Ph.D., Antioch University, 2023, Leadership and Change

There are systematic barriers to educational equity in the U.S. higher education system, and the system overwhelmingly fails Latinx undergraduates more often than other students. It is crucial that evidence-based methods be used to reduce the existing postsecondary student success inequities. Scholars have linked specific educational practices to positive learning effects. A growing body of evidence has suggested these educational practices, coined high-impact practices (HIPs), provide amplified benefits to historically underserved students (HUS) and may be an effective tool for advancing equity and closing achievement gaps. The extant literature has neither adequately explained the reason(s) that HIPs provide an academic boost to HUS nor described their lived experience. Such qualitative research is important for understanding how HIPs contribute to HUS' learning and engagement, better support student success, and address inequities. Through narrative inquiry and inductive/emergent analysis, this study explored the lived experience of Latinx in HIPs at a 4-year public university. Deductive/a priori analysis drew from two theoretical frameworks: validation theory and cultural capital. This study investigated several guiding questions: In which curricular experiences do Latinx undergraduates experience the deepest learning and engagement? To what elements or aspects of the experiences do Latinx undergraduates attribute the learning and engagement? What are the key validating experiences or experiences that recognize/reward cultural capital? Findings revealed five major course elements as associated with deep learning and engagement: professor behaviors or traits, real-world and relevant content, preparation for future or career, relationships with peers, and diverse perspectives. A key implication for practice is that faculty are central to student success and through the application of teaching and curricular elements, every academic course can ensure deep learning (open full item for complete abstract)

Committee: Laurien Alexandre PhD (Committee Chair); Jon Wergin PhD (Committee Member); Marisol Clark-Ibáñez PhD (Committee Member) Subjects: Education; Educational Leadership; Higher Education; Higher Education Administration; Hispanic Americans; Teaching

Master of Science, The Ohio State University, 2022, Psychology

News coverage of scandalous behavior in the present-day media is at an all-time high and being caught up in a widely covered scandal can have serious consequences for those involved. In most cases, being exposed as a perpetrator of scandal can do untold damage to one's career and reputation. But for some, being caught in a scandal has few consequences, leaving the perpetrator virtually unaffected by news of their behavior going public. While previous research has pointed to several predictors than can produce different reactions to scandal, like the gender of the perpetrator and the type of scandal, they do little to explain why such variables produce different reactions. The current research seeks to bridge this gap and explores people's pre-existing expectations about whether a scandal will occur as a driving factor for why some reactions to scandal are more severe than others. Two theories predicting separate and opposite mechanisms by which expectations would affect the perceived scandalousness of a scandalous behavior were considered. Specifically, it was predicted that the more a transgresssion was expected to occur, (1) the more unsurprised people would be by news of the scandal rendering judgments of it less scandalous (decision affect theory) and (2) the more they expected the transgression to occur, the more they would believe that the scandal actually occurred, rendering judgments of it more scandalous (self-validation theory). Results of two studies provide evidence that both of these hypothesized pathways play a role in determining how scandalous a perpetrator's behavior is perceived to be.

Committee: Richard Petty (Advisor); Russ Fazio (Committee Member); Duane Wegener (Committee Member) Subjects: Psychology; Social Psychology

Master of Science, Miami University, 2025, Mechanical and Manufacturing Engineering

Reduced order modeling (ROM) methods, such as those based upon Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD), offer data-based turbulence modeling with potential applications for flow control. While these models are often cheaper than numerical approaches, their results require validation with source data. Within the literature, the metrics and standards used to validate these models are often inconsistent. Chabot (2014) produced a data-driven framework for validating these ROMs that used surrogate Markov models (SMMs) to compare how the system dynamics evolved rather than how any single metric evolved. These SMMs were constructed by clustering the flow data into different states of suitably similar flow fields, and the Markov model then mapped how likely each state was to transition into another. While this method was successful, there persisted an amount of uncertainty in how the outlier states within this clustering scheme were determined. Additionally, the study only examined the application of this procedure to POD-Galerkin ROMs. This study aims to tie the outlier state determination directly to the models' parent data. The study will also apply this procedure to ROMs generated from DMD to investigate how this framework's effectiveness carries over to different classes of ROMs.

