 | |
The Roles of Evaluation for Vocational Education and Training: Palin Talk on the Field of Dreams
Grubb, W.N. & Ryan, P.
ILO, 1999
Individually sponsored training
We move now to private sponsorship of training, starting with the training that individuals sponsor for themselves, ie VET conducted largely or wholly apart from employment. We include in self-sponsored training all vocational courses taken in post-secondary education, such as secretarial vocational courses at American community colleges, technical institutes and proprietary schools, Brevet de technicien courses at French colleges, teacher training courses in British universities, and upgrading courses for Meister qualifications in German Fachschulen.
Two features require comment: the scale of activity and the age of participants. The importance of self-sponsored training varies greatly across countries. In the United States it accounts for nearly three-quarters of the private sector training experiences reported by young adults, dominating training for the 'mid-skilled' labour market in particular1. In the United Kingdom, by contrast, only one-fifth of privately sponsored training episodes are sponsored by individuals; the rest are sponsored by employers. Second, sponsorship of one's own training is a predominantly youth affair in many countries, its incidence declining sharply with age2 - though in the United States there has been a steady increase in numbers of older students, particularly in community colleges.
The evaluation of self-sponsored training has relied heavily on quasi-experimental methods and single outcome criteria. The standard human capital earnings function applies more directly to self-sponsorship than to other categories of training. Individual pay and employment rates are related statistically to the various types of post-secondary VET, using regression analysis to remove the effects of other individual attributes such as race and sex. Although the selection biases potentially associated with unobserved attributes - notably ability and motivation - have occasionally been addressed3, the effectiveness of such 'controls' remains dubious and, in any case, most studies simply ignore selection issues.
American research suggests private returns to two-year Associate degrees of the order of 20 to 30 per cent- somewhat lower than to a four-year first degree - and returns to one-year certificates of 10 to 20 per cent. The benefits are higher in particular areas, notably business and technical fields for men and the health field for women. When students complete occupational subjects in community colleges, the private returns tend to be higher than for the equivalent general subjects if they find employment related to their area of training, and lower (even zero, particularly for women) if their employment is unrelated. A smattering of evidence indicates that the returns to the highly vocational courses offered by proprietary schools and technical institutes are lower than those in comprehensive community colleges. Finally, people who take but fail to complete post-secondary programmes, taking only a few courses, receive small net returns, generally less than 10 per cent4.
Such findings are probably affected by selection bias. For example, lower returns to vocational than to general subjects in community colleges may reflect lower student ability in the vocational category.
One study found that correction for selection bias removed the earnings effect of vocational Associate degrees for American men, though not for women5. Another found that the marked effects of self-sponsored training on the pay of young American adults survived both the two step and the fixed effects corrections for selection bias6.
The parallel French literature on the links between qualifications and employment suggests similarly that the possession of vocational post-secondary qualifications, notably the Brevet de technicien and the Diplome d'ecole d'ingenieur, confers substantial private benefits, in terms of access to employment, relative to the less vocationally qualified at the same level of attainment. During 1970-93 such training reduced the incidence of unemployment, and of employment on fixed-term contracts, during the first year of labour market experience, as well as increasing life-cycle earnings powerfully. The findings survived attempts to, for unobserved selection processes7. They are not, however, capable of indicating whether such training is a success on wider efficiency criteria.
Finally, evidence has begun to emerge on the private returns to formal upgrade training in Germany, which we treat as sponsored by individuals rather than by employers. Skilled German workers who wish to upgrade their qualifications from Facharbeiter to Meister or Techniker level do so typically by taking part-time courses outside working hours, relying on local public technical colleges8. The training might be expected to pay substantial dividends, given large pay differentials between these qualifications, together with strong employer demand for higher vocational qualifications. Quasi-experimental evaluations of adult training in Germany unfortunately tend not to distinguish between upgrade and further training (Fortbildung, Weiterbildung), on the one hand, and retraining (Umschulung), on the other9. Moreover, those who make the distinction have studied effects on employment rather than pay. In any event, positive effects on employment have been inferred by two studies. The first used a comparison group to analyse adult training in Hamburg, finding even stronger effects for upgrade training than for retraining10 . The second dealt with eastern Germany, finding that part-time adult courses are associated with lower exposure to unemployment, and full-time ones with the opposite. To the extent that upgrade training is conducted part time, and retraining full time, the implication is that upgrade training reduces the probability of entering unemployment. However, this conclusion too is endangered by selection problems11.
In sum, in its highly varied national forms, individually sponsored training appears to improve markedly participants subsequent labour market fortunes, even allowing for unobserved positive selection into such training. That is perhaps no surprise - and it is just as well that it is the case - given that the individuals who take the training have to invest their own time and money while doing so. It is not easy to establish the merits of such training from the efficiency standpoint, as most studies use only partial outcome measures and ignore costs. Its merits from an equity standpoint are intrinsically limited, in so far as individuals must typically already possess resources in order to undertake such training. Public subsidies do, however, reduce the costs to individuals of some forms of VET, notably community college courses in the United States, helping less advantaged individuals to undertake training.
[Top]