Analysis of Factors Affecting Migration of Freshmen Applicants at Selected Public Institutions of Higher Education in Connecticut

With enrollments declining nationally, and with Connecticut losing such a large portion of its potential college population, higher education in Connecticut may suffer profoundly from diminishing enrollments in the not-too-distant future. The need for information on migrating students has become more critical to the New England Region for planning and decision making due to the substantial out-migration patterns in this area, the diminishing number of traditional students in the 18- to 21-year-old age group and the increasing concerns on the viability of growth in adult and nontraditional enrollment. This study sought to ascertain whether differences existed between three groups of accepted applicants to specific institutions of higher education in relation to selected personal, academic, and socioeconomic variables. Method: For each of three populations, the University of Connecticut, the University of Connecticut Branches, and Central Connecticut State College, there were three groups of accepted applicants those students choosing to enroll (Group 1), those individuals remaining in Connecticut but turning down admission to the specified institutions (Group 2), and those students enrolling in out-of-state institutions (Group 3). Multivariate discriminant analysis was chosen as the most appropriate statistical technique for analyzing whether the respective groups could be differentiated from each other on the basis of the variables constituting the multivariate discriminant space. The minimization of Wilks' Lambda was selected as the discriminant criterion and all hypotheses were tested at the P<.01 level of significance. Results: For each of the three populations, the University of Connecticut, the University of Connecticut Branches, and Central Connecticut State College, both the first and second discriminant functions were significant beyond the .001 level of confidence The probability of obtaining group differences this large on the basis of samples drawn from the multivariate discriminant space would be less than one in a thousand for each population. Therefore, since there are significant group differences in each of the three populations, the null hypotheses that group centroids are equal was rejected. Additionally, since the Lambdas associated with all functions were significant, it can be concluded that there are significant differences among the groups within each of three populations along both axes which defined the overall discriminant space. Since the three general multivariate null hypotheses were rejected, a stepwise discriminant analysis was performed for each population to determine which of the 41 variables comprising the total discriminant space were associated with each of the discriminant functions. In each of the populations, the first discriminant function primarily accounted for factors differentiating between enrolling students (Group 1) and non-enrolling students (Groups 2 and 3) The variables contributing heavily to this function related most closely to considerations involving prestige and elements of choice entering into college selection; hence, it was identified as a Prestige/Preference function. The second function appears to differentiate the two groups of non-enrolling students from each other, those students choosing other Connecticut institutions (Group 2) from migrant students (Group 3). A singular interpretation of this function was generally not possible since it appeared to be comprised of several components including prestige and preference variables, socioeconomic variables, and external influences. The most compelling and consistant result of the present study involved the concept of prestige. The assessment of institutional prestige clearly dominated the results of the three major discriminant analyses and consistantly appeared as a major factor in the decision-making behavior of students. Students who elected to enroll in the institutions included in this study reported that prestige v/as a major consideration in their decision. Similarly, students electing not to accept offers of admission reported that institutions they had chosen to attend were more prestigious.