Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13
Topics
Discussion
Announcements
 

Do Community College Students Benefit from Federal Work Study Participation?


by Hongwei Yu, Lyle McKinney & Vincent D. Carales - 2020

Background: Prior studies suggest that Federal Work-Study (FWS) participation is positively associated with student learning, persistence, and academic achievement at four-year institutions. Limited research, however, has evaluated whether FWS participation improves academic success among students attending community colleges.

Purpose and Research Questions: The purpose of this study was to determine whether and how FWS participation impacted academic performance and enrollment outcomes among a racially/ethnically diverse sample of students attending a large, urban community college (UCC) system in Texas. There were two research questions: (1) What are the characteristics of students at UCC who participated in FWS, compared with their peers who did not participate? (2) After controlling for self-selection bias, are there significant differences in academic success (i.e., cumulative GPA; credential attainment and/or four-year transfer) among UCC students who did and did not participate in the FWS program?

Research Design: The longitudinal data set (fall 2010 through summer 2016) analyzed in this study was built using detailed student-level transcript data records. The full sample included 8,837 students who had filed a Free Application for Federal Student Aid (a necessary step to receive FWS funding), but the primary focus was on the subsample of FWS participants (n = 260). Descriptive analysis was performed to compare the demographic and academic characteristics of FWS participants with nonparticipants. To assuage self-selection bias, propensity score matching (nearest neighbor matching algorithm) was used to match similar students who did and did not participate in FWS. We employed multiple regression and logistic regression techniques on the matched data to investigate whether FWS participation was associated with students’ academic outcomes.

Results: Relative to their non-FWS peers, FWS participants at this community college were more likely to be female, African American, 24 years of age or older, very low income, and academically underprepared. After successfully matching FWS participants with similar non-FWS participants, results indicated that FWS participation was associated with a higher cumulative GPA and significantly higher odds of credential completion and/or vertical transfer.

Conclusions: There are important equity implications in our findings; the results suggest that the FWS program can improve educational outcomes for student populations that are often marginalized and underserved by the higher education system. We describe several ways that the FWS program could be redesigned and expanded to better meet the needs of community college students.

Working during college is the norm for most community college students, but historically, only a small percentage of students in this sector participate in the Federal Work-Study (FWS) program. This disconnect is surprising given that the purpose of FWS is to expand employment opportunities as a means of increasing college access and completion among low-income students. Consistent with historical trends, in 2016–17, less than 10% of all FWS participants were enrolled at a community college (College Board, 2017). One possible explanation for the low participation rate is that the types of jobs available through FWS simply do not align with community college students’ employment preferences or income needs. However, there is increasing evidence the FWS program, as currently designed and delivered, disproportionally penalizes community colleges and constrains participation among students in this sector (Kelchen, 2017; O’Sullivan & Setzer, 2014; Scott-Clayton, 2017; Scott-Clayton & Zhou, 2017).


Similar to federal grants, the FWS program is intended to supplement students’ personal financial resources and minimize their need to incur debt. Although they provide financial benefit to the student, FWS awards do not directly reduce college expenses; rather, they expand students’ employment opportunities to earn income to pay for college. In theory, the FWS program objectives are well aligned with the needs of community college students. Recent figures suggest that approximately 78% of part-time students and 43% of full-time students in this sector hold employment while in college (National Center for Education Statistics, 2017). Moreover, 28% of community college students view themselves as “an employee who studies” rather than a “student who works” (Velez et al., 2018). But factors associated with holding off-campus employment, such as an unexpected change in one’s job hours and work schedule, frequently disrupt academic progress among community college students (Mukherjee et al., 2016).


Considering that FWS provides flexible hours that work around the student’s class schedule, expanding FWS opportunities at community colleges could address these students’ employment needs while facilitating their academic progress. However, expanding FWS participation among community college students would require modifications to the current formula used to allocate FWS funds, which has disproportionally benefited four-year, private institutions (Kelchen, 2017; Scott-Clayton & Zhou, 2017). These modifications to the FWS program design and funding formula should be based on empirical evidence. Promisingly, several recent studies using quasi-experimental designs have found that FWS has a positive effect on persistence and early labor-market outcomes for students at four-year institutions (e.g., Scott-Clayton, 2011; Soliz & Long, 2016). But there is very little evidence on whether FWS benefits community college students. Our study addresses this gap in the literature and, in doing so, explains how expanding FWS participation at community colleges could significantly increase degree completion among our nation’s low-income and racial/ethnic minority students.


The purpose of our study was to determine how FWS participation impacted academic success among a sample of highly diverse students attending a large, urban community college system in Texas. Urban Community College (UCC; a pseudonym) provides an ideal case to examine the equity implications of the FWS program, because the vast majority of UCC students are low income and students of color. Given that prior studies have found that the effects of FWS can vary across institutional contexts, we were particularly interested in determining whether FWS worked for a diverse student population such as UCC’s. Two research questions guided our study:


1.

What are the characteristics of students at UCC who participated in FWS, compared with their peers who did not participate?

a.

Are there significant differences in participation in terms of students’ race/ethnicity, income-status, and/or academic preparedness?

2.

After controlling for self-selection bias, are there significant differences in academic success (i.e., cumulative GPA; credential attainment and/or four-year transfer) among UCC students who did and did not participate in the FWS program?


FWS BACKGROUND AND POLICY CONTEXT


Established in 1964 as part of the Economic Opportunity Act, the FWS program was designed to mitigate the costs of college by providing access to on- and off-campus part-time employment for low-income students (Wisconsin HOPE Lab, 2016). The FWS program has been operational for more than five decades, surpassing both the Pell Grant and Stafford Loans as one of the earliest forms of federal student aid. Students are typically paid FWS wages on an hourly basis, with the minimum set at the current federal minimum wage. However, institutions located in states or local entities with minimum wage requirements above the federal minimum must pay the higher hourly wage (U.S. Department of Education, 2017). Institutions that elect to participate in the FWS program determine the number of hours a student is allowed to work based on the student’s financial need, which is calculated from the Free Application for Federal Student Aid (FAFSA). There are no statutory limits on the number of FWS hours per week a student may work, but most institutions set the limit at 20 hours. The total amount a student can earn through FWS is determined by the number of hours worked, the duration of the FWS appointment, the hourly wage rate, and the amount of other financial aid made available to the student (U.S. Department of Education, 2017).


U.S. Department of Education (2017) guidelines state that “to the maximum extent practicable, a school must provide FWS jobs that complement and reinforce each recipient’s educational program or career goals” (p. 6-39). But in reality, this is often not the case; a recent study found that only 5% of FWS students reported their employment to be relevant to their academic and career goals, compared with 23% of students in non-FWS jobs (Wisconsin HOPE Lab, 2016). Offices on campus where FWS students are commonly employed include enrollment services, financial aid, career services, student affairs, library services, and human resources. In addition, some FWS positions allow students to serve as peer guidance counselors, academic tutors, or teaching assistants. Although the vast majority of FWS positions are based on campus, some colleges establish partnerships with area organizations that result in FWS jobs within local government agencies, nonprofits, and for-profit businesses.  


Relative to other forms of employment, FWS offers several distinct benefits for students. First and foremost, wages earned through the FWS program are not included in the earned income calculation to determine the expected family contribution (EFC) when students complete the FAFSA in subsequent years. Second, FWS employers are more likely to work around students’ course schedules when assigning work hours and be more lenient when students need time off to complete class assignments or study for exams. Students in FWS jobs were more likely to report being able to do schoolwork on the job and to have a more consistent work schedule from week to week compared with their counterparts in non-FWS jobs (Wisconsin HOPE Lab, 2016). In addition, FWS positions have the potential to improve students’ chances of degree completion by strengthening their social and academic integration into campus life (Beeson & Wessel, 2002; Hossler et al., 2008; Wisconsin HOPE Lab, 2016).


Historically, funding for the FWS program has accounted for a small proportion of the total federal financial aid disbursed to college students. In 2016–17, FWS represented less than 1% of the total $102 billion disbursed to undergraduate students through federal student aid programs (College Board, 2017). A total of 619,000 recipients, which includes graduate students, were allocated $990 million through the FWS program in that same year, representing a 14% funding decrease over the last decade (College Board, 2017). The lower level of funding for FWS, relative to other federal student aid programs, limits the program’s ability to increase college access and persistence among low-income students. Given that most FWS dollars now go to students attending private four-year institutions, there are concerns that the program does not benefit students who need it most, such as those attending community colleges (Kelchen, 2017; O’Sullivan & Setzer, 2014; Scott-Clayton, 2017). In addition, although the higher education system has evolved considerably since the FWS program was first established, little has changed in the formula that determines how much and which institutions receive FWS funding, which is a key reason that some have advocated for a complete overhaul of the program (O’Sullivan & Setzer, 2014).


At the time of this writing, several federal policy proposals have been introduced that could potentially result in significant changes to the FWS program. The Trump administration’s 2018 federal budget recommendations included a request to reduce FWS funding by almost 50% (Scott-Clayton, 2017). In contrast, other congressional leaders have proposed expanding FWS by rolling existing Federal Supplemental Educational Opportunity Grant dollars into the FWS program. This proposed legislation, called the Promoting Real Opportunity, Success and Prosperity Through Education Reform (PROSPER) Act, also seeks to eliminate FWS eligibility for graduate students, noted as an area of concern by the American Council on Education and other higher education stakeholders in a letter sent to the authors of the bill (American Council on Education [ACE], 2017a). The National Association of Student Financial Aid Administrators supports the PROSPER Act’s elimination of caps on private sector jobs and set-asides for community service positions, providing more flexibility for institutions to place students in positions where jobs might be more readily available (ACE, 2017b). The PROSPER Act would also preserve the requirement that jobs in the private sector be academically relevant to the student’s program of study (ACE, 2017a).


In response to the House Republicans’ passage of the PROSPER Act, House Democrats introduced their own version of a higher education reauthorization bill, the AIM Higher Act. Most of the provisions to the FWS program are in both bills; however, the AIM Higher Act proposes to increase overall funding for FWS and maintains graduate student eligibility. Both bills also propose moving away from historic funding allocations to a fair-share formula based on overall student need and the number of Pell Grant recipients that an institution enrolls (Cooper, 2018). The AIM Higher Act goes one step further by allocating a set-aside of additional funds for institutions that improve completion rates among Pell Grant recipients (Cooper, 2018).


