Browsing by Subject "Hierarchical linear modeling"

Now showing 1 - 4 of 4

Assessing sample size and mobility limits for a two-level multiple membership random effects model
(2017-05) Wheelis, Meaghann Marie; Pituch, Keenan A.; Beretvas, Susan N; Hersh, Matthew; Pustejovsky, James; Whittaker, Tiffany
The purpose of the present study was to examine the minimal sample sizes and mobility rates needed for accurate estimation with the multiple-membership random effects model (MM-REM), and the conditions in which the MM-REM provided improved estimation over a traditional multilevel model (MLM). The mobility rate, level one and level two sample size, and the ICC for the level one predictor were manipulated for both a traditional MLM and an MM-REM. Relative parameter bias, relative standard error bias, and credible interval coverage were evaluated across 36 conditions for both methods. Standard error estimates of the intercept were negatively biased and credible interval coverage was low for both methods, and estimation improved as the level one sample size increased and as the ICC for the level one predictor decreased. Additionally, for MLM estimates, standard error bias decreased and credible interval coverage improved as the mobility rate decreased. Negative relative parameter bias was found for estimates of the level two coefficient, which was found to increase as the mobility rate increased for both methods. The level two variance component was overestimated with the MM-REM and underestimated with the MLM, and credible interval coverage was low for both methods. Estimation improved for MM-REM estimates as the level two sample size increased, and as the mobility rate decreased for MLM estimates. The results from the study suggest that, if applied researchers are interested primarily in estimates of the regression coefficients associated with predictors, both the MLM and the MM-REM provide accurate estimates of the level one coefficient, and accurate credible intervals for estimates of the level two coefficient. When applied researchers are interested in the variance components, however, the MM-REM should be used over the MLM when mobility exceeds 10% and the level two sample size is 40 or greater. The results of the study for each condition, in addition to study limitations and recommendations for applied researchers, are discussed.
Bayesian estimation of a longitudinal mediation model with three-level clustered data
(2015-12) Israni, Anita; Beretvas, Susan Natasha; Hersh, Matthew; Pituch, Keenan; Roberts, Gregory; Whittaker, Tiffany
Longitudinal modeling allows researchers to capture changes in variables that take time to exert their effects. Furthermore, incorporating mediation into a longitudinal model allows for researchers to test causal inferences about, for example, how an independent variable might affect growth in an outcome variable through growth in a mediating variable. In scenarios in which multiple variables are measured over time, the parallel process model can be used to model the inter-relationships among the measures’ trajectories where both processes are modeled to have their own separate but related growth parameters. The hierarchical linear modeling (HLM) framework can be used to model a parallel process model and allows for easy extensions to handle multiple levels and non-hierarchical data, such as cross-classified or multiple membership data structures, in clustered data. This study assessed a three-level parallel process model couched in the context of longitudinal mediation where treatment was assigned at the cluster level, matching a longitudinal cluster randomized trial design. The treatment’s effect on growth in an outcome is modeled as mediated by the growth in a mediating variable at the cluster and individual level, resulting in a cross-level and cluster-level mediated effect. A simulation and real data analysis study were conducted using a fully Bayesian analysis. In the simulation study, the following four factors were manipulated to assess the recovery of the parameters of interest: mediated effect size, random effects variance component values, number of measurement occasions, and number of clusters. Overall, relative parameter bias and statistical power improved for higher values for each of the four factors. The cross-level mediated effects were less biased and had greater statistical power than the cluster-level mediated effects. For the mediated effects that were truly zero, coverage rates based on the highest posterior density intervals showed mostly acceptable rates for the cross-level mediated effect and when path b was zero paired with a non-zero path a for the cluster-level effect. For conditions with a true value of zero for the cluster-level mediated effect with a path a of zero, the cluster-level coverage rates provided over-coverage. Results are discussed along with clarification of study limitations and suggestions for future research. Recommendations for applied researchers are also noted.
