General diagnostic models (GDMs) and Bayesian networks are mathematical frameworks that cover a wide variety of psychometric models. Both extend latent class models, and while GDMs also extend item response theory (IRT) models, Bayesian networks can be parameterized using discretized IRT. The purpose of this study is to examine similarities and differences between GDMs and Bayesian networks using both simulated data and real test data sets. The performances of the two frameworks in data generation and estimation under various possible conditions are investigated. Several indices for accuracy and precision are examined as well as the agreement between the GDM and Bayesian network for simulated data and a real data set in this study. Both have problems with identifiability and high-level proficiency variables. Bayesian network slightly better with small samples and can use correlations among proficiency variables to stabilize estimates for scales with few items. / A Dissertation submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester, 2013. / November 1, 2013. / Bayesian network, Cognitive Diagnosis Model, Evidence-centered Assessment Design, Evidence Model, General Diagnostic Model, Proficiency Model / Includes bibliographical references. / Russell Almond, Professor Directing Dissertation; Diana Rice, University Representative; Betsy Becker, Committee Member; Valerie Shute, Committee Member.

I examined ways pre-service teachers' motivational and cognitive variables (i.e., endogenous/exogenous instrumentality, intrinsic/extrinsic motivational goal orientation, metacognitive self-regulation, and cognitive learning strategies) contributed to explaining their use of extensive knowledge integration strategies within teacher education courses. Future time perspective (FTP) theory was used as a framework to study pre-service teachers' future goals to become teachers and the effect that future goals have on students' initiation and maintenance of their motivation and learning. A total of 197 college students in teacher education courses participated in this research. To obtain answers for specific research questions, data were analyzed through multiple regression analysis and path analysis. The multiple regression result of this research indicated that students' ratings on use of elaboration strategies (β = .27, p < .01) and critical thinking strategies (β = .25, p < .01) made statistically significant contributions to the prediction of students' ratings on their use of extensive knowledge integration strategies. Additionally, results from the hierarchical regression analysis showed that, the final model indicated that prediction of students' extensive knowledge integration significantly improved; R2 = .19, change in R2 = .10, F(3, 188) = 7.49, p < .01. The path analysis demonstrated that three variables had direct effects on students' use of knowledge integration strategies; specifically, their use of critical thinking strategies (β = .25, p < .01), elaboration strategies (β = .29, p < .01), and rehearsal strategies (β = -.17, p < .05). The total effects on students' extensive knowledge integration with statistically significant p values (p< .05) were endogenous instrumentality (β = .10), rehearsal strategies (β = -.17), intrinsic motivation (β = .19), metacognitive self-regulation (β = .22), critical thinking strategies (β = .25), and elaboration strategies (β = .29). Results suggested that understanding of the relationship between current academic tasks and their future goals, intrinsic motivational goal orientation, metacognitive self-regulation, and use of deep learning strategies were necessary for pre-service teachers' effective learning that could promote the development of their teaching expertise. / A Dissertation submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester, 2013. / July 24, 2013. / Expertise, Extensive Knowledge Integration, Perceived Instrumentality, Pre-service Teachers, Self-Regulation, Teacher Education / Includes bibliographical references. / Jeannine Turner, Professor Directing Dissertation; Sherry Southerland, University Representative; Alysia Roehrig, Committee Member; Yanyun Yang, Committee Member.

Deane (2011) proposed a multi-layer cognitive writing model. The 2009 Cognitively Based Assessment of, for, and as learning (CBAL) Writing pilot assessment was designed to support that multi-layer model of writing. One of the features of the assessment was that the keystroke activity of students writing essays were logged to computer files. The preliminary analysis (Almond, et al., 2012) developed an algorithm to classify the pauses in writing based on the keystroke and suggested the distribution for pause events is a mixture of lognormal distribution. This early research was a promising effort to tie the mixture components to the layers in the multi-layer writing model. However, the preliminary analysis with sample size of 68 needed to be repeated with the large data set. Moreover, the data needed to be hierarchically modeled so that the data can fit mixture components when the number of pause events is small per essay. To address these problems, the first part of this thesis aims to fit the large data set (CBAL Writing pilot 2009) in a mixture of lognormal distribution. Then, a distributional analysis was carried out to evaluate the fit of the model to the data. The result confirms the preliminary analysis result. Namely, the two-component mixture model provides an adequate description of the data. The second part of the thesis aims to estimate mixture parameters by using the Bayesian hierarchical model suggested in the preliminary analysis (Almond, et al., 2012). In the Bayesian framework, the hierarchical model is useful when the number of observed pause events is small per essay. Therefore, the proposed model becomes a mixture of k univariate lognormal heteroscedastic components in the Bayesian hierarchical framework. The evaluation of Bayesian model implementation primarily requires the convergence of the Markov chain Monte Carlo (MCMC) sampler. Diagnostic tools were used to assess the convergence of the MCMC sampler. The results show that the MCMC sampler converged for both the two- and the three-component models. / A Thesis submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Master of Science. / Fall Semester, 2013. / June 26, 2013. / Bayesian statistics, Mixture Model, Timing Data, Writing Assessment / Includes bibliographical references. / Russell George Almond, Professor Directing Thesis; Young-Suk Kim, Committee Member; Yanyun Yang, Committee Member.

