Ultrahigh-dimensional variable selection is of great importance in the statistical research. And independence screening is a powerful tool to select important variable when there are massive variables. Some commonly used independence screening procedures are based on single replicate data and are not applicable to longitudinal data. This motivates us to propose a new Sure Independence Screening (SIS) procedure to bring the dimension from ultra-high down to a relatively large scale which is similar to or smaller than the sample size. In chapter 2, we provide two types of SIS, and their iterative extensions (iterative SIS) to enhance the finite sample performance. An upper bound on the number of variables to be included is derived and assumptions are given under which sure screening is applicable. The proposed procedures are assessed by simulations and an application of them to a study on systemic lupus erythematosus illustrates the practical use of these procedures. After the variables screening process, we then explore the relationship among the variables. Graphical models are commonly used to explore the association network for a set of variables, which could be genes or other objects under study. However, graphical modes currently used are only designed for single replicate data, rather than longitudinal data. In chapter 3, we propose a penalized likelihood approach to identify the edges in a conditional independence graph for longitudinal data. We used pairwise coordinate descent combined with second order cone programming to optimize the penalized likelihood and estimate the parameters. Furthermore, we extended the nodewise regression method the for longitudinal data case. Simulation and real data analysis exhibit the competitive performance of the penalized likelihood method. / Doctor of Philosophy / Longitudinal data have received a considerable amount of attention in the fields of health science studies. The information from this type of data could be helpful with disease detection and control. Besides, a graph of factors related to the disease can also be built up to represent their relationships between each other. In this dissertation, we develop a framework to find out important factor(s) from thousands of factors in longitudinal data that is/are related to the disease. In addition, we develop a graphical method that can show the relationship among the important factors identified from the previous screening. In practice, combining these two methods together can identify important factors for a disease as well as the relationship among the factors, and thus provide us a deeper understanding about the disease.

Since the days of Booker T. Washington and W. E. B. Du Bois, the argument has flourished relative to the value of vocational education for Black youth. This study, using data from the "High School and Beyond 1980 Sophomore Cohort Third Follow-Up (1986)" survey, investigated three basic areas, namely: (a) the demographic, personological, and educational profile of Black youth enrolled in vocational education, and the manner in which this profile varied in relation to their vocational concentration patterns, (b) the profile of these youth in terms of their employment outcomes, educational expectations, and civic and political participation > practices after completion of their secondary schooling, according to their concentration patterns, and (c) the changes over time among these youth within their vocational concentration patterns, with regard to aptitude, educational and vocational expectations, and employment status. Major findings of this study have been presented for the students by concentration patterns. Some of the major overall findings were: Students with greater concentration in vocational education course work tended to come from urban areas, the southern region of the United States, and the lowest socioeconomic status quartile. Both educational and occupational expectations were unrealistic in terms of Standardized test performance and grades. A large percentage waS not in the labor force and a very small percentage was participating in civic or political activities. Findings for outcome and change over time variables were presented for the three vocational participation patterns, Concentrators, Limited Concentrators, and Samplers. / Ed. D.

Longitudinal data from 113 dual-career couples are used to explore the relationship between perceptions of marital equity and well-being and distress in marital, professional, and parental roles, how perceptions of equity change over time, and to examine efforts to restore equity. Differences in well-being and distress in roles are identified by gender and equity group. study results indicate that for both spouses, perceptions of inequity are associated with lower marital well-being and higher marital distress; however, under benefited wives reported higher professional well-being and under benefited husbands reported higher parental well-being. Couple perceptions of equity decreased between 1986 and 1990; under benefited husbands were more likely to use threats and bargaining to negotiate for relationship changes while wives sought counseling. Results demonstrate the importance of using multiple measures of well-being and distress and including gender and equity type in investigations of marital equity. / Ph. D.

Team cohesion represents arguably the most studied team construct as it has been consistently shown to be associated with improved performance. However, although cohesion is now understood to be an emergent state—as it develops over a team's life cycle—research has yet to uncover the dynamic nature of cohesion. The current study was designed to particularly test the impact of team personality composition both on the initial status of cohesion, and on changes in cohesion over time. 80 newly formed teams performed a highly interdependent team task, and team cohesion was measured over six time points. Personality was measured prior to the task and calculated at the team level, as both an average and a variability score. After performing longitudinal hierarchical linear modeling, results indicated that team personality impacts cohesion differently at initial status and over time. In particular, higher team agreeableness predicted greater slopes of cohesion, but not initial cohesion levels. Also, higher extraversion predicted greater initial status of cohesion, but not greater slopes. These results present important boundary conditions for understanding the role of team personality composition on team cohesion. / Master of Science

Parenting behaviors and a variety of behavioral problems in children covary. The current study first aimed to examine how and why parenting and child behavioral problems are linked in middle childhood. In particular, a longitudinal design (1364 children assessed from 54 months to 5th grade) was used to examine whether the developmental link between parenting and child behavioral problems were reciprocal. A twin design (131 pairs of monozygotic and 173 pairs of dizygotic twins assessed from 6 to 8 years of age on average) was used to examine the underlying genetic and nongenetic etiology of this link. In addition, using these two samples, the current study also aimed to examine whether parental attributes, including negative affect, executive function, and social cognitive factors, modulate the link between parenting and child behavioral problems. Results across these two studies suggested that parenting and child behavioral problems mutually influenced the development of each other over time, potentially through both evocative and passive gene-environment correlation processes and environmental transmissions. In addition, maternal dispositional anger modulated the effects of child behavioral problems on changes in maternal parenting quality over time. Finally, implications of the current study were also discussed. / Ph. D.