Considering that 32% of all Pell Grant recipients attend community colleges (National Association of Student Financial Aid Administrators, 2018), these proposed policy changes to the FWS program could have a significant impact on community college student success. But there is little evidence on whether FWS participation improves academic performance and graduation rates among this student population. The findings from our study provide empirical evidence that can be used to inform current and future policy decisions about the FWS program. As Scott-Clayton (2017) suggested, before making cuts to FWS that would likely hinder student success, “energy would be better spent innovating, experimenting, and rigorously evaluating this half-century old program that still appears to have a valuable place in a modern college completion and career readiness agenda” (p. 4).


LITERATURE REVIEW


A large body of empirical scholarship has examined the extent to which different types and amounts of financial aid influence college student success. Most research examining the effects of financial aid has focused on Pell Grants (e.g., Schudde & Scott-Clayton, 2014; Singell & Stone, 2007; Turner, 2014), state need-based aid programs (e.g., Castleman & Long, 2016; Goldrick-Rab et al., 2016), institutional aid (e.g., Waddell & Singell, 2011), and loans (e.g., Dowd & Coury, 2006; Kim, 2007; McKinney & Backscheider Burridge, 2015; Museus, 2010). Empirical research on the FWS program is comparatively scarce, which is somewhat surprising given the program’s longevity and the sizable federal financial investment made in the program annually.


Recognizing the paucity of scholarship on FWS specifically, the first section of this review provides a synthesis of research that has examined how employment—a fundamental feature of FWS—impacts college students’ academic progress and educational outcomes. The second section provides a detailed review of the few key studies that have specifically examined the impact of FWS participation on student success. The closing section identifies the gaps in the literature and describes the contributions of our study to advancing research, policy, and practice related to the FWS program.


THE IMPACT OF EMPLOYMENT ON STUDENT OUTCOMES


Nearly 70% of today’s college students work while attending school (Carnevale & Smith, 2018). The increase in working students has been attributed to multiple factors, including the escalating costs of college attendance, debt aversion, and an uptick in working among traditional-age students who enroll full time (Broton et al., 2016; Perna, 2010). Students work to earn income to pay for college and living expenses, and to gain practical work experience that can better prepare them for their career (Soliz & Long, 2016). Not surprisingly, holding employment and working more than 20 hours per week is most common among lower-income students. But while low-income students often have little choice but to work through college, a recent report found that compared to higher-income working students, low-income workers had poorer academic and early labor market outcomes (Carnevale & Smith, 2018).


Research has examined the relationship between student employment and its impact on a wide range of outcomes, including academic performance, cognitive and social development, leadership skills, student debt, peer and faculty interaction, student engagement, persistence, degree completion, and labor market outcomes (see Pascarella & Terenzini, 2005; Perna, 2010; Riggert et al., 2006; Tuttle et al., 2005). A notable theme from this literature is that the complexity of students’ working behaviors and choices presents many methodological challenges for researchers. For example, employment may take place off campus or on campus, students may hold multiple jobs simultaneously, and/or the number of hours worked may change from week to week. Many data sets do not contain reliable information about the amount of income earned from off-campus jobs during college or account for local labor market conditions that can influence students’ working behaviors. These factors may help explain why research on the effects of working while attending college, regardless of the type or intensity, has generated mixed and inconclusive results (Riggert et al., 2006).


Given the goals of our study, we were primarily concerned with findings from research examining the effects of working on students’ academic performance and persistence in college. A central issue examined across the student employment literature is how the number of hours worked per week influences key outcomes (Choi, 2018). In general, moderate working (less than 20 hours) appears to be positively associated with academic performance and student engagement (Dundes & Marx, 2006; Pike et al., 2008), whereas working 20 or more hours seems to adversely influence GPA and course completion (Torres et al., 2010), as well as persistence (Bozick, 2007; Choi, 2018; Orozco & Cauthen, 2009). Students working 25 or more hours per week were more likely to report that employment had negatively impacted their grades (T. King & Bannon, 2002). This negative impact increased for students who worked full time.


In contrast, other research suggests that working a limited number of hours (i.e., less than 20) has relatively no positive or negative consequences on academic success (Dadgar, 2012; Darolia, 2014; Nonis & Hudson, 2006). Analyzing data from the 1997 National Longitudinal Survey of Youth, Darolia (2014) compared grades and credit completion for full-time and part-time college students who worked and found a small yet statistically insignificant impact of working on grades for both groups. Darolia also found that increasing work hours had a small negative impact on credit completion for full time-students at four-year institutions but minimal impact for students attending two-year institutions. In one of the few studies to examine the impact of employment specifically for community college students, Dadgar (2012) analyzed a unique administrative data set from Washington State during 2001–02 that included students’ transcripts, as well as state employment records. The results suggested a small negative impact of working on student grades but that working an additional 1–10 hours per week was associated with increased credit accumulation. The author hypothesized that working more hours may have generated income that enabled students to register for additional courses.


With regard to persistence and degree completion, some studies have found a positive correlation with employment during college. Using survey data from undergraduates at a single Midwestern university, Kulm and Cramer (2006) found that persistence increased when the number of hours employed increased and attributed this positive correlation to the possibility that these students were more focused toward the end goal of graduating on time. Conversely, researchers have found that working on campus did not impact persistence to graduation (Beeson & Wessel, 2002), whereas others found a negative association between timely degree completion and the number of hours worked (Mendoza, 2012; Titus, 2010). More specifically, in her analysis of data from the 2007–08 National Postsecondary Student Aid Study (NPSAS:08), Mendoza (2012) reported that students working less than 30 hours a week were 1.4 times more likely to graduate in six years than those who worked more than 30 hours a week. It is important to note that all these studies were focused on students attending four-year institutions.


A smaller body of literature has examined how working during college impacts a range of interpersonal and career-related outcomes. Some have argued that work inhibits students’ ability to learn because it prevents them from fully engaging in the classroom, while also compromising their engagement in civic learning, community service, and other extracurricular activities (T. King & Bannon, 2002). On the other hand, studies have found that positive aspects of working include intellectual development, planning and time management, social skills and self-image, career connections, enhanced work habits, and broader support networks that help students overcome academic and personal challenges (Curtis & Nimmer, 1991; McCormick et al., 2010; Pusser, 2010; Scott-Clayton, 2011). Greater gains in leadership development skills were also reported among students who were employed, compared with their peers who were not (Salisbury et al., 2012).


A particularly noteworthy theme from the student employment literature is that working often has differential impacts for students from different age groups, racial/ethnic backgrounds, socioeconomic statuses, and institutional types (Canabal, 1998; Choi, 2018; Perna, 2010; Pusser, 2010; Torres et al., 2010; Wood et al., 2016). Using data from 11 postsecondary institutions collected through the Assessment of the Status of Minorities in Education Project in Illinois, Canabal (1998) found a negative relationship between grades and employment among Black and Hispanic students. Analyzing data from the National Survey of Youth 1997, Choi (2018) found a negative relationship between persistence and working more than 20 hours a week for students from low socioeconomic backgrounds. However, Choi concluded that working during college was actually less detrimental to those from more disadvantaged backgrounds. Some research suggests that community college students may be more likely to benefit from working than their peers at four-year institutions in terms of estimated returns to in-school work (Molitor & Leigh, 2005). But other findings indicate that working among community college students can be a barrier to their academic performance and persistence (Dadgar, 2012; Levin et al., 2010; Mukherjee et al., 2016; Wood et al., 2016).


FEDERAL WORK-STUDY PARTICIPATION AND STUDENT OUTCOMES


Compared with research on grants and loans, research that explicitly focuses on FWS is limited, and early studies on the program’s impact generated inconsistent findings (see Pascarella & Terenzini, 2005). For example, early studies reported null (e.g., Curtis & Nimmer, 1991), negative (e.g., Clotfelter, 1991; Kaltenbaugh et al., 1999; Paulsen & St. John, 2002), and positive (e.g., Adelman, 1999; Heller, 2003; St. John, 1990) effects of FWS participation on academic outcomes. As Scott-Clayton (2017) explains, the FWS program has never been evaluated via a randomized controlled trial design, making it difficult to assess the program’s causal impact on academic and labor market outcomes. But promisingly, more recent work on FWS has used quasi-experimental techniques (e.g., Scott-Clayton, 2011; Scott-Clayton & Minaya, 2016; Soliz & Long, 2016) in an attempt to reduce selection bias associated with FWS participation and obtain more accurate estimates of the program’s impact. In general, recent studies have found that FWS participation had a small negative impact on students’ first-year GPA (e.g., Scott-Clayton, 2011; Soliz & Long, 2016), but this may be overshadowed by a positive association between FWS participation and longer term outcomes such as persistence, degree completion, and employment after college (Scott-Clayton, 2017; Scott-Clayton & Zhou, 2017).


The following paragraphs provide a detailed review of the handful of studies that have applied quasi-experimental methods to examine the impact of FWS employment (and on-campus work more broadly) on student outcomes. Using an administrative data set (1989–1997) from Berea College, a private four-year institution in Kentucky, Stinebrickner and Stinebrickner (2003) examined the relationship between hours worked on campus and academic performance for full-time students. The authors clearly explained that the generalizability of their results may be limited because Berea is unique in that the college has a mandatory work-study program, and all students receive full-tuition scholarships as part of the college’s mission to serve high-achieving, lower income students. Berea’s work-study program requires students to work a minimum of 10 hours per week, but students can elect to work additional hours. The study carefully describes how unobserved factors, such as students’ level of motivation, can influence the number of hours a student chooses to work and therefore should be controlled for during data analysis to reduce bias. Accordingly, the researchers adopted an instrumental variable (IV) approach. Whereas results from simple regression and fixed-effect approaches suggested a positive association between working more hours on campus and students’ GPA, results from the IV approach suggested a negative association.


