THEORY, PREMISES AND EMPIRICAL RELATION: DIFFERENCE-IN-DIFFERENCE IN POLICY DESCISIONS

Abstract

When determining the efficiency of welfare programs, development initiatives and public policies impact evaluation of such is important. Impact evaluation measures the casual relationship between the dependent and independent variable. In impact evaluation estimates can be experimental or quasi experimental. Experimental studies are not possible everywhere due to its financial and logistical constraint, so quasi experimental studies plays a vital role there. One of the quasi experimental technique of impact evaluation is Difference-in-Difference .The Difference-in-Difference (DID) approach has become one of the most popular quasi-experimental methods for estimating causal effects in non-experimental settings. A treatment group exposed to a policy intervention and a control group that is not impacted by the intervention are compared for changes in outcomes over time using the DID method. The approach accounts for common time trends that impact both groups and unobserved factors that remains constant over time by concentrating on differences across both time and groups. In the context of impact evaluation, this paper offers a thorough analysis of the DID approach. It covers the method's conceptual framework, mathematical formulation, fundamental presumptions, and real-world application. Additionally, a fictitious example of how DID uses interaction terms in a regression framework to estimate treatment effects is presented in the paper. Also, a critical analysis of the DID approach's advantages and limitation is discussed. The review emphasizes that although DID offers an effective and user-friendly tool for evaluating the impact for policies, its validity is primarily dependent on the satisfaction of crucial presumptions like the parallel trend assumption. The relevance of DID in current empirical research and its value in assessing development interventions are highlighted in the paper's conclusion.