Policy Design Analysis and Evaluation

PN. The course's content can undergo some improvements. The consolidated version will be available on the Teams channel.

Class 00 - Map
The first class will be for introducing ourselves and recapping (a) the course's topics and goals; (b) available resources; (c) the learning path and assessment strategies.
Backing:
A Schneider, H Ingram, Systematically pinching ideas, 10.1017/S0143814X00006851
A Schneider, H Ingram, Behavioral assumptions of policy tools, 10.2307/2131904

___Module A

We understand public policies as solutions X to collective problems Y. As such, they build on two intertwined hypotheses.
The first maintains that problem Y arises as some actors make a particular decision in situations Φ. The second considers that intervention X on the situation can trigger some mechanism M that changes actors' behaviors to solve or alleviate the problem.
The module offers analytic tools to develop a whole policy theory, then focuses on using information as an instrument of intervention.

At the end of this module, you will be able to:
- Render effectiveness as a Y through proper gauges,
- Conceive some micro-foundations M of policy effectiveness,
- Conceptualize policy interventions X as tools geared toward changing the gauge's values by triggering M,
- Consider the setting Φ of policy interventions and how they can twist the tools' rationale.


___Module B

In module A we have learned essential frameworks and practices to define XΦMY and render the core of a policy theory.
But how can we know that such theory 'holds'?
A long-honored answer maintains that a theory holds when its shape allows for valid reasoning about the world.
The module introduces the criteria for moving from conversational theories to valid structured arguments and amenable to empirical probation.


At the end of this module, you will be able to:
- identify Reasons and Conclusions in written texts,
- pinpoint those elements that remain implicit,
- organize Reasons and Conclusions into inferential structures,
- recognize basic logical fallacies in those structures, if any,
- consider which evidence could prove the structure sound.


__Module C

Either explicitly or implicitly, most empirical research aims to interpret the co-occurrence of interesting phenomena causally. Addressing causality, however, has been notoriously difficult without the luxury of experimental data. This course will introduce methods that allow making convincing causal claims without working with experimental data. We will look at four such designs:
1. Instrumental Variables
2. Matching
3. Difference-in-Differences estimation
4. Regression Discontinuity Design
For each method, an in-depth theoretical explanation will be followed by a short review of the mathematical intuition behind the model. Furthermore, we will have a hands-on lab section for each method.

By the end of the module, students will be in position to:
- Critically read and evaluate statements about causal relationships based on some data analysis.
- Apply a variety of design-based easy-to-implement methods that will help them draw causal inferences in their own research.
- More broadly, think about policy proposals and their outputs or outcomes under the logic of causal inference.


___Module D

Qualitative Comparative Analysis (QCA) allows establishing the tenability of complex causal hypotheses as Boolean models.
The module introduces the special assumptions and strategy of the explanatory usage of QCA with the aid of actual examples.

At the end of the module, you'll be able to:
- Conceive of deterministic causal models,
- Establish classification rules for the units of a population,
- Identify underspecified models,
- Get rid of irrelevant components in over-specified explanatory models.