30234

SoSe 22: Experimental Approaches in Social Inequality and Stratification Research

Stefan Liebig

Comments

Experiments are considered as the gold standard of scientific inquiry. They enable researchers to identify causal relationships and provide scientific explanations of various phenomena. From a sociological point of view, experiments have the problem that they are mainly conducted in the lab and–as consequence–their participants are in most of the cases students which reflect only a small subsample of the entire population of a society. Although there may be a high internal validity of the results of experiments, the external validity seems to be comparable low–what we observe with students in the lab may not be true for ordinary folks outside the golden cage of a university.

This course is structured in three parts: In the first part, the epistemological and methodological foundations of experiments and the history of experimental research in sociology are introduced. In the second part we concentrate on the problem of “external validity” of experiments and discuss various “solutions” of the problem. One methodological solution is the factorial survey design, which will be discussed in more detail. This method–originally developed by Peter Rossi in the late 1940s– became more and more popular over the last 25 years. It is used mainly in attitude related research and to study decision-making processes – e.g. to identify the factors that are important to take a job-offer or accept applicants for a job or which factors should be considered for a fair income at the workplace or a fair punishment in a trial. With its multifactorial experimental design and its capacity to integrate it into population surveys, factorial surveys promise to free scholars from making trade-offs between internal and external validity. Moreover, it pledges to overcome some fundamental methodological shortcomings of the classical item-based survey research. close