Students who work full-time are also more likely to drop out of school. For example, the available evidence is consistent with a roughly 10 percentage point differential in graduation rates between full-time and part-time workers. In 2000, nearly 830,000 full-time college students worked full-time. Because of the adverse effects of such full-time work, tens of thousands of these college students are likely to drop out of school and fail to receive a college degree.
Working a limited number of hours (e.g. 10 hours a week) at an on-campus job appears to have positive impacts on student performance, while working a significant number of hours (e.g. 35 hours or more per week) has adverse consequences. It is unclear at what point student employment moves from being beneficial to being counterproductive. But the difference between graduating from college and not graduating from college may involve a change in work schedules that would have a modest impact on student earnings relative to the lifetime gains from completing college. For example, reducing hours worked by 10 hours (from 35 hours per week to 25 hours per week) would reduce a student's annual earnings during the school year by roughly $2,250. Such potential earnings pale in comparison to the lifetime gains from completing college.
Since full-time work appears to have negative effects on student enrollment rates and perhaps also on academic performance, it is therefore of particular concern that full-time work among full-time college students has risen sharply over the past 15 years. For these students, the research suggests that, if possible, it may be prudent to find other ways of financing college so they can complete their degrees, maintain their academic performance levels, and thereby reap the long-term benefits of a college education.
LEARNING AND EARNING: WORKING IN COLLEGE
Jonathan M. Orszag, Peter R. Orszag, and Diane M. Whitmore1
The financial burden of college tuition is significant and rising. In light of the increasing price of college, many families are facing significant challenges in financing their children's education.2 The evidence shows that as one response to the financial burden of college tuition, students are working more while in college. Over 60 percent of college students report that their parents now expect them to work during the school year to help cover expenses.3 More students are working, and more are working longer hours.
Today, more than half of college students have a job. In 1999, on average, working students earned roughly $7.50 per hour. These earnings undoubtedly help to alleviate some of families' financial struggle to pay for college in the short run. But the extent of working while in college raises important questions. In particular, what is the overall effect of work? Does it have a beneficial effect in the long run by building discipline and a strong work ethic in students, or does it have a deleterious effect by diverting students' efforts from schoolwork? The evidence suggests that the answers depend on how much a student works and what type of job she has.
Characteristics of Student Employment
An increasing number of students work while in college. Since 1984, the fraction of college students aged 16 to 24 who also work full- or part-time has increased from 49 to 57 percent.4 The share of full-time college students working has also increased; in October 2000, a majority of full-time college students was employed. Figure 1 displays the increase in employment rate among college students.
Not only are students more likely to work today, but the share of students working full-time while going to school full-time also has increased dramatically. The percentage of full-time college students working full-time has nearly doubled, rising from 5.6 percent in 1985 to 10.4 percent in 2000 (see Table 1).5 The data indicate that the increase in the employment rate is largely due to increases in work among full-time college students; the share of part-time college students working has remained relatively constant over the past 15 years.
The Department of Education undertook a special survey during the 1995-96 school year to provide an in-depth picture of how undergraduates finance college through work.6 Working students can be categorized into two groups: those who primarily identify themselves as students but who work in order to pay the bills, and those who are first and foremost workers who also take some college classes. According to the survey, almost two-thirds of undergraduates who work considered themselves "students who work," while about a third considered themselves "workers who study."7
The average number of hours worked by students during the school year varies by the type of institution and by whether the student is part-time or full-time. As Table 2 shows, in the 1995-96 school year, employed students worked an average of 25 hours per week. About one-quarter of students work 35 or more hours per week, and another quarter of students work 15 hours or less. Students at four-year colleges are more likely to work a smaller number of hours per week; over one-third of such students work fewer than 15 hours.
On average, working college students earn roughly $7.50 per hour.8 Although hourly earnings vary somewhat for college students, three-quarters earned less than $8.00 per hour in 1999. More than one-third of students earn within $1.00 of the minimum wage. Table 3 also shows that part-time college students earn more per hour (on average) than full-time college students.