Using a large-scale administrative data set from West Virginia, Scott-Clayton (2011) conducted the first quasi-experimental examination of FWS on student outcomes. The goal of the study was to determine whether FWS participation had a causal effect on students’ GPA, credit accumulation, and degree completion. The study combined an instrumental variable and difference-in-difference approach that allowed for comparison of FWS funding allocations to institutions in the sample (using a low and high classification system), as well as actual student participation rates in the program. Overall, results for the full sample, which were reinforced through multiple forms of robustness checks, indicated that FWS did not significantly improve any of the academic outcomes under investigation. There were several notable findings from the subgroup analyses; women had significantly lower outcomes than men across all the measures. Additionally, FWS participants who delayed enrollment into college had better outcomes than their younger peers. The results did not reveal significant differences between students attending two-year or four-year institutions. In summarizing the main findings, Scott-Clayton explained “that at least in the state of West Virginia, Federal Work-Study participation does not generally improve academic outcomes, and in fact, it may cause some students some harm” (p. 525). Importantly, the author explains the generalizability of the findings to broader contexts should be interpreted cautiously, given that West Virginia is a nationally unrepresentative state (e.g., more than 90% of the students in the study’s sample were White).    


Polson and Weisburst (2014) examined the impact of FWS on community college students across Texas, and the findings are particularly relevant given the context and goals of our study. The authors used administrative data from the Texas Education Resource Center to determine the relationship between FWS participation and enrollment outcomes for students graduating high school in 2008 and 2009. The sample was restricted to FWS participants who had also received a Pell Grant. Results indicated that these FWS students were more likely to be male, be African American, and have lower college placement exam scores; Hispanic students were less likely to participate in FWS. Regression results suggested that FWS participation was associated with a 12%–15% increase in the odds of persistence to the second year, and 3%–4% higher odds of transferring to a four-year institution by 2012. Regarding limitations, the authors acknowledged potential differences in the types of students selecting into the FWS program across colleges and the importance of using a data set that can determine FWS students’ enrollment outcomes over a longer time period (e.g., six years). Nonetheless, the study provides evidence that FWS may be beneficial for community college students.  


Using two waves of nationally representative data sets from the Beginning Postsecondary Students Longitudinal Study (BPS: 96/01 and BPS: 04/09), Scott-Clayton and Minaya (2016) examined the effects of FWS on academic and early labor-market outcomes for students attending four-year institutions. Results suggested that FWS participation increased bachelor’s degree attainment (by 3.2 percentage points) and employment rates (by 2.4 percentage points) six years after initial college enrollment. But notably, the use of a conditional counterfactual analytic approach revealed important heterogeneous effects of the program. Relative to other students who would have worked during their first year, FWS participation seemed to positively affect their academic outcomes but not their postcollege employment outcomes. Conversely, compared with students who would have not worked, FWS participation appeared to positively impact postcollege employment outcomes but not academic outcomes. Particularly relevant to our study, the researchers found that the positive effects of FWS participation were greater for lower income students.


Most recently, Soliz and Long (2016) used an instrumental variables differences-in-differences model to determine the impact of FWS participation on a broad range of first-year academic outcomes for first-time, full-time freshmen in the Ohio public higher education system. Similar to prior studies (Scott-Clayton, 2011; Scott-Clayton & Minaya, 2016; Stinebrickner & Stinebrickner, 2003), the authors found that FWS had a small but statistically significant negative effect on students’ first- and second-semester GPAs. However, FWS participation had a positive impact on the total number of credits earned by the end of the first year. Soliz and Long (2016) also found less of a negative effect of FWS on academic outcomes for students who were financially independent of their parents, suggesting that adults older than “traditional” college age (who represent a large proportion of the sample analyzed in our study) may experience added benefits from FWS participation.


SUMMARY AND CONTRIBUTIONS OF THE PRESENT STUDY


Collectively, research on the effects of working while attending college, regardless of the type or intensity, has generated mixed results. Multiple studies seem to suggest that working 20 hours per week or less is typically not a barrier to academic success, whereas working more hours can be detrimental to persistence. The vast majority of student employment studies have sampled students at four-year institutions, not community colleges, which is surprising given the higher rates of employment among community college students. Research has largely focused on the number of hours worked or the placement of work (e.g., on or off campus), with less attention given to the specific type of work students are engaged in. There is considerable evidence that working during college can exert a differential effect on educational outcomes depending on students’ age, race/ethnicity, income status, and enrollment intensity.


Studies specifically examining the impact of FWS participation have largely focused on four-year students or a combined sample of two- and four-year students. Much of the research on FWS is now dated, but several recent studies have applied a quasi-experimental research design to advance knowledge about the effects of FWS on student outcomes. There is evidence that FWS participation may initially result in a lower first-year GPA as students adjust to balancing work and academic responsibilities. However, this short-term negative outcome appears to be offset by findings indicating that FWS improves students’ likelihood of degree completion and early labor-market success. Recent literature also makes clear that the effects of FWS participation are heterogeneous and dependent on factors such as the state context, student demographics, and institutional sector.  


Our study advances research on the FWS program by focusing specifically on a sample of FWS participants at a large, urban, and highly diverse community college district in Texas. In addition to identifying the profile of students who participate in FWS, the study uses a quasi-experimental design to estimate the impact of FWS on students’ GPA and enrollment outcome. If low-income and independent students (e.g., age 24 and older for federal financial purposes) stand to benefit the most from FWS, as several recent studies suggest, then allocating a larger proportion of FWS funding to community colleges holds promise for increasing persistence and completion rates in this sector. Our study addresses the equity implications of the FWS program and provides evidence that can inform current policy proposals aimed at redesigning FWS to better achieve its goal of increasing college access and completion among low-income students.


CONCEPTUAL FRAMEWORK


Many studies examining the impact of employment during college have drawn on seminal work by Bean and Metzner (1985) and conceptualized off-campus employment as an “external pull factor” that draws the student away from academics and campus life (Arbona & Nora, 2007; J. Chen & Hossler, 2017; Mukherjee et al., 2016; Nora, 2003). The ability to successfully balance work and school across multiple years is a difficult undertaking. When forced to choose between the two, many lower income students have little choice but to prioritize earned income over the immediate costs associated with pursuing a college degree. Among community college students, researchers have found that working off campus—particularly full time—is a factor that often hinders persistence (Levin et al., 2010; Mukherjee et al., 2016).


But a notable distinction between FWS and other common types of student employment is that, rather than pulling the student away from campus, FWS keeps them there. Consequently, multiple studies have described FWS as a mechanism for increasing student engagement on campus (e.g., Flowers, 2010; McCormick et al., 2010; Nuñez & Sansone, 2016; Pike et al., 2008; Umbach et al., 2010). Although engagement is sometimes considered a more salient concept for four-year students, numerous studies show that engagement matters at community colleges. Researchers have found a relationship between increased engagement and key community college student outcomes such as sense of belonging (e.g., Wood & Harris, 2015), higher GPA (e.g., McClenney et al., 2012), and degree completion (e.g., Center for Community College Student Engagement, 2013; McClenney et al., 2012; Price & Tovar, 2014). Moreover, there is evidence that increased engagement can be particularly beneficial for low-income students and students of color (Flowers, 2010; Long, 2016), which has important implications given the demographic profile of our sample.


For the purposes of our study, we hypothesized that FWS participation increases students’ level of engagement at the community college, which in turn increases their academic performance and persistence. Because of their obligations outside of school, the classroom is the primary—and sometimes the only—source of campus engagement for many community college students (Levin et al., 2010). But working on campus can naturally increase the frequency of a student’s interactions with faculty, academic advisors, and other college staff who can support the student’s academic success. FWS supervisors, relative to off-campus employers, are also more likely to allow students to modify their work hours around their class schedule each semester (Soliz & Long, 2016). In addition, transportation to and from campus each week is a common barrier to persistence among community college students (Mukherjee et al., 2016). By working at the same physical location where they attend classes, FWS can decrease students’ transportation costs and reduce the number of times they must commute to campus each week. Collectively, these factors associated with on-campus FWS employment are hypothesized to increase students’ study time and overall engagement in campus life.


Prior research corroborates that the effects of working, as well as financial aid, can vary significantly depending on students’ demographic characteristics. Therefore, in addition to student engagement literature, we used an integrated conceptual model proposed by R. Chen (2008) for examining the effects of financial aid on college student persistence. Chen’s model underscores the importance of examining possible heterogeneous effects of financial aid types (e.g., FWS) by students’ race/ethnicity and income status. Given the diversity of our sample, we do not anticipate that FWS participation will exert a uniform effect on academic outcomes for all students. Informed by prior research (Flowers, 2010; McCormick et al., 2010; Pike et al., 2008), we instead hypothesize that the increased campus engagement provided by FWS is particularly beneficial for students of color and low-income students.


Chen’s model incorporates perspectives from five major frameworks (i.e. psychological, sociological, organizational, interactionalist, economic) used in studying college student persistence. Sociological dimensions of the model are relevant for understanding socioeconomic stratification across the U.S. higher education system. Students from higher income families and those who possess greater access to social and cultural capital are overrepresented in resource-rich institutions (e.g., private four-year, Ivy League schools) with high graduation rates, whereas students who do not have these luxuries are overrepresented in institutions that are perpetually underfunded and underresourced (e.g., community colleges). Applied to the FWS program, we argue that the current allocation formula (in which significantly more resources are awarded to private four-year students than community college students) is a policy mechanism that perpetuates gaps in educational attainment by socioeconomic status and places historically marginalized groups at a distinct disadvantage.    


Chen’s work provides guidance on how to apply the conceptual model during data analysis. Based on a critical review of the literature, Chen proposed eight constructs for independent variables: student background, educational aspiration, precollege preparation, financial factors, college experience, institutional characteristics, interaction effects, and time in college. Chen advised that researchers apply statistical techniques when studying financial aid that can assuage the potential selection bias associated with students’ choices regarding financial aid use. In conjunction with the student engagement literature, Chen’s conceptual model and proposed analytic approach guided our selection of variables and overall research design.  


METHOD

DATA SOURCE AND ANALYTIC SAMPLE


We analyzed longitudinal student-level transcript records from a large, urban community college district in Texas. Urban Community College district (UCC, a pseudonym) is one of the largest and most racially/ethnically diverse community college systems in Texas, currently serving more than 80,000 students annually. In 2017, the UCC student body was 71% racial/ethnic minority, 37% of students were older than 25 years, and 40% were low income, and 46% of incoming first-time-in-college and transfer students were referred to developmental education. A key goal of our study was to understand if FWS benefited this type of community college student population.


