KAFP 6530 - Policy Analysis and Policy Evaluation - Beginner

The course aims at providing students with the knowledge and skills to develop a research protocol to rigorously assess the impact of a policy programme. We will examine different research designs for policy evaluation: pre-post comparisons, matched case-control studies, differences-in-differences, randomized experiments, regression discontinuity designs. Moreover, we will consider the benefits of the integration between methods of policy impact evaluation and qualitative methods to analyse the implementation and reception of a programme.

The course examines the different steps of data collection: targeting, sampling, questionnaire design, choice of the data collection method, randomization. The course makes very limited use of mathematical abstraction and does not involve data analysis. The emphasis is instead on the strengths and limitations of different designs of impact evaluation and learning survey research methods. Randomized controlled experiments and educational evaluations receive particular attention in this course but a variety of examples from other domains are considered as well. Throughout the course, students will work in groups to analyse a case study and deliver an evaluation protocol based on a randomized controlled trial. They will learn how to carry out each stage of this research process.

Carlo Barone is a sociologist working on educational policies and, particularly, on interventions aimed at reducing educational inequalities. He has carried out several randomized interventions in collaboration with public authorities and private foundations.

Students are expected to have attended an introductory course on descriptive statistics and regression analysis. However, some basic statistical notions are rapidly reintroduced and additional materials are made available for students without any prior knowledge.

• A midterm test ("galop") accounting for 40% of the final grade. -
• A group project to be submitted by early December accounting for 40% of the final grade (including 10% for oral presentation)

-MCQ with few short open answer questions after lecture 12 (20% of the final grade).