Implications of an initial empirical victory for the truth of the theory and additional empirical victories |
| |
Authors: | David Trafimow |
| |
Affiliation: | Department of Psychology, New Mexico State University, Las Cruces, NM, USA |
| |
Abstract: | Psychologists take two propositions for granted. Specifically, empirical verification of predictions derived from a theory (a) support that the theory is more likely to be true and (b) support that additional predictions derived from the theory have an increased probability of being sustained if subjected to empirical testing. In contrast, I argue that both propositions depend strongly on whether auxiliary assumptions are taken into account. When auxiliary assumptions are not taken into account, the first proposition is valid but the second is not. When auxiliary assumptions are taken into account, the first proposition is not valid, and the second proposition encounters additional problems. I use Venn diagrams and Bayesian principles to demonstrate these conclusions. |
| |
Keywords: | Auxiliary assumptions conditional probability empirical victory theory theory probability |
|
|