Friday, August 10, 2012

Comparisons: One Way to Add Flesh to the Bones

What does it really mean to "put your ancestors in historical context"? Especially when you have no letters or first-person accounts of them at all?

One approach for 19th-century folks involves that most commonplace of genealogy sources, the census. Use it differently. Compare your people with the neighbors (cross-sectional analysis) and compare them with themselves over time (longitudinal analysis). Sure, we'd all love to have a diary or a set of revealing letters. But lacking that, go for the numbers.

In 1850, 1860, and 1870 census enumerators were supposed to write down the value of any real estate owned (plus figures for personal property in 1860 and 1870). In those years, my great-great grandfather's widowed sister-in-law, Cynthia Balentine Thrall, lived in Wabash County, Illinois. Her husband Aaron had died before 1850, so the family was already a bit shadowy. The census population schedules reported that her real estate was worth $2,000 in 1850 and $10,000 in 1860. The 1855 state census showed her livestock worth  $450, and the 1865 state census showed her farm produce all together was worth $2350.

Of course you can convert these numbers into 2012 dollars, but that process is fraught with measurement problems and uncertainty. The census process itself is fraught with uncertainty (we don't know who provided the numbers, and from other research it's not at all clear to me that she owned five times as much in 1860 as in 1850), so I'm not crazy about a longitudinal comparison here either. Tax records would be a good corrective but this is a burned and tornadoed county.

My preferred way to make some human sense of these figures is to set them against those of her near neighbors in each census year: the adjacent five pages on both sides in 1850 and 1860, and the adjacent one page on both sides in 1855 and 1865 (which had only one line per household). These relative cross-sectional rankings were a bit more consistent than the raw numbers:

In 1850 five of her 70 near census neighbors had more real estate than her $2000, placing her (conservatively) in the top 10 percent.

In 1855, thirteen of her 86 near census neighbors had more livestock than her $450, placing her in the two 20 percent.

In 1860, only one of her 82 near census neighbors had property worth more than $10,000, placing her in the top 3 percent.

In 1865, five of her 117 near census neighbors had farm products worth more than her $2350, placing her in the top 6 percent.

These figures should not be taken as precise. I rounded the percentages up to give a more moderate result and to allow for poor-quality information and the randomness of which neighbors were visited. But it's clear that her family was better off than most of their neighbors -- maybe in the top 6-10 percent if we discard the outliers.

There are at least two ways to take this further: the agricultural schedules and overall county averages (perhaps a fairer comparison than immediate neighbors). As luck would have it, in 1860 she did appear in the agriculture schedule, and that was the year for which a diligent census-bureau employee compiled elaborate county-level statistics, obviously by hand (the book was several years in the making). So I was able to learn that the agriculture schedule had a much lower value for her farm in 1860 than the population schedule had.

The agriculture schedule shows that she had 120 acres of improved land; the median sized farm in the county (probably including unimproved land) was a little over 50 acres. In the preceding year her farm had produced 2000 bushels of Indian corn, more than triple the county average (mean). The household's production of butter and hay and buckwheat was also well above average. In terms of basic farm power, Cynthia had five horses (county average 3.5) and $200 worth of farm implements and machinery (county average $101). Some of these numbers can be qualified because she had significantly more land than average to work with. On yield-per-acre basis, for instance, her corn production was likely not so far above the average as the raw number of bushels would suggest.

These figures are reflected elsewhere in their lives. Her son and daughter who lived to have offspring both married into families who were better-off than the Wabash County average (although I haven't finished quantifying that casual observation yet!).

For this particular process, it helps if your folks didn't move around too much, and it helps to be comfortable with numbers and the difference between mean and median when working with "averages." This is just one approach among many possibilities.

Joseph C. G. Kennedy, Agriculture of the United States in 1860; Compiled from the Original Returns of the Eighth Census (Washington: Government Printing Office, 1864), pp. 34-37, line 93 (Wabash County totals), and p. 197, line 93. Also available on GoogleBooks.

1860 US Census, Wabash County, Illinois, agriculture schedule, no subdivision named, p. 25, line 26, Cynthia Thrall;
NARA microfilm publication T1133, “Illinois Nonpopulation Census 1850-1880,” “1860 Agr.: Vermillion [sic] pt.)-

1850 US Census, Wabash County, Illinois, population schedule, no subdivision named, pp. 404-9 (stamped), pp. 805-15
(penned), families 185-254; digital images, ( : accessed 30 October 2011), citing
NARA microfilm publication M432, roll 130.

1855 Illinois State Census, Wabash County, pp. 15-17, “Township 1”; digital images, (http://www. : accessed 3 November 2011), citing Record Series 103.008, roll 2196; Illinois State Archives, Springfield.

1860 US Census, Wabash County, Illinois, population schedule, Bonpas Precinct, pp. 143-53, families 1015-96; digital
images, ( : accessed 30 October 2011), citing NARA microfilm publication M653,
roll 234.

1865 Illinois State Census, Wabash County, pp. 11-13; digital images, (http://www. :
accessed 3 November 2011), citing microfilm of Record Series 103.010, roll 2185, Illinois State Archives, Springfield.

Harold Henderson, "Comparisons: One Way to Add Flesh to the Bones," Midwestern Microhistory: A Genealogy Blog, posted 10 August 2012 ( : accessed [access date]). [Please feel free to link to the specific post if you prefer.]


Debbie V. said...

Good thinking Harold.
Hey - do you know of any source for census records already in Excel spreadsheet form?
Thanks for all your great posts lately.

Harold Henderson said...

Thanks, Debbie. I don't know, but hopefully another commenter will!

Melanie D. Holtz, CG said...

Interesting analysis, Harold. Thinking beyond the obvious is a difficult skill to master.

Melanie D. Holtz, CG