Two paradigms of behavioral decision research (BDR) are criticized for their methodological problems. The Structural Modeling approach suffers from the use of indices of questionable validity as well as from conceptual confusions concerning the distinction of theories about behavior and the statistical tool to test these theories. The Process Tracing approach, on the other hand, suffers from often unjustified assumptions and the use of ad hoc statistics as measures for noncompensatory decision making. The lack of an explicit error model prevents the development of rational similarity criteria to classify individual behavior as "compensatory" or ,.noncompensatory". Furthermore, results from both approaches seem to imply contradictory conclusions, questioning the validity of at least one of the paradigms. While some of these criticisms have been acknowledged before, a more fundamental methodological analysis attributes these problems to the inductive nature of both approaches. It is recommended to replace this inductive research mode by a more deductive and theory-testing methodology that includes the derivation of expected data patterns given a specific theory of decision making.
Dieser Datensatz wurde nicht während einer Tätigkeit an der Universität Mannheim veröffentlicht, dies ist eine Externe Publikation.
Search Authors in