The majority of students who work also receive financial aid to attend college. As Table 4 shows, 56 percent of workers receive some other type of aid in the form of grants or loans. Nearly 70 percent of students who work 15 hours or fewer also receive another form of financial aid. Students who work more hours per week are less likely to be recipients of financial aid.
Workers on financial aid received an average award of $5,988 in 1995-96. Students receiving financial aid reported that they earned $5,197 - including work-study and other employment, but excluding summer employment - while enrolled that year.9
There is some evidence that parents count on students' earnings to finance college. Students were asked whether their parents expected them to work, and if so how many hours per week.10 Over 60 percent of undergraduate students reported that their parents expected them to work. On average, these parents expected students to work more than 20 hours per week. Generally, as students worked more hours, they were more likely to report that their parents expected them to work (see Table 5).
Evidence on the Effects of Working in College
The evidence suggests that the effects of working while in college varies by the type of job held: full-time versus part-time, and on-campus versus off-campus. Negative effects typically arise because hours spent at work take time away from studying - which may lead to lower grades and less attractive post-college opportunities. Working may contribute to students dropping out of college, or taking a longer time to graduate. But student employment can also be a positive experience. For example, some workers may gain experience from their job that helps them in the classroom, or in the labor market after college. On-campus jobs may also build connections to academic departments or the community at large, which in turn may make students more likely to stay in school.
According to the Department of Education, 85 percent of student workers are employed off-campus. Indeed, only 15 percent of student workers have an on-campus job. Students working exclusively on campus tend to work fewer hours: on average, they work 16 hours per week, compared to 27 hours per week for those who work exclusively off campus.11
Impact on drop-out rates
The empirical evidence suggests that full-time employment may cause students to drop out of school, but that part-time, on-campus employment may have beneficial effects in encouraging students to remain in school.
In an extensive study of what causes students to drop out of college, Tinto (1987) summarizes the conventional wisdom on in-college employment.12 Although employment generally harms persistence rates, the effect depends on hours of work and the degree to which employment removes the student from the campus community. For example, Tinto finds that full-time employment is "clearly more harmful" than part-time, and off-campus work is more detrimental than on-campus work. But he also finds that some employment can be beneficial. On-campus, part-time work was found to improve a student's probability of graduating - possibly because such a job helps a student to integrate socially and intellectually into the campus community. Tinto concludes that such positive integrative experiences, especially those that bring students and faculty into contact outside of the classroom, can play a crucial role in retaining students.13
Ehrenberg and Sherman (1987) examine the connections between work and drop-out rates using statistical techniques to control for other attributes of students.14 Using longitudinal data from the 1970s, they find that working 20 hours per week increases the likelihood that a student drops out after his first year by 3.2 percentage points for four-year college students and 6.6 percentage points for two-year college students. After four years, college students who had worked 20 hours per week at an off-campus job were 8.7 percentage points less likely to have graduated than non-workers. However, Ehrenberg and Sherman also find evidence that on-campus employment has positive effects, possibly because such jobs might be more career-related or spark interest in further academic programs.
Another study uses data from the 1980s and finds that working is related to dropping out.15 In any given semester in a student's college career, 7.3 percent of students who have worked in every previous semester drop out, compared to 5.7 percent
of students who have never worked. Cumulatively, this means that a never-employed student has a 37.5 percent chance of dropping out before graduating, and an always-employed student has a 45.5 percent probability of dropping out.
Survey evidence lends further support to the notion that working may be an important cause of dropping out. When college dropouts were asked to cite their main reasons for dropping out, 20.8 percent reported part-time employment as a reason, and 11.5 percent reported full-time employment.16 The study also finds that students who work continuously take somewhat longer to graduate than non-workers: Student workers took an average of 9.2 semesters to graduate, compared to 8.9 semesters for those who have at least one non-working semester.
