Academic journal article Journal of Applied Research in the Community College

Application of Survival Analysis to Study Timing and Probability of Outcome Attainment by a Community College Student Cohort

Academic journal article Journal of Applied Research in the Community College

Application of Survival Analysis to Study Timing and Probability of Outcome Attainment by a Community College Student Cohort

Article excerpt

This study applies competing risks survival analysis to describe outcome attainment for an entire cohort of students who first attended a Midwestern community college in the Fall Semester 2001. Outcome attainment included transfer to a four-year institution, degree/certificate attainment from the community college under study, and transfer to a different two-year college. The analysis also shows how the technique can be used to categorize students who have not realized a graduation or transfer outcome. Outcome probability at different points in time (13 semesters) is calculated and discussed. The effects of demographic and academic variables on timing and type of outcome attained are also discussed. Findings from the study show that outcome attainment was widely distributed over chronological time, there were three discernable stages of outcome attainment across number of semesters enrolled, and cumulative grade point average (GPA) was the most prominent effect on outcome.

Introduction

Community colleges serve a majority of postsecondary students in the United States. The need for community colleges to demonstrate their effectiveness in terms of quantifiable outcomes continues to be a fundamental expectation of policymakers and the public. However, understanding outcome attainment is often difficult for many reasons. Among these reasons, limited access to enrollment records at institutions other than the institution under study, long and irregular enrollment histories, and unapparent student intentions can impede a comprehensive and accurate representation of outcomes.

One issue in studying student outcomes pertaining to community colleges is the fact that, unlike four-year institutions, student goal is often something other than degree attainment. Common goal categories are transfer to a four-year institution (with or without degree/ certificate attainment at the two-year institution), education for discrete skill acquisition, or education for personal interest without degree or transfer intent.

A second research problem involved in studying student histories that is especially applicable in the community college context is the phenomenon of stopout. The standard retention model for higher education is based on the traditional four-year university in which a student enrolls and persists from term to term at the same institution until graduating, and this remains the predominant paradigm for the structure of federal and state reporting and research studies. Yet it is very common for enrollment behavior to vary from this model, especially for students who attend community colleges (Adelman, 2005), which raises the issue of how to summarize cohort behavior for students who have not (yet) attained transfer or graduation outcomes and who were not currently enrolled at the time of the study. A related concern is also noted by Voorhees, Smith, and Luan (2006), who cite the need to study the complex array of community college student transitions.

The purpose of this study is to explore the application of survival analysis to enhance the representation of outcomes in the context of community colleges. The potential merit of this method is particularly relevant to community colleges because of the multiplicity of student outcomes, variety of enrollment patterns, and diversity of community college students. Survival analysis models the status of subjects at discrete time points in a way that accounts for changes in the underlying number of subjects who could realize outcomes at different stages. For this reason, it may provide a richer narrative of enrollment and outcomes than a simple representation that is limited to calculating outcomes as a percentage of the number of students who are present at the beginning of a study period.

To our knowledge there is no existing published research that applies survival analysis for the primary purpose of understanding the paths to outcome attainment of entire student cohorts over time from a single community college, and that is the distinct focus of this paper. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.