| Abstract: | Outcome research in counselling and psychotherapy has come to rely heavily on the use of self‐report questionnaire measures of anxiety, depression and general symptomatology. The validity of these measures in the context of psychotherapy research has, in general, been taken for granted. This paper proposes that, despite the success of self‐report questionnaires in personality research and personnel selection, there are significant problems involved in the use of these methods in research into therapeutic change. Specifically, three areas of difficulty are reviewed. First, questionnaires are completed in a social setting that itself changes as a result of therapy. Second, the experience of psychotherapy changes the way that clients make sense of questionnaire items (the ‘response shift’ phenomenon). Third, the conceptualisation of the person implicit in questionnaire design is not compatible with the conceptualisation of persons espoused by most contemporary therapeutic approaches. It is argued that the continuing popularity of self‐report questionnaires can be understood in terms of Kurt Danziger's model of the social construction of psychological knowledge. |