This multiple-baseline, single subject study examined the impact of a Video Self-Modeling (VSM) intervention on the letter-naming knowledge of four preschool students identified as developmentally delayed (DD). This study adds to the literature by investigating the effectiveness of efficient school-based interventions for improving the pre-reading skills of preschool students with DD. The following research questions were identified for the study: 1. Does Video Self-Modeling improve the letter-naming knowledge of pre-kindergarten children with developmental delays? 2. What is the level of teacher satisfaction of VSM use in a pre-kindergarten classroom? Study participants were preschool students previously identified as having a developmental delay. Following baseline, participants observed a video of themselves correctly identifying letters of the alphabet. The DIBELS Letter-Naming Fluency (LNF) probe was utilized to assess progress throughout the intervention phase. Percentage of Non-Overlapping Data (PND) indicated that the VSM intervention was effective for three of four participants. Teacher satisfaction was high across three of four areas assessed. A discussion of the findings, limitations, and implications of the study for future research and practice are presented. / A Dissertation submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester, 2013. / July 25, 2013. / Developmental Delay, Letter-naming, Preschool, Video Self-Modeling / Includes bibliographical references. / James Sampson, Professor Directing Dissertation; Deborah Ebener, University Representative; Angela Canto, Committee Member; Jeannine Turner, Committee Member.

The number of students with disabilities attending postsecondary institutions is increasing as greater access and support services are provided on campus. Increasing demand requires the need to offer support to such students not only in terms of accommodations, but also with adjusting to their disability as well as increasing their quality of life (QOL). Positive psychology is gaining credibility and being utilized under the premise that a person's strengths are of equal, if not of more importance than focusing on a person's weaknesses or illness. Research demonstrates the many benefits of positive psychology, including being optimistic, not only for the general population, but also for persons with chronic illnesses and disabilities (CID). Studies are sparing in terms of incorporating optimism in relation to adaptation to disability and quality of life, particularly with the college student population. This research used a convenience sample of 229 college students with disabilities attending postsecondary institutions in the southeast in order to address the following two research questions: (1) What is the relationship between optimism, adaptation to disability, and quality of life among college students with disabilities? and (2) Are there any differences in optimism, adaptation to disability, and quality of life between college students with low perceived severity of disability and college students with high perceived severity of disability? The following null hypotheses are associated with the first research question: (a) there is not a significant correlation between adaptation to disability and quality of life and (b) there is not a significant moderating effect of optimism on the relationship between adaptation to disability and quality of life. The first null hypothesis was rejected as this research demonstrated a significant relationship between adaptation to disability and quality of life. The second null hypothesis was not rejected as optimism did not significantly moderate the relationship between adaptation to disability and quality of life. However, optimism, along with adaptation to disability, did predict quality of life. A MANOVA was used to test the following null hypothesis associated with the second research question: there are no significant differences in optimism, adaptation to disability, and quality of life between college students with low perceived severity of disability and high perceived severity of disability. The null hypothesis was not rejected as this research indicated significant differences in optimism, adaptation to disability, and quality of life between college students with low and high perceived severity of disability. This research supports the tenets behind Bishop's Disability Centrality model and demonstrates the relationship and similarities between adaptation to disability and quality of life. This study also demonstrates the importance of continued research regarding functional limitations and perceived severity of disability. Additionally, this research provides evidence that further study of individual differences for college students with disabilities is necessary in order to better understand how to foster successful adaptation to disability and increase quality of life. / A Dissertation submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester, 2013. / April 4, 2013. / Includes bibliographical references. / Deborah Ebener, Professor Directing Dissertation; Mary Francis Hanline, University Representative; Angel Canto, Committee Member; Debra Osbron, Committee Member.

