dc.contributor.author	Fitzgerald, Barrie Dwain
dc.coverage.spatial	United States	en_US
dc.date.accessioned	2024-04-29T18:40:56Z
dc.date.available	2024-04-29T18:40:56Z
dc.date.issued	2024-04-16
dc.identifier.other	62c51223-e34a-427e-a375-63d32a8d05ff	en_US
dc.identifier.uri	https://hdl.handle.net/10428/7166
dc.description.abstract	Regional comprehensive universities offer accessible and diverse undergraduate educational programs, while grappling with funding cuts and affordability. The study’s first research question underscores the enduring importance of factors such as student characteristics, pre-college characteristics, and financial situations. The findings highlight high school GPA's (HS GPA) pivotal role in academic performance. Higher HS GPAs correlate with successful academic performance resulting in higher retaining likelihoods; conversely, lower HS GPAs are associated with academic struggles and increased departure likelihoods. HS curriculum variables also impact academic performance, notably in extreme gradient boosting (XGBoost) models. The second research question centers on the algorithms’ predictive power. XGBoost and random forest models consistently outperform the other models in predicting GPAs. Prioritizing area under the curve values for retention, both XGBoost and random forest models are statistically comparable for developing predictive algorithms, despite facing challenges with low specificity rates. Only slight enhancements in predictions were detected in the upsample ensemble learning models. Implications for practice underscore the importance of targeted interventions through leveraging data science techniques and machine learning algorithms to identify and allocate support resources for at-risk students. This research significantly contributes to the discussion on student success in higher education by providing practical insights and guiding evidence-based practices. As education evolves, integrating data science into strategic planning becomes pivotal for shaping the trajectory of student success initiatives.	en_US
dc.description.tableofcontents	Chapter I: INTRODUCTION 1 -- Statement of Problem 6 -- Purpose of the Study 7 -- Research Questions 8 -- Research Methodology 10 -- Significant of the Study 16 -- Theoretical Framework of the Study 17 -- Limitations of the Study 18 -- Definitions of Terms 20 -- Organization of the Study 24 -- Chapter II: LITERATURE REVIEW 25 -- Regional Comprehensive Universities 25 -- Growing concerns for regional comprehensive universities. 27 -- Decline in traditional-age student population. 27 -- Changing demographics 29 -- Public’s perception of postsecondary education 29 -- Development of Attrition Theories 30 -- Spady’s undergraduate dropout process model. 31 -- Tinto’s institutional departure model. 31 -- Bean’s student attrition model. 32 -- Astin’s student involvement theory. 33 -- Characteristics Impacting Academic Performance 33 -- Student characteristics. 34 -- Gender. 34 -- Race and ethnicity. 38 -- Family educational background. 43 -- Locale 45 -- Education in rural areas. 48 -- Precollege characteristics. 52 -- High school curriculum quality. 52 -- High school GPA 56 -- Grade inflation concerns 59 -- Admission test scores 62 -- Financial situations. 65 -- Family financial income or situations 65 -- Financial aid 69 -- HOPE scholarship 76 -- Major declaration and grouping. 78 -- Institutional financial expenditures. 82 -- Data Science 85 -- Data mining and its application in higher education 87 -- Cross-validation methods 89 -- Model evaluation methods 90 -- Summary 93 -- Chapter III: METHODOLOGY 95 -- Research Design 97 -- Independent variables 97 -- Dependent variables 100 -- Participants 100 -- Instrumentation 101 -- USG institutional data 102 -- High schools’ CCRPI data 102 -- High schools’ EOC data 103 -- Institutions’ financial expenditures 105 -- Data collection 106 -- Data Analysis 107 -- Data preparation. 107 -- Descriptive statistics 108 -- Inferential statistics 108 -- Linear regression 109 -- Logistic regression 112 -- Support vector machine for regression and classification 114 -- Random forest 16 -- Extreme gradient boosting 117 -- Ensemble learning 18 -- Data science approach 118 -- Statistical Considerations and Assumptions 120 -- Considerations 121 -- Assumptions 122 -- Data imbalance 126 -- Data leakage 127 -- Data preparation 127 -- Summary 129 -- Chapter IV: RESULTS 131 -- Population Characteristics 134 -- GA public high school represented. 