Doing Sociology in the Age of ‘Evidence-Based Research’: Scientific Epistemology versus Political Dominance

Abstract

Social science history from its beginnings has witnessed periods of confrontation between – generally speaking – qualitative and quantitative paradigms, even talking of ‘war’, ‘wrestlers’ and ‘warriors’. And, again from the very beginnings, our discipline has been forced to relate with funding agencies. Sometimes, the two paths – the scientific one and the financial one – cross: we may think at the role of the private foundations in financing a certain type of investigation, notably surveys against case studies or qualitative research. Nowadays, we see an increased attention by federal agencies and private foundations on a particular sector of research, randomized controlled trials (RCTs), focusing on techniques seen intrinsically superior from the methodological and epistemological point of view. This article will analyze the recent increase of the randomized controlled trials as the new “gold standard” for social research; the call for the “experimenting society” (Campbell American Psychologist, 24(4), 409–429, 1969), willing to import the randomized controlled trials approach into the field of social policy and planning, is not new, if we think that yet in 1963 Campbell and Stanley wrote that “a wave of enthusiasm for experimentation dominated the field of education in the Thorndike era, perhaps reaching its apex in the 1920s” (Campbell and Stanley 1963/1966: 2). Many problems of validity with RCTs soon came to be recognized – even by Campbell, who stated that he has “held off advocating an experimenting society until they can be solved” (Campbell Evaluation Practice, 15(3), 291–298, 1994: 294). In the following years a different set of evaluation strategies were developed, but today there’s a new effort to re-introduce the experimental approach in the academic arena. What is the difference? As we will see, the scientifically-based research is now established and codified by law, funding is linked to a particular way of doing research, and the consequences on scientists work are yet to be explored.