LN9+Randomized+Experiments+(and+Intro+to+Fixed+Effects)

LN9+Randomized+Experiments+(and+Intro+to+Fixed+Effects) -...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
Empirical Methods II (API-202) Kennedy School of Government Harvard University 1 Lecture Notes 9 Randomized Experiments (and Intro to Fixed Effects) I – INTRODUCTION Today : Analyze validity of randomized experiments and introduce the concept of fixed effects in the context of the Star Experiment Our goal: Counterfactual Closest we can get: Properly designed and conducted randomized experiment Methodological strength of experiments well done comes from the fact that by randomly assigning subjects to treatment and control groups, we are forcing the error term to be uncorrelated with the treatment variable(s) by design: E( u |X)=0 But experiments in the real world tend to differ from the ideal textbook-version of an experiment in various ways This means we must still assess the validity of an experiment with the same rigor we apply to assess the validity of any observational study As an example, we are going to look at the Tennessee Star Experiment in more detail Many of the questions we will analyze are valid for any impact evaluation (i.e. whether we are using experimental or non-experimental methods) OPTIONAL (but recommended for your own edification): Read the paper Alan B. Krueger, 1999. "Experimental Estimates Of Education Production Functions," The Quarterly Journal of Economics, MIT Press, vol. 114(2), pages 497-532, May. o “This paper provides an econometric analysis of the only large-scale randomized experiment on class size ever conducted in the US, the Tennessee Student/Teacher Achievement Ratio experiment, known as Project STAR.” o Good complement to this LN. o Most important pages: 497-517.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Empirical Methods II (API-202) Kennedy School of Government Harvard University 2
Background image of page 2
Empirical Methods II (API-202) Kennedy School of Government Harvard University 3 II – VALIDITY Validity represents a set of criteria by which research may be judged. o There are degrees of validity – more valid or less valid – not just “valid” or “not valid” o Comes in two flavors: internal and external validity o A key goal of any research study should be to achieve high validity Internal Validity A study is internally valid if it estimates the causal effect of interest for the population from which sample is obtained. o To what extent does the evidence presented support a causal link between X and Y? o Important questions to consider while making this assessment: Are there confounding factors that could be partly responsible for the observed association between X and Y? Does Y also cause X? o Generally internal validity is not a problem in a properly conducted randomized experiment. External Validity External validity refers to the extent to which findings can be generalized to other settings (i.e. people, times, programs, etc.) Example: Are STAR results valid for policy design in Guatemala? Key question to ask: Are the findings generalizable to…
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 4
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/12/2009 for the course HKS API202A taught by Professor Levy during the Spring '09 term at Harvard.

Page1 / 16

LN9+Randomized+Experiments+(and+Intro+to+Fixed+Effects) -...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online