Logo: to the web site of Uppsala University

Aims: We examined how nondirective counselling skills affect the evaluation of counsellors and the observed working alliance, and how evaluators' social skills and cognitive ability moderate evaluations. Method: In a first analogue experiment, counselling transcripts containing more or less restatements, reflections and open-ended questions were audio-taped and rated by evaluators (n=52) who self-reported social skills. In a second experiment, a different sample of evaluators (n=63) rated transcripts and self-reported social skills and were tested on cognitive ability. Results: Restatements and reflections improved ratings of the counsellors and the observed working alliance, but open-ended questions did not. Ratings were moderated by evaluators' social skills and cognitive ability. Discussion: The results agree with previous evidence on restating–reflecting and open-ended questions. That certain evaluators preferred open-ended questions more than others can be understood as an issue of matching between conceptual level, social skills, and counselling structure.