Why it is Important to Look Closely at What Happens When Therapy Clients Complete Symptom Measures John McLeod (bio) A concern for evidence can be viewed as a fundamental aspect of human existence. The biological structure of our bodies evolved during over thousands of years in which survival was predicated on a capacity to interpret small signs, such as crushed grass, smells, and sounds as evidence of the whereabouts of prey. The emergence of modern science and medicine was built on the ability to learn about what counted as evidence for what, and to observe it reliably. Evidence is information that matters. Drinking a glass of water because one feels thirsty is not evidence, unless there is a suspicion of an underlying condition for which water intake is a symptom. Appraisal of evidence is a complex task. It relies on the existence of some kind of theory that predicts and explains a link between what is observed and some broader state of affairs that it signifies. It requires agreement on criteria for accepting an observation as relevant. Finally, given the likelihood that multiple sources of evidence are available, there needs to be some kind of procedure through which different types of evidence are weighed against each other. It is inevitable, therefore, that the search for evidence tends to be characterized by differences in opinion, continual revision of theories and criteria, and skepticism around efforts to impose premature closure. Psychotherapy is an activity in which the search for reliable evidence represents a constant challenge. The actual practice of therapy involves client and therapist appraisal of the meaning and evidential significance of a very wide range of information that occurs within therapy sessions: thoughts, feelings, patterns of behavior, gestures, ways of talking, and so on. Weaving through the process of therapy is the question, "Is it making a difference?" The issue of therapy outcome or effectiveness opens out into an even broader set of perspectives, beyond the views of the client and therapist themselves, on what counts as evidence. For example, do members of their family view the client's therapy as having been worthwhile? Does the government health agency that paid for the client's course of treatment regard it as having been cost-effective? Research into the effectiveness of psychotherapy has been carried out since at least the 1950s [End Page 133] (see, e.g., Schjelderup, 1955). The first phase of this global research initiative was marked by a high degree of methodological pluralism in which different types of data (e.g., questionnaire ratings, interviews, responses to projective tests) were collected from different sources (e.g., the client, therapist, independent observer, and significant others such as family members) and subjected to single case analysis as well as trends across groups (Strupp & Hadley, 1977.)The next phase, from about 1980, was marked by a shift in the direction of evaluating therapy outcome almost entirely through comparing scores on self-report symptom measures completed by the client at the start of therapy and at the end. At the level of scientific discourse, this narrowing of outcome focus was driven by a desire to emulate the achievements associated with the use of randomized clinical trial methodology in medicine, which depends on the prior identification of a primary outcome measure. It was also influenced by the relative simplicity and ease of administration of self-report measures, and the possibility of assembling very large data sets. In the background, these rationales fitted well with a broader neoliberal sociopolitical agenda that advocated the adoption of a markets and competition in public services, rather than collaborative consensus. In recent years, following more than three decades of psychotherapy outcome research based on evidence from client self-report data generated during randomized clinical trials, a further shift has become apparent. There is growing dissatisfaction with the quality of scientific knowledge and policy-making associated with established methodologies, which has fueled determined efforts to build alternative strategies. These initiatives have mainly focused on greater exploration of the qualitative and single-case evidence in relation to therapy outcome, and the development of meta-analytic techniques for combining findings from studies using different methodologies. An important contribution has involved philosophical analysis of...