The document discusses differences-in-differences (DID) estimation and provides examples of its application. It first reviews DID and its identifying assumptions. It then analyzes a study that uses DID to examine whether governments allocate more resources to councils controlled by their party. Next, it uses DID graphs and linear regression models to estimate the causal effects of electronic voting on voter turnout and spoiled votes in Kyoto, Japan. Standard errors in DID estimation and the synthetic control method are also briefly discussed.