Compulsory Schooling Legislation and School Attendance in Turn-of-the-Century America: A "Natural Experiment" Approach

Recent research by Joshua Angrist and Alan Krueger has used information on exact dates of birth in the 1960 to 1980 federal censuses to study the impact of compulsory schooling laws on school attendance. This paper modifies their methodology to analyze similar data in the 1900 federal census to measure the impact of turn-of-the-century compulsory schooling laws. Using data on 14-year olds from the 1900 census public use microdata sample we compare attendance rates of children born after January 1, 1900 with those born before, across states with and without compulsory schooling laws. In states that combined school-leaving with child labor laws, we find that compulsion significantly raised attendance rates.