This document discusses heteroskedasticity and various methods for testing and correcting for it. It begins by explaining how heteroskedasticity violates the assumption of homoskedasticity and impacts standard errors and tests in ordinary least squares regression. It then describes the Breusch-Pagan test and White test for detecting heteroskedasticity by regressing the squared residuals on the regressors. Finally, it outlines how to implement heteroskedasticity-robust standard errors and tests to obtain valid inference even when heteroskedasticity is present.