PHYSICS 102 Lecture Notes - Lecture 4: Multilevel Model, Randomized Experiment, Panel Data
Document Summary
Econometric methods for causal evaluation of education policies and practices: a non-technical guide: schlotter, g. schwerdt, l. woessmann, intro. Reap returns in labor market (economist view) Goal: provide empirical strategies to identify the causal impact of different reforms on any kind of educational outcomes. Reverse causality: outcome has a causal effect on treatment. Omitted variables: another variable influences both the treatment and outcome, but is not included in the model. Unobserved heterogeneity (= variables that affect both the treatment and the outcome) ex: ability. Endogeneity problem: treatment cannot be viewed as exogenous to the model, but rather endogenously biased results. = estimate the effects of multiple variables on the outcome at the same time: ordinary least squares (ols, multilevel modeling (or hierarchical linear models hlm) = find matching pairs of treated and untreated individuals who are as similar as possible in terms of observed (pre-treatment) characteristics.
