As a genealogist conducts research to fill in data gaps or to flesh out the basic information with more facts and history, conflicts in data often occur. Experienced genealogists sort through these conflicts by evaluating the quality of the data and the probably source. Birth dates seem to be most susceptible to conflicts.
Examples of Conflicting Birth Data
My own genealogy research has turned up two excellent examples of how footprints left by an ancestor can result in conflicting birth dates. Seth Boynton Cheney was one of my wife’s great-grandfathers. Family lore said he was born in upstate Vermont, went west as a 49er at age 16, worked his way east to the Texas Panhandle and eventually to southwest Kansas, where he died in 1907. if the family lore (passed down to descendants of two of his children) was correct, that puts his birth year around 1832-1833.
But what does the evidence indicate? The attached figure shows all the data available: Vermont town records, Federal and state censuses, voter registration, death certificate, obituary, and tombstone. Based on his age on the various documents, Seth could have been born any time between 20 October 1832 and 1837. The different documents do not indicate any convergence as to a probably date. His death certificate is a government document. We would like to believe it. But how to reconcile this versus other data?
The same situation exists with Seth’s wife, Sarah Matilda Best. This figure shows the data available for her birth date. Note that her death certificate has not yet been obtained. The variety of dates are as varied as they are for her husband, with no convergence toward a probable date.
Evaluate Data – Time Between Event and Record
Normally the document created closest to the event is considered the most likely to be accurate. Thus, when trying to pin down an ancestor’s birth date, a birth record by a government agency would be expected to be more accurate than a death notice. For Seth Cheney, the birth record entered in the town record book was 14 years after his birth. The first census that lists him, 1860, was 28 years after the event, and the death certificate was more than 70 years after. Thus, a birth year of 1832 should be considered most probable.
For Sarah Best, the 1860 census is the record closest to her birth. Later censuses and the tombstone were created much after the event, and would be considered less accurate. So a birth year of 1858 should be considered most probable.
Primary Sources vs. Secondary Sources
The definition of primary sources and secondary sources in genealogy vary. Some say a primary source is one close to an event, while a secondary source is far removed from the event. So by these definitions a tombstone would be a secondary source (except for an infant) and a birth certificate or Bible record a primary source.
Another definition is that a primary source is an original document while a secondary source is transcribed from a primary source. By these definitions, a census book would be a primary source, while a transcription of a census book into a compiled database would be a secondary source.
By whatever definition, a primary source should always be given more weight than a secondary source. The figures accompanying this article are obviously secondary sources. I could easily have made transcription or typing errors. So although I give much credence to them, since I did the research and know how meticulous I was in compiling them, another genealogist should use them with caution. In the examples, those documents that are primary sources should be given the most weight in pinpointing a birth year.
Evaluate Probable Source of the Data
One other piece of information that can be used to evaluate the quality of the data is who probably put the information on whatever document you are looking at. In the case of a death certificate, it would most likely be the surviving spouse of one of the children. It could also be a neighbor if no relatives were at hand. In the case of the census, it would likely be either the husband or wife. For a Bible record or government-issued birth certificate, it is almost always a parent (for the either) or, in more modern times, a doctor.
Looking at the discrepancies for the birth of Seth Cheney, the information on his death certificate was probably given by Sarah, who would only know what Seth told her. In the case of his government birth record, it was most likely a parent. A check of other documents shows that on the same day that 14 year old Seth’s birthday was recorded, his father was at the courthouse to record a deed of sale. He was about to move to another town, so it seems he decided to have the city clerk record the births of all his children born in the town he was leaving.
The time duration between the event and documentation, and the person likely providing the information on the document, are two powerful considerations when deciding between conflicting data.
David A. Todd - David Todd is a civil engineer, a genealogist, a citizen concerned with the environment, and a writer by passion.
Join the Conversation