Committee: Edgar Caraballo (Advisor); Andrew Sommers (Committee Member); Mehdi Zanjani (Committee Member) Subjects: Aerospace Engineering; Fluid Dynamics; Mathematics; Mechanical Engineering; Statistics

Doctor of Philosophy, The Ohio State University, 2023, Communication

Perceived character depth is a concept relevant for understanding and predicting audience responses to narrative media, yet it has been largely unexplored in the field of media psychology. Through a careful review of diverse literatures, the current work offers a formal conceptualization of character depth as the extent to which a character's textual exposition evokes a detailed and multi-faceted mental conception of a character's psyche, behavior, and experience. After devising a series of items to measure character depth, this work then presents a series of experimental studies designed to test various aspects of validity of the proposed measurement scale and to test a causal path model of the relationship between character depth and processes specified by affective disposition theory (see Zillmann, 2000). Study 1 uses a known-groups approach and confirmatory factor analysis to test the predictive validity and measurement model of a 20-item proposed perceived character depth scale. Selective item retention results in a 6-item scale with excellent model fit. Studies 2 and 3 lend additional support to the validity of this 6-item scale's measurement model through tests of the scale in two different narrative contexts, both of which result in excellent model fit. Across Studies 1-3, evidence emerges of the convergent and discriminant validity of the scale in relation to other character perception variables. Study 4 applies this new measure in a 2 X 3 between-subjects experimental design that manipulates both character depth and character moral behavior independently. Results show that character depth impacts disposition formation and anticipatory responses above and beyond audience reactions to moral behavior. This finding has important theoretical implications for affective disposition theory (Zillmann, 2000), indicating that perceived character depth may serve as an additional predictor of disposition formation not specified in the original theory. Future research d (open full item for complete abstract)

Committee: Matthew Grizzard (Advisor); Emily Moyer-Guse (Advisor); Nicholas Matthews (Committee Member) Subjects: Communication; Mass Communications; Mass Media; Psychology

Doctor of Philosophy, The Ohio State University, 2023, Educational Studies

The publication of the Family Engagement Core Competencies by the National Association for Family, School, and Community Engagement in 2022 provided a body of knowledge of the work of educators engaging with families, a domain not well-understood previously. Situated in the pandemic recovery phase in U.S. education, this dissertation study aimed to explore the nature of efficacy beliefs of educators in family engagement, a latent construct grounded in social cognitive theory. To achieve this aim, a survey measure was developed, and an interpretation and use argument (IUA) anchored the process of the validation of the interpretation and use of its scores. The study was conducted in five phases. In the first phase, the Family Engagement Efficacy Beliefs of Educators (FEEB-E) survey was drafted based on a review of literature, existing instruments, and subject-matter expert and user feedback. In the second phase, the FEEB-E was piloted with a national sample of educators, and exploratory factor, parallel, and reliability analyses were conducted. In the third phase, the FEEB-E was revised based on findings from the pilot study. In the fourth phase, the FEEB-E was administered to a sample of teachers in two Ohio urban school districts, and confirmatory factor, correlation, and reliability analyses were conducted. In the fifth phase, the IUA was constructed, and the appropriateness of the interpretation and use of the scores from the FEEB-E was determined. The study found that survey items were valid for interpretation as representative of their target domain of self-efficacy beliefs for family engagement. There were no extraneous sources of variability in wording of items and directions, order of items, or the scoring scale. The survey items measured the intended population by reflecting a range of efficacy beliefs around multiple dimensions of family engagement and did so reliably (α = .917). Further, the FEEB-E was found to be valid for use as a research instrum (open full item for complete abstract)

Committee: Karen Stansberry Beard (Advisor); Minjung Kim (Committee Member); Roger Goddard (Committee Member); Barbara Boone (Committee Member) Subjects: Educational Psychology; Families and Family Life; Teaching

Doctor of Philosophy, The Ohio State University, 2022, Mechanical Engineering

Due to the complexity of the operational design domain of Automated Driving Systems, the industry is trending towards the use of simulation-based methods for their verification and validation (V\&V), which rely on the use of models of the vehicles, sensors, vehicle environments, etc. Depending on the testing requirements, computational capabilities, modeling effort, and other such factors, these models can vary in fidelity. However, this variation in fidelity has an effect on the excitation of the control systems under test, and can therefore affect the results of the tests themselves. Moreover, since every model is an approximation of the actual physical system it represents, there is uncertainty associated with its output. Therefore, we need to be able to compare uncertain system models in order to understand the effect of model fidelity variation on test accuracy. The existing metrics such as Hankel Singular Values compare asymptotic behavior of the models, whereas simulation studies are over finite time. Although some of these metrics may be applied over finite time, they rely on hyper-parameters like weights on time or frequency. In this study, we propose an approach for computing a (pseudo)metric based on the literature for comparing the predictive performance of two models, called finite-time Kullback-Leibler (KL) rate. For any two general state space models with a general additive uncertainty, we first discuss an approach for propagating a general additive uncertainty (represented as a Gaussian Mixture Model (GMM) ) through a linear time-invariant system. We then apply this propagation approach to linear representations of nonlinear systems obtained through Dynamic Mode Decomposition (DMD). We illustrate this combined approach for the comparison of two lateral vehicle dynamics models over an obstacle avoidance maneuver to measure the effect of fidelity on the predictive performance of each model. We also apply this to a \vv problem, wherein we compare (open full item for complete abstract)

