Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
6 results
Search Results
Item Investigating Uncertainty with Fungible Parameter Estimate Analysis(2020) Prendez, Jordan Yee; Harring, Jeffrey R; Measurement, Statistics and Evaluation; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Researchers need methods for evaluating whether statistical results are worthy of interpretation. Likelihood functions contain large amounts of information regarding the support for differing estimates. However, maximum likelihood estimates (MLE) are typically the only set of estimates interpreted. Previous research has indicated that these alternative estimates can often be computed and represent data approximately as well as their MLE counterparts. The close fit between these alternative estimates are said to make them fungible. While similar in fit, fungible estimates are in some cases different enough (from the MLE) that they would support alternative substantive interpretations of the data. By calculating fungible parameter estimates (FPEs) one can either strengthen or weaken one’s inference by exploring the degree in which diverging estimates are supported. This dissertation has two contributions. First, it proposes a new method for generating FPEs under a broader definition of what should constitute fungible parameter estimates. This method allows for flexible computation of FPEs. Second, this method allows for an exploration of research inquiries that have been largely unexplored. What are the circumstances in which FPEs would convey uncertainty in the parameter estimates? That is, what are the causes of uncertainty that are measured by FPEs. Understanding the causes of this uncertainty are important for utilizing FPEs in practice. This dissertation uses a simulation study in order to investigate several factors that might be encountered in applied data analytic scenarios and affect the range of fungible parameter estimates including model misfit. The results of this study indicate the importance of interactions when examining FPEs. For some conditions, FPE ranges indicate that there was less uncertainty when the model was correctly specified. Under alternative conditions, FPE ranges suggest greater uncertainty for the correctly specified model. This example is mirrored in several results that suggest that a simple prediction of the level of uncertainty is difficult for likelihoods characterizing real world modeling scenarios.Item Exploring the Influence of Urban Form on Travel and Energy Consumption, using Structural Equation Modeling(2012) Liu, Chao; Ducca, Frederick W; Urban and Regional Planning and Design; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This dissertation has contributed to the current knowledge by gaining additional insights into the linkages of different aspects of the built environments, travel behavior, and energy consumption using Structural Equation Modeling (SEM) that provides a powerful analytic framework for a better understanding of the complex relationships of urban form, travel and energy consumption. Several urban form measurements (density, mixed land use index, street network connectivity, regional accessibility, and distance to transit) were gathered from multiple external sources and utilized for both trip/tour origins and destinations. This dissertation also contributed to the analysis framework by aggregating trips into tours to test whether the tour-based analysis generates better results than the trip-based analysis in terms of model fit, significance, and coefficient estimations. In addition to that, tour-based samples were also stratified into three different classification schemes to investigate the variations of relationship of urban form and travel among auto and transit modes and among various travel types.: (1) by modes (i.e. auto and transit); (2) by travel purposes (i.e. work, mixed, and non-work tours); and (3) by modes and purposes (first by modes, then by purpose). Stratification by purposes and modes provided an in-depth investigation of the linkages of urban form and travel behavior. The research findings are many: (1) urban form does have direct effects on travel distance for all tour types modeled; (2) urban form at the destination ends has more influence than on the origin ends; (3) Urban form has indirect effects on travel distance and energy consumption through affecting driving patterns, mode choice, vehicle type and tour complexity; (4) People tend to drive when they have complicated travel patterns; (5) The effects of intermediate variables (driving patterns, tour complexity, mode choice, and vehicle type) are stronger than the direct effects generated from urban form; (6) Tour-based analyses have better model fit than trip-based analysis; (7) Different types and modes of travel have various working mechanisms for travel behavior. No single transportation technology or land use policy action can offer a complete checklist of achieving deep reductions of travel and energy consumption while preserving mobility of driving.Item A Comparison of Methods for Testing for Interaction Effects in Structural Equation Modeling(2010) Weiss, Brandi A.; Harring, Jeffrey R.; Hancock, Gregory R.; Measurement, Statistics and Evaluation; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The current study aimed to determine the best method for estimating latent variable interactions as a function of the size of the interaction effect, sample size, the loadings of the indicators, the size of the relation between the first-order latent variables, and normality. Data were simulated from known population parameters, and data were analyzed using nine latent variable methods of testing for interaction effects. Evaluation criteria used for comparing the methods included proportion of relative bias, the standard deviation of parameter estimates, the mean standard error estimate, a relative ratio of the mean standard error estimate to the standard deviation of parameter estimates, the percent of converged solutions, Type I error rates, and empirical power. It was found that when data were normally distributed and the sample size was 250 or more, the constrained approach results in the least biased estimates of the interaction effect, had the most accurate standard error estimates, high convergence rates, and adequate type I error rates and power. However, when sample sizes were small and the loadings were of adequate size, the latent variable scores approach may be preferable to the constrained approach. When data were severely non-normal, all of the methods were biased, had inaccurate standard error estimates, low power, and high Type I error rates. Thus, when data were non-normal, relative comparisons were made regarding the approaches rather than absolute comparisons. In relative