UNT Theses and Dissertations - 113 Matching Results

Search Results

A Predictive Model of Hispanic Participation in Texas Higher Education: Inferences Drawn from Institutional Data in Prevalent Hispanic States

Description: In Texas, Hispanic populations (people of Mexican, Puerto Rican, Cuban, Central or South American or other Spanish culture or origin, regardless of race) have increased from 6.7 million in 2000 to 7.4 million in 2005, or by approximately 10.5%. This growth trend is expected to continue with estimates that Hispanics will represent approximately 37% of the state's population by 2015. The problem this research addressed is that participation in higher education by Texas Hispanics is not keeping pace with the growth in the Texas Hispanic population. If allowed to continue, the state could be in danger of realizing devastating economic and societal consequences. The present study utilized regression analysis to determine how well four institutional characteristics explained the variance in Hispanic enrollment and graduation percentages of students attending public 4-year institutions in states with prevalent Hispanic populations. Findings indicate that while local Hispanic population is a strong, positive predictor of Hispanic enrollments, it has a negative impact on Hispanic graduation rates. The independent variables of average cost of attendance and average financial aid package are the strongest predictors of Hispanic graduation percentages. Implications for the state of Texas include stress on public 4-year institutions in coping with Hispanic population increases, possible enrollment overflows at the community college level, and need for additional allocations to state and institutional financial aid programs.
Date: August 2009
Creator: Haynes, Robert Michael
Partner: UNT Libraries

The relationship of student characteristics, help seeking behavior, academic and environmental variables with student course completion in community college online courses: An application of a conceptual model.

Description: The purpose of this study was to examine differences and relationships in student definition and background characteristics, help seeking behaviors, academic and environmental variables between and among community college students at a single institution who successfully completed and those who did not complete online courses during a single term. An adapted version of Bean and Metzner's conceptual model of nontraditional student attrition provided the theoretical framework for the study. The results of data analysis revealed statistically significant differences between completers and noncompleters on the basis of definition, gender, ethnicity, experience and prior GPA. Statistically significant relationships were found between definition, ethnicity, gender, experience, prior GPA, orientation and completion and noncompletion. No statistically significant interactions were found between definition and experience and help seeking behaviors. No statistically significant differences, relationships or predictor variables were found by degree seeking, preassessment, or technical help seeking. Additional analyses by defining characteristics revealed statistically significant differences between completers and noncompleters on the basis of residency, age and enrollment status. Predictor variables found to be significant were definition, gender, experience, prior GPA and orientation. The odds of completion increased with nontraditional definition, female gender, higher prior GPA, and orientation participation. The odds of completion decreased with experience.
Date: December 2009
Creator: Schumann, Sherry Haskin
Partner: UNT Libraries

An Investigation Into the Factors Leading to the Closure of 40 Private Four-Year Colleges between 1965 and 2005

Description: This study searches for a set of common indicators that contributed to the ultimate closure of 40 colleges and universities between 1965 and 2005. From research on related literature, a set of 31 contributing factors was identified by published experts and observers in higher education. That set of indicators was then used as a list of 31 questions answered by data found in newspaper articles, professional journals, published research work, published institutional records, data taken from the Department of Education, data taken from IPEDS, data published in historical recounts of the colleges of interest, etc. The data was accumulated in the form of yes/no responses to the 31 questions. Although the study involved only 40 colleges and universities this population represents the majority of institutions that pass the restrictions of limitations and delimitations described in the full document. The complete data set was processed using SPSS which produced ANOVA tables and level of statistical significance for each indicator question. The results indicate that out of the 31 original indicator questions there were two groups of statistically significant indicators. The larger group of indicators having statistical significance at the .05 level encompassed the smaller group having statistical significance at the .001 level. There were ten indicators in the first group with significance at the .05 level and seven in the second group with significance at the .001 level. Both groups conform to Bowen's revenue theory of cost associated with the operations of colleges and universities. The first group also has a cultural values component observed by a number of the experts cited in this study. The second is very tightly associated with Bowen's revenue theory of cost and Bates and Santerre's for profit theory of economics. Future research needs to be done to investigate the effect of such use of those ...
Date: December 2009
Creator: Province, Terry Paul
Partner: UNT Libraries