The Validity of Laboratory Research in Social and Behavioral Science

The validity of artificial situations is often questioned, and particularly so the possibility of transfer of findings to the real world. Such questions, or doubts, may stem from a rigid distinction between real and artificial situations or from too strict a notion of representation. This article will argue that `the real world' does not provide unambiguous criteria for representation and that, moreover, many experiments and simulation games do not have to represent `the real world' in any direct way. Both issues are usually treated under the heading of external validity, which means compliance to conventions that dominated thinking about validity over decades. These conventions need to be reconsidered. Quality standards for research must not be rigid, nor should be applied in a way that ignores the characteristics of a particular research project. Fixed notions about validity may prevent a researcher from adapting validation procedures to the circumstances at hand. The article takes issue with a conception of external validity as surface resemblance between artificial and real situations, advocates an active, non-routine approach to validity questions, and encourages individual researchers to develop a line of reasoning on these questions instead of adhering to standards that may not suit their particular research.