Stern, D., Finkelstein, N., Stone, J. R., III, Latting, J., & Dornsife, C. (1994). Research on School-to-Work Programs in the United States (MDS-771). Berkeley: National Center for Research in Vocational Education, University of California.

Cooperative Education

The 1990 Perkins Amendments define cooperative education as

a method of instruction of vocational education for individuals who, through written cooperative arrangements between the school and employers, receive instruction, including required academic courses and related vocational instruction by alternation of study in school with a job in any occupational field. Such alternation shall be planned and supervised by the school and employers so that each contributes to the student's education and to his or her employability. Work periods and school attendance may be on alternative half days, full days, weeks, or other periods of time in fulfilling the cooperative program.

Cooperative education has been recognized by federal authority since the regulations implementing the 1917 Smith-Hughes Act. In the 1960s and 1970s, federal vocational education statutes provided specific categorical support for cooperative education, but current law no longer does. In fact, although the 1990 Perkins Amendments continue to define cooperative education, it is otherwise virtually silent on the subject, mentioning it in passing only twice (Sections 365(2)(B) and 404(b)(2)). In comparison, apprenticeship, which enrolls far fewer students, is mentioned at least six times. The venerable co-op method seems almost to be taken for granted.

The lack of attention to cooperative education in the latest federal law on vocational education is symptomatic. Until quite recently, when there has been some renewed interest, cooperative education has been virtually absent from policy discussions about preparing the workforce. During the 1980s, it was an idea with no cachet.

Lack of interest has also meant lack of evaluation, especially for cooperative education in high schools. The Higher Education Act has provided funds for research and evaluation of cooperative education in postsecondary institutions, but there has been no money earmarked for the secondary level. As a result, in recent years there have been only one or two significant inquiries into high school cooperative education.

Research Findings

Research before 1990 was reviewed by Stern, McMillion, Hopkins, and Stone (1990). For convenience, that review will be briefly recapitulated here. One of the most thorough studies was by Herrnstadt, Horowitz, and Sum (1979). They interviewed 427 male seniors in cooperative vocational education, regular vocational education, work-study, and the "general academic" program in a northeastern metropolitan high school. The sample was resurveyed three times, the last time between 17 and 21 months after graduation. The researchers described their findings as "mixed." Co-op students did not experience higher rates of labor force participation, employment, or wages. On the other hand, co-op students were more likely to value the jobs they held in high school, receive on-the-job training from their employers while in high school, claim that their high school programs favorably affected their decision to stay in school, attend classes during senior year, obtain a full-time job immediately following graduation, obtain a job related to their high school program, and express more satisfaction with their final jobs. In short, co-op students were more positive about school and its relationship to employment, but they were no more successful than other students in finding work or earning high wages.

A similar pattern of results emerged from a large, federally sponsored, longitudinal evaluation of several programs that combined school and work, including cooperative vocational education (Walsh & Breglio, 1976). Co-op students were more likely to say that their satisfaction with school had improved after taking their jobs and that these jobs fit with their career interests. However, in the postschool follow-up two years later, students who had participated in school-supervised employment, including co-op, were less likely to be employed and more likely to be looking for work. Among those who were working, however, the ones who had been in school-supervised jobs two years earlier usually did report higher earnings.

Analysis of the three major national longitudinal surveys has usually failed to find positive economic outcomes for high school co-op students. Lewis, Gardner, and Seitz (1983) looked at early (through 1980) follow-up data from the National Longitudinal Survey of Youth Labor Market Experience (NLSY), which started in 1979 with individuals ages 14 to 21. They found no higher earnings or lower unemployment during the first few years after high school for former students who had been in school-supervised jobs--which included co-op, work-study, or in-school programs sponsored by the Comprehensive Employment and Training Act of 1973 (CETA). Bishop, Blakemore, and Low (1985) analyzed October 1973 data from the National Longitudinal Study of the High School Class of 1972 (NLS72) and spring 1982 data from 1980 seniors in the High School and Beyond (HSB) survey. They found participation in co-op or work-study was positively related to postschool wages and fraction of time employed in the NLS72 data, but the opposite was true for the HSB sample.

To summarize, this review found evidence that participation in co-op was associated with more positive attitudes toward school and a stronger perceived connection between school and work, but no consistent association between participation in co-op and subsequent success in the labor market. A review of other studies by Leske and Persico (1984) reached the same conclusions. Taken at face value, these results seem to imply that creating a stronger connection between school and work for high school students does not improve their prospects in the labor market after high school. They raise a clear warning flag for proponents of current efforts to expand school-to-work programs.

