Help with Creating Treatment Variables for Policy Impact Analysis

[Jonathan Thorsell]

I'm conducting a study replicating an analysis on the impacts of contraception and abortion access on male socioeconomic outcomes. My datasets include the IPUMS dataset (with individual-level data from 1970, 1980, 1990, and 2000 Census samples) and the Myers dataset (state-level policy data on contraception and abortion access from 1960-1976). I'm facing challenges in merging these datasets due to different structures and time spans. IPUMS contains individual data with birth years ranging from 1934 to 1982, while Myers provides policy data only from 1960 to 1976. I need to merge them based on state and the years when individuals were 16 and 18, to assess if they were 'treated' under the new laws at these ages. However, I'm struggling to find a common variable for merging, as the direct merge is not feasible. Any advice on how to effectively merge these datasets or alternative approaches to analyze the policy impact on individuals would be greatly appreciated.

[Clyde Schechter]

Example data, using -dataex-, from both data sets, please.