Research variables in psychology and the logic of their creation

In their book The Language of Psychology, Mandler and Kessen (1964) have concluded that research psychologists should limit themselves to studying variables that are defined in "response invariant" terms. The present author agrees with this general conclusion. The first part of this article will be a review of Mandler and Kessen's concept of "response invariance" and a probe of the reasoning that suggests such a conclusion. This analysis will prepare the way for outlining three kinds of methodological problems in psychology and corresponding solutions for each. This methodological analysis will be used also to relate two traditions in psychological research: the objective operational tradition that has been concerned, primarily, with the rigor of psychological research, and the phenomenological tradition that has been concerned with elucidating subtle, psychological profundities that are mysteriously entangled in human values and perspective, delicate dimensions of human experience that do not yield easily to rigorous and effective research designs.