NON-COMPLETERS ARE NON-ENTITIES

Reduced attrition has long been identified as a major benefit of coop­erative education (Smith, 1965; Porter, 1982). The linking of theoretical learning with practical experience contributes to higher academic degree completion through increased levels of motivation (Wilson and Lyons, 1961; Gore, 1972; Tyler 1981), career clarification (Wilson, 1971; Martello and Shelton, 1980), and realism of instruction (Wilson and Lyons, 1961; Wilson, 1978). Remuneration also makes financing college easier (Wilson and Lyons, 1961). Co-op research has focused on those retained-students currently enrolled in institutions of higher education or those having graduated. The literature contains nothing regarding benefits that may accrue to those leaving college prematurely. Research focuses on program successes. Non-completers are viewed as failures and quickly discounted.

This article seeks to rectify this. It is the third in a series of papers link­ing human capital ideas to the measurement of monetary (investment) benefits from co-op participation. The first article argued for the use of eco­nomics and regression analysis to isolate the impact of cooperative educa­tion on such outcomes as wage rate (Siedenberg, 1989). A theoretical model associating wage rate with personal characteristics, home background, human capital accumulation (including co-op participation), cognitive and affective traits, and existing labor market conditions was posited. The sec­ond paper employed those ideas and presented empirical results (Siedenberg, 1990). Cooperative education participation proved significant in raising the wage potential of alumni from a nontraditional four-year col­lege. This article has two major objectives: (1) to identify a large group of individuals - non-completers - who may benefit from their involvement in cooperative education and (2) to apply, economics and education concepts to evaluate quantitative outcomes for these non-completers.

The Problem

The extent of collegiate attrition is documented in the U.S. Department of Education's Digest of Education Statistics 1992. Findings suggest that in Spring 1987 (the latest data available), 23.3 percent of the total population eighteen and over had earned an associate degree or higher, compared to 17.6 percent that report some college, with no degree or certificates earned. In 1980 49.3 percent of high school graduates enrolled in college, but six years later only 25.4 percent of them had earned a degree. In 1991 62.4 percent of high school graduates enrolled in college. In absolute terms, this is more than two million first year students. Attrition rates mirroring the prior decade will yield even more individu­als having spent time in college with no degree as the outcome.

Assuming that most of these non-completers enter the work force or increase their labor force participation, a significant question arises: Does co­op participation have the same positive impact on wage rates for non-com­pleters as has been documented for alumni, or is there a type of "diploma effect," where positive returns can only be realized for completing a degree? In pursuing an answer to this question, a sample of non-completers was analyzed analogously to that employed on alumni (Siedenberg, 1990). This approach uses economics and regression analysis to isolate the effect of co­op participation on wage rates and controls for personal, human capital and labor market variables that affect wage rates and differ among individuals. The focus is on a single explanatory variable, co-op participation.

Site, Sample, and Survey

The subjects for the study were drawn from a sample of 1980-1985 non-completers of Lehman College of the City University of New York. As an independent, open admission, public, liberal arts college within CUNY, Lehman, like many colleges that serve a nontraditional student body, has a high attrition rate. During the last decade only 42-43 percent of students returned five semesters after being admitted. During the same period, only 4-9 percent graduated in eight consecutive semesters, and 27-32 per­cent in ten consecutive semesters (Phelps, 1986). Co-op was instituted at Lehman in 1980. The program is selective, using a parallel mode of work placements surrounded by three courses.

All non-completers who had participated to any extent in the co-op program were selected for analysis. The non-co-op non-completer cohort was selected randomly from all Lehman College non-completers between 1980 and 1985. Non-completers were operationally defined as former stu­dents that left on their own accord before graduation. Individuals that had academic or disciplinary problems that led to their dismissal were not included. Sample numbers and response rates are found in Table 1.

In examining survey responses concerning initial expectations regard­ing college, reasons for leaving Lehman, and future plans with respect to higher education, certain generalizations among non-completers emerged. In terms of expectations, only two-thirds of these students initially planned to stay at Lehman through graduation. Over 40 percent said that financial matters had been important in their decision to leave. Other reasons given included changes in personal circumstances, inconvenience of the campus and course hours, and majors not available at the College. Low expectations coupled with financial, personal, location, and program problems accounted for the attrition. A majority ( 67.2%) had not transferred to another school and this underscored the hypothesis that leaving Lehman was directly correlated to increasing labor force participation.