One possible reason why co-op students apparently do not obtain any significant advantage in the labor market in the first few years after high school is that they do not receive any formal certification that is recognized by other employers. Any gain in knowledge, skill, or work habits resulting from the co-op experience may fail to pay off in the short run if former co-op students do not keep working for their co-op employer because other employers cannot readily recognize these gains. Therefore, if they are not given a permanent job by their co-op employer, the former co-op student may have no particular advantage over other former students in the labor market. This possible explanation can be tested by comparing the labor market experience of former students who do and do not obtain permanent jobs with their high school employers. In a test using Colorado data, Stern and Stevens (1992) found that former co-op students who kept working for their co-op employers did experience significantly higher earnings than other former students who also continued working for their high school employers, while among students who changed employers, former co-op students had no advantage.

It is also possible that existing studies have underestimated the effects of co-op participation on labor market outcomes. Methodologically, the studies summarized above shared certain strengths and weaknesses. Strong points were the use of longitudinal data and multivariate analysis that controlled for measured characteristics of students in trying to estimate the separate effect of participation in co-op. One weak point was that most of these studies lumped together co-op and other school-supervised work experience programs despite the fact that co-op has a relatively rigorous format, as described below. They also merged specialized co-op--in which the teacher in a particular field supervises only students working in that field--with diversified co-op--where students working in various fields are supervised by a coordinator who does not necessarily teach in any of those fields. Another weakness has been failure to correct for possible bias associated with unobserved variables that may affect the probability of participation in co-op.

In addition, an important limitation of these studies has been their failure to account for variation in the quality of students' jobs. The implicit assumption has been that differences among co-op jobs are negligible compared to the difference between co-op and non-co-op jobs. This seems unlikely. Even in well-run co-op programs, some placements offer more complexity, autonomy, and opportunity to learn than other placements do. Ignoring this variation may make it more difficult to tell whether being in co-op makes a difference. It also makes it more difficult to explain why co-op makes a difference, if it does.

A recent study by Stone, Stern, Hopkins, and McMillion (1990) was designed to measure qualitative characteristics of students' jobs. On average, co-op students are more likely than classmates employed in non-co-op jobs to report that their work provides opportunities to learn a variety of new things, that they use reading and writing on the job in addition to other things they have learned in school, that they have contact with adults on the job and good relationships with their supervisors, that the job is related to their desired career, and that the work is meaningful and motivating. Notwithstanding these average differences, there are still some non-co-op students who describe their jobs more favorably than co-op students do. It should therefore be possible to test whether or not co-op participation has an effect on students' subsequent performance separately from whether or not if affects the quality of students' work experience while they are in school.

Two other recent studies also should be mentioned. One is a short-term longitudinal study conducted by the New York State Department of Education (1990). Secondary students who had completed occupational education programs offered by school districts or BOCES (Boards of Cooperative Educational Services, providers of various specialized programs including occupational education) were surveyed approximately six months after leaving high school. Comparing co-op to non-co-op students, the survey found co-op students were more likely to report working as their primary activity (53% versus 40%) but less likely to report postsecondary education (36% versus 47%).

A similar pattern was discovered by the U.S. General Accounting Office (GAO) (1991), which reanalyzed 1982 data from the 1980 HSB seniors. Among co-op participants, 74% reported they were working for pay (not necessarily their primary activity), compared to 57% of all seniors in Fall 1981. But only 33% of former co-op students were in some kind of postsecondary education (25% in two- or four-year colleges and 8% in vocational or technical schools), compared to 53% of all seniors (46% in colleges and 7% in vocational or technical schools). This is consistent with the fact that a large proportion of co-op students identified themselves as being in the vocational track while they were still in high school. Compared to other seniors, co-op students also came from lower socioeconomic backgrounds and had lower scores on the HSB test of vocabulary, reading, and math.

These two studies reflect the fact that co-op programs in high school have been tied to vocational education and have not been seen as part of the college-prep curriculum, though they have kept the college option open for some students. However, lacking statistical controls for other characteristics of students, these two studies were not designed to measure the extent to which co-op adds to or detracts from students' subsequent success in school or work.

The GAO study does signal a rekindling of interest in co-op among policymakers. The study estimated how many high school students currently participate in co-op. As of 1989-1990, a survey of state directors of cooperative education yielded an estimate of "about 430,000" students nationwide. This was less than 4% of total high school enrollment, but it represents about 8% of the juniors and seniors, who comprise most of the co-op participants. These estimates correspond closely to findings from the 1992 NAVE survey of secondary schools, which estimated that co-op enrollment in 1990-1991 represented 3.7% of all students in grades 9-12, or 7.7% of juniors and seniors (Stern, 1992). The NAVE survey further indicates that another 1.8% of students in grades 9-12, or 3.7% of juniors and seniors, were participating in other school-based work experience programs, which often possess the characteristics of co-op (see Table 2 below). It would therefore be accurate to say that approximately 5% of students in grades 9-12, or 10% of seniors, were in programs that were either called co-op or had all the main features of co-op.

The GAO study also included a review of practices in eleven high school sites (and eight community colleges) chosen from a list of programs recommended by co-op researchers and practitioners. Based on these site visits, GAO concluded that well-run co-op programs apparently help students both in finding permanent jobs and in preparing for further education. Employers apparently benefit from the extra motivation of co-op students and also from the chance to "try out" possible permanent employees. Schools reportedly gain from increased student interest and connections with employers.

