Abstract
As part of an investigation into criteria used by counsellor trainers to distinguish between good and bad students of counselling, experienced counsellor trainers were asked to complete a repertory grid. Each trainer was asked to list five good counsellor trainees and five not so good or bad counselling trainees they had worked with, as elements for the grid. Repertory grid forms were provided by the researcher with spaces for ten constructs which were elicited from the trainers in the standard triadic manner. For each elicited construct subjects were asked to rate each trainee on a five point scale. Twenty-seven correctly completed grids were collected. The 262 constructs elicited were conflated into 22 semantically similar categories by the researcher and independent raters. The data was subjected to statistical analysis using the Statistical Package for the Social Services (SPSS). Overall the constructs that were most frequently used in describing trainees were open-closed, personable/aloof, secure-insecure, professionally skilled-unskilled. The mean differences for scores of good and bad trainees for each of the 22 conflated constructs were calculated. The greatest mean difference related to professional competence. The 22 conflated constructs were subsumed into three categories, and the mean of the difference between ratings for good and bad trainees within each category was calculated. Counselling related competence showed the greatest mean difference. Implications for counselling course curricula and the selection and assessment of counsellor trainees are discussed.