This study examined the relationships between performance and appraisals of efficacy beliefs and presentational concerns within performing dyads. The expansion of a performing unit from one to two athletes is reflected in a structure of cognitions expanding to appraisals about oneself as well one's partner. Self-efficacy, other-efficacy, and collective efficacy were examined. While evidence for self-efficacy and collective efficacy have been supported in sport psychology research, other-efficacy (Lent & Lopez, 2002), the belief in a partner's ability, has recently been examined. Self-presentation, other-presentation, and collective presentation were also assessed in a similar relational structure. Sixty-six cheerleading (male-female) dyads completed questionnaires appraising the dyad's ability and individuals within the dyad to perform. Performance by each individual and the dyad unit were assessed by four judges. Actor-Partner Interdependence Models (APIM; Kenny et al., 2006) were used to model the interdependence and mutual influence among the partners. Interaction effects were significant for the actor's self-efficacy and actor's collective efficacy. That is, dependent on position (male base or female flyer) one's self-efficacy and collective efficacy were significantly related to one's own performance. This evidence along with correlational findings were in support of the importance of relational cognitions between partners and the importance of recognizing the relationship between two athletes. / A Thesis submitted to the Department of Educational Policy and Learning Systems in partial fulfillment of the requirements for the degree of Master of Science. / Fall Semester, 2013. / November 5, 2013. / Collective Efficacy, Dyad, Other-Efficacy, Performance, Self-Efficacy, Self-Presentation / Includes bibliographical references. / Robert C. Eklund, Professor Directing Thesis; Gershon Tenenbaum, Committee Member; Insu Paek, Committee Member.

Over the years, there have been conflicting findings concerning the question of whether athletes are at increased risk for disordered eating. While there have been many studies examining the relationship between sports demands and eating disorders, the specific risk factors underlying the development of these disorders have been debated. The purpose of this study was to expand on the existing literature and Petrie, Anderson, and Neumann's model (2011) in particular, by examining the notion of perfectionistic self-presentation in relation to various other risk factors leading to the development of disordered eating attitudes and behaviors within the female running population. This study was conducted under the assumption that competitive females participating in lean-type sports (e.g. runners) are under the influence of sport pressures, along with societal pressures (Petrie, Anderson, & Neumann, 2011). It was predicted that these pressures would be subsequently related to greater internalization of the thin ideal. Moreover, it was expected that the greater the internalization, the lower one's body satisfaction, and the higher one's self-presentation scores. It was hypothesized that the combination of all these risk factors would relate to a higher rate of disordered eating attitudes and behaviors. A second hypothesis was that, with age, female runners would exhibit less concern about body image and self-presentation, thus producing a lowered incidence of disordered eating attitudes and behaviors. Structural equation modeling analyses were conducted to test the proposed model. Overall, the constructs within the respecified structural model were supported, although some of the initially proposed hypothesized pathways were not. The greater one's internalization of the thin ideal was, the higher one's self-presentational scores, and the lower one's body satisfaction, which then related to more disordered eating attitudes and behaviors. These findings provide direction for future research and interventions that could reduce female runners' risk of developing disordered eating attitudes and behaviors by reducing the importance placed on messages promoting the thin ideal, and instead focusing on communicating the importance of physical health over physical appearance and weight-loss. / A Thesis submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Master of Science. / Spring Semester, 2013. / April 1, 2013. / Includes bibliographical references. / Robert C. Eklund, Professor Directing Thesis; Gershon Tenenbaum, Committee Member; Beth Phillips, Committee Member.

