These should, at minimum: - provide for the creation of independent monitoring systems to analyze and review progress in policy, and services development and implementation, with clear terms of reference, work programmes and working methods. Regular public reporting on progress must be ensured; - put in place procedures to target and monitor funding allocated to minorityrelated projects (i.e. regular reports including a breakdown on expenditure); - allocate funding for monitoring projects carried out by minority non-governmental organizations; and - provide support for capacity-building of minority organizations, so that they will be able to play an active role in monitoring the implementation of projects designed to benefit them. PART I The ombudsman should have a role in monitoring progress of governmental action programmes and policy. Only a few minority programmes have moved beyond general aspirations and set specific and quantified targets, which provide a basis for monitoring progress. Moreover, in many cases there is confusion between monitoring and evaluation, and the scope of these activities is not defined or is unclear. There is a tendency to consider that internal monitoring carried out by the implementation agencies is sufficient. The issues of monitoring by specially created independent agencies and by civil society is too often disregarded. Each of these forms of monitoring has its own rules and conditions which should be examined and then embedded in the policy-making process. The ombudsman should be involved in evaluating programmes and projects supporting minorities. There is a pressing need to evaluate the impact of projects and programmes for minorities. Since it is change on the ground that matters, the real impact of changes in attitudes, actions and policies needs to be assessed in a way which is both credible and useful. Neither the amount of funding nor the compliance with budget lines and expenditures is a criterion for the adequacy of a project. Impact is what matters and for it to be assessed there is a need for quality base line studies, which will lay the foundation for reliable and credible impact assessments. In order to assess the impact of projects and programmes, a methodology is needed which addresses the issues of validity, reliability, credibility and attribution. For a primarily qualitative approach to be consistent with the overall aims of minority projects it must also address the issues of participation, ownership and empowerment by the minority themselves. So far, analyses of projects and funding are too often based on either a few intuitive assessments, randomly collected qualitative information, or on items of quantitative data which amount to a description of what has been done and not what has changed. Planning the implementation of policies should be completed with detailed chapters on evaluation, which would permit impact assessment, independent of any governmental structure and based on the views of the target groups and participating individuals. The whole spectrum of types of evaluations should be taken into consideration: ex ante (preparatory and feasibility studies, appraisals); mid-term (during the implementation of the project); end term (at the completion of the project); ex post (some time after completion). During the implementation, attention should be paid to formative evaluation – where the evaluator is a member of the implementation team whose role is to continu33