Will They Stay or Will They Go? Predicting the Risk of Attrition at a Large Public University
2007; American Association of Collegiate Registrars and Admissions Officers; Volume: 83; Issue: 2 Linguagem: Inglês
ISSN
0010-0889
Autores Tópico(s)Evaluation of Teaching Practices
ResumoTHE PROJECT described in this article seeks to utilize a complex set of variables and characteristics, including data mined from an administration of the College Student Experiences Questionnaire, to determine the specific risk of attrition of individual college students prior to their matriculation, permitting specific and personal interventions. Student persistence is a matter of concern at almost all institutions of higher education, as well as to legislators, government officials, and the public. While some student attrition is not necessarily a negative outcome, the wasted resources and perceptions of failure that are associated with unacceptable attrition levels are problematic for college and university administrators. Many institutions have implemented strategies intended to enhance student persistence. However, most involve the broad application of programs, services, or instructional initiatives that affect large groups of students. Because it is reasonable to expect that some portion of any group of students would have been retained regardless of the intervention, such a broad approach is, at best, inefficient. The project described in this article seeks to utilize a complex set of variables and characteristics to determine the specific risk of attrition of individual students prior to their matriculation. For those determined to be most at risk, an intervention specific to the individual student and his or her risk factors will be implemented. BACKGROUND A number of studies of student persistence have examined the interaction between students and institutions. Tinto (1975) proposes a predictive model based upon principles of the student's level of academic and social integration. The model appears very useful in explaining attrition in the second year of college and beyond. However, decades of evidence suggest that attrition is greatest during the first year of college, as demonstrated by Iffert (1958), Marsh (1966), andEckland (1964). Pascarella and Terenzini (1980) applied Tintos principles of academic and social integration of students in the first year of college. They found evidence to support the basic aspects of the Tinto model - particularly the value of interaction between students and faculty members. Tintos research and that of Pascarella and Terenzini are useful models for predicting attrition based upon student characteristics and the nature and extent of student interaction with the institution. Chapman and Pascarella (1983) studied differences in student social and academic integration, central principles in Tintos work, across various types of institutions. The researchers controlled for differences in student characteristics and found that there were differences across institutional type regarding both social and academic integration. The results suggest that particular types of institutions foster different sorts of integration or interaction opportunities for students. Robbins, Allen, Cassilas & Peterson (2006) studied the effect of student self-reported psychosocial factors on college outcomes in the first year and also found differences across institutional type. Braxton, Vesper, and Hossler (1995) studied students' expectations of the college experience and the relationship of those expectations to students' intention to persist across a number of institutions. The study demonstrated that the extent to which student expectations of their experience are met has an effect on their plans to continue their matriculation. Heiland, Stallings, and Braxton (2002) also studied how the fulfillment of student expectation relates to social integration and student departure. Their study at a single institution concluded that the satisfaction of student expectations plays a substantial role in student departure. In a study at Canisius College, a smaller institution, Glynn, Sauer, and Miller (2003) developed a model for predicting attrition based upon pre-matriculation characteristics and opinions. …
Referência(s)