The unemployment statistics of the 1930s' Great Depression have often been used as a way of estimating the endurance and severity of the economic crisis in different countries. The Swedish unemployment data have, among other things, been used in comparison with the American data to describe the different economic development that these two countries experienced during the twentieth century. Sweden's 'work approach' to unemployment during the Great Depression has also been highlighted in earlier research as an answer to the question of why Sweden managed the Great Depression quite well and was able to kick-start the development of the welfare state. Unfortunately, the official unemployment data mainly covered men; female unemployment has been neglected both in the official records and in earlier research, and the praised 'work approach' only targeted male unemployment. In this paper, a new and previously neglected source of unemployment data is brought into the light in the form of poor relief statistics. From this information, a new estimation of the female unemployment during the Great Depression is made, which questions the grand conclusions that have been reached regarding the connections of the crisis and the later welfare state construction.