Every researcher runs into conflicting evidence at some point. This is one reason we must know where our information has come from to compare the reliability of the sources. The degree of reliability of a source is usually a factor of:
(1) How soon after the event the record was created.
(2) The credibility of the person who made the record.
(3) The reason why the record was made.
(4) Corroboration with other evidence.
I personally have three forms that I use when working with my sources. The first form is a Research Calendar which lists the Researcher, the Ancestor, Locality, Time Period, Problem Stated, Search Date, Where Available, Call Number, Title/Author/Publisher/Year or Record Identification Information, Notes, and Page numbers. The second form is a Research Log which lists the Problem, Action Needed, Institution, Types of Record Needed, Results and Suggestions. The third form is Analysis Sheet, Name of Ancestor, Birth Date, Reference Number: What I already Know ...And This Brings Me To Conclude that... This way I do not waste my researcher time nor the person that is helping me to find my sources for my conclusion.
The transcript of a death certificate is not as reliable as a photo of the original because it was made later and there is the possibility of an error in the transcription. If two people tell the same story but they all learned it from the same original source then they do not corroborate each other. A marriage record signed by a minister or priest is excellent evidence of the date of marriage but, if it includes the ages of the couple, that information is less reliable because it is based on what someone told him, not on his own knowledge.
Tombstones and censuses are very useful sources of information they also have errors and mistakes. This again are records that are based on what someone told him, not on his own knowledge. Some census is based on uncorroborated statements by people who may not know when they were born or simply didn't care, or don't want to admit their true age. In any event the useful data they contain needs to be corroborated by the original documents.
Again, I cannot express how important it is to always document your work and look for several (rule of thumb) at least three sources.
Communication among family researchers sometimes has a tendency to the overuse of acronyms in their research. Again, this can be quite confusing when documenting your sources. It is best not to use these. Not everyone knows what they mean and most are not in the dictionary and could pose problems for others using your documentation. Here are a few that have caused problems for myself: LNU - Last Name Unknown, FNU-First Name Unknown, NMN-No Middle Name, ILMO-In Loving Memory of, SKS-Some Kind Soul, NSF-No Spouse Found.
I welcome comments, queries, and suggestions at: firstname.lastname@example.org attn: Ms. Sylvia.