Educational Reform, Ability, and Family Background

Top Cited Papers
Publisher Website
Google Scholar
Abstract
In the period between 1950 and the mid- 1970s many European countries carried out major educational reforms that resulted in increases in the number of compulsory years of education, in the introduction of national cur- ricula, and in the abolition, or delay, of the placement of more able students into separate schools at an early age (streaming). Examples of such countries are the United Kingdom, France, and the Scandinavian countries.1 Gen- erally, it is very difficult to evaluate the impact of such reforms, because they are implemented nationwide simultaneously and thus evaluations have to rely on before-and-after comparisons which may confound the effects of the policy with other macro-aggregate or cohort effects.2 In Sweden, major educational reform was designed in the late 1940s. The main elements of this reform were to (a) increase compulsory schooling to nine years from seven or eight years; (b) abolish placement based on academic achievement into an academic or nonacademic stream after grade six, i.e., at age 12 or 13; and (c) impose a nationally unified curriculum. An attractive feature of this reform is that it was preceded by a social experiment, albeit not ran- domized, where school reform was imple- mented gradually across municipalities. This permits an evaluation approach similar to those that have been used in a number of U.S. studies, which have examined the cross-state and cross- time variation in compulsory schooling laws and child labor laws to estimate their impact on educational attainment or to estimate the returns from education.3 In this study we evaluate the effect of the reform on final educational attainment and earn- ings. We have survey data on two cohorts of pupils: those born in 1948 and those born in 1953. For a substantial portion of the munici- palities, these two cohorts were assigned to dif- ferent school systems: the 1948 cohort to the old system and the 1953 cohort to the new one. In some municipalities, however, both cohorts were assigned to the old system, while in others both cohorts were assigned to the new, re- formed system. This allows us to evaluate the reform using a differences-in-differences meth- odology, comparing outcomes across cohorts and municipalities. The data for the two cohorts contain information on parental background, IQ test scores, and achievement in school (grades by subject) in grade six. Educational attainment and earnings are merged into the original survey data from the national education register and the 1985-1996 tax records, respectively. The distinctive feature of this policy experi- ment, combined with the data, is the ability to compare individuals working in the same labor