Time series analysis for single-subject designs (original) (raw)

Psychological Bulletin, 1982

Abstract

1. Suggests that the result of most interest in many single-S designs is change in behavioral trends. Trend lines, however, would differ if the experiment were repeated. Recently developed theory on combining least squares regression and time series modeling provides (a) accurate description of trends and other patterns and (b) tests of generality for observed trends. An illustrative application resulted in adequate models for 2 experimental phases and strong evidence that the direction of behavior was reversed. Restrictions on ...

William B. Ware hasn't uploaded this paper.

Let William B. know you want this paper to be uploaded.

Ask for this paper to be uploaded.