This thesis examines the effect of the Revised Family Law implemented in Ethiopia in 2000, using a difference-in-difference approach. The main data used is from the Demographic and Health Surveys (DHS) for Ethiopia from 2000 and 2005. Utilizing the different regional timing of the implementation of the law, the effect of the law change is measured clear from other underlying trends. Focusing on the outcome variables for education and violence against women, the main results show that there is no increase in years of education attained, but that there is a positive change in norms as fewer women answer that it is justifiable to hit a wife. It is important to quantify the relations between policy and outcomes in order to give policy recommendations that work and that will improve women's lives. Because cultures and norms differ widely, it is important to study empirical data in the country concerned and not only extrapolate knowledge from studies done other places in the world.