UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Can countertransference manifestations be identified during counselling sessions? De Vita, Elsie Lorna


This study investigated two main research questions: first, could countertransference manifestations, operationalized as counsellor over-involvement and under-involvement, be reliably identified by independent judges observing videotapes of actual therapy sessions; and second, was there evidence to support the contention that counsellor over-involvement and under-involvement were valid indicators of countertransference behaviour. A multiple case study research design was employed to research this phenomenon. In order to respond to the second research question, this study had to determine whether the first research question could be answered with confidence. Thus, a methodology was employed for the first research question that maximized the reliability of measuring counsellor over-involvement and under-involvement. A generalizability (G) study was conducted to assess the dependability o f the behavioural measure of countertransference. The G study helped to design the decision (D) study (e.g., how many counsellor-client dyads, sessions, and judges would be needed to obtain a reliable measure o f counsellor over-involvement and under-involvement). The D study included two counsellor-client dyads across eight therapy sessions. Three judges used videotapes and transcriptions of the sessions to rate each counsellor response for over-involvement and under-involvement using a 7-point Likert scale (i.e., - 3 = under-involved; -2 = somewhat under-involved; -1 = possibly under-involved; 0 = empathically involved; +1 = possibly over-involved; +2 = somewhat over-involved; +3 = over-involved). This study confirmed that counsellor over-involvement and underinvolvement could be reliably observed by independent judges. The average intra-class correlation across eight therapy sessions was .76 for Counsellor One and .79 for Counsellor Two. A moving averages graphing procedure was used to identify episodes where each counsellor's over or under-involved response departed from their individual baseline, using sessions with reliability coefficients greater than .75. These episodes were used as the focus for research question two, investigating indicators of countertransference. A Qualitative (Q) study was conducted to respond to the second research question. Data was collected from multiple information sources (e.g., episodes of over and under-involvement from session transcripts, counsellor session notes, supervision notes, and counsellor and supervisors ratings). These data were then analyzed qualitatively by triangulating the data and identifying themes. The results suggested that counsellor over-involvement and under-involvement are interpretable as valid indicators of countertransference behaviour.

Item Media

Item Citations and Data


For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.