In recent decades, United States public school education has moved toward standards-based curricula. However, performance on standardized tests may not be representative of subject literacy or workforce preparedness. This misalignment may be particularly true in the sciences, where low science literacy and gender-related workforce shortfalls are evident. This study was an exploration of how well standardized test scores and other academic metrics reflected progression to a science major, by gender. This exploratory study used longitudinal data from the Virginia Department of Education, prepared by the Virginia Longitudinal Data System, for students who graduated from Virginia public schools from 2004-2016 (N=1,089,389). Students' standardized assessment scores, science course grades, demographics, and post-secondary major were analyzed using correlation analysis, logistic regression, principal component analysis, and hypothesis testing. Overall, 9% of high school completers enrolled in a post-secondary science major, with approximately half of those students attending 4-year schools. Seventy percent of science majors were female; females were most prevalent in health-related majors and least prevalent in physical sciences. Logistic regression identified the following factors significantly related to enrolling in a post-secondary science major: gender, high school science grades, and the high school's percent of students who majored in science. A student's status as economically disadvantaged or an underrepresented minority was significantly related to enrolling in a 2-year science major. In comparisons among academic metrics, standardized test scores and science grades were uncorrelated, and science grades differed significantly among demographic subgroups. Overall, demographic and school-level factors were more closely related to majoring in science than were academic factors. For both genders and for biological, physical, and health sciences, the percent of students majoring in science doubled from 2005-2015. Standardized test scores and course grades measured different aspects of learning, and higher science grades were related to majoring in science. However, the designation of "science major" is so broad as to be uninformative in a research context; more specificity would be needed to develop academic profiles. From these findings, one can conclude that demographic and cultural factors – rather than academic factors – were more closely related to whether students pursued a science pathway. / Doctor of Philosophy / In recent decades, United States public school education has moved toward standards-based curricula. However, performance on standardized tests may not represent subject knowledge or job preparedness, particularly in the science fields. This study was an exploration of how well standardized test scores and other academic measured were related to majoring in science, for male and female students. This exploratory study used data from the Virginia Department of Education, prepared by the Virginia Longitudinal Data System, for students who graduated from Virginia public schools from 2004-2016. Students' standardized test scores, science course grades, demographics, and college major were analyzed. Overall, 9% of high school completers enrolled in a science major after high school, with approximately half of those students attending 4-year schools. Seventy percent of science majors were female; females were most prevalent in health-related majors and least prevalent in physical sciences. The following factors were significantly related to enrolling in a science major: gender, high school science grades, and the high school's percent of students who majored in science. A student's status as economically disadvantaged or an underrepresented minority was significantly related to enrolling in a 2-year science major. In comparisons among academic measures, standardized test scores and science grades were not related to each other, and science grades differed among demographic groups. Overall, demographic and school-level factors were more closely related to majoring in science than were academic factors. For both genders and for biological, physical, and health sciences, the percent of students majoring in science doubled from 2005-2015. Standardized test scores and course grades measured different aspects of learning, and higher science grades were related to majoring in science. However, the designation of "science major" is so broad as to be uninformative in a research context; more specificity would be needed to develop academic profiles. From these findings, one can conclude that demographic and cultural factors – rather than academic factors – were more closely related to whether students pursued a science pathway.

Early childhood marks a time when word learning is accompanied by rapid growth in the cognitive processes that underlie self-modulated and goal-directed behavior (i.e., executive functions (EF)). Although there is empirical evidence to support the association between EF and vocabulary development in childhood, inconsistent findings have been reported regarding the extent to which early EF abilities predict later vocabulary outcomes and vice versa. Thus, the first aim of the present study was to employ a stringent analytic approach to examining the longitudinal relations between EF and vocabulary across multiple waves in early childhood (i.e., at ages 2, 3, 4, and 6). Among the studies that have documented a link between children's early and later EF/vocabulary skills, the underlying mechanism(s) that can account for this association have yet to be identified. As such, the second and third aims of this study were to investigate children's private speech and visual attention skills as potential mediators of the hypothesized link between early and later EF/vocabulary. The results indicate that after controlling for maternal education, a unidirectional cross-lagged panel model best fit the data. That is, across all measurement waves, children's vocabulary scores at one timepoint were positively predictive of their EF performance at the following timepoint. Although no evidence of mediation was detected, a significant and novel association emerged between children's early vocabulary scores and their later private speech production. Moreover, this study was able to replicate the well-established link between visual attention and receptive vocabulary among a sample of older children. / Doctor of Philosophy / Across the first few years of life, young children are quickly learning to understand/say new words (vocabulary development) at the same time that they are improving upon their ability to create plans, solve problems, and control their thoughts/actions (executive functioning (EF) development). These early skills act as developmental "building blocks" because they play an essential role in shaping more complex abilities later in childhood and even adulthood. Given their importance, the first goal of this study was to examine whether children's early abilities in one area are related to their later abilities in the other area at multiple ages across the early childhood period. For example, does vocabulary size at age 2 or age 3 relate to children's EF scores at age 3 or age 4? The results show that at each age measured in this study, children's early word knowledge predicted their later EF performance. The second goal of this study was to identify research-based, mediating variables in order to better understand why the expected association between vocabulary and EF exists in childhood. The results show that children with larger vocabularies at age 2 think out loud more when completing a difficult task at age 4, and that children who are skilled at finding requested objects when viewing a cluttered picture at age 4 have larger vocabularies at age 6. Collectively, these findings suggest that children's language and cognitive skills are closely related starting in toddlerhood and continuing into the elementary school years.

