February 2, 2021 3:14 PM

Pages 2728-2734
Whole Number 129

PERSONS NAMED SPARKS APPEARING IN CONNECTICUT CENSUS RECORDS, 1790-1840

As has been noted on numerous occasions in the past, for reasons that we cannot explain, a considerably larger number of persons named Sparks immigrating to America settled in the South rather than the North. Nevertheless, the Sparks name does appear in the early records of New England as the following census records from Connecticut prove.

In the Quarterly for June 1981, Whole No. 114, we published a record of all persons named Sparks (including Spark and Sparkes) found on the 1850 census of Connecticut. there was a total of 143 individuals. Here we provide a record of persons named Sparks (including Spark and Sparkes) found listed on the federal censuses of Connecticut prior to 1850, from the first such census taken in 1790 through 1840. Although a listing of the Sparkses in 1790 appeared in the very first issue of the Quarterly (March 1953) and that for 1800 on page 2041 of the September 1978 issue (Whole No. 103), we are repeating those here to make the present record more complete.

The first census taken by the U.S. government was that for 1790. This counting of the population has continued at the close of each decade since then. Until the 1850 census was taken, only the individual considered to be head of a household was recorded by name. Then, following the person's name, all members of that household, including the head, were enumerated by sex and age groups.

In the first census of 1790, all females were counted in one category, but males were divided by age: Those over 16 and those under 16. In 1800 and again in 1810, however, both males and females were enumerated in five different age groupings. In 1820, there was a sub-category provided for males between 16 and 18, although any male of that age was also supposed to be counted again in the category of 16 to 26. (Many census takers were confused by this and failed to list such males in both columns; the intent of the government was, of course, to measure the potential military strength of the country.)

For the census of 1830 and again for 1840, 13 different columns were provided for enumerationg both males and females, the first four by five-year age groupings, the rest by decade.

When using these early census records for genealogical research, it is important to note that the person named as head of the household was also enumerated. Thus, as we look at the 1810 census of Connecticut, for example, we know that the Reuben Sparks in Glastonbury, in Hartford County, was the same male who was then enumerated as "45 years of age and upwards" since there was no other male in the household. In the case of David Sparks in East Hartford that same year, on the other hand, we cannot be sure whether he was the male aged between 26 and 45 or whether he was the one over 45. The likelihood is that he was the one over 45, but we know, for example, that when an aged father was living with a son, the son was likely to be considered as head of the family. We must remember, also, that it was the "household" that was enumerated and that this could include domestics and "hired men" employed by the head of the household as well as roomers, relatives, in-laws, neighbor children taken in as orphans, step-children, etc. In most instances, however, when we find an adult male and an adult female of similar age enumerated with several children, we can hypothesize that the adults are the parents of those children. Without other sources of evidence, however, the census record itself does not prove these relationships.

Males headed most households in the decades included here. Where a woman's name does appear, we can make a reasonable guess that she was a widow, especially if there were children enumerated in her household. If the heads of two households bearing the same surname were listed by the census taker on the same page, the chances are great that they were related, especially if the surname is unusual. But again, we can only speculate in the absence of other supporting evidence. We must also keep in mind that, while the census taker was paid by the U.S. government to cover a specified geographical area and that he was charged with being thorough and accurate, mistakes were made. Whole households were sometimes missed, and sometimes the same household was counted twice. Names were sometimes misspelled, especially if they were unusual or if the census taker were accustomed to a different spelling from that used by a given individual (Sparkes vs. Sparks, for example), and individual family members were sometimes omitted or incorrectly enumerated. The census taker normally had to rely on the information provided by whom ever was home on the day that he visited a given household. Even a parent might not remember whether a child was over 16 or under 16.

While census data must be interpreted with care, and with many conjectures and tentative conclusions, these records still constitute a most valuable source for genealogical research.

SPARKS HOUSEHOLDS LISTED ON THE 1790 CENSUS OF CONNECTICUT

(Editor's Note: From other sources, we know that the two men living in Killingley Town were named Sparks, not Spearks.)

SPARKS HOUSEHOLDS LISTED ON THE 1830 CENSUS OF CONNECTICUT

(Editor's Note: For an explanation of the 1850 census of the United States and the information taken from that census for publication in the Quarterly, see the explanation provided on pages 1712 -1713 of this issue of the Quarterly.)