De politieke relevantie van correlaties

Samenvatting

It is argued that C. Jencks's model (Inequality, a Reassessment of the Effect of Family and Schooling in America, New York: Basic Books, 1972) has not yet been sufficiently corroborated & that results using this model have a purely descriptive, not causal value. In consequence they cannot be used as a basis for political decisions. The value of the results is further restricted by the poverty of the data & statistical limitations. The main value of the research by J. Dronkers & U. de Jong ('Jencks en Fagerlind op zijn Hollands: Een aangezet tot de studie van de relaties tussen milieu, intelligentie, onderwijs, beroop en inkomen' [Jencks and Fagerlind in a Dutch Way: A Contribution to the Study of the Relations between Milieu, Intelligence, Education, Occupation, and Income], [an abstract of which appears in this section]) is to show that Jencks's results are not usable in the Netherlands. Unfortunately, the same is true of their own results. In Weerwoord of De Leeuw 'Sociale wetenschappen tussen relevantie en methode' (Answer to De Leeuw 'Social Sciences between Relevance and Method') J. Dronkers & U. de Jong agree with most of the methodological criticisms of De Leeuw, but they argue that their conclusions still have a limited practical value. They point out that the social sciences are forced to find compromise solutions, which are scientifically moderately sound, & politically relevant. Modified HA.