This dissertation investigates the interactive or joint influence of autocorrelative processes (autoregressive-AR, moving average-MA, and autoregressive moving average-ARMA) and sample weights present in a longitudinal panel data set. Specifically, to what extent are the sample estimates influenced when autocorrelation (which is usually present in a panel data having correlated observations and errors) and sample weights (complex sample design feature used in longitudinal data having multi-stage sampling design) are modeled versus when they are not modeled or either one of them is taken into account. The current study utilized a Monte Carlo simulation design to vary the type and magnitude of autocorrelative processes and sample weights as factors incorporated in growth or latent curve models to evaluate the effect on sample latent curve estimates (mean intercept, mean slope, intercept variance, slope variance, and intercept slope correlation). Various latent curve models with weights or without weights were specified with an autocorrelative process and then fitted to data sets having either the AR, MA or ARMA process. The relevance and practical importance of the simulation results were ascertained by testing the joint influence of autocorrelation and weights on the Early Childhood Longitudinal Study for Kindergartens (ECLS-K) data set which is a panel data set having complex sample design features. The results indicate that autocorrelative processes and weights interact with each other as sources of error to a statistically significant degree. Accounting for just the autocorrelative process without weights or utilizing weights while ignoring the autocorrelative process may lead to bias in the sample estimates particularly in large-scale datasets in which these two sources of error are inherently embedded. The mean intercept and mean slope of latent curve models without weights was consistently underestimated when fitted to data sets having AR, MA or ARMA process. On the other hand, the intercept variance, intercept slope, and intercept slope correlation were overestimated for latent curve models with weights. However, these three estimates were not accurate as the standard errors associated with them were high. In addition, fit indices, AR and MA estimates, parsimony of the model, behavior of sample latent curve estimates, and interaction effects between autocorrelative processes and sample weights should be assessed for all the models before a particular model is deemed as most appropriate. If the AR estimate is high and MA estimate is low for a LCAR model than the other models that are fitted to a data set having sample weights and the fit indices are in the acceptable cut-off range, then the data set has a higher likelihood of having an AR process between the observations. If the MA estimate is high and AR estimate is low for a LCMA model than the other models that are fitted to a data set having sample weights and the fit indices are in the acceptable cut-off range, then the data set has a higher likelihood of having an MA process between the observations. If both AR and MA estimates are high for a LCARMA model than the other models that are fitted to a data set having sample weights and the fit indices are in the acceptable cut-off range, then the data set has a higher likelihood of having an ARMA process between the observations. The results from the current study recommends that biases from both autocorrelation and sample weights needs to be simultaneously modeled to obtain accurate estimates. The type of autocorrelation (AR, MA or ARMA), magnitude of autocorrelation, and sample weights influences the behavior of estimates and all the three facets should be carefully considered to correctly interpret the estimates especially in the context of measuring growth or change in the variable(s) of interest over time in large-scale longitudinal panel data sets.

More power semiconductors are applying to the aircraft power system to make the system smaller, lighter and more reliable. Average models provide a good solution to system simulation and can also serve as the basis to derive the small signal model for system-level study using linear control theory. A new average modeling approach for three-phase and nine-phase diode rectifiers with improved ac and dc dynamics is proposed in this dissertation. The key assumption is to model the load current using its first-order Taylor Series expansion throughout the entire averaging time span. A thorough comparison in the time domain is given of this model and two additional average models that were developed based on different load current assumptions, using the detailed switching models as the benchmark. The proposed average model is further verified by experimental results. In the frequency domain, the output impedance of a nine-phase diode rectifier is derived, and the sampling effect in the average model is investigated by Fourier analysis. The feeder's impedance before the rectifier is modeled differently in the output impedance in contrast in the equivalent commutation inductance. The average model is applied to the resonance study in a system composed of a synchronous generator, a nine-phase diode rectifier and a motor drive. The Thevenin's and Norton's equivalent circuits are derived to construct a linearized system. The equivalent impedance are derived from the average models, and the source are obtained from the switching circuit by short-circuit or open-circuit. Transfer functions are derived from the harmonic sources to the bus capacitor voltage for resonance study. The relationship between the stability and the resonance is analyzed, and the effect of controllers on the resonance is investigated. Optimization is another system-level application of the average model. A half-bridge circuit with piezoelectric actuator as its load is optimized using genetic algorithm. The optimization provides the possibility to design the actuator and its driving circuit automatically. / Ph. D.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

In this educational study, the student population at one of the state’s 15 community colleges was the target over a 5-year period (FY 2004-FY2009). Four variables (gender, race, socioeconomic status, and enrollment status were studied to predict if they had any affect on a student’s transfer grade point average. In 4 out of the 5 years in the study, this institution had the highest transfer grade point average as compared to native students at the state’s 8 universities. The purpose of this study was to examine the student population and look at four student variables to see if any were related to transfer grade point average. Over a 5-year period for this study, the number of women who have attended this community college has been 2 to 1. The number of students who receive a Pell Grant through financial aid is high. The ethnicity of the student population has also changed within the 5 years of this study.The first research question examined whether gender or race could predict a student’s transfer grade point average. The second research question explored whether socioeconomic status could predict a student’s transfer grade point average. The third research question asked whether a student’s enrollment status could predict the transfer grade point average. The fourth question examined which of the set of four variables had the most impact, and which one had the least impact. Race and sex were shown to have stronger relationships to transfer GPA. These variables only explain about 9% of grade variance; therefore, there are other factors that explain differences in the transfer GPA. The research concluded with a summary of the findings along with limitations of the study. Recommendations for practitioners and policy makers along with recommendations for future research were to study more variables, use other institutions, and perhaps to do a survey of the student population at community colleges.

While most public community colleges today advocate that they are open door and have liberal registration policies, there is little current research on the effects of late registration on student performance at the community college level. Community colleges need sound evidence in order to implement institutional practices and policies that will benefit students. The purpose of this study was twofold: (1) to examine the effects of late registration on student success at a rural Mississippi community college and (2) to identify reasons that students register late. In examining the effects of late registration on student success the study focused on the success measures of student GPA, course withdrawal, and persistence. Data for the first study purpose were obtained from the records of students enrolled at the respective college during the fall 2011, 2012, and 2013 semesters. For the second study purpose data were obtained using a self-developed survey that was emailed to students who late registered during the fall 2014 semester. Independent samples t-test, chi-square, frequencies, and percentages were used for data analysis. Results of the study indicate that late registration has a significantly negative effect on student success. Results of the statistical analysis are presented in narrative and table form to answer the 4 research questions. The study concludes with a summary of findings and a discussion of the limitations of the study. Recommendations for practitioners and policymakers are discussed along with recommendations for future research.