 feedback sessions drawn from a larger randomized controlled trial of the Family Check-Up (FCU). Coders rated fidelity and caregiver in-session engagement at the age 2 feedback session. The composite fidelity and caregiver engagement scores were tested using path analysis to examine whether they predicted parenting behavior at age 3. Twenty percent of the sessions were double coded to assess interrater reliability. The interrater reliability and predictive validity of fidelity scores and caregiver engagement did not significantly differ across rating modality. However, caution must be used in interpreting these results because the interrater reliabilities in both conditions were low. Possible explanations for the low reliability, limitations of the current study, and directions for future research are discussed.
ContributorsRudo-Stern, Jenna Simone (Author) / Wolchik, Sharlene A (Thesis advisor) / Mauricio, Anne M (Committee member) / Grimm, Kevin J. (Committee member) / González, Nancy (Committee member) / Arizona State University (Publisher)
Created2019