Committee: Punit Tulpule (Advisor); Shawn Midlam-Mohler (Advisor); Marcello Canova (Committee Member) Subjects: Mechanical Engineering

Doctor of Philosophy, University of Toledo, 2019, Educational Research and Measurements

Validity is a foundational aspect of learning assessments; the process of validation is vital. Validation is context-specific and needs to be examined for test-scores from each learning assessment that will be used in decision-making--especially high-stakes decision-making which may affect progression in a pharmacy education program, graduation, and/or licensing. Kane's Framework for Validation has four inferences (scoring, generalization, extrapolation, and implications), of which generalization is an important first step. Generalizability Theory (G-Theory) can do this though it has been used very rarely in pharmacy education. This dissertation is focused on demonstrating use of G-Theory for reliability evidence (generalization inference) with three common applications of learning assessments in pharmacy education--a performance-based assessment, multiple examinations, and multiple quizzes in preparation for an examination. Reliability was reported for each application. As well, variance was found, and optimization of test parameters was explored for each. Rigor can and should be a focus for every high-stakes assessment, including (and especially with) performance-based assessments. In the unique educational context and resources of each college/school of pharmacy, reliability and the impact of alterations in test parameters should be examined. Following from this, the impact on reliability in varying the number of exam questions, and varying the weight of quizzes was also demonstrated. At each institution and for various learning assessments, G-Theory provides validation evidence that can show rigor for use of learning assessment scores. G-Theory also allows exploration for customization of test parameters. It would be prudent for G-Theory to evolve to much wider use in pharmacy education.

Committee: Gregory Stone PhD (Advisor); Gregory Stone PhD (Committee Chair); Noela Haughton PhD (Committee Member); Kenneth Cor PhD (Committee Member); Spencer Harpe PhD (Committee Member) Subjects: Education; Higher Education; Pharmacy Sciences

Doctor of Philosophy, The Ohio State University, 2019, Psychology

This dissertation introduces a new model evaluation tool - the RORME index - that can be used to select an item response model among competing alternatives. This criterion assesses the out-of-sample predictive performance of candidate models using a k-fold cross-validation procedure. The validity of RORME is evaluated with several simulation studies, which conclude that the proposed index performs well under a multitude of conditions. RORME often follows a similar selection pattern as the AIC; however, unlike the AIC or BIC, it does not rely on underlying model assumptions or likelihood-based estimation. The RORME index is also flexible in how a researcher would prefer a model to be evaluated - the specic performance metric used to calculate RORME can be changed and investigations into local misfit (e.g. item- or person-specic mist) can be conducted. The RORME index is of particular interest when utilized to compare item response tree, or IRTree, models with non-IRTree alternative models. IRTree models are a particular application of the item response theory framework that allow interesting, novel research questions to be addressed. One of these novel explorations is assessing the most appropriate process that a respondent undertakes when responding to an item. This type of research question naturally requires comparing several candidate models and eventually endorsing the most appropriate alternative. Traditional model selection criteria, such as AIC or BIC, may not be appropriate for these types of model comparisons, since IRTree models require the underlying data to be transformed. In this project, the RORME index is developed and applied to directly compare IRTree and non-IRTree models. Simulation results suggest that the new metric can successfully determine the appropriate model when applied in this context, even when criteria like the AIC or BIC are invalid.

Committee: Paul De Boeck (Advisor); Andrew Hayes (Committee Member); Jolynn Pek (Committee Member) Subjects: Quantitative Psychology

Doctor of Philosophy, Miami University, 2015, Educational Leadership

One in six students at American four-year universities is a first-generation student. First-generation students, defined in this study as those whose parents did not attend a four-year university, encounter unique tensions as they navigate the disparate worlds of home and school. This constructivist narrative study of first-generation student experiences explores the success stories of eight diverse first-generation seniors at Miami University – a selective, four-year, residential institution in Oxford, Ohio. The study resists a deficient perspective on first-generation students, instead telling the stories of participants who have successfully navigated tension between the worlds of home and school, and accomplished their goal of graduating from college. Though nearly all research in this area focuses on first-generation students' transitions to college, this study found that tension between home and school persisted throughout participants' college experiences. This persistent tension is identified and described in three non-linear realms: tension of transition, tension of identity, and tension of success. The study describes how participants experienced a temporal dimension of tension between their past and future worlds within these realms, and identifies participants' commitment to a college future as a powerful source of motivation for persistence amidst home/school tension. Implications for practice underscore the importance of validation for first-generation students from campus peers, faculty, and staff; the value of safe space for building authentic connections with others in their college world; and the responsibility of four-year institutions to transform institutional culture in order to provide such validation and support for first-generation students as they negotiate tension between the worlds of home and school. Particular implications for practice include targeted outreach for first-generation students in the areas of pre-semes (open full item for complete abstract)