Comparison of the Groups

Inferential statistics were used to contrast the two groups (co-op vs. non-co-op) of non-completers. Chi-square and t-test analyses were used to compare the subsamples on ninety-seven personal, home background, human capital, college experience, work history, and labor market variables.

Several differences between the groups proved statistically significant. Co-op non-completers were younger at the time they left Lehman-by approximately six and one-half years on average (23.7 vs. 30.1 years old). The co-op group also had nearly four years less of full-time previous work experience (1.4 vs. 5.1 years). The co-op group had more single individuals (85.2% vs. 55.0%), many of whom lived with their parents. Initially, while in college, the co-op group offered potential employers younger, less experi­enced individuals, who probably possessed lower levels of confidence. This may account for the significant advantage in hourly wage rate held by the non-co-op student group - $2.50 per hour vs. $1.76 per hour in real terms (1967 is used as the base year because it is employed in most government publications for controlling changes in the purchasing power of wages earned in different years - controlling for inflation).

Upon leaving Lehman, the co-op non-completer was more likely to quit his or her college job. Only 20 percent of co-ops continued their col­lege jobs compared to 57 percent of non-co-op non-completers. Having spent more time and money on job-search ($272 and nine weeks for co-ops vs. $33 and four weeks for non-co-ops), the co-op student ended with a position that had a pay rate lower, but not significantly less, than that of the non-co-op leaver ($2.04 vs. $2.59). Statistically there was no significant difference between the wage rates of the groups. Something had inter­vened to cause the co-op non-completer to be more adventurous in the labor market and emerge with a "relative wage gain" from being statisti­cally behind in wages in college to statistically equal in wages at gradua­tion. To investigate the significance of co-op participation in causing this "increase," both the determinants of wage rate earned while at Lehman and after leaving the College were examined.

Regression Results

Several forms of a model used to explain real wage-rates (controlling for inflation) were studied.
Basic Model: W = f(H,P,L)

The model, assumes that wage rate (W) is dependent on three groups of independent variables: (P) personal characteristics and home background (e.g., sex), (H) human capital investment (e.g., work experience) and (L) labor market conditions (e.g., current unemployment rate). The indepen­dent variable CO-OP, representing participation or non-participation in the co-op program, was included (H) and of primary concern. Multiple regres­sion analysis would show co-op's unique contribution in explaining wage ­rates when other deterministic variables in the equation were held constant. The results from analysis of one such estimating equation, used to explain wage rate earned while attending Lehman, can be found in Table 2.

Years of previous work experience, hours of college job, and living away from parents were statistically significant in explaining college wage rates. However, the crucial finding was that after controlling for other prominent explanatory variables, a wage disadvantage was held by co-op students, of approximately $0.45 per hour in real terms. This result was never statistically significant at the .05 level (the significance level most often employed in social science research for testing hypotheses comparing subsamples), but consistently significant at the .10 level (the extended significance level some­times employed in social science research to suggest association).

Various forms of the general model were then run for wage rate received by students immediately after leaving Lehman. The variable co-op never proved to be significantly different from zero. See Table 3. Shortcomings in certain desired employee traits (e.g., maturation and confidence) that attracted younger, less experienced students to co-op and instrumental in causing lower college wages had been erased by the time the students exited the college prematurely. Initial deficiencies may have caused many individuals to opt for co-op participation. They were rewarded.

Conclusions

At Lehman, the significance of the relationship between co-op partici­pation and wage-rates for non-completers is not as unequivocal as was found for alumni (Siedenberg, 1990). Among graduates, co-op students made up an initial disadvantage (partially caused by traits reflected in the proxy variable co-op) by the time they got their first job after graduation (Siedenberg, 1990). With regard to non-completers, such an ambitious state­ment cannot be made. At an expanded level of significance of ten-percent, there is an indication that traits captured by the co-op variable (e.g., human relations skills) that proved disadvantageous to wage rates while in college, were compensated for during college by participation in co-op. These char­acteristics no longer proved to be negative factors on wages after leaving Lehman. Thus, co-op appears to enhance wage potential for non-completers.

were compensated for during college by participation in co-op. These char­acteristics no longer proved to be negative factors on wages after leaving Lehman. Thus, co-op appears to enhance wage potential for non-completers. Despite educator efforts to reduce attrition and increase graduation rates, non-completion among college students persists. There is some sat­isfaction in knowing that the provision of cooperative education helps these non-completers in the workplace.