A recent study uses data from college attendance in the 1990s to study differential college persistence rates between African American and white students.17 The authors found that an extra $1,000 in income from an on-campus work-study job increased whites' persistence by four percentage points, while blacks' persistence increased by a smaller and statistically insignificant amount. Full-time work, however, decreased blacks' persistence by almost six percentage points, and whites' persistence by one percentage point. In other words, the beneficial effects of part-time on-campus work appear to be smaller, and the adverse effects of full-time work seem to be larger, for African American students than white students. African American students are also more likely to stay in school if financial aid is increased - specifically if the grant portion is increased or if a loan forgiveness program is established. In this paper and related work, the authors find that large levels of debt are particularly harmful to dropout rates among African Americans, the poor, and two-year college students.18
Another recent study in the state of Washington examined undergraduates' off-campus employment (excluding summer work).19 The study found that 69 percent of enrolled resident college students worked off campus during the 1996-97 school year, and 17 percent of them worked full-time.20 Overall, 86 percent of non-workers remained enrolled, while 85 percent of those working less than 20 hours per week persisted. Those who worked more hours were less likely to persist - only 79 percent of those working more than 20 hours per week persisted.21
The research thus suggests that part-time employment and on-campus employment does not significantly reduce college enrollment, and may even be beneficial, but that longer hours and off-campus employment are associated with higher drop-out rates. As Astin (1993) states, "the largest negative effect on retention [staying in school] is working full time as a student."22 Because of deleterious effects of full-time work on enrollment rates, tens of thousands of full-time college students will fail to graduate. For example, the evidence cited above is consistent with a roughly 10 percentage point differential in graduation rates between full-time and part-time workers. In 2000, 828,000 full-time college students worked full-time.
Impact on learning
In addition to its effects on drop-out rates, working while in school may affect academic performance. When students are asked what effect working has on their studies, the most common response is "no effect." Not surprisingly, students' reports of the impact on their academic performance are highly correlated with the average number of hours they work in a week (see Table 6).23 Over 80 percent of those working few hours report no negative effect on their schoolwork. By contrast, 55 percent of those working 35 or more hours per week report a negative impact.
Students were also asked what types of academic limitations work imposed. Overall, working students reported the following liabilities: 40 percent report that work limits their class schedule; 36 percent report it reduces their class choices; 30 percent report it limits the number of classes they take; and 26 percent report it limits access to the library.24
On the other hand, it is possible that working forces students to be more effective managers of their time. A survey of time-use of undergraduates at five universities found only one difference between non-working students and part-time workers: 51 percent of non-working students reported watching television at least three hours a day, compared to 34 percent of those who worked part-time. There were no other major differences between the groups in activities such as volunteering, cultural activities, and reading for pleasure.25
One study found no significant overall effect of working on GPAs. It also found that those who work 10 hours or less per week had slightly higher GPAs than other students, even after controlling for other attributes of the student, while those who work 30 or more hours per week had slightly lower GPAs. The study concludes that "a student who begins a 30 hour a week job can expect his or her GPA to fall by a mere 0.05 points."26 Table 7 shows the average GPA by work status (without correcting for other attributes of the student).
Work is becoming increasingly common among college students. Research suggests that such work may not be harmful, and may even be helpful for undergraduates if it is limited. For example, part-time on-campus work appears to have no negative effects on students' enrollment rates or GPAs, and it may even have a positive effect. Students similarly generally perceive that limited work does not have a negative effect on their academics.
Full-time work, on the other hand, does appear to have negative effects on student enrollment rates and academic performance. It is therefore of particular concern that full-time work among full-time college students has risen sharply over the past 15 years. For these students, the research suggests that, if possible, it may be prudent to find other ways of financing college so they can complete their degrees, maintain their academic performance levels, and thereby reap the long-term benefits of a college education.
