Effects of Part-Time Faculty Employment on Community College Graduation Rates
Jacoby, Daniel, Journal of Higher Education
Over the past three decades, one of the most significant changes in the delivery of postsecondary education involves the dramatic increase in the use of contingent or part-time faculty. The pattern is particularly pronounced at community colleges, where part-time faculty provide virtually half of all instruction. Despite this, little systematic attention has been given to the effect of this phenomenon upon student persistence and attainment. Until very recently the literature on part-time faculty concentrated almost entirely on faculty equity (American Association of University Professors [AAUP], 2003; Jacoby, 2001), human resource policy (Antony & Valadez, 2002; Gappa & Leslie, 1993; Schuetz, 2002), or the corporatization of academia (Aronowitz, 2000). Even with its varied focus, much of this scholarship makes the tacit assumption that reliance upon a system of part-time faculty employment harms college students. This article examines whether student graduation rates at community colleges decrease when part-time faculty employment increases.
In most prior research on persistence, students are the unit of analysis. This is most likely the case because institutional-level data, particularly for the community colleges, have some shortcomings. Nonetheless, in analyzing the effects of part-time faculty there are at least three reasons to conduct research in which colleges are the unit of analysis. Foremost among these is the fact that institutions are being held accountable for their graduation rates, making essential research that contributes to understanding institutional performance. Second, colleges are a natural unit to study because employment decisions are generally made at the institution level. Finally, the most accurate data regarding part-time faculty are to be found at the institutional level. This study uses institutional data, including graduation rates, provided by the National Center for Educational Statistics (NCES) within its Integrated Postsecondary Data System (IPEDS).
This article first summarizes related theory and evidence regarding student graduation, learning outcomes, and the use of part-time or contingent faculty. I then discuss the limits that existing data impose upon this study. Next, I present an analytical model and report the results generated from it. Finally, I address the results and their significance and make suggestions for further study.
Theory and Prior Evidence
Research that connects the separate literatures on part-time faculty to the extensive literature on student persistence is scarce. Bailey and Alfonso (2005) have pointed out that research on persistence has not been particularly effective in identifying programs and policies that improve student outcomes at community colleges. Two things have been lacking: good data that would allow effective comparisons among community colleges and, in the case of past single-institution studies, adequate controls that would have helped determine causality. NCES publication of institutional graduation rates goes some distance towards remedying this situation (Bailey, Alfonso, et al. 2005; Bailey, Calcagno, et al., 2005).
Bailey and Alfonso (2005) observed that much of the research on student persistence and graduation is now more than a decade old and suggested that it be updated. While counseling, advising, and developmental education have been identified as "crucial" to community college students, current studies have not identified "the most effective design and organization" for these services (Bailey & Alfonso, 2005, p. 2). Faculty at community colleges typically assume greater responsibilities for these services than do those at four-year institutions where research expectations are greater.
College graduation and persistence studies emphasize the vital importance of student integration or engagement. Typically, such studies have been grounded in research conducted at four-year colleges attended by traditional students (Astin, 1993; Bailey & Alfonzo, 2005; Pascarella & Terenzini, 1991, 2005). …