The institutional records were made available as part of a longstanding research-to-practice partnership between the authors and UCC. The full data set included a cohort of degree-seeking students who enrolled at UCC during fall 2010 (n = 13,569); their experiences were followed through summer 2016, yielding data from six academic years. In addition to providing detailed information about students’ demographic characteristics and academic behaviors, the data set was linked to data from the National Student Clearinghouse to include transfer outcomes. Financial aid records were merged into the data set and provided information about the type and amount of financial aid that a student received each semester. Although longitudinal—transcript-level records are most commonly used for institutional reporting and planning purposes—these data can be particularly powerful for researchers aiming to deconstruct the complex enrollment patterns and academic behaviors of community college students (Bahr, 2013; Crosta, 2014; Hagedorn & Kress, 2008).


To be eligible for federal financial aid, including FWS, a student must first complete the FAFSA. Prior studies have shown there are often notable differences in the types of students at community colleges who do and do not file the FAFSA (McKinney & Novak, 2013, 2015). Therefore, consistent with our analytic goal of matching similar FWS and non-FWS students, we limited the sample to include only those students who had filed a FAFSA during their first year at UCC (n = 8,837). International students were not included in the sample because they are ineligible for federal student aid. Our primary subsample of interest was students in this cohort who participated in the FWS program at any point during the six years captured in the data set (n = 260). The relatively low rates of FWS participation at UCC are consistent with national rates at community colleges. However, the large size of UCC’s entering student cohorts, coupled with the colleges’ commitment to expanding FWS opportunities, is a rather unique institutional context that provided a sufficient number of FWS cases for the purposes of our analysis.


FEDERAL WORK-STUDY AT UCC


A description of how FWS operates at UCC is useful in contextualizing the results from the data analysis. To facilitate FWS appointments, each semester, UCC hosts a series of FWS job fairs that allow interested students to connect with potential employers. Students who have been deemed eligible for FWS receive targeted emails to preregister for a job fair. Hiring managers at UCC, as well as community partners, attend the fair and talk with students about the jobs available. Students then choose which employers to interview with for FWS positions. As students are hired, they are referred to a financial aid coordinator to begin the application process for the position, which is completed online. After clearing a background check, students complete an I-9 form with the UCC human resources office and begin their job assignment.


During the period captured in our data set, the vast majority (86%) of FWS jobs at UCC were on-campus positions. UCC divisions where FWS students are commonly employed include enrollment services, financial aid, student services, ability services, career services, human resources, facilities, faculty affairs, and information technology. In recent years, UCC has also expanded the number of off-campus FWS positions within outside organizations, including K–12 school districts, community and social services, philanthropic organizations, and child care facilities. At the time of this writing, on-campus FWS positions at UCC paid $9.50 per hour, and off-campus FWS jobs paid $10 per hour. Students are capped at working 19.5 hours per week, but the actual number of hours worked varies based on student availability and employer needs. The UCC financial aid director explained that compared with four-year students, UCC students are more likely to be older and have additional financial responsibilities that require them to work full time. Therefore, for many UCC students, it is not financially feasible to take an FWS position and reduce their employment status from full time to part time.


VARIABLES


Guided by R. Chen’s conceptual model (2008) and the student engagement literature, the independent variables reflected student demographic characteristics (i.e., gender, race/ethnicity, age, EFC) and academic/college experiences (e.g., high school academic preparation, dual enrollment, developmental education status, program of study, financial aid received). Prior research suggests that each of these variables can play a role in shaping the academic progress and success of community college students. With the exception of the developmental education and financial aid measures (described next), all the independent variables were captured during the students’ first academic year at UCC (i.e., 2010–2011).


We performed data cleaning and recoding prior before the multivariate analysis. Table 1 presents the coding and levels for each of the independent variables. High school academic preparation was recoded into two categories: earned a high school diploma, and GED/other. The total number of dual credits earned in high school was included to reflect students’ academic momentum before UCC enrollment. Regarding program of study, students were classified as either being enrolled in an academic-focused (i.e., transfer) or a vocational/technical (i.e., workforce preparation) program. The developmental education variables captured the breadth and depth of students’ academic preparation across three subject areas: English, reading, and math. Rather than relying on placement test scores or referrals, the transcript data allowed us to identify students’ actual enrollment in developmental courses across all years in the data set. FWS recipients are eligible to receive additional types of financial aid to offset the total costs of college. Therefore, we examined FWS students’ complete financial aid records to determine whether they had ever received other financial aid in the form of Pell Grants, state need-based aid, and/or federal student loans.



Table 1. Descriptive Statistics of Pre- and Postmatching Samples of Urban Community College Students

 

 

Prematching (N = 8,837)

 

Postmatching (N = 780)

Covariates

Categories/intervals

FWS (n = 260)

Non-FWS (n = 8,577)

Mean difference

FWS

(n = 260)

Non-FWS (n = 520)

Mean difference

t test/chi-square (p values)

Distance

 

0.044

0.029

0.015

 

0.044

0.043

0.000

 

Gender

Male

0.350

0.388

-0.038

 

0.350

0.379

-0.029

0.431

Female

0.650

0.612

0.038

 

0.650

0.621

0.029

 
          

Race/Ethnicity

White

0.104

0.148

-0.044

 

0.104

0.106

-0.002

 

Hispanic

0.212

0.306

-0.066

 

0.212

0.233

-0.021

0.505

African American

0.542

0.440

0.102

 

0.542

0.527

0.015

0.685

Asian

0.115

0.083

0.032

 

0.115

0.117

-0.002

0.937

Other Race/Ethnicity

0.027

0.023

0.004

 

0.027

0.017

0.010

0.372

          

Age

< 19

0.262

0.372

-0.111

 

0.262

0.260

0.002

 

19–24

0.281

0.289

-0.008

 

0.281

0.271

0.010

0.777

25 or Older

0.458

0.339

0.119

 

0.458

0.469

-0.012

0.761

          

Expected Family Contribution

Total EFC

0.279

1.227

-0.948

 

0.279

0.320

-0.041

0.693

          

Precollege Preparation

GED/Other

0.204

0.138

0.066

 

0.204

0.212

-0.008

 

High School Diploma

0.796

0.862

-0.066

 

0.796

0.789

0.008

0.803

          

Developmental Education

College-Level Reading

0.635

0.771

-0.136

 

0.635

0.633

0.002

 

Developmental Reading I

0.235

0.155

0.080

 

0.235

0.235

0.000

1.000

Developmental Reading II

0.131

0.075

0.056

 

0.131

0.133

-0.002

0.940

         

College-Level English

0.662

0.739

-0.078

 

0.662

0.656

0.006

 

Developmental English I

0.154

0.121

0.033

 

0.154

0.146

0.008

0.776

Developmental English II

0.185

0.140

0.045

 

0.185

0.198

-0.014

0.654

         

College-Level Math

0.269

0.408

-0.138

 

0.269

0.269

0.000

 

Developmental Math I

0.100

0.136

-0.036

 

0.100

0.089

0.012

0.600

Developmental Math II

0.200

0.133

0.067

 

0.200

0.177

0.023

0.434

Developmental Math III

0.431

0.323

0.108

 

0.431

0.465

-0.035

0.360

          

Dual Credits Earned

Total Dual Credits

0.542

0.608

-0.066

 

0.542

0.390

0.152

0.394

          

Program of Study

Vocational or Technical

0.350

0.239

0.111

 

0.350

0.350

0.000

 

Academic Transfer

0.650

0.761

-0.111

 

0.650

0.650

0.000

1.000

          

Financial Aid

Pell Grant

1.209

0.956

0.254

 

1.209

1.239

-0.030

0.699

Non–Pell Grant

0.077

0.095

-0.018

 

0.077

0.069

0.009

0.641

Federal Student Loan

1.417

1.049

0.367

 

1.417

1.495

-0.079

0.584

Notes: Financial aid variables are divided by 1,000. For numerical variables, we reported p values for t test under the assumption of unequal variance (Satterthwaite method). For categorical variables, we reported p values for chi-square statistic. Insignificant p values suggest no significant difference between the FWS and non-FWS group after matching.



Consistent with prior studies (e.g., Scott-Clayton, 2011; Scott-Clayton & Minaya, 2016), we examined how FWS participation impacted students’ academic and enrollment outcomes. There were two key dependent variables of interest: (a) academic performance measured by cumulative GPA (0.0–4.0 scale) as of the student’s last semester of enrollment at UCC, and (b) whether a student had earned a UCC credential (i.e., associate’s degree or certificate) and/or four-year transferred within six years of initial enrollment (coded 0 = No, 1 = Yes).


DATA ANALYSIS


To answer the first research question, we used descriptive statistics to compare the characteristics of FWS students with those of their peers who did not participate in FWS. Recognizing that students “self-select” to participate in the FWS program, we used propensity score matching (PSM) to reduce selection bias. PSM allows researchers to match a student’s probability of enrolling in the treatment group (FWS participation) with his or her probability of enrolling in the control group (non-FWS participation) (Rosenbaum & Rubin, 1983). Under this technique, each student was assigned a propensity score generated by a logistic/logit regression model. Students’ propensity of participating in the FWS program was then used to match students in the FWS and non-FWS groups.


The propensity score matching procedure was implemented using MatchIt package within R statistical software. Each student in FWS was matched with two students in the non-FWS group using a nearest neighboring matching algorithm. Prior research suggests that either a 2:1 or 1:1 ratio can achieve optimal estimation of treatment effects. Using a 2:1 ratio provided a larger control group (i.e., non-FWS students) from which to find a strong match for each FWS student (Austin, 2010). Through this matching process, we obtained a matched sample with little difference between students in FWS and non-FWS groups in terms of selected covariates (i.e., nonsignificant p values for either t or chi-square statistic). Therefore, the observed differences between these groups in student outcomes are more likely to be associated with FWS participation. After obtaining the matched data set, we employed multiple regression and logistic regression techniques to investigate whether FWS participation was associated with students’ cumulative GPA and likelihood of UCC credential completion and/or vertical transfer.