terms, the marginal-maximum likelihood approach performed the least poorly of the methods for estimating the interaction effect, but requires sample sizes of 500 or greater. However, when data were non-normal, the latent moderated structure analysis resulted in the least biased estimates of the first-order effects and had bias similar to that of the marginal-maximum likelihood approach. Recommendations are made for researchers who wish to test for latent variable interaction effects.Item THE IMPACT OF PRELIMINARY MODEL SELECTION ON LATENT GROWTH MODEL PARAMETER ESTIMATES(2010) Wang, Hsiu-Fei; Hancock, Gregory R; Measurement, Statistics and Evaluation; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)In latent growth modeling (LGM), model selection and inference are treated as separate stages of data analysis, but they are generally conducted on the assumption that the model is known a priori and thus model selection and inference are performed on the same data set. This two-step process ignores the effects of model uncertainty on parameter estimation and statistical inference, and thus may result in problems which ultimately lead to misleading or invalid inferences. This present study was thus designed to investigate the possible problems arising from the use of the two-step process in LGM. The goals of this study were: (1) To examine the subsequent impact of preliminary model selection using information criteria on LGM parameter estimates; (2) To assess the data splitting method as a possible way to mitigate the effects of model uncertainty. To achieve these goals, two Monte Carlo simulation studies were conducted. Study1 conducted both model selection and parameter estimation using the same data set, to investigate the possible impact of preliminary model selection in terms of relative parameter biases, and coverage rate. Study 2 conducted model selection and parameter estimation using different split-data sets, in order to assess the data splitting method as a possible way to mitigate the effects of model uncertainty. The major finding of this study was that inference based on the AIC or the BIC model selection leads to additional bias in the estimates and overestimates the sampling variability of the parameters estimates. The results of the simulation study showed that the post-model-selection parameter estimator has larger relative parameter biases, larger relative variance biases, and smaller coverage rate of confidence interval than those of the pre-model-selection estimator. These post-model-selection problems due to model uncertainty, unfortunately, still existed when the data splitting method was applied.Item A TALE OF TWO GROUPS: DIFFERENCES BETWEEN MINORITY STUDENTS AND NON-MINORITY STUDENTS IN THEIR PREDISPOSITION TO AND ENGAGEMENT WITH DIVERSE PEERS AT A PREDOMINANTLY WHITE INSTITUTION(2009) Hall, Wendell Diedrik; Cabrera, Alberto F.; Education Policy, and Leadership; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The purpose of this study was to examine the extent to which minority students and non-minority students differ in their predispositions to engage in campus-based diversity activities upon entering college and engagement with diverse college peers during college. These ethnicity-based interactional differences were examined under a revised version of the Transition to College Model (Locks et al., 2008). The Diverse College Student Engagement Model accounts for the joint influence of student pre-college characteristics along with collegiate experiences, in shaping engagement with racially diverse peers at a predominantly White college. Using Structural Equation Modeling (SEM) and Latent Means Modeling (LMM), this dissertation examined direct and indirect effects of factors that influence engagement with diverse students in college. Findings indicated that engagement with diverse peers does not take place in a vacuum; conditions and mechanisms that facilitate engagement also matter. Several pre-college variables and college variables were shown to influence predisposition to engage in diversity-related activities and engagement among diverse peers in college. Findings from testing the proposed model indicate that minority students were significantly higher in the latent factor Predisposition to Engage when entering college; however, no significant differences were found in the latent factor Engagement after the sophomore year of college. The differences appear to have been attenuated by some of the campus mechanisms the University of Maryland has in place to foster engagement among diverse students.Item The influence of maternal sensitivity and maternal stimulation on later development of executive functioning via structural equation modeling(2007-04-25) Emick, Jessica; Strein, William; Counseling and Personnel Services; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This study investigated the relations between early maternal behaviors, maternal sensitivity and maternal stimulation, and the later development of executive function. It was hypothesized that maternal behaviors could influence the development of executive function either directly or indirectly by influencing a child's language or attentional abilities. This study attempted to model these relationships using archival data from phase I and phase II from the Study of Early Child Care and Youth Development (SECCYD). Structural equation modeling was used with data from 470 participants on measures of SES, maternal sensitivity, maternal stimulation, language, attention, and executive function. From existing literature three nested models were proposed to examine how maternal behaviors influenced the later development of executive function. While there were significant differences between the three proposed models it is important to recognize the overall poor fit of the models. The differences between the models suggest that maternal sensitivity and maternal stimulation do not directly influence executive functioning in the 1st grade but instead influence the development of executive functioning through assisting the child in development of attention and language skills. Interestingly, the model also indicated verbal ability played an important role in the development of executive function. Secondly the study attempted to examine multi-group differences in the proposed models (Caucasian and African American). While small sample size precluded this analysis, examining the effect size differences between the two groups indicated that within the current sample ethnicity, language ability, and SES are deeply entangled. The results of the current study highlight the potential role of language ability in the development of executive function and the need for cleaner measures of executive function that are developmentally appropriate.