The Committee on Social Statistics of the American Statistical Association, in conjunction with the Committee on Sociology and Social Work of the American Sociological Society, held a joint session in connection with the meetings of the two Societies in Cleveland, on December 31, 1930, devoted to a discussion of the teaching of social statistics to prospective social workers. This session was in the form of a round table, led by Ralph G. Hurlin, Director of the Department of Statistics of the Russell Sage Foundation, and was presided over by the writer. The discussion centered around three questions: (1) the purpose, (2) content and (3) method, which should characterize a beginning course in social statistics to future social workers. Dr. Hurlin, who has had a great deal of experience in the last few years in dealing with social work statistics in connection with the material which he has been gathering from family welfare and other social agencies throughout the country, with respect to the amount of relief given, number of cases handled, etc., based his discussion on his experience in gathering, treating and interpreting the data, and also on his experience in teaching prospective social workers in the Training School for Jewish Social Work. He raised the question as to what should be the purpose of a course in social statistics for social workers. Should it be looked upon as discipline? Should it aim to give the case workers a critical point of view? Or should it aim to give them tools for treating quantitative data? He made it clear that the content of the course would depend very largely upon the aim set for it. In like manner would the method of teaching vary with the purpose which the course or courses are to serve. Thus if the aim is to provide the students with discipline, a great deal of laboratory work and drill would be necessary. If, however, only critical ability and a point of view are to be developed, then drill would be of less importance and emphasis would have to be placed on analysis of statistical studies. Similarly, if the course primarily aims to give intensive training in statistical procedures, it would have to be built up along lines different from those necessary if only a critical point of view is aimed at. In the discussion which developed the following points were emphasized: First, that since it is impossible to tell what type of work the