LIMITATIONS


There are several limitations of our study that warrant discussion. Only those students who filed a FAFSA were included in the sample used to match FWS and non-FWS participants.

Therefore, our sample is not representative of the entire community college student population.  Transcript-level data allow researchers to deconstruct the complex academic and enrollment behaviors of community college students, but these data sets do not typically capture cultural factors (i.e., funds of college knowledge, primary spoken language) that can influence students’ educational outcomes. Additionally, a common limitation in studying student employment during college is that few data sets capture information about the quality of the work experience or the specific nature of work. Although we know that the students held a FWS job, UCC’s financial aid records captured in our datafile do not identify the specific division or office where the student worked. The possibility of omitted variable bias is important to acknowledge because propensity score matching is less effective at reducing selection bias when the model is missing important variables (Padgett et al., 2010). This limited our ability to infer a causal relationship between FWS participation and student outcomes.  


To obtain a sufficient number of FWS cases for our quasi-experimental design, our sample included students who had participated in FWS at any point during their time at UCC. This approach was consistent with our goal of providing a foundational understanding of how FWS participation influences academic outcomes for a highly diverse group of community college students. But we recognize that participating in FWS during the first academic year could, for example, exert a differential effect as compared with waiting until the second year to begin FWS employment. Relatedly, our sample size also limited our ability to address a possible dosage effect of FWS, whereas there could be different effects on student outcomes depending on how many total semesters the student participated in FWS. A future study could use a larger sample of FWS students to investigate the temporal effects of program participation on student outcomes.


Many of the described limitations have been well documented in prior research. The methodological complexities in studying the effects of financial aid on student persistence in general (e.g., R. Chen, 2008; Dowd, 2008), and on FWS more specifically (e.g., Scott-Clayton, 2017), have likely contributed to the scarcity of empirical research on FWS. Research has shown that working can have varying effects for different types of students across different institutional sectors, which underscores the value of institutional-level research on FWS. The large enrollment size of UCC provides a somewhat unique opportunity to study FWS within the community college context. Considering that our analysis focused on FWS participants at a large, urban community college district in Texas, our findings may not be generalizable to other institutional contexts or environments (e.g., four-year public or private institutions), particularly those much less racially/ethnically diverse than UCC. Relatedly, the findings could be partially explained by UCC polices and processes regarding FWS, given that institutions have some discretion in how FWS allocations are distributed across campus. But despite these potential limitations, the findings from our quasi-experimental design provide new empirical evidence about the benefits of FWS for a highly diverse population of community college students.  


RESULTS


WHO PARTICIPATED IN FWS AT THIS COMMUNITY COLLEGE?


We presented descriptive statistics for two groups: (1) students who received FWS and (2) students who did not receive FWS (see Table 1). Overall, only 2.9% (260 out of 8,837 students) of the full sample of FAFSA filers participated in FWS during their time at UCC, a figure consistent with national FWS participation rates at community colleges. There were several notable differences between the two groups in terms of demographic characteristics. Relative to their non-FWS counterparts, females were overrepresented in the FWS group (65%). In terms of race/ethnicity, African American students were overrepresented in the FWS group and constituted the largest proportion of all FWS participants (54.2%). On the other hand, Hispanic students were underrepresented in FWS (21.2%) relative to their percentage in the non-FWS group (30.6%). Students 25 years of age and older were overrepresented in the FWS group (45.8%) in comparison with their percentage of the non-FWS group (33.9%). In addition, students who received FWS aid had a significantly lower EFC ($279, SD = 1303) than their non-FWS counterparts ($320, SD = 1479).


In terms of students’ academic and collegiate experiences, compared with their non-FWS peers, a higher percentage of FWS students took developmental coursework in reading (36.5% vs. 22.9%), English (33.8% vs. 26.1%), and math (73.1% vs. 59.2%). A higher percentage of students (43.1% vs. 32.3%) in the FWS group were referred to the very lowest level of developmental math. Regarding financial aid receipt, the average total FWS amount received by FWS students was $893 (SD = 1191). FWS recipients, on average, received relatively smaller amounts of Pell Grant funding ($1,209, SD = 995) relative to their non-FWS peers ($1,239, SD = 1036) and had slightly lower average federal loan debt ($1,417, SD = 1834) than non-FWS students ($1,495, SD = 1995).


WAS FWS PARTICIPATION ASSOCIATED WITH IMPROVED ACADEMIC OUTCOMES?


The descriptive analysis of prematching samples revealed key differences in the characteristics of UCC students who do and do not select to participate in the FWS program. After nearest neighbor matching, we obtained a sample of FWS students and non-FWS students who were similar in terms of demographic characteristics and their academic experiences (indicated by small mean differences and correspondingly insignificant p values). The matched sample was then used for subsequent multivariate analyses. The analyses of the matched sample (both multiple and logistic regression) suggested that FWS participation at UCC was positively associated with students’ academic performance (measured by cumulative GPA) and likelihood of credential completion from UCC and/or four-year transfer.


As indicated by Table 2, relative to their non-FWS peers, students enrolled in the FWS program had a .224 higher cumulative GPA as of their last semester of enrollment at UCC (β = .224, p < .0001). In addition, several covariates were related to student cumulative GPA. In the matched sample, African American students had lower GPAs when compared with White students (β = -.194, p = .001). Students with a high school diploma had a higher GPA relative to their peers who earned a GED (β = .119, p = .001). Students placed in level 1 of developmental English had lower GPAs relative to their peers not placed in developmental English (β = -.117, p = .001). Being placed in the lowest level of developmental math adversely impacted students’ cumulative GPA (β = -.090, p = .044), underscoring the critical connection between developmental math and student academic performance at community colleges. Students with higher federal loan amounts tended to have lower cumulative GPAs (β = -.113, p = .004), as did students with higher amounts of Pell Grants (β = -.076, p = .043). These results may reflect the reality that the lowest income students are more likely to need to borrow and are eligible for higher Pell Grant amounts.


Table 2. Multiple Regression Results: Student Cumulative GPA (Propensity Score Matched Sample)

Parameter

Estimate

Standardized

estimate

SE

t Value

Pr > |t|

Intercept

2.417

0.000

0.161

14.980

<.0001***

FWS Participation

0.475

0.224

0.068

7.030

<.0001***

Gender

-0.144

-0.069

0.068

-2.120

0.034*

Asian American

0.253

0.081

0.141

1.790

0.074

African American

-0.389

-0.194

0.111

-3.500

0.001**

Hispanic American

-0.163

-0.068

0.124

-1.320

0.187

Other Race/Ethnicity

0.186

0.026

0.247

0.750

0.452

High School Graduation

0.293

0.119

0.084

3.490

0.001**

Academic Major

-0.136

-0.065

0.070

-1.930

0.054

Age1 (19–24)

-0.059

-0.026

0.093

-0.640

0.526

Age2 (25 or Older)

0.299

0.149

0.090

3.330

0.001**

EFC

0.002

0.003

0.024

0.090

0.928

Developmental Read 1

0.148

0.063

0.088

1.690

0.092

Developmental Read 2

-0.017

-0.006

0.111

-0.150

0.878

Developmental English 1

-0.329

-0.117

0.097

-3.380

0.001**

Developmental English 2

-0.176

-0.070

0.100

-1.770

0.077

Developmental Math 1

0.241

0.070

0.125

1.930

0.055

Developmental Math 2

-0.027

-0.010

0.100

-0.270

0.787

Developmental Math 3

-0.181

-0.090

0.090

-2.020

0.044*

Total dual credits

0.021

0.047

0.015

1.410

0.158

Total Pell

-0.074

-0.076

0.037

-2.030

0.043*

Total Non–Pell Grant

-0.110

-0.029

0.123

-0.890

0.371

Total Loan

-0.058

-0.113

0.020

-2.920

0.004**

Note: F  = 10.77, p < .0001, R2 = 0.238, adjR2 = 0.216. Financial aid variables are divided by 1,000.

*p < .05. **p < .01. ***p < .001.



When examining the relationship between FWS and enrollment outcomes, we found that program participation significantly improved students’ odds of credential completion from UCC and/or four-year transfer. Specifically, as revealed in Table 3, FWS participants were more than twice as likely to achieve a successful enrollment outcome compared with their non-FWS peers (β = 2.024, p < .0001). Students placed in level 2 (β = .555, p = .013) or level 3 (β = .451, p = .0002) developmental math courses were significantly less likely to achieve a successful enrollment outcome than students who were college-ready in math. Relative to GED holders, students who earned a high school diploma were twice as likely to achieve a positive enrollment outcome (β = 2.091, p = .0003). Other covariates reflecting demographic characteristics (e.g., gender, race/ethnicity, age), precollege academic preparation (e.g., total dual credits), and college experiences (e.g., program of study, financial aid) were not associated with students’ likelihood of credential completion from UCC and/or four-year transfer.  


Table 3.  Logistic Regression Results: Likelihood of UCC Credential Completion and/or Four-Year Transfer (Propensity Score Matched Sample)

Variable (Reference Group)

 Categories/Intervals

Log Odds

Odds Ratio

95% Confidence Limits

p value

FWS Participation

 

0.705

2.024

1.471

2.784

<.0001***

Gender (Female)

Male

-0.285

0.752

0.547

1.035

0.081

       

Race/Ethnicity (White)

Asian American

0.319

1.376

0.709

2.671

0.345

African American

0.166

1.181

0.703

1.984

0.530

Hispanic American

-0.295

0.745

0.417

1.331

0.320

Other Race/Ethnicity

0.158

1.171

0.375

3.658

0.786

       

Age

Age (< 19 vs. 19–24)

-0.217

0.805

0.521

1.244

0.329

Age (< 19 vs. 25 or Older)

-0.037

0.964

0.633

1.468

0.865

       

Expected Family Contribution

Total EFC

0.062

1.063

0.953

1.187

0.272

       

High School Preparation

(GED/Other)


High School Diploma

0.738

2.091

1.402

3.119

0.0003***

       

Dual Credits Earned

Total Dual Credits

-0.019

0.981

0.916

1.051

0.593

       

Developmental Education Placement

Developmental Reading  (1 vs. 0)

-0.195

0.823

0.545

1.240

0.351

Developmental Reading  (2 vs. 0)

-0.372

0.689

0.410

1.159

0.160

Developmental English    (1 vs. 0)

-0.025

0.976

0.619

1.538

0.916

Developmental English    (2 vs. 0)

0.050

1.051

0.659

1.677

0.833

Developmental Math        (1 vs. 0)

0.241

1.272

0.697

2.322

0.433

Developmental Math        (2 vs. 0)

-0.588

0.555

0.349

0.884

0.0130*

Developmental Math        (3 vs. 0)

-0.797

0.451

0.296

0.686

0.0002***

       

Program of Study

(Technical/Vocational)

Academic or Transfer

0.197

1.218

0.875

1.695

0.329

Financial Aid

Pell Aid

0.052

1.053

0.888

1.249

0.554

Non–Pell Grant Aid

0.243

1.275

0.718

2.265

0.407

Loan Aid

0.074

1.076

0.980

1.182

0.123

Notes. Financial aid variables are divided by 1,000. Developmental Reading 0 means college-level reading; Developmental English 0 means college-level English; Developmental Math 0 means college-level math.