A large body of research has examined if early psychopathology symptoms predict continued social difficulties into adolescence; however, few studies have examined whether early social competence predict which children will develop and maintain psychopathology symptoms across childhood and adolescence. Utilizing the National Institute of Child Health and Human Development Study of Early Child Care and Youth Development dataset, this study included a multi-method assessment of social competence in preschoolers and examined whether these social competency measures predicted internalizing and externalizing psychopathology symptom trajectories throughout childhood and adolescence. Additionally, it examined whether social competency measures predicted narrowband measures of attention-deficit/hyperactivity disorder, oppositional defiant disorder, depression, and psychopathy in 3rd - 9th grade. Exploratory factory analyses could not identify underlying factors among the social competency measures; as such, these were examined individually. Two to four classes were identified using Growth Mixture Modeling for parent- and teacher-reported internalizing and externalizing trajectories. Generally, children who had worse preschool social competence were more likely to be in the various moderate and high psychopathology symptom trajectory classes. Some measures of social competency were predictive of parent and teacher-reported ADHD and ODD and self-reported psychopathy, with differing relations for males and females; no measures were predictive of self-reported depression. This study has important implications with regard to early identification of and intervention for at-risk youth. Identifying these youth during the preschool years can result in improved developmental trajectories and significantly decreased societal costs. / M.S. / A large amount of research has looked to see if early atypical mental or behavioral symptoms (known as psychopathology symptoms) predict continued social difficulties into adolescence; however, only a few studies have looked at whether early social skills and performance (social competence) predict which children will develop and continue to show psychopathology symptoms across childhood and adolescence. Using the National Institute of Child Health and Human Development Study of Early Child Care and Youth Development dataset, this study included multiple ways to assess social competence in preschoolers and wanted to see if these measures predicted psychopathology symptoms over time as the child grows up. Additionally, it looked to see whether social competency measures predicted symptoms of attention-deficit/hyperactivity disorder, oppositional defiant disorder, depression, and psychopathy in 3rd - 9th grade. Generally, after running analyses on the data, children who had worse preschool social competence were more likely to have moderate and high psychopathology symptoms over time. Some measures of social competency were shown to be related to parent and teacher-reported ADHD and ODD and self-reported psychopathy, with different connections for males and females; no measures were connected to self-reported depression. This study has important possibilities when looking at early recognition of and treatment for at-risk youth. Finding these youth during the preschool years can lead to better paths over time.

Many common clustering methods cannot be used for clustering multivariate longitudinal data when the covariance of random variables is a function of the time points. For this reason, a copula kernel mixture model (CKMM) is proposed for clustering such data. The CKMM is a finite mixture model that decomposes each mixture component’s joint density function into a copula and marginal distribution functions, where a Gaussian copula is used for its mathematical traceability. This thesis considers three scenarios: first, the CKMM is developed for balanced multivariate longitudinal data with known eigenfunctions; second, the CKMM is used to fit unbalanced data where trajectories are aligned on the time axis, and eigenfunctions are unknown; and lastly, a dynamic CKMM (DCKMM) is applied to unbalanced data where trajectories are misaligned, and eigenfunctions are unknown. Expectation-maximization type algorithms are used for parameter estimation. The performance of CKMM is demonstrated on both simulated and real data. / Thesis / Candidate in Philosophy

Yes / To review the longitudinal trajectories – and the factors influencing their development – of mental health problems in children with neurodevelopmental disabilities. Systematic review methods were employed. Searches of six databases used keywords and MeSH terms related to children with neurodevelopmental disabilities, mental health problems, and longitudinal research. After the removal of duplicates, reviewers independently screened records for inclusion, extracted data (outcomes and influencing factors), and evaluated the risk of bias. Findings were tabulated and synthesized using graphs and a narrative. Searches identified 94,662 unique records, from which 49 publications were included. The median publication year was 2015. Children with attention deficit hyperactivity disorder were the most commonly included population in retrieved studies. In almost 50% of studies, trajectories of mental health problems changed by / Swedish Research Council (2018-05824_VR)