Date of Award

8-1-1981

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Counseling Psychology & Community Services

Abstract

The purpose of this study was to investigate the characteristics which differentiated freshmen who persisted beyond the fail semester of 1980 at the University of North Dakota from those who did not.

The research population consisted of all "undergraduate regular"— "new freshman" enrolled during the fall semesters of 1979 (M » 5560) and 1980 (N » 1551). A random sample of 400 students drawn from \he 1980 research population were sent a mail questionnaire, A total of 336 (84X) usable responses were received.

Variables Included in this study were of two types: (a) pre-college characteristics from students' ACT Assessments and the University's Student Record System, and (b) measures of early college experiences from the mail questionnaire. Three discriminant models were developed in an attempt to classify students—eased on first semester GPA and second term enrollment status—into four persistence groups: (a) successful persisters, (b) unsuccessful persisters, (c) successful non- persisters, and (d) unsuccessful non-persisters. Two models used only pre-college characteristics as predictors and yielded "hit" rates of 78.9% and 86,8%, A third model included additional measures of early college experiences as predictors and resulted in a hit rate of 94.2%.

Although considerable predictive accuracy was achieved with pre- college characteristics alone, a chi-square analysis showed that the additional measures of early college experiences significantly improved the predictive efficiency of the discriminant aa- lei. For those models using only pre-college chat acteristics as predictors, the ANQVA revealed that variables relating to academic aptitude/achievement had the greatest differences in persistence group means. The model which included all variables had the greatest differences across means on the measures of early college experiences. The survey item, "what are your plans for the second semester . . , revealed the single greatest difference between persistence groups.

A further analysis of the data showed that the variable "presence" or "absence" of ACT Assessment data was a useful predictor of persistence for students who were otherwise dropped from the discriminant analyses because of that missing data.

Suggestions for campus intervention programs targeted at probable non-persisters were given.

Download

Share

COinS