*p < .05. **p < .01. ***p < .001.


DISCUSSION AND IMPLICATIONS


Our study builds on several recent quasi-experimental investigations of the FWS program by focusing specifically on a highly diverse sample of community college students. After matching, FWS participants at UCC had significantly higher cumulative GPAs than their non-FWS peers and were more likely to have earned a community college credential (associate’s degree or workforce certificate) and/or transferred to a four-year institution. There are important equity implications of our findings; students who were African American, who were 25 years of age and older, who were the least academically prepared for college, and who had the lowest EFCs were overrepresented among FWS participants. Consistent with recommendations from several recent studies (e.g., Kelchen, 2017; Polson & Weisburst, 2014; Scott-Clayton, 2017; Scott-Clayton & Minaya, 2016), our results suggest that the FWS program could be more effectively leveraged to improve college success rates among low-income and historically marginalized students.    


We attribute the higher academic performance and completion/transfer rates among FWS participants at UCC to an increased level of engagement that directly resulted from working on campus. FWS likely provided students the opportunity to develop supportive relationships with UCC personnel, increase their study time, and reduce transportation barriers associated with traveling from work to school multiple times per week. The earned income from FWS enabled students to cover some of the costs associated with attending college. Several recent studies largely focusing on four-year students found a small negative impact of FWS participation on students’ first-year GPA (Scott-Clayton, 2017; Scott-Clayton & Minaya, 2016; Soliz & Long, 2016). However, we found a positive association between FWS and students’ cumulative GPA as of their final semester at UCC. These differences could be attributed to the timing at which GPA was measured, as well as the characteristics of the student and institutional samples examined in the analyses. However, our finding may suggest that FWS is effective at increasing longer term academic engagement, and thus performance, among community college students. Our finding that FWS students were more likely to earn a UCC credential and/or four-year transfer is aligned with results from several recent studies (e.g., Polson & Weisburst, 2014; Scott-Clayton & Minaya, 2016; Soliz & Long, 2016) indicating that FWS participation has the potential to improve college persistence and completion rates.     


Promisingly, in our study, many of the student groups at heightened risk for nonsuccess in college were among those benefiting from FWS. R. Chen’s (2008) conceptual model posits that students’ race/ethnicity and income status influence their use of different types of financial aid. African American students represented about half of all FWS students in the sample. Given that racial/ethnic minorities are more likely to face discrimination when seeking employment in the labor market (Darity & Mason, 1998; Pager & Shepherd, 2008; Ziegert & Hanges, 2005), school-sponsored FWS jobs may provide a particularly important employment option for students of color. Additionally, the very lowest income students at UCC were overrepresented among FWS participants, and income from FWS was likely essential for many of these students to attend college. Older adult students and those placed in the lowest levels of developmental education were also overrepresented in FWS. Collectively, these results suggest that the FWS program holds the potential to improve educational outcomes for student populations that are often marginalized and underserved by the higher education system.  


Over time, the FWS program has drifted away from its stated purpose of providing employment opportunities and financial assistance to college students with the greatest financial need (Kelchen, 2017; O’Sullivan & Setzer, 2014; Scott-Clayton & Zhou, 2017). In 2015–2016, the private nonprofit sector represented 18% of all undergraduate student enrollment (FTE) but received 41% of FWS funds (College Board, 2017). In contrast, community colleges represented 32% of undergraduate enrollment (FTE) but only received 18% of FWS allocations. A natural consequence of the current FWS allocation formula is that only a small proportion of community college students, most of whom have high levels of financial need, are able to benefit from FWS. Our study should be replicated in other community college contexts, but the findings suggest that directing additional FWS funding to open-access colleges that predominantly serve our nation’s less advantaged student populations is a policy strategy that deserves serious attention.


Modifications to the FWS allocation formula are necessary and should prioritize funding for institutions that serve the student populations that truly need the employment experiences and income provided by FWS. To better align FWS resources with the program’s stated purpose of assisting low-income students, the updated funding formula should allocate funds based on the number of Pell Grant recipients an institution serves. Kelchen (2017) analyzed national data to explore how reallocating FWS funds—in combination with funds from the smaller campus-based Supplemental Education Opportunity Grant (SEOG) program—using this approach would change the distribution across institutional sectors. He found that


if FWS and SEOG funds were allocated solely based on the number of students in each sector receiving Pell Grants, community colleges would receive 45% of total aid dollars in the 2013-14 award year, public 4-year colleges would receive 31%, private 4-year colleges would receive 11%, and proprietary institutions would get 12%. These allocations would result in a distribution that more closely reflects enrollment by sector than the current SEOG or FWS allocations. (p. 464)


We strongly support provisions in current federal policy proposals (i.e., PROSPER Act, AIM Higher Act) that would (a) allocate FWS funding based on the number of Pell Grant recipients at institution serves and (b) increase overall funding for FWS by incorporating funds from the SEOG program.  


Other changes to FWS may also be necessary to significantly increase participation among community college students. On average, community college students are older than students in the nonprofit four-year sectors and are more likely to have additional financial obligations (e.g., mortgage, children) that require full-time employment (Malcom, 2013). But most institutions currently cap students at 20 FWS hours per week based on federal guidelines (U.S. Department of Education, 2017). At the current FWS hourly wage rate (e.g., $9.50 for on-campus jobs at UCC), many low-income community college students simply cannot afford to participate in FWS. There are at least two possible strategies to address this issue. The first is to increase the hourly wage for FWS jobs, making on-campus employment more attractive and financially viable. The second, and less desirable, approach would be to increase the maximum FWS hours a student can work (e.g., from 20 to 30) that remain untaxed. Although additional financial resources would be required, these policy changes could significantly increase FWS participation rates and begin to transform the nature of student employment within the community college sector.


Many positive spillover effects would accompany policy changes that broaden FWS participation among low-income students. In addition to increasing persistence rates, research suggests that on-campus employment can enrich students’ engagement, learning, and career preparation (Scott-Clayton, 2017; Scott-Clayton & Minaya, 2016). These are important externalities that should not be overlooked in current policy discussions about FWS, which have largely centered on the financial costs of program expansion. As Soliz and Long (2016) explained,


If funds can be used to provide students with work experience that enriches their course of study and gives them a foothold when they enter the labor market, while at the same time meeting the labor needs of the colleges and universities where they attend, then this would seem to be an efficient use of scarce government resources. (p. 28)


Numerous studies indicate that the nature and quality of the work experience play a critical role in shaping college students’ learning, skill development, and early labor market success (Broton et al., 2016; J. E. King, 2005; Perna, 2010; Stern et al., 1997). Although transferrable skills can be gleaned from many different types of jobs, ideally a student’s FWS job will be closely related to his or her academic major and intended career field. But unfortunately, this is not the case for most students (Wisconsin HOPE Lab, 2016). Strategies that better connected the FWS job with the student’s career goals could substantially improve the overall effectiveness of the program. For example, matching a student studying computer programming with an FWS job in the college’s information technology department could generate numerous benefits for the student and college. Recognizing that it is not possible to create such an ideal match for all students, maximizing the effectiveness of the FWS program will likely require colleges to create and expand partnerships with off-campus employers.     


Our study advances knowledge on the impact of FWS for community college students, but there is a need for additional research on this topic. Future studies may consider examining other factors associated with FWS participation at community colleges, such as time to degree/transfer and students’ early labor market outcomes. Research underscores the value of examining the effects of FWS within specific institutional contexts, but a study using a recent nationally representative data set (e.g., BPS, NPSAS) to specifically examine the impact of FWS at community colleges would represent an important contribution to the literature. The vast majority of FWS studies are quantitative in nature, but qualitative methods could provide a deeper understanding of how FWS influences students’ campus engagement and academic progress. Researchers may also consider exploring potential differential impacts of FWS as a function of the specific type of appointment the student holds (e.g., on campus or off campus, related or unrelated to the student’s academic major).

CONCLUSION


As one of the few studies to explicitly examine the impact of FWS for community college students, our study contributes to current policy discussions about how FWS can be redesigned and leveraged to improve success among low-income and racial/ethnic minority students. The current FWS allocation formula disproportionally penalizes community colleges and consequently denies important participation benefits to the types of students the program was originally designed to help. Community college students, more so than students in any other sector, are highly dependent on gainful employment to achieve their educational goals. Allocating FWS funds based on the number of Pell Grant recipients an institution serves is a policy change that could dramatically improve student engagement, degree completion, and career preparation at community colleges. The time has come to redesign FWS so that it truly works for our nation’s low-income students.  


References


Adelman, C. (1999). Answers in the toolbox: Academic intensity, attendance patterns, and bachelor’s degree attainment. U.S. Department of Education. Washington, DC: Office of Educational Research and Improvement.