134 -- Institutional expenditures per FTE 135 -- Demographic characteristics 136 -- Descriptive statistics 139 -- Data Splitting and Imbalance 141 -- Preliminary Considerations and Assumptions 142 -- Considerations 142 -- Missing data. 142 -- Outliers. 148 -- Assumptions 152 -- Observation independence. 152 -- Linearity. 153 -- Normality. 160 -- Homogeneity of variance. 165 -- First Research Question 167 -- First-fall GPA 168 -- Linear regression 169 -- Assumptions for linear regression 172 -- Support vector machine with linear kernel. 177 -- Support vector machine with polynomial kernel. 179 -- Support vector machine with radial basis function kernel. 181 -- Random forest. 182 -- Extreme gradient boosting. 184 -- Variable importance comparison for first-fall GPA. 186 -- First-year GPA. 189 -- Linear regression. 189 -- Assumptions for linear regression 192 -- Support vector machine with linear kernel. 197 -- Support vector machine with polynomial kernel. 199 -- Support vector machine with radial basis function kernel. 201 -- Random forest. 203 -- Extreme gradient boosting. 205 -- Variable importance comparison for first-year GPA. 207 -- One-year retention. 209 -- Logistic regression 210 -- Assumptions for logistic regression. 214 -- Sampling modifications. 215 -- Testing data set. 222 -- Support vector machine with linear kernel. 226 -- Support vector machine with polynomial kernel. 227 -- Support vector machine with radial basis function kernel. 231 -- Random forest. 233 -- Extreme gradient boosting. 236 -- Variable importance comparison for one-year retention status. 239 -- Second Research Question 243 -- First-fall GPA 244 -- First-year GPA 249 -- One-year retention status 254 -- Original model 255 -- Downsample model. 260 -- Upsample model. 266 -- Summary 271 -- Chapter V: SUMMARY, DISCUSSION, AND CONCLUSIONS 280 -- Overview of the Study 281 -- Related Literature 281 -- Data science. 282 -- Predictive factors. 283 -- Methodology 283 -- Participants 284 -- Variables Studied 284 -- Student characteristics. 285 -- Pre-college characteristics. 285 -- Financial situations. 286 -- Program of study. 287 -- Institutional expenditures. 287 -- Procedures 287 -- Summary of Findings 288 -- First research question. 289 -- First-fall GPA. 289 -- First-year GPA. 291 -- One-year retention status. 292 -- Second research question. 294 -- First-fall GPA. 294 -- First-year GPA. 296 -- One-year retention status. 299 -- Discussions of Findings 307 -- First research question. 307 -- Second research question. 309 -- Limitations of the Study 11 -- Implications for Future Research 315 -- Recommendations for Practice 316 -- Conclusions 318 -- REFERENCES 320.	en_US
dc.format.extent	1 electronic record. PDF/A document, 527 pages, 12747027 bytes	en_US
dc.format.mimetype	application/pdf	en_US
dc.language.iso	en_US	en_US
dc.rights	This dissertation is protected by the Copyright Laws of the United States (Public Law 94-553, revised in 1976). Consistent with fair use as defined in the Copyright Laws, brief quotations from this material are allowed with proper acknowledgement. Use of the materials for financial gain with the author's expressed written permissions is not allowed.	en_US
dc.subject	Dissertations, Academic--United States	en_US
dc.subject	Education	en_US
dc.subject	Educational leadership	en_US
dc.subject	Education, Higher--Administration	en_US
dc.subject	Academic achievement	en_US
dc.subject	Statistics	en_US
dc.subject	College dropouts--Prevention	en_US
dc.title	Assessing Secondary Curriculums' Impact on Postsecondary First Year Academic Performance Using Data Science Techniques	en_US
dc.type	Dissertation	en_US
dc.contributor.department	Department of Leadership, Technology, and Workforce Development of the Dewar College of Education and Human Services	en_US
dc.description.advisor	Nobles, Kathy
dc.description.committee	Pate, James
dc.description.committee	Brockmeier, Lantry
dc.description.degree	Ed.D.	en_US
dc.description.major	Education in Curriculum and Instruction	en_US