Int J Psychoanal. 1997 Oct;78:959-73.
The International journal of psycho-analysis
S C Vaughan, R Spitzer, M Davies, S Roose
Although analytic process (AP) is a core concept in psychoanalytic theory and practice and has emerged as an important variable in outcome studies, there is no consensus regarding its definition and operationalisation. This paper describes the development and validation of the Columbia Analytic Process Scale (CAPS), a rating scale developed to evaluate the presence or absence of AP in a single psychoanalytic session transcript for purposes of an outcome study. Definitions of interrater reliability and construct validity are reviewed and two studies designed to evaluate these important aspects of the CAPS are presented. The results demonstrate that the CAPS has adequate interrater reliability (kappa = .5). To establish construct validity the plan was to compare the CAPS rating of AP to clinical consensus. However, when a group of ten senior training and supervising analysts at Columbia were asked to rate five psychoanalytic session transcripts, no clinical consensus could be established. Statistical analysis of the pattern of the analysts' clinical ratings showed that the largest portion of the variance was accounted for by the error term of a two-way ANOVA. The implication of this finding is that the construct of AP itself is ill-defined. The results of this study suggest that the commonly used term AP has less consensually held meanings than analysts tend to believe; the impact of lack of definition of key terms on clinical and research pursuits within psychoanalysis is discussed.