Читать книгу Economic Evaluation in Education - Henry M. Levin - Страница 19

Source: Adapted from Stern, Dayton, Paik, and Weisberg (1989, Table 6).

Notes: Adjusted to 2015 dollars. Rounded to nearest ten.

The first step of the evaluation is to estimate the additional costs of each academy, beyond what would have been spent on a traditional high school education. The second column gives the total costs for the 3-year (Grades 10 through 12) program delivered to the 1985–1986 cohort. The cost ingredients included personnel (teachers, aides, and administrators), facilities and equipment, and the cost of time donated by local employers. Academy costs tended to be higher because of their relatively smaller class sizes and extra preparation periods that were given to some teachers.

The evaluators then estimated the benefits produced by lowering the number of high school dropouts. To do so, the authors employed a quasi-experimental design. Prior to initiating the program, a comparison group of observationally equivalent students attending the traditional high school was selected. After 3 years, the academy dropout rates were compared with those of the comparison group. In most schools, the academies reduced dropouts (indicated by a positive number of “dropouts averted” in the table), although in a few schools, the dropout rate was higher in academy schools (indicated by a negative number). To monetize these changes—that is, to transform them into benefits—the authors calculated the lifetime income gain for a high school graduate over a dropout. Their estimate of $172,000 is a present value, discounted to reflect for the differential timing of the benefits received (a complete discussion of discounting is given in Chapters 3–6). The number of dropouts averted is multiplied by this value to derive each academy’s total benefit.

The final step is to subtract costs from benefits. The net benefits column suggests that the program is worthwhile in academies C, D, E, F, and K (i.e., the benefits are greater than the costs). In academies A, G, and H, however, the costs outweigh the benefits. Across all eight academies, the overall benefits of the program exceed the costs. Another metric for comparing benefits and costs is the benefit-cost (BC) ratio. Ratios greater than 1 suggest that benefits are greater than costs. Despite the favorable results, the final results are heavily influenced by a single academy (C).

This analysis assumes we have exhaustively catalogued the relevant costs and benefits. However, we may have excluded important benefits—for example, the savings incurred because more-educated adults are less likely to become incarcerated. The authors also present evidence that academy schools are effective at producing other outcomes such as higher grades. Some measures of effectiveness may be difficult to monetize and so to include in BC analysis (although they might be usefully included in a cost-effectiveness [CE] analysis).

Source: Adapted from Stern et al. (1989).

BC analysis can be a useful way to gauge the overall worth of a program or policy. If the program costs are greater than its benefits, it should not be implemented. Also, we can judge a project by the overall size of the net benefits—that is, by how much benefits exceed costs. Further, to the degree that other educational endeavors and those in other areas of public expenditure (such as health, transportation, environmental improvement, or criminal justice) are evaluated by the BC method, it is possible to compare any particular educational alternative with projects in other areas that compete for resources.

The disadvantage of this method is that benefits and costs must be assessed in pecuniary terms. It is not often possible to do this in a systematic and rigorous manner. For example, while the gains in earnings attributed to increased graduation rates might be assessed according to their pecuniary worth, how does one assess benefits such as improvement in citizen functioning of the educated adults or their enhanced appreciation of reading materials? This shortcoming suggests that only under certain circumstances would one wish to use BC analysis. Those situations would occur when the preponderance of benefits could be readily converted into pecuniary values or when those that cannot be converted tend to be unimportant or can be shown to be similar among the alternatives that are being considered. That is, if the decision alternatives differ only on the basis of those benefit factors that can be converted to pecuniary values, the other aspects can be ignored in the BC calculations. Or, if those dimensions of benefits that cannot be assessed in pecuniary terms are considered to be trivial, one can limit the BC comparison to the factors that can be evaluated with monetary measures. However, in those cases in which the major benefits are difficult to assess in pecuniary terms, some other mechanism for assessment must be found.