Committee: Elisa Abes (Committee Chair); Kathleen Goodman (Committee Member); David Perez II (Committee Member); John Jeep (Committee Member) Subjects: Education; Educational Leadership; Families and Family Life; Higher Education; Higher Education Administration

PHD, Kent State University, 2014, College of Education, Health and Human Services / School of Foundations, Leadership and Administration

This study sought to determine whether the items that were in use for the TRAILS K-12 information literacy classroom assessments at the 3rd, 6th, 9th, and 12th grade levels could be used to develop a valid and reliable test for each grade level. To determine whether the assessments could be reengineered into valid tests of information literacy, the entire item banks for each grade level were administered to students recruited across the United States. The data gathered were then used to answer three research questions. First, the study sought to determine whether the current item banks could be used to create an efficient and reliable test at each grade level. The Rasch model of Item Response Theory was used to develop each test based on overall scale reliability, item fit to scale, distractor function, lack of bias based on differential item functioning analysis, item difficulty spread, and content coverage. It was determined that reliable tests could be created at each grade level (3rd N = 20; 6th N = 25; 9th N = 30; 12th N = 35) with generally strong psychometric properties at the item level across all four tests. The study then sought to gather evidence of construct validity through two methods. First, content experts at each grade level were presented each item in the draft test and were asked to rate the degree to which it measured its associated TRAILS objective. The results of this study identified items on each test for further examination but found general endorsement at the item and scale level for objective measurement. Additionally, the amount of reading and item difficulty were examined through a correlation study. No relationship was found on any of the tests, reducing the likelihood that reading affects item difficulty. Finally, the study sought to determine the most generally agreed upon proficiency score for each test by using a modified bookmarking standard-setting procedure. This process utilized expertise at each grade level th (open full item for complete abstract)

Committee: Schenker Jason PhD (Committee Co-Chair); Fitzgerald Shawn PhD (Committee Co-Chair); Harper Meghan PhD (Committee Member) Subjects: Educational Tests and Measurements; Library Science

Doctor of Philosophy, The Ohio State University, 2004, Statistics

Statistical models are used to describe collections of data and the processes that generated them. Since models are used to draw conclusions about research questions, it is important to determine if a proposed model is an adequate approximation of the true process. When there are competing models, we ask which of the models provides the best approximation. This research addresses the issue of how to best assess model fit using the approach of cross-validation. We suggest that cross-validation should mimic as closely as possible the intended use of the model. This idea motivates a new way to implement cross-validation, covariate included cross-validation. We show that covariate included cross-validation provides better estimates of model error than traditional cross-validation in the appropriate model setting. We also point out a major di®erence between classical cross-validation and a Bayesian version of crossvalidation; for Bayesian models, the apparent error of the model may be larger than a cross-validatory estimate of error. We give an example where this occurs and provide results indicating when the apparent error will not be larger than the leave-one-out cross-validation estimate of error. The latter part of the dissertation looks at the assessment of item response models used for educational assessment. We implement Bayesian cross-validation for Bayes item response models and show how out-of-sample measures of model fit are particularly relevant for nonparametric item response models.

Committee: Steven MacEachern (Advisor) Subjects: Statistics

Master of Education, The Ohio State University, 2002, Physical Activity and Educational Services

The purpose of this study is to complete a construct validation of a physical activity intervention on the subjects' knowledge and use of 8 self-regulation strategies: self-monitoring, goal setting, social support, environmental aid, self-reinforcement, time-management, self-efficacy, and tailoring. Using a one group pre-test post-test design, a physical education instructor delivered a 5-week intervention to all sixth (n=20), seventh (n=28) and eighth (n=24) graders at a private school in Pittsburgh, Pennsylvania. A self-report self-regulation instrument was administered at baseline and immediately post-intervention. Students also completed a knowledge test at the end of the intervention. Seven of the eight subscales as well as global self regulation for physical activity were found to be significantly changed from pre-test to post-test (p<.006). The mean score on the knowledge test was 17.5 out of 22 points.

Committee: Rick Petosa (Advisor) Subjects: Education, Physical