); ga('send', 'pageview');
Benno Herzog
GlobalH-M (eds). Global Health and Migration: Interdisciplinary Tools to Tackle Health Inequalities. Valencia: Consortium GlobalH-M: 63-69
Publication year: 2013


Statistics seem to be the quintessence of scientific methods. A large number of cases which can be analysed by mathematical methods independently of the subjective position of the researcher and of the casualty of data production seem to allow neutral or objective social science. Indeed, quantitative methods are tightly linked to the birth of social science and its attempt to give social phenomena objective, scientific explanations. In the beginning of the last century social scientists therefore openly copied the strategies of natural sciences which celebrated break-throughs in quick succession. For politicians and decision makers statistically prepared data promises to condense quite diverse information into objective numbers, independent of ideologies and interests. And also for the media, citing numbers and percentages is the most common way of bridging the gap between “abstract” social science and citizens. The ease with which statistical data can be cited leads also to higher citation indices and therefore often to a higher reputation for articles, journals, researchers and research groups who use quantitative methods instead of qualitative ones.

The myth of objectivity and neutrality of statistics is based on ignorance at both ends of social research: The production and selection of the data on the one side, and the interpretation and use of the data on the other. Nevertheless, production, selection, interpretation and use are part of the scientific enterprise and therefore should be part of critical reflections on our research methods. In what follows, I will show how the embededness of statistical methods in the research process requires diverse decision making which is not free of normative implications.