Teachers’ motivation and emotion during professional development : antecedents, concomitants, and consequences
(2017-05) Osman, David J.; Schallert, Diane L.; Patall, Erika A; Pustejovsky, James E; Schutz, Paul A
Professional development (PD) opportunities are offered to teachers as means for them to develop their knowledge and teaching practices, with the hope of improved learning outcomes for students. However, PD experiences often do not improve teacher knowledge or lead to changed teacher practices. Research exploring how teachers interact with professional development can serve as a powerful tool and help to outline further the landscape of professional development. Specifically, understanding the intersections of motivation, emotion, and teacher learning may inform our understanding of why teachers do or do not implement what they learn in PD and contribute to theories about the motivation-emotion-learning connection. Theoretical frameworks influencing this work include Expectancy-Value theory of motivation (Eccles et al., 1983), with the idea that the theory may help with explaining teachers’ motivation during PD by way of teachers’ expectancies for successful implementation, value for implementing, and perceived costs of implementing influence their intentions to implement what they learned in PD. In addition to motivation, this study considers teachers’ emotional experiences during professional development. Emotion theories, as formulated by Pekrun (2006) and Fredrickson (2001), frame emotions as the product of cognitions, and emotions as antecedents to future cognition. In this way, emotions can support or hinder teachers’ learning during PD. As teaching is an emotionally laden profession (Hargreaves, 1998), the consequences of teachers’ emotions during PD are especially important to understand why and how teachers’ learn and implement professional development. In this descriptive study, I measured the antecedents and consequences of teachers’ motivational and emotional experiences during PD. Educator participants (n = 673) were sampled from 64 summer professional development experiences. Participants completed two questionnaires, one immediately following the summer PD experience and a second in the following fall semester. Data were analyzed using hierarchical linear modeling. Results indicated that participants’ motivation to implement what they had learned in PD and the degree to which they had experienced pleasant affect during PD predicted their intentions to implement what they had learned. Participants’ motivation to implement was also predicted by their teaching self-efficacy. Implications for research and practitioners are discussed.
Unpacking the complex relationship between land use, vehicle travel, and transportation greenhouse gas (GHG) emissions
(2016-09-14) Choi, Kwangyul; Zhang, Ming, 1963 April 22-; Paterson, Robert; Whittaker, Tiffany; Bhat, Chandra; Jiao, Junfeng
This dissertation research aims to disentangle the relationship between land use, vehicle travel, and transportation greenhouse gas (GHG) emissions. A great number of studies have paid attention to the impact of land use on transportation GHG emissions using vehicle miles traveled (VMT) as a substitute. Most studies equated VMT reduction with reduction of transportation GHG emissions. Few have examined in depth the varying components that affect transportation GHG emissions in vehicle travel operational dimensions. Moreover, few have applied the use of larger geographic-level land use. These studies, however, have limitations in examining a comprehensive relationship between land use and transportation GHG emissions. This dissertation research therefore focuses on the links between land-use measures at various geographic levels and household vehicle travel characteristics impacts on transportation GHG emissions. In doing so, this dissertation research consists of the three closely related research questions. Using the 2009 National Household Travel Survey (NHTS), this research first examines whether neighborhood-level land use attributes proportionally affect household daily VMT and transportation GHG emissions (CO2e). A series of multiple regression models developed in Chapter Four address the impact of land use on household vehicle travel characteristics and transportation GHG emissions. Results suggest that land use strategies at the neighborhood level such as densification, a mixture of land use, and improvement of road connectivity can play a significant role in reducing vehicle travel. However, these land use changes may cause traffic delays in the area. Chapter Five focuses on the impact of multiple geographic-level land use (i.e., neighborhood, county, and MSA) on both household VMT and transportation GHG emissions by applying hierarchical linear modeling. Results suggest that the effectiveness of similar strategies can vary by geographic scales at which those strategies are implemented. Chapter Six examines the intervening effects of vehicle travel characteristics on transportation GHG emissions by employing structural equation modeling. Results suggest that land use at various geographic levels influence not only household VMT but also vehicle travel speed and vehicle trip frequency, which together in turn affect household transportation GHG emissions. Finally, this research presents a case study of the Austin, TX region using the 2006 Austin Travel Survey (ATS) in Chapter Seven. Applying a path model similar to the one developed in the preceding chapter, this study scrutinizes the role of land use in reducing transportation GHG emissions in both regional and local contexts. Results suggest that densification and a mixture of land use are still effective land use strategies to reduce region-wide emissions. However, design improvement can be a double-edged sword because of its unintended effect of reduced vehicle travel speed. Overall, the findings contend that both travel demand management and mobility management at various geographic levels should be fully discussed in the early stages of planning. In addition, the role of metropolitan planning organizations (MPOs) in controlling regional development should be extended. The expansion of authorities and responsibilities of MPOs may enable the region at all levels to be developed more sustainably.