information system data, programme monitoring statistics or other forms of systematic information that are not necessarily research-orientated. According to Patton, programme evaluation differs fundamentally from research as regards the purpose of data collection and the standards for judging quality. Basic scientific research is undertaken to discover new knowledge, test theories, establish the truth, and generalise across time and space. Programme evaluation, however, is undertaken to inform decisions, clarify options, identify improvements, and provide information about programmes and policies within contextual boundaries of time, place, values and politics. Ideas proposed by both Rossi et al. (2004) and Patton (1997) inform the way in which evaluation is understood and utilised in this manual. Purposes of evaluation The two approaches discussed above in fact converge with respect to the purposes of evaluation. Their first point of convergence is to distinguish two categories of evaluation: summative and formative. They describe summative evaluation as an exercise that judges the overall effectiveness of a programme and is particularly important in making decisions about continuing or terminating a programme or project. They posit that summative evaluation can be contrasted with formative evaluation, which focuses on ways of improving and enhancing programmes rather than rendering a definitive judgement about effectiveness. Summative evaluation, by contrast, provides data to support a judgement about a programme’s worth so that a decision can be made about the merit of continuing with the programme. Be that as it may, there are three main purposes of evaluation that Patton (1997, pp. 65–70) identifies: Judgement-orientated evaluation Judgement-orientated evaluation is evaluation that is aimed at determining the overall merit, worth or value of a project or programme. ‘Merit’ refers to the value of a programme, for example how effective it is in meeting the needs of those it is intended to help. ‘Worth’ refers to the extrinsic value for those outside the programme, for example the larger community or society. Judgement-orientated evaluation approaches include summative evaluations aimed at deciding if a programme is sufficiently effective to be continued or replicated. Questions asked include: Did the programme work? Did it attain its goals? Should the programme be continued or ended? Were desired client outcomes achieved? Improvement-orientated evaluation Improvement-orientated forms of evaluation include formative evaluation, quality enhance-ment and responsive evaluation, among others. These approaches gather varieties of data about strengths and weaknesses, with the expectation that both will be found and each can be used to inform an ongoing cycle of reflection and innovation. According to Patton (1997, p. 68), improvement-orientated evaluations ask: What are the programme’s strengths and weaknesses? To what extent are participants progressing towards the desired out-comes? Which types of participants are making good progress and which types are not doing so well? What kinds of implementation problems have emerged and how are they being addressed? What is happening that was not expected? How is the programme’s external environment affecting internal operations? What new ideas are emerging that can be tried out and tested? Knowledge-orientated evaluation The evaluation findings contribute to knowledge and may involve: clarifying a programme’s model; testing theory; distinguishing types of interventions; determining how to measure outcomes; generating lessons learnt; and/or elaborating policy options. This is sometimes described as ‘enlightenment’, where the evaluation findings broaden the knowledge base. 128 Community Healing: A training manual for Zimbabwe