Interpreting the Tick Marks on Federal Censuses
– Elizabeth Shown Mills, CG, CGL, FASG
Tick marks are a too-common cause of errors in the interpretation of census data. From 1850 forward, unexplained markings are rampant on the microfilmed and digitized pages that we regularly consult. Worse, the fact that we are using filmed images or digitized images made from the film, rather than the original manuscripts, makes it harder to discern exactly what was written by the enumerator and what markings might have been added later by some other party.
The problem is not so hard to counter, if we understand two things: why the tick marks exist and what instructions were given to each year's census taker.
Why Tick Marks?
From 1850 to 1880, the census marshals were required to make three copies. One copy was to be deposited locally as a public record, one was filed at the state level, and one was sent to the federal government's Census Bureau. The federal copy, with few exceptions, is the one that was filmed and distributed by the National Archives. Thus, as a rule, it is the one we consult at research libraries and the one that Ancestry.com and others have used for digitization.
Unfortunately, the federal copy, from a researcher's standpoint, has been adulterated. Beginning in 1850, the purpose of the population schedule was greatly expanded. Rather than being just a document that counted heads for the apportionment of congressional seats, it was designed to collect statistics by which America could define the social and economic characteristics of its population. That change gave birth to the tick marks.
Using the federal copies, bureau statisticians tallied every type of data. These analyses were then published by congress, for each census year, in statistical compendiums that examine age patterns, migration, occupations, marriage trends, crop production, etc. In the process of making these analyses, the statisticians added many marks to the original pages to prevent tallying errors. In most cases, they made their marks in places and ways that are clearly extraneous. In other cases, however, they added their marks in columns also used by the enumerators.
The Marshall County, Mississippi, return for 1860 illustrates the kind of mischief those tick marks cause. (Readers may wish to call up this census at Ancestry.com or another website to follow the case described here.) Until recently, one Internet site, which the creator commendably removed when he became aware of his misinterpretation, carried a paper entitled “The Marshall Mulattoes: An Index to Free Individuals with Non-Standard Racial Designations on the Federal Census for Marshall County, Mississippi, 1860.” This paper presented roughly one thousand individuals who were, supposedly, “listed as ‘free white mulattoes'--a non-standard racial designation.” To support that conclusion, the author commendably described his analysis of the census data. Below is a summary of that analysis and his conclusion.
1. The three race options from which enumerators could choose in 1860 were white, mulatto, and black.
2. No free blacks lived in that enumeration district.
3. The enumerator made no marks in the racial column if an individual was white.
4. For “free white mulattoes,” he made tick marks in the racial column.
5. Some of these “free white mulattoes,” surprisingly, had considerable wealth.
6. In tallying racial classes at the bottom of the page, the enumerator occasionally crossed out figures for whites and moved those individuals into the totals column for “colored males” and “colored females.”
In listing these one thousand or so individuals as “free white mulattoes,” the enumerator “was attempting to account for people of mixed red and white ancestry on a form that presumed everyone was black or white.”
Points 1, 2, and 3 are correct. All other points err--including the crucial conclusion that one thousand or so specific individuals had Indian ancestry.
The Source of the Confusion
If you examine the Marshall County pages from which the paper was compiled, you will see several types of tick marks in various columns. In particular, the race column carried quite tiny backward ticks, while other columns carried check marks and quite long forward slashes.
To properly interpret the census data, we need to do two things:
- know the instructions given to the enumerator, including the types of marks and abbreviations that were authorized for that particular year; and
- carefully distinguish between the data the enumerator recorded and the quite different “tick marks” the statisticians added.
Understanding the Instructions
Researchers will find the basic instructions outlined in two guides to the 1790–1930 federal censuses. Both are available in most Government Document departments of college, university, and major urban libraries. The current (2002) guide is available from the Census Bureau itself. The rules pertinent to the Marshall County problem are these:
Instructions for Tick Marks. In 1850 and 1860, the only columns for which enumerators were authorized to insert anything resembling a tick mark were columns 10, 11, and 12. If column 10 (married during the preceding 12 months) applied to an individual, the enumerator was to “make a mark, or dash.” If column 11 applied (attended school within those 12 months), the enumerator was to “make a mark, thus (1).” That is, he was to write one, using an Arabic numeral. If column 12 applied (over 20 and unable to read or write), the enumerator again was to insert a 1. It is important to emphasize here that the proper marks were to be dashes or ones, not tick marks.
Instructions for the Race Column. Explicitly, in 1850 and 1860, the enumerators were told: "Under heading 6, entitled "Color," in all cases where the person is white, leave the space blank; in all cases where the person is black, insert the letter B; if mulatto, insert M. It is very desirable that these particulars be carefully regarded."
On the microfilmed 1860 return for Marshall County, you will see that the enumerator followed instructions exactly. He recorded race for only a handful of individuals (“mulattoes”) because everyone else in his jurisdiction was white.
What you will not see, anywhere, is the term “free white mulattoes” that was said to be "applied” to that thousand or so individuals. That term was devised by the author of “Marshall Mulattoes” in an effort to describe not only the tick marks but also a corresponding “correction” that he had trouble interpreting.
Puzzled by the tick marks in the racial column for his ancestral entries, the author commendably sought an explanation in the “totals” that appear at the bottom of the page. There, four categories appear, positioned in this manner:
No. white males, ___ No. colored males. ___
No. white females, ___ No. colored females, ___
On many pages of the census, the enumerator erred in his bottom-of-the-page totals. Corrections were added. In analyzing these corrections, you will want to distinguish between the handwriting of the original enumerator and that of the individual making the alterations.
