Introduction: In this article, Jessica Edwards gives tips for evaluating the accuracy of the information you uncover in your family history research. Jessica has had a lifelong interest in her family’s history – especially on her father’s side, which goes back to the first settlers in Pennsylvania, Jamestown and New England – and has documented and added more than 27,000 people to her family tree!
A problem genealogists face is evaluating the accuracy of their sources of information. To help with this, I have put together a list of the problem types and ways to possibly avoid them.
Introduction to Source Problems in Genealogy
Information found in historical or genealogical sources can be unreliable, and it is good practice to evaluate all sources with a critical eye. The knowledge of the informant (or writer); the bias and mental state of the informant (or writer); the passage of time; and the potential for copying and compiling errors, are all factors influencing the reliability of genealogical information.
What Is the Knowledge of the Informant about the Event?
What exactly is the knowledge of the informant? The informant is the individual who provided the recorded information, and genealogists must carefully consider who provided the information and what they knew. In many cases the informant is identified in the record itself.
For example, a death certificate usually has two informants: a physician who provides information about the time and cause of death; and a family member who provides the birth date, names of parents, etc. When the informant is not identified, one can sometimes figure out the information about the identity of the person by careful examination of the source. One should first consider who was alive (and nearby) when the record was created (such as neighbors on the census). When the informant is also the person recording the information, the handwriting can be compared to other handwriting samples.
Who Was the Informant?
When a source does not provide clues about the informant, genealogists should treat the source with caution. These sources can be useful if they can be compared with independent sources.
For example, a census record by itself is not completely reliable because the informant is unknown. (Was it a family member? A neighbor? If the latter, how well did they know the family you are researching?)
However, when censuses for several years concur on a piece of information, it is likely that the information in these censuses was provided by a family member or other informed person. On the other hand, information from a single census cannot be confirmed by information in an undocumented compiled genealogy, because the genealogist may have used the census record as their only source – and might therefore be dependent on the same misinformed individual the census taker relied upon.
What Was the Motivation of the Person Supplying the Information?
The motivation of the informant is a consideration. In some cases, individuals who had knowledge of the true facts intentionally or unintentionally provided false or misleading information. A person may have lied in order to obtain a government benefit (such as a military pension), avoid taxation, or cover up an embarrassing situation (such as the existence of a non-marital child). A person with a distressed state of mind may not have been able to accurately recall information. Many genealogical records were recorded at the time of a loved one’s death, and so genealogists should consider the effect that grief may have had on the informant of these records.
When Was the Information Recorded?
The passage of time often affects a person’s ability to recall information. Therefore, as a general rule, data recorded soon after the event is more reliable than data recorded many years later. However, some types of data are more difficult to recall after many years than others. For example, one type of information especially prone to recollection errors are exact dates.
Was It an Original Source or a Derivative Source?
Genealogists must consider the effects that copying and compiling errors may have had on the information in a source. For this reason, sources are generally categorized into two types: original and derivative. An original source is one that is not based on another source, and a derivative source is information taken from another source.
This distinction is important because each time a source is copied, information about the record may be lost and errors may creep in from the copyist misreading, mistyping, or miswriting the information. Common types of derivatives include: photocopies, transcriptions, abstracts, translations, extractions, and compilations. In addition to copying errors, compiled sources (such as published genealogies and online pedigree databases) are susceptible to identification errors and incorrect conclusions based on circumstantial evidence (I can’t count the number of family trees I have seen that list grandfathers being born to their sons, or daughters marrying their great-grandfathers or uncles).
Identity errors usually occur when two or more individuals are assumed to be the same person. Circumstantial or indirect is evidence that does not explicitly answer a genealogical question, but either may be used with other sources to answer the question, suggest a probable answer, or eliminate certain possibilities. Compilers sometimes draw hasty conclusions from circumstantial evidence without sufficiently examining all available sources, without properly understanding the evidence, and without appropriately indicating the level of uncertainty.
Whenever possible, choose the information from an original source over a derivative one and be sure to DOCUMENT, DOCUMENT, DOCUMENT your sources. If there is a difference between my sources, I will list what the differences are and why I went with the one I did (this allows someone else examining the documents to decide which they will go with). An example of this just happened with one cousin who had a birthdate on his tombstone that was one year and two months different than his obituary and death certificate and Social Security Death Index (I went with the date that occurred most often).