The objectives of this exploratory study into the nature of empathy were two-fold. First, to investigate the association between two measures of counsellor empathy, the Barrett-Lennard Relationship Inventory (BLRI), rated from the perspectives of counsellors and clients and the Truax/Carkhuff Accurate Empathy Scale (TCAE) rated from the perspectives of observers, counselling supervisors and counsellors. Secondly, to investigate whether there was a relationship between empathy and dyadic compatibility (as measured by the FIRO-F scales), or empathy and the occasions, nature and degree of counsellor and client perceived similarity to the other (as described in post-counselling interviews).
Sixty counsellor-client dyads (24 dyads across 1 - 3 sessions per dyad) were videotaped during counselling. Counsellors and clients were then interviewed independently, subsequent to each counselling session.
The results of the study were described in three sections. Part 1 reported that while the BLRI and the TCAE scale both exhibited a high degree of reliability, they appeared to be measures of different constructs. Results of a factor analysis suggested that not only must the two instruments be treated as independent measures of empathy, but that ratings made from differing counselling perspectives (counsellor, client, external judge) using the same instrument, must also be treated independently.
Client ratings of perceived similarity to their counsellors, and client ratings of counsellor (BLRI) empathy were highly correlated although there was no correlation between counsellors' perceived similarity to clients, and counsellor, observer or supervisor ratings of empathy.
There was no correlation between FIRO-F dyadic compatibility and either perceived similarity or empathy.
Part II involved an intensive case 3tudy comparison of specific dyadic rating patterns across measures of empathy, with patterns obtained from a content analysis of counsellor interview responses to questions on the occasions, nature and degree of similarity perceived between self and other. A complex matrix of findings was obtained which suggested that the link between empathy and counsellor perceived similarity to clients was not uniform and differed across both levels of empathy and perceived similarity.
Part III presented an historical review of the nature of empathy, undertaken in order to clarify and interpret earlier reported findings. It was suggested that conceptualizations of empathy have altered markedly over time, largely as a result of researcher's and counselling trainer's need for observable, measurable and trainable processes within counselling. An eight phase model of the process by which empathic understanding may occur was presented, and the manner and extent to which the model could be deemed accountable for the findings of the present study was discussed.
It was suggested that the divergent state of research associated with empathy may be largely attributable to the emphasis placed on one or several phases of the process to the exclusion of other phases. Suggestions for further research were offered, and the question was raised as to whether either the BLRI or the TCAE scale were in fact tapping any of the phases of empathy described in the present study.
