Early identification of persisting and nonpersisting college students

The purpose of this study was to determine the potential usefulness of information contained in the college freshman\u27s American College Testing Program (ACT) Assessment for determining those factors or combination of factors that would identify students have a high probability of being a persisting or nonpersisting college student. The study compared the ACT Assessment results for persisting and nonpersisting students who enrolled in the fall semester of 1982-83, 1983-84, and 1984-85. There were 7720 students in the study who met the selection criteria. There were 1606 nonpersisting students, those students who did not enroll at the University of Nebraska-Lincoln the fall semester immediately following the freshmen year, and 6114 persisting students. The independent variables in the study were the scores and the 161 variables on the ACT Assessment. Descriptive statistics were assessed for the two groups of students, persisting and nonpersisting, followed by factor analysis. After a factor analysis was done to decrease the number of variables into seven construct variables was completed, a discriminate analysis was performed on the seven construct variables. Because of the conflict in the literature as to using dichotomous variables in factor analysis, a second discriminate analysis based on all variables was done. A third discriminate analysis was done using only the seven major variables of the 54 variables found in the second discriminate analysis. The seven major factors were high school GPA, plan to be employed while in college, ACT composite score, prefer to attend UNL, plan to join a fraternity/sorority, amount of high school math, and high school accomplishments in music. The descriptive statistics presented differences in responses to the ACT variables by persisting and nonpersisting students. Some of the findings from the descriptive statistics were that persisting students tend: to come from smaller high schools; to have better high school grades; to come from greater distances; to come from the upper quartiles of their high school class; to have participated in more activities in high school; to have worked fewer hours in employment; to come from smaller high schools and towns, and to come from smaller families with higher incomes. In essence, the descriptive statistics presented a difference in the responses to the ACT Assessment by persisting and nonpersisting student groupings, but the difference in responses between the two groups was statistically so insignificant as to prevent the classification of a student as persisting or nonpersisting based on the student\u27s responses to the variables