Total amount: € 0,00
Official Journal of the Italian Society of Social Psychiatry
Indexed/Abstracted in: EMBASE, e-psyche, PsycINFO, Scopus
Mangini E., Onofri F., Piovan C., Dello Buono M., Caon F.
Background. The authors present the second part of a study whose object is the formulation of an operative definition of the concept of interpretation, with regard to psychoanalytic psychotherapies, which could be useful in the empirical verification of the psychotherapeutic process. In this phase of the study the concept of operative definition is assessed according to reliability and construct validity criteria, as they are evaluated in empirical research in psychotherapy.
Methods. 22 independent judges, divided into three groups according to their clinical experience, were required to read 20 extracts selected from the verbatim transcriptions of audiorecorded sittings of brief psychoanalytic psychotherapies and to single out and distinguish the interpretations referring to the definitions for three different types of interpretation (A: Simple interpretation; B: Transference interpretation; C: Transference interpretation based on separation) formulated by a supervision group. Data were analyzed in four different steps: 1) analysis of agreement between judges according to Kappa coefficient of reliability; 2) analysis of casual-effects variance in order to control the stability of the main factors considered in the research (type of rating and group of judges); 3) analysis of deviance according to a log-linear model in order to assess the weight of three different factors (type of rating, group of judges and item) and their interaction; 4) a qualitative analysis based on a comparative assessment of the ratings of judges and supervision group according to the content of the singled out interventions.
Results. Parametric and non-parametric statistical methods showed an highly significant agreement between judges and confirmed the results of the previous study. In particular, ANOVA and log-linear analysis showed, on the one hand, that the main factors (type of rating and group of judges) are stable and, on the other, that 90% of variability is due to interaction between the factors type of rating and item; so the good approach of the study is confirmed. In this research, like in the previous pilot study, it appeared that judges could not equally recognize the three types of interpretation and they did differently from the research group. Finally, it appeared that clinical experience didn't differentiate the rating, that is similar in the three groups of judges.
Conclusions. The results seem to confirm that the definitions have a good level of operativeness. Nevertheless, if, from the point of view of reliability, the definitions are good enough, from the point of view of construct validity, definitions would get better if we could find other variance factors not yet controlled in this research plan.