Integrating the Study of Shine's Actualized and Pure Single-Subject Behavior Functions

Abstract

The purpose of the present paper is to illustrate how an integrated study of Shine's actualized and pure single-subject behavior functions can produce more information than studying only one behavior function would. After summarizing the mathematics behind Shine's viewpoint, an actual experimental example is used to illustrate such an integrated study. The results of an ordinary regression analysis approach to the actualized behavior function are integrated with the results of a time-series analysis approach to the pure behavior function. This integration provides an estimate of the degree of experimental control over sources of systematic variation for the experiment. The estimate indicates an apparent poor degree of control. Further analysis leads to the tentative identification of a major contributor to the apparent poor degree of control over sources of systematic variation.