Now Available!

2017 UA Data User Meeting Presentations

Data 101

  • Designed for new users
  • Subsets of our best variables
  • Tailored for popular research topics

Examination (data forthcoming)

To examine potential selection bias caused by those men rejected from military service, researchers collected a 20 percent sample of all men examined for service under the draft law between September 1864 and April 1865. Collected from Record Group 110 in the Records of the Provost Marshal General’s Bureau at the National Archives and Records Administration in Washington, D.C., the final sample totals 70,800 examination records that include data on:

The records cover people enrolled for the draft and note whether or not they were exempted from service. In order to more closely reflect the Union Army (UA) sample demographically, researchers further selected Examination sample records based on the congressional districts in which UA sample veterans enlisted. Thus the Examination sample is directly proportional to men enrolled from each congressional district represented in the UA sample.

During collection it became clear that fewer men were rejected for service than originally thought. Researchers decided to collect information on the entire sample, not just rejectees. Researchers collected a 20 percent sample of all men enrolled and examined for service, proceeding state by state, and inputting every 5th line, regardless of examination result.

This collection effort resulted in a final sample of 70,800 men. To examine their similarity to UA sample veterans researchers linked 10,000 rejected men to the 1860 and 1870 censuses as no military records exist for men who did not serve in the army. A further subsample of 1,000 of these men with uncommon names (which are more easily linked) was searched and linked to all available census and death records.