The correction made to totals on page 24 seems to be the basis for the author's conclusion that all individuals with tick marks were “free white mulattoes.” Even here, however, he recognized that this conclusion contradicts itself. Regarding page 24, he stated:
"This page had all persons totaled as whites, then scratched out and totaled with the coloreds. This page totaled 18 colored males and 22 colored females [i.e., 18 and 22 had tick marks], for a total of 40 people, meaning even people who were not marked as mulattoes were totaled with the coloreds."
As you examine the page yourself, you see that the totals for “white males” and “white females” are not “scratched out.” More importantly, they are in a hand totally different from that of the enumerator. The mistake that the census taker made on this one page was to put his totals in the “colored” column rather than the “white” column.
When the statistician at the Census Bureau processed that return, he corrected all of the totals where the original enumerator made a counting error. Those corrections appear in much lighter ink, using a pen with a finer point and different penmanship. On the troublesome page 24, the enumerator put his “white” totals into the “colored” column whose blanks fall just under the race column. Therefore, the Bureau statistician inserted the correct totals in the “white” columns. However, he did not scratch out the totals that the enumerator haphazardly placed in the “colored” column. Elsewhere, in the correction of “white” totals, the lack of space for those corrections caused the Bureau employee to write his “white” corrections atop part of the “colored” column.
The Federal Census Bureau's corrected statistics for the total number of free people of color in Marshall County, 1860, was not “a thousand” but eight.
Understanding the Ticks in the Race Column
The one remaining issue is the crucial question that initially puzzled the author of “Marshall Mulattoes.” Why the tick marks in the race column for “about a thousand” of Marshall County's roughly thirty thousand residents?
As you analyze the pages more closely, you will see that the tick marks in that blank column correspond to every instance in which the individual was born outside the state of Mississippi. Those born within Mississippi have no tick mark. This, of course, explains the riddle the author raised: Why would so many families have some individuals with tick marks and some without? It was not their ethnicity that was “mixed.” It was their states of birth.
Watching for "Red Flags"
Two of the minefields that all researchers encounter are the ambiguity of original records and the appearance of error when something does not conform to our expectations. However, “red flags” often exist to help us spot the anomalies. When reading censuses, for example, we frequently encounter situations in which the enumerator, in writing a birthplace as Mississippi uses the old-style, long-tailed double-s. The result appears to be Mipipippi. Applying common sense, we decide that a different handwriting convention must be at play here. In the present case, several other “red flags” exist.
Playing devil's advocate, I suggest the following:
- A conclusion that 1,000 residents of a single rural county in 1860 are racially misclassified is a whopping error rate that calls for evidentiary support from other records before a conclusion is justified.
- The fact that those supposedly misclassified individuals come from many different states rules out the possibility that we are dealing with a distinct ethnic community. Therefore, we bear the burden of explaining how and why 1,000 such individuals from many different states ended up in one corner of one small Mississippi county--and how the enumerator would know the ethnic background of all these people from such diverse areas.
- Finally, the fact that two of those individuals were born in Scotland and England is, on its face, a reason to rethink any hypothesis that the enumerator “was attempting to account for people of mixed red and white ancestry.”
In order to properly interpret puzzling situations on census schedules, we must know the instructions that were given to the enumerators. A “half-way” knowledge does not suffice. For a problem such as this one, we need to know not only that three racial designations were authorized, but also that census takers were instructed not to record race for whites.
Genealogical research can be an expensive hobby. Faced with a choice between investing our limited resources in guidebooks or in materials that give us actual data, it is tempting to choose the data and forego the guides. The present example, in which one small misunderstanding has created widespread confusion for genealogists working in Marshall County, Mississippi, illustrates why it is important for every serious researcher to invest in a basic library of essential guides.
1 . From 1850 to 1870 the original copy was to be filed locally and all three copies were to be identical. In 1880, to better preserve privacy, the local copy had radically abbreviated data.
2 . These publications are available in the Government Documents department of most college, university, and large urban libraries.
3. Bureau of the Census, Twenty Censuses: Population and Housing Questions, 1790–1980 (Washington: Government Printing Office, 1979); and Jason G. Gauthier, Measuring America: The Decennial Censuses from 1790 to 2000 (Washington: U.S. Department of Commerce/Census Bureau, 2002). Gauthier's expansion of the 1979 publication deletes some valuable portions of the original work but adds informative background discussions.
4. Twenty Censuses, 14 (1850) and 16 (1860).
5 . In some other years, the enumerator was instructed to insert in a particular column an “affirmation mark,” resembled a long slash, which one might generically call a tick mark.
6 . Twenty Censuses, 14.
7 . University of Virginia Library, Geospatial and Statistical Data Center, Historical Census Browser, GeoStat Center: Collections.
Elizabeth Shown Mills, CG, CGL, FASG, is author of the best-selling Evidence! Citation and Analysis for the Family Historian, and the creator and principal lecturer of the popular Advanced Methodology track of the Samford University Institute of Genealogy and Historical Research. Her first historical novel, Isle of Canes, is an epic account of a legendary Louisiana Creole family, and will be released by Ancestry in May 2004.
Copyright 2004, MyFamily.com.