 

American Council on Education. (2017a, December 17). Renewing the Higher Education Act: Letters to the House on the PROSPER ACT [Letter signed by 37 higher education associations on the PROSPER Act]. Retrieved from http://www.acenet.edu/Pages/Renewing-the-Higher-Education-Act.aspx

 

American Council on Education. (2017b, December 8). Renewing the Higher Education Act: Letters to the House on the PROSPER ACT [National Association of Student Financial Aid Administrators]. Retrieved from http://www.acenet.edu/Pages/Renewing-the-Higher-Education-Act.aspx

 

Arbona, C., & Nora, A. (2007). The influence of academic and environmental factors on Hispanic college degree attainment. The Review of Higher Education, 30(3), 247–269. 10.1353/rhe.2007.0001

 

Austin, P. C. (2010). Statistical criteria for selecting the optimal number of untreated subjects matched to each treated subject when using many-to-one matching on the propensity score. American Journal of Epidemiology, 172(9), 1092–1097.

 

Bahr, P. R. (2013). The deconstructive approach to understanding community college students’ pathways and outcomes. Community College Review, 41, 137–153.

 

Bean, J. P., & Metzner, B. S. (1985). A conceptual model of nontraditional undergraduate student attrition. Review of Educational Research, 55(4), 485–540.

 

Beeson, M. J., & Wessel, R. D. (2002). The impact of working on-campus on the academic persistence of freshmen. Journal of Student Financial Aid, 32(2), 37–45.

 

Bozick, R. (2007). Making it through the first year of college: The role of students’ economic resources, employment, and living arrangements. Sociology of Education, 80(3), 261–285.

 

Broton, K. M., Goldrick-Rab, S., & Benson, J. (2016). Working for college: The causal impacts of financial grants on undergraduate employment. Educational Evaluation and Policy Analysis, 38(3), 477–494.

 

Canabal, M. E. (1998). College student degree of participation in the labor force: Determinants and relationship to school performance. College Student Journal, 32(4), 597–605.

 

Carnevale, A., & Smith, N. (2018). Balancing work and learning: Implications for low-income students. Washington, DC: Georgetown University Center on Education and the Workforce. Retrieved from https://cew.georgetown.edu/cew-reports/LearnAndEarn/

 

Castleman, B. L., & Long, B. T. (2016). Looking beyond enrollment: The causal effect of need-based grants on college access, persistence, and graduation. Journal of Labor Economics, 34(4), 1023–1073.

 

Center for Community College Student Engagement. (2013). A matter of degrees: Engaging practices, engaging students (high-impact practices for community college student engagement). Austin: University of Texas at Austin Community College Leadership Program.


Chen, R. (2008). Financial aid and student dropout in higher education: A heterogeneous research approach. In J. C. Smart (Ed.), Higher education: Handbook of theory and research (Vol. 23, pp. 209–239). Dordrecht, the Netherlands: Springer.

 

Chen, J., & Hossler, D. (2017). The effects of financial aid on college success of two-year beginning nontraditional students. Research in Higher Education, 58(1), 40–76.

 

Choi, Y. (2018). Student employment and persistence: Evidence of effect heterogeneity of student employment on college dropout. Research in Higher Education, 59(1), 88–107.

 

Clotfelter, C. T. (1991). Explaining the demand. In C. T. Clotfelter, R. G. Ehrenberg, M. Getz, & J. J. Siegfried (Eds.), Economic challenges in higher education (pp. 89–123). Chicago, IL: University of Chicago Press.

 

College Board. (2017). Trends in student aid 2017. Trends in Higher Education. Washington, DC: Author.

 

Cooper, M. A. (2018, August 1). A pivotal moment in higher education policy: Reauthorization of the Higher Education Act. Washington, DC: Institute for Higher Education Policy. Retrieved from http://www.ihep.org/sites/default/files/uploads/prosper-aha_ihep_analysis_final.pdf

 

Crosta, P. M. (2014). Intensity and attachment: How the chaotic enrollment patterns of community college students relate to educational outcomes. Community College Review, 42(2), 118–142.

 

Curtis, G., & Nimmer, C. E. (1991). To work or not to work: That is the question. Journal of Student Financial Aid, 21(3), 16–26.

 

Dadgar, M. (2012). Essays on the economics of community college students’ academic and labor market success (Unpublished doctoral dissertation). Teachers College, Columbia University.

 

Darity, W. A., & Mason, P. L. (1998). Evidence on discrimination in employment: Codes of color, codes of gender. Journal of Economic Perspectives, 12(2), 63–90.

 

Darolia, R. (2014). Working (and studying) day and night: Heterogeneous effects of working on the academic performance of full-time and part-time students. Economics of Education Review, 38, 38–50.

 

Dowd, A. C. (2008). Dynamic interactions and intersubjectivity: Challenges to causal modeling in studies of college student debt. Review of Educational Research, 78(2), 232–259.

 

Dowd, A. C., & Coury, T. (2006). The effect of loans on the persistence and attainment of community college students. Research in Higher Education, 47(1), 33–62.

 

Dundes, L., & Marx, J. (2006). Balancing work and academics in college: Why do students working 10 to 19 hours per week excel? Journal of College Student Retention: Research, Theory & Practice, 8(1), 107–120.

 

Flowers, L. A. (2010). Effects of work on African American college students’ engagement. In L. W. Perna (Ed.), Understanding the working college student: New research and its implications for policy and practice (pp. 213–233). Sterling, VA: Stylus.


Goldrick-Rab, S., Kelchen, R., Harris, D. N., & Benson, J. (2016). Reducing income inequality in educational attainment: Experimental evidence on the impact of financial aid on college completion. American Journal of Sociology, 121(6), 1762–1817.

 

Hagedorn, L. S., & Kress, A. M. (2008). Using transcripts in analyses: Directions and opportunities. New Directions for Community Colleges, 143, 7–17.

 

Heller, D. E. (2003). Informing public policy: Financial aid and student persistence. Boulder, CO: Western Interstate Commission for Higher Education.

 

Hossler, D., Ziskin, M., Kim, S., Cekic, O., & Gross, J. P. K. (2008). Student aid and its role in encouraging persistence. In S. Baum, M. McPherson, & P. Steele (Eds.), The effectiveness of student aid policies: What the research tells us (pp. 101–115). New York, NY: College Board.

 

Kaltenbaugh, L. S., St. John, E. P., & Starkey, J. B. (1999). What difference does tuition make? An analysis of ethnic differences in persistence. Journal of Student Financial Aid, 29(2), 21–31.

 

Kelchen, R. (2017). Campus-based financial aid programs: Trends and alternative allocation strategies. Educational Policy, 31(4), 448–480.

 

Kim, D. (2007). Multilevel analysis of the effect of loans on students’ degree attainment:  Differences by student and institutional characteristics. Harvard Educational Review, 177(1), 64–100.

 

King, J. E. (2005). Academic success and financial decisions: Helping students make crucial choices. In R. S. Feldman (Ed.), Improving the first year of college: Research and practice (pp. 3–26). Mahwah, NJ: Erlbaum.

 

King, T. & Bannon, E. (2002). At what cost? The price that working students pay for a college education. The State PIRGs’ Higher Education Project. Retrieved from the United States Public Interest Research Group website: http://www.pirg.org/highered/atwhatcost4_16_02.pdf 

 

Kulm, T. L., & Cramer, S. (2006). The relationship of student employment to student role, family relationships, social interactions and persistence. College Student Journal, 40(4), 927–938.

 

Levin, J. S., Montero-Hernandez, V., & Cerven, C. (2010). Overcoming adversity: Community College students and work. In L. W. Perna (Ed.), Understanding the working college student: Implications for policy, administrators, academic affairs, and institutional support (pp. 43–66). Sterling, VA: Stylus.

 

Long, A. (2016). Overcoming educational racism in the community college: Creating pathways to success for minority and impoverished student populations. Sterling, VA: Stylus.

 

Malcom, L. E. (2013). Student diversity in community colleges: Examining trends and understanding the challenges. In J. S. Levin & S. T. Kater (Eds.), Understanding community colleges (pp. 21–44). New York, NY: Routledge.

 

McClenney, K., Marti, C. N., & Adkins, C. (2012). Student engagement and student outcomes: Key findings from Community College Survey of Student Engagement. Austin: University of Texas at Austin Community College Leadership Program.


McCormick, A. C, Moore, J. V., III, & Kuh, G. D. (2010). Working in college: Its relationship to student engagement and educational outcomes. In L. W. Perna (Ed.), Understanding the working college student: Implications for policy, administrators, academic affairs, and institutional support (pp. 179–212). Sterling, VA: Stylus.

 

McKinney, L., & Backscheider Burridge, A. (2015). Helping or hindering? The effects of loans on community college student persistence. Research in Higher Education, 56(4), 299–324.

 

McKinney, L., & Novak, H. (2013). The relationship between FAFSA filing and persistence among first-year community college students. Community College Review, 41(1), 63–85.

 

McKinney, L., & Novak, H. (2015). FAFSA filing among first-year college students: Who files on time, who doesn't, and why does it matter? Research in Higher Education, 56(1), 1–28.

 

Mendoza, P. (2012). Should I work or should I borrow? Counterfactual analysis on the effect of working while enrolled and debt on baccalaureate completion. Journal of Student Financial Aid, 42(2), 25–59.

 

Molitor, C. J., & Leigh, D. E. (2005). In-school work experience and the returns to two-year and four-year colleges. Economics of Education Review, 24(4), 459–468.

 

Mukherjee, M., McKinney, L., Hagedorn, L. S., Purnamasari, A., & Martinez, F. S. (2016). Stretching every dollar: The impact of personal financial stress on the enrollment behaviors of working and non-working community college students. Community College Journal of Research & Practice, 41(9), 551–565.

 

Museus, S. D. (2010). Understanding racial differences in the effects of loans on degree attainment: A path analysis. Journal of College Student Retention: Research, Theory & Practice, 11(4), 499–527.

 

National Association of Student Financial Aid Administrators. (2018). National Student Aid Profile: Overview of 2018 federal programs. Retrieved from https://www.nasfaa.org/uploads/documents/2018_National_Profile.pdf

 

National Center for Education Statistics. (2017, May). The condition of education: College student employment. Washington, DC: U.S. Department of Education. Retrieved from https://nces.ed.gov/programs/coe/indicator_ssa.asp

 

Nonis, S. A., & Hudson, G. I. (2006). Academic performance of college students: Influence of time spent studying and working. Journal of Education for Business, 81(3), 151–159.