ABSTRACT The purpose of this study was to examine the effects of argumentation scaffolding on problem-solving outcomes and learner motivation in a Problem-Based Learning course. Undergraduate students enrolled in a mandatory introductory logic course were randomly assigned to either the experimental (argumentation scaffolding) or control (verbal guidance) condition. Students in each condition were randomly assigned to small groups of five students within the class. Other than the argumentation scaffolding treatment, the remaining activities were the same for the experimental and control conditions. Problem-solving outcomes were measured using scores from a pretest and posttest that consisted of five ill-structured questions. The pretest was administered prior to the study, and the posttest was administered after the treatment on the last day of the study. Prior to administering the posttest, participants completed a self-reported survey to measure their level of motivation at the end of the study. Results revealed there was a statistically significant difference in mean problem solving scores – from pretest to posttest – between the experimental and control conditions F(1, 95) = 206.37, p < .05. Participants in the argumentation scaffolding condition scored significantly higher (M = 28.36) than students in the verbal guidance condition (M = 21.13) on the posttest. Results also revealed that there was not a significant difference in learner motivation between the experimental and control conditions F(1,95) = 0.01, p > .05. Participants in the argumentation scaffolding condition (M = 3.37) did not score significantly higher on motivation than students in the verbal guidance condition (M = 3.39). While the findings suggest argumentation scaffolding is an effective instructional strategy for the development of problem-solving skills, the results suggest no effect on learner motivation. Researchers and educators can apply this information and use it to structure research conditions so that argumentation can have a more positive effect on participants. / A Dissertation submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester 2015. / June 29, 2015. / Argumentation Scaffolding, Learner Motivation, Problem-Based Learning / Includes bibliographical references. / Valerie Shute, Professor Directing Dissertation; Jay Rayburn, University Representative; Russell Almond, Committee Member; Allan Jeong, Committee Member.

An influx of veterans have enrolled in college since the passage of the Veterans Educational Assistance Act in 2008. Since then, there have been a number of recommendations about how to provide support for transitioning student veterans; however, there is limited research about the experiences of student veterans with these support services and their adjustment to college. Therefore, this study's author investigated the relationship between institutional support, social support, and college adjustment via a cross-sectional survey administered to student veterans nationwide. The final sample size was 130 student veterans. Student veterans completed the following instruments: a demographic questionnaire, the Perceived Social Support Scale (Procidano & Heller, 1983), a measure of perceived institutional support, and the Student's Adjustment to College Questionnaire (Baker & Siryk, 1999). Regression analyses and ANOVAS were conducted to determine statistical significance and effect sizes amongst the variables of age, gender, combat status, length of enrollment in college, perceived institutional support, perceived social support, and college adjustment. Results revealed that perceived institutional support significantly predicted college adjustment (R¬2 = .171, p = .00) but that perceived social support did not. One-way ANOVAs conducted for gender and combat status were not statistically significant in this study. These results suggest that institutional support (both general and veteran-specific) is important to student veterans' college adjustment. / A Dissertation submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester 2015. / June 23, 2015. / Adjustment, College, Students, Veterans / Includes bibliographical references. / Deborah J. Ebener, Professor Directing Dissertation; Bruce A. Thyer, University Representative; Angel Canto, Committee Member; Debra S. Osborn, Committee Member.

The process of evaluating teachers has existed for many years and in a myriad of designs. Teacher evaluations are often a rating of instruction in a single observation. Teachers and other stakeholders strive to ensure that students are receiving instruction that promotes learning growth, so assessing those who provide instruction to students is essential. In the past few years, teacher evaluations are not only assessing classroom instruction, but are also measuring teachers' effects on student learning growth. In 2011, the Florida legislature passed the Student Success Act, which revised the requirements for teachers' annual evaluations. One component of the evaluation, the Florida value-added model (FL VAM), is a statistical model that computes a value-added score for teachers. This score represents the value added by a teacher to student learning. I investigated the new evaluation system from teachers' perspectives. Teachers from school districts across Florida completed a newly developed survey designed to gauge teachers' knowledge and perceptions about the new evaluation system. Evidence from my findings indicated that Florida teachers are willing to admit their lack of knowledge about FL VAM, and teachers, in general, feel negatively toward this assessment. As well, they lack confidence in the model to produce accurate value-added scores, and they are concerned that their scores may adversely impact their careers as teachers. Due to the rise in value-added assessment of teachers, research should continue to monitor the impact of this evaluation approach on teachers. / A Dissertation submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Summer Semester, 2014. / June 24, 2014. / Teacher Evaluation, Value-Added Assessment / Includes bibliographical references. / Betsy Jane Becker, Professor Directing Dissertation; Thomas Blomberg, University Representative; Russell Almond, Committee Member; Yanyun Yang, Committee Member.