1 Jonathan M. Orszag (firstname.lastname@example.org) is Managing Director of Sebago Associates. He previously served as Director of the Office of Policy and Strategic Planning at the Department of Commerce, and was an Economic Policy Advisor on the National Economic Council at the White House. Peter R. Orszag (email@example.com) is President of Sebago Associates. He is also a Senior Fellow in Economic Studies at the Brookings Institution. He has previously served as Special Assistant to the President for Economic Policy, as Senior Economist on the Council of Economic Advisers, and as a member of the economics faculty at the University of California, Berkeley. Diane M. Whitmore (firstname.lastname@example.org) is Economic Counselor at Sebago Associates, Inc. and a Ph.D. candidate in economics at Princeton University. She previously served on the staff of the Council of Economic Advisers.
2 Joseph E. Stiglitz, Laura D. Tyson, Peter R. Orszag, and Jonathan M. Orszag, "The Impact of Paying for College on Family Finances," Sebago Associates, Inc. November 2000, available at http://www.sbgo.com.
3 U.S. Department of Education, National Center for Education Statistics (NCES), "Profile of Undergraduates in U.S. Postsecondary Educational Institutions: 1995-96." NCES 98-084.
4 Data are from the October Current Population Survey (CPS).
5 In 2000, 828,000 full-time college students worked full-time; in 1985, 366,000 such students worked full-time.
6 U.S. Department of Education, National Center for Education Statistics (NCES), "Profile of Undergraduates in U.S. Postsecondary Educational Institutions: 1995-96." NCES 98-084.
7 The following section limits analysis to those who primarily identify themselves as students.
8 These figures exclude tips and overtime. The amount of earnings from tips is not available, but 17.8 percent of employed students report receiving tips, overtime pay, or commission.
9 Authors' tabulation of the 1996 NPSAS.
10 The results are limited to undergraduates who are still financially dependent on their parents.
11 Authors' tabulation of the NPSAS.
12 See especially Chapter 3 in Vincent Tinto's Leaving College: Rethinking the Causes and Cures of Student Attrition (Chicago: University of Chicago Press, 1987).
13 Although Tinto downplays the effect of finances directly on persistence, he thinks that finances play a larger role in a student's decision whether to attend college in the first place.
14 Ronald Ehrenberg and Daniel Sherman, "Employment While in College, Academic Achievement and Postcollege Outcomes: A Summary of Results," Journal of Human Resources, 22(1), Winter 1987, pp 1-21.
15 Philip M. Gleason, "College Student Employment, Academic Progress, and Postcollege Labor Market Success," Journal of Student Financial Aid, 23(2), Spring 1993, pp 5-14.
16 Respondents could report multiple reasons for dropping out. Other reasons included: indecision about career (38.9%), school workload too great (28.6%), failure (26.6%), cost too high (26.5%), program not relevant (19.1%), and family reasons (11.3%).
17 Patricia Somers, Margaret Martin Hall, Jim Cofer, and Jim Vander Putten, "The Persistence of African American College Students: How National Data Inform a Hopwood-proof Retention Strategy," University of Arkansas at Little Rock, Unpublished Working Paper, 2000.
18 See also Patricia Somers and James Cofer, "Mortgaging Their Future: Student Debtload in the U.S." Testimony before the Senate Governmental Affairs Committee, February 10, 2000.
19 Edie Harding and Laura Harmon, "Higher Education Students' Off-Campus Work Patterns," Washington State Institute for Public Policy, January 1999.
2 Forty-nine percent of those in four-year colleges worked off campus. Full-time numbers were not reported.
21 Only aggregate hours were available for the 13-week term, so the variables were defined as more or less than 260 hours per term.
22 Alexander Astin, What Matters in College? (San Francisco: Jossey-Bass, 1993).
23 Table is limited to dependent undergraduates who work.
2 4Students could report multiple responses.
25 Study by Scott Schnackenberg, reported in Anne-Marie McCartan, "Students Who Work: Are They Paying Too High a Price?" Change, 20(5), Sept-Oct 1988, pp. 11-20.
26 Philip M. Gleason, "College Student Employment, Academic Progress, and Postcollege Labor Market Success," Journal of Student Financial Aid, 23(2), Spring 1993, page 8.