 

Nora, A. (2003). Access to higher education for Hispanic students: Real or illusory? In J. Castellanos & L. Jones (Eds.), The majority in the minority: Expanding the representation of Latina/o faculty, administrators and students in higher education (pp. 47–70). Sterling, VA: Stylus.

 

Nuñez, A. M., & Sansone, V. A. (2016). Earning and learning: Exploring the meaning of work in the experiences of first-generation Latino college students. Review of Higher Education, 40(1), 91–116.

 

Orozco, V., & Cauthen, N. K. (2009). Work less, study more, & succeed. New York, NY: Demos.

 

O’Sullivan, R., & Setzer, R. (2014). A federal work study reform agenda to better serve low-income students. Washington, DC: Young Invincibles.

income students. Washington, DC: Young Invincibles.

 

Padgett, R. D., Salisbury, M. H., An, B. P., & Pascarella, E. T. (2010). Required, practical, or unnecessary? An examination and demonstration of propensity score matching using longitudinal secondary data. In T. A. Seifert (Ed.), New Directions for Institutional Research, Assessment Supplement: S2. Longitudinal Assessment for Institutional Improvement (pp. 29–42). San Francisco, CA: Jossey-Bass.

 

Pager, D., & Shepherd, H. (2008). The sociology of discrimination: Racial discrimination in employment, housing, credit, and consumer markets. Annual Review of Sociology, 34, 181–209.

 

Pascarella, E. T., & Terenzini, P. T. (2005). How college affects students: A third decade of research (Vol. 2). San Francisco, CA: Jossey-Bass.

 

Paulsen, M. B., & St. John, E. P. (2002). Social class and college costs: Examining the financial nexus between college choice and persistence. Journal of Higher Education, 73(2), 189–236.

 

Perna, L. W. (2010). Understanding the working college student. Academe, 96(4), 30–33.

 

Pike, G. R., Kuh, G. D., & Massa-McKinley, R. C. (2008). First-year students’ employment, engagement, and academic achievement: Untangling the relationship between work and grades. NASPA Journal, 45(4), 560–582.

 

Polson, C., & Weisburst, E. (2014). Work-study financial aid and student outcomes: Evidence from community colleges in Texas. Retrieved from Ray Marshall Center website: 

https://raymarshallcenter.org/files/2011/08/WDQI_Workstudy_PolsonWeisburst.pdf

 

Price, D. V., & Tovar, E. (2014). Student engagement and institutional graduation rates: Identifying high-impact educational practices for community colleges. Community College Journal of Research and Practice, 38(9), 766–782.

 

Pusser, B. (2010). Of a mind to labor: Reconceptualizing student work and higher education. In L. W. Perna (Ed.), Understanding the working college student: Implications for policy, administrators, academic affairs, and institutional support (pp. 134–155). Sterling, VA: Stylus.

 

Riggert, S. C., Boyle, M., Petrosko, J. M., Ash, D., & Rude-Parkins, C. (2006). Student employment and higher education: Empiricism and contradiction. Review of Educational Research, 76(1), 63–92.

 

Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70(1), 41–55.

 

Salisbury, M. H., Pascarella, E. T., Padgett, R. D., & Blaich, C. (2012). The effects of work on  leadership development among first-year college students. Journal of College Student Development, 53(2), 300–324.

 

Schudde, L., & Scott-Clayton, J. (2014). Pell Grants as performance-based aid? An examination of satisfactory academic progress requirements in the nation’s largest need-based aid program (Working paper). New York, NY: Center for Analysis of Postsecondary Education and Employment.

 

Scott-Clayton, J. (2011). The causal effect of federal work-study participation: Quasi-experimental evidence from West Virginia. Educational Evaluation and Policy Analysis, 33(4), 506–527.

 

Scott-Clayton, J. (2017). Federal Work-Study: Past its prime, or ripe for renewal? In Evidence Speaks Reports (Vol. 2, No. 16). Washington, DC: Brookings Institution. Retrieved from https://www.brookings.edu/wp-content/uploads/2017/06/ccf_20170622_scott-clayton_evidence_speaks.pdf

 

Scott-Clayton, J., & Minaya, V. (2016). Should student employment be subsidized? Conditional counterfactuals and the outcomes of work-study participation. Economics of Education Review, 52, 1–18.

 

Scott-Clayton, J. E., & Zhou, Y. (2017). Does the Federal Work-Study program really work—and for whom? CAPSEE Research Brief. Retrieved from Center for Analysis of Postsecondary Education and Employment website: https://capseecenter.org/does-the-federal-work-study-program-really-work-and-for-whom/

 

Singell, L. D., & Stone, J. A. (2007). For whom the Pell tolls: The response of university tuition to federal grants-in-aid. Economics of Education Review, 26(3), 285–295.

 

Soliz, A., & Long, B. T. (2016). Does working help or hurt college students? The effects of Federal Work-Study participation on student outcomes (Working paper). Retrieved from Center for Analysis of Postsecondary Education and Employment website: https://capseecenter.org/does-working-help-or-hurt-college-students/

 

Stern, D., Finkelstein, N., Urquiola, M., & Cagampang, H. (1997). What difference does it make if school and work are connected? Evidence on co-operative education in the United States. Economics of Education Review, 16(3), 213–229.

 

Stinebrickner, R., & Stinebrickner, T. R. (2003). Working during school and academic performance. Journal of Labor Economics, 21(2), 473–491.

 

St. John, E. P. (1990). Price response in persistence decisions: An analysis of the high school and beyond senior cohort. Research in Higher Education, 31(4), 387–403.

 

Titus, M. A. (2010). Understanding the relationship between working while in college and future salaries. In L. W. Perna (Ed.), Understanding the working college student: Implications for policy, administrators, academic affairs, and institutional support (pp. 251–281). Sterling, VA: Stylus.

 

Torres, V., Gross, J. P., & Dadashova, A. (2010). Traditional-age students becoming at-risk:

Does working threaten college students’ academic success? Journal of College Student Retention: Research, Theory & Practice, 12(1), 51–68.

 

Turner, L. J. (2014). The road to Pell is paved with good intentions: The economic incidence of federal student grant aid (Working paper). Retrieved from University of Maryland Department of Economics website: http://econweb.umd.edu/~turner/Turner_FedAidIncidence.pdf

 

Tuttle, T., McKinney, J., & Rago, M. (2005). College students working: The choice nexus. Indiana Project on Academic Success (IPAS) topic brief. Indiana University. Retrieved from https://www.academia.edu/10702156/The_Choice_Nexus

 

Umbach, P. D., Padgett, R. D., & Pascarella, E. T. (2010). The impact of working on undergraduate students’ interactions with faculty. In L. W. Perna (Ed.), Understanding the working college student: New research and its implications for policy and practice (pp. 234–260). Sterling, VA: Stylus.

 

U.S. Department of Education. (2017). Federal student aid handbook, 2017–2018. Washington, DC: U.S. Department of Education, Office of Federal Student Aid. Retrieved from https://ifap.ed.gov/fsahandbook/attachments/1718FSAHdbkVol6Ch2.pdf

 

Velez, E. D., Bentz, A., Arbeit, C. A., & Simone, S. (2018). Working before, during, and after beginning at a public 2-year institution: Labor market experiences of community college students. Stats in Brief. NCES 2018-428. Washington, DC: National Center for Education Statistics. Retrieved from https://nces.ed.gov/pubs2018/2018428.pdf

 

Waddell, G. R., & Singell, L. D. (2011). Do no-loan policies change the matriculation patterns of low-income students? Economics of Education Review, 30(2), 203–214.

 

Wisconsin HOPE Lab. (2016). What we’re learning: Work-study program: A data update from the Wisconsin HOPE Lab. Wisconsin HOPE Lab Data Brief 16-06. Retrieved from http://wihopelab.com/publications/

 

Wood, J. L., & Harris, F., III. (2015). The effect of academic engagement on sense of belonging: A hierarchical, multilevel analysis of Black men in community colleges. Spectrum: A Journal on Black Men, 4(1), 21–47.

 

Wood, J. L., Harrison, J. D., & Jones, T. K. (2016). Black males’ perceptions of the work–college balance: The impact of employment on academic success in the community college. The Journal of Men’s Studies, 24(3), 326–334.

 

Ziegert, J. C., & Hanges, P. J. (2005). Employment discrimination: The role of implicit attitudes, motivation, and a climate for racial bias. Journal of Applied Psychology, 90(3), 553–562.

 

 

 






Cite This Article as: Teachers College Record Volume 122 Number 1, 2020, p. 1-36
https://www.tcrecord.org ID Number: 23182, Date Accessed: 10/22/2021 10:20:41 PM

Purchase Reprint Rights for this article or review
 
Article Tools
Related Articles

Related Discussion
 
Post a Comment | Read All

About the Author
  • Hongwei Yu
    Texas State University
    E-mail Author
    HONGWEI YU serves as a graduate faculty member at Texas State University. His research interests include student persistence and academic achievement, academic misconduct, and organizational leadership. His research appears in peer-reviewed journals such as Ethics and Behavior and Review of Higher Education.
  • Lyle McKinney
    University of Houston
    E-mail Author
    LYLE MCKINNEY is an associate professor in the Higher Education Leadership & Policy Studies program at the University of Houston. His research focuses on improving degree completion rates among community college students. His recent publications include “Giving Up on a Course: An Analysis of Course Dropping Behaviors Among Community College Students,” published in Research in Higher Education, and “Performance-Based Funding for Community Colleges: Are Colleges Disadvantaged by Serving the Most Disadvantaged Students?,” published in The Journal of Higher Education.
  • Vincent Carales
    University of Houston
    E-mail Author
    VINCENT D. CARALES is an assistant professor in Higher Education Leadership & Policy Studies at the University of Houston. His research interests center on understanding the experiences and educational outcomes of first-generation, Latino, low-income, and community college students. He is also interested in examining federal, state, and institutional policies related to diversity, equity, and college affordability. His research appears in peer-reviewed journals such as Mentoring & Tutoring, Community College Journal of Research and Practice, and Community College Review.
 
Member Center
In Print
This Month's Issue

Submit
EMAIL

Twitter

RSS