This article investigates contributions and limitations of empirical educational research at the interface between science and politics using the example of large scale assessment-studies (LSA). Within the meta-theoretical frame of the divergent rationalities of science and politics and with reference to Goldthorpe’s (2001) concept of causation as a generative process, it describes the contributions of LSA in three fields: the theoretical and empirical conceptualization of domain-specific achievement dispositions, the description and explanation of social and ethnic disparities, and the identification of adolescents at risk of social exclusion at the transition to vocational training and the labor market. With reference to these examples, the article discusses whether it is theoretically and empirically advisable in social sciences to distinguish different forms of scientific knowledge according to their relevance for political decision-making. The article addresses the problem of how long-term communication can be established between science and politics despite the differences in their functional rationality and draws attention to the risk of not taking into account these differences and making promises that the social sciences cannot fulfill. © 2016, The Author(s).
Author Keywords
Index Keywords
