Over time, the predominant tendency of many governments' agencies has been to evaluate a programme or policy investing large amount of resources in supporting policy evaluation. However, recommendations suggested by policy evaluators are not always taken up. Moreover, there is relatively little evidence of the extent of policy evaluation effectiveness (i.e. the influence of evaluation on the programme evaluated) and the factors which have significant impact on it. This dissertation aims to shed light on this issue by focusing on the Mexican and Chilean experiences of policy evaluation in the health care sector. It provides a detailed analysis of the extent to which evaluations have led to changes in policies and programmes and reveals a rather limited effectiveness of policy evaluation in these countries. I argue that shortcomings in the effectiveness of policy evaluation can be explained by institutional and political factors, primarily the nature of Intra Governmental Relations (IGR), but also the quality of bureaucracy, the level of democracy, the autonomy of policy evaluators, and the type of policy evaluation framework. While all of these factors seem to have some influence, the relationship between the executive and legislature is clearly the key determinant of the take up of recommendations. Thus, the findings of this thesis suggest that strengthening coordination between the different parts of government is needed to enhance the effectiveness of policy evaluation. In addition, the analysis also suggests that policy evaluation is likely to be more effective when it incorporates budgetary incentives.