Certain problems also were mentioned in the U.S. GAO (1991) study. One is the "negative reputation that many people have of co-op programs and vocational education in general" (p. 34). This contributes to reluctance of nonvocational teachers to recommend students for co-op. It also repels parents who fear that co-op will detract from their children's chances of attending college. The challenge here is to make co-op part of the integration of vocational and academic education that is envisioned by Perkins legislation.

GAO also emphasized the lack of uniform certification for co-op completers. Co-op is like a short-term apprenticeship, but unlike traditional apprenticeship, however, it does not result in any kind of skill certification that is recognized by employers at large. Current national efforts to create skill standards for specific occupations and industries should help provide a target for which co-op programs can aim. Common skill standards might facilitate employment of co-op graduates who do not find permanent jobs with their co-op employers. As suggested above, this might resolve in practice the puzzle revealed by research: co-op succeeds in creating a strong connection between school and work, but former co-op students generally do not obtain any significant advantage in employment or earnings in the first few years after high school.

Elements of Good Practice

The GAO described what it considered to be the elements of a "high quality" co-op program, based on observations and interviews at the selected program sites. This wisdom of practice has emerged from decades of experience and has also been articulated elsewhere (Leske & Persico, 1984; National Child Labor Committee, 1984). We offer a brief distillation here:

• A written training agreement between the school and each employer sets forth the expectations for each party. The employer will provide a job with opportunities to learn. The school will monitor students' performance.
• A written training plan for each student is at the heart of the co-op program. It specifies what the student is expected to learn on the job. Learning objectives may be linked to vocational or academic courses. The plan also specifies who will judge whether the student has achieved the stated objectives. It is usually signed by the student, the job supervisor, the co-op coordinator, and sometimes by a parent.
• The co-op coordinator may be the teacher in a related class (e.g., business or marketing) with responsibility for supervising students only in that field. Alternatively, in a diversified co-op program, the coordinator supervises students from several fields. The coordinator also may have special training and certification as a co-op specialist. The coordinator's responsibilities include
  • Finding suitable job placements. Some programs provide summer salary for coordinators to come back early for this purpose. In well-established programs, employers initiate placements by calling the school and requesting a co-op student. It is considered desirable for the coordinator to visit the worksite before placing a student there to make sure the employer understands the responsibility entailed in hiring a co-op student. The training agreement embodies this understanding.
  • Identifying suitable students for each placement. It is considered advisable to refer more than one student for the employer to interview. This gives students practice in job interviews and gives the employer some choice.
  • Negotiating training plans for all students.
  • Monitoring students on the job. Coordinators must have released time during the day for this purpose. By visiting the job site, the coordinator can ensure that both the student and the job supervisor are satisfied with the way things are going. If there are problems, the coordinator can try to mediate or, if that fails, can remove the student from the job. At the end of the student's placement, the coordinator is responsible, usually in collaboration with the job supervisor, for evaluating students' performance according to their training plans.
  • Offering related instruction. If the co-op program is offered in conjunction with a regular class, the coordinator would normally be the teacher in that class, which would naturally deal with issues from students' jobs. In a diversified program, there may be a special class for co-op students, which would deal with more generic issues about work.

Data on the actual prevalence of some of these characteristics are available from the National Assessment of Vocational Education (NAVE) survey of high schools and secondary vocational centers (Stern, 1992). Table 2 shows the weighted percentages of schools indicating that their co-op programs had the specified features. The large majority of responses are affirmative, except that in only about half the co-op programs do the coordinators find jobs for students and have some paid time during the summer for that purpose. These exceptions point to the fact that paying school staff to find jobs for students is expensive--but this is an expense that will have to be incurred by any school-to-work program that does not rely on students to find their own jobs.

Also shown in Table 2 are the percentages of schools indicating that this same list of features are present in two types of work experience program that are similar to co-op. One is in-field work experience (IFWE), defined in the survey as a program that does not qualify as cooperative education but allows students to earn school credit in conjunction with paid or unpaid employment in their vocational field of study. The second is other work experience (OWE), which allows students to earn school credit in conjunction with paid or unpaid employment outside their vocational field of study. Schools indicate that both of these programs usually share the defining features of co-op, though to a somewhat lesser extent. The biggest differences are in whether students are required to complete a separate course of related instruction and whether coordinators supervise only in their own field of specialization--co-op programs more often possess both of these features. However, the fact that many of these other school-supervised work experience programs do share the features of co-op means that, in effect, there are students who participate in co-op even though they use another name for it.

Stern, D., Finkelstein, N., Stone, J. R., III, Latting, J., & Dornsife, C. (1994). Research on School-to-Work Programs in the United States (MDS-771). Berkeley: National Center for Research in Vocational Education, University of California.