The U.S. population censuses of 1860 and 1870 report less information about prisoners than does the census of 1850. The censuses of 1860 and 1870 did not report prisoners by sex. The censuses of 1860 and 1870 also did not group prisoners by the type of penal institution in which the prisoners were confined.

Textual comments in the census report for 1870 indicate that census officials put considerable effort into producing prisoner statistics. The report observed:

In the preparation of the following tables {tables of prisoners in 1850, 1860, and 1870} the effort has been made to reduce the material to something like consistency and uniformity. … The numbers reported … as in prison on the 1st of June, 1870, are regarded as quite accurately determined. Errors may be found to exist, but an extensive correspondence on the part of the Census Office has established their substantial accuracy and completeness.

The report also noted:

From the number in prison have been excluded the inmates of houses of refuge, houses of correction, and institutions of kindred character.^

These remarks suggest that the figures for the prisons included were of relatively good quality. However, prisoners held in local, short-term confinement (jails, houses of correction) were not included.

The reported prisoner figures for the censuses of 1860 and 1870 exclude a significant number of prisoners. The census of 1870 reported 6737 prisoners for 1850. This figure is the sum of prisoners reported on the schedules of social statistics from the census of 1850. The schedule of social statistics included prisoners other than those elsewhere categorized as in state prisons and penitentiaries. Nonetheless, in 1850 the number of prisoners reported on the schedule of social statistics was about 50% less than the estimated total number of prisoners. Since the prisoner figures reported in the censuses of 1860 and 1870 were also based on the schedules of social statistics, those figures most likely also significantly undercounted the total number of prisoners.

Study of the archival records of the population returns for 1860 and 1870 might provide more information about prisoners in 1860 and 1870. However, reasonable, comprehensive estimates of total prisoners in the U.S. for 1860 and 1870 can be made from published data. Comparing those estimates with subsequent census data on prisoners provides a reasonably comparable decennial series for total prisoners in the U.S. from 1850.