How should I record and report information which conflicts (e.g. differing birth dates for what I believe to be the same person)?
I have spent a lot of time thinking about conflicting and incorrect data and have come to the conclusion that conflicting data is not a bad thing. Indeed, the conflicting data may be the very thing that leads to more information about the individual.
Given the example of conflicting birthday dates, consider the source of the conflict. It may by a simple error of two people entering the date differently, a lot of sources are hard to read after all. Another possibility is that there are different documents with the conflicting dates. Lastly, one of the date made be forged, made up, best guess etc.
Regardless of the source of the conflict, the conflict exists and is codified in documentation and research to one degree or another. Once you come to a conclusion which is the right date, the temptation is to toss the "incorrect" one. This is wrong for a few reasons.
First, you may decide down the road that the one deemed incorrect is really the correct one causing you to have to go back and possibly research again. Second, some records will have one date and some the other. By keeping both dates it makes relating to these other records easier.
It's very important to keep track of all sources, and to understand whether these sources are primary or secondary. A good primary source for a birth is the birth certificate or a record of a birth. A death certificate is a secondary source for a birth because those facts were not necessarily provided by a reliable source.
Sometimes the same original information is republished in many places, which does not make those sources independent. So you can't trust a majority vote either.
If at some point you do find conflicting information (and you will!) having kept the source references for each fact will help you interpret the new (or the old) information. Many tools for creating and maintaining your family trees provide mechanisms to import or to create source information for facts associate with people in your trees. You should familiarize yourself with how these work, and record as much as you can. Relying on memory is not a viable option.
Sometimes "conflicting information" can be evidence of something else and worthy of recording in its own right.
I have a date of birth for my father from his Naval enlistment that does not match any other document. Like thousands of other men, he lied about his age to go to war. I recorded both dates as "birth" events with appropriate annotations on each. Knowing that he falsified his birth details tells me something important about his life.
More significantly, I now have an original "Certified Copy of Entry in the Register Book of Births" with a date of issue between the time he applied to join and when he shipped out. I infer that someone (his mother?) was attempting to prove that he was not eligible to serve.
At face value, the d.o.b. on his service record was simply "wrong". In fact it opens up several rich stories in Family History. I need to preserve those opportunities.
There's a mention somewhere in this thread about "multiple entries for the same fact". I think this terminology helps confuse the core, and often asked, question.
Your research may be citing multiple sources. Each source will usually yield one value for each datum (e.g. date of birth, age, personal name, place of birth) but none of these should be described as facts. They're all items of evidence that may not agree 100%. Small variations are common as in ages on a census or misspellings of a personal name. Other variations may be deliberate or accidental errors, or it could mean that you're actually dealing with more than one distinct person. All that evidence is therefore important and should be kept, together with references to the associated sources.
What you record as the primary data for your person are really conclusions formed by assessing the sum of the evidence available. Taking date-of-birth as a specific case, you will probably want just a single date-of-birth to be displayed. This doesn't mean that the conflicting dates or ages have been lost or discarded - they're the evidence from which you selected your primary value. Hopefully, you would document how you reached your conclusion if there was any conflict.
A more difficult situation is where you may have formed a finite set of conclusions rather than a single one. For instance, someone born in 'November of 1853 or 1854' but the evidence is not yet sufficient to chose one. This could be handled by picking an arbitrary one, or selecting an enclosing range, but adding a very important note about the actual possibilities.
When dealing with conflicting information the number of source you have will matter as well as the source itself.
For example if you have two conflicting dates from two sources, source A is from a birth certificate and source two is from a newspaper article, I would list the birthdate on the birth certificate and in the notes on the person I would list the newspaper article and the 2nd possible date. The birth certificate being a more authoritative source I would trust it over the newspaper.
If you have three newspapers, two of which list the same date and a third listing a different date I would list the date from the first two papers and in the notes of the person I would list the 3rd newspaper.
As beachbuddah mentioned its not just about the number of sources listing the date. If I found a birth certificate and three newspapers, all of the news papers listing a different date from the birth certificate I would list the date on the certificate and keeps notes on the newspapers.
It is important to always keeps notes on all the sources you find. If in the future you find the birth certificate to be erroneous in some way or perhaps another document that would be more such as a church record that would match others in your notes you may change what date you list for the persons birth date.
Never throw anything out either. If you find a source to be false, keep the source in your notes and mark it as false with the reasons for doing so.
Some software used to maintain family trees have features to support multiple entries like this, each with its own source. Some software known to support this includes:
- Mac Family Tree
- Family Tree Maker
I have found that I can't rush the past. If I have conflicting information, it's incumbent on me to continue my research. I can't get overly excited about a majority of a given date. Just because there are more of one than another, that doesn't make them right. Could be that they were all copied from the same erroneous source.
And that's the key, finding the source closest to the original date you're trying to pin down. The further from that yesterday and the closer to today, the more likely that the date may not be the correct one.
Hope this helps.