Most standard regression techniques focus on estimating how the conditional expectation of an outcome variable ($Y$) depends on a set of predictor variables ($X$). Quantile regression goes beyond mean effects, to estimate the impact of $X$ on any quantile or quantiles of $Y$. This enables researchers to assess many interesting questions like: what is the effect of smoking on infants with the lowest birth weight? How does a job market training program affect those at the bottom of the ability distribution? Or does smaller class size benefit the stronger or weaker students more?