Defining Source Categories

#Gigatrees supports adding #source categories to your existing sources using the <Sources> option. Gigatrees also supports source categories when defined by #RootsMagic using the non-standard GEDCOM quality records ( _QUAL._SOUR and _QUAL._INFO ). When defined, source categories are shown on the source pages, in all source reference tooltips, and in the sources table of contents.

The purpose of defining source categories is to help determine the possible types of bias that may overshadow the claims the source makes. Source categories are used internally by Gigatrees to help calculate Reliability Assessments.

The #Genealogy Proof Standard supports 5 source categories: original, primary, secondary, derivative and negative. Gigatrees supports 7 additional source categories: dna, transcript, copy, abstract, research, memoir and unknown. The source categories are translatable so that you can enter values in your native language.

The source categories concerning concurrency, ( primary, secondary ) represent when the source document was created in relation to when the documented event occured. Primary represents that the document was created at the time of the recorded events (such as a birth certificate) and secondary represents that the document was created some significant amount of time after the events occurred (such as a transcription of a birth certificate).

The source categories concerning authority ( dna, original, transcript, copy, abstract, research, memoir, derivative, unknown and negative ) represent the authority of the source document recorder. When adding a source containing yDNA or mtDNA test results that prove a biological relationship, the dna category should be used. It would be inappropriate to use this for autosomal DNA test results unless those results were combined with an in-depth analysis showing why they could be used as proof (See the <DNA> option). Original represents a document in its original form (such as a birth certificate). Transcript is a direct unaltered transcription of an original source. Copy is a subsequent copy of a transcription or of another copy (i.e. copy of a copy). Abstract is any abstraction or summary of the original source (not in its original form). Research is an authoritative collection of claims derived from multiple sources. In order for the collection to be considered authoritative, the researcher or genealogist should have included the relevant source citations. Memoir is a collection of authoritative claims taken from memory, such as an autobiography, letters, notes, etc. Derivative is a non-authoritative collection of claims derived from multiple primary and secondary sources. This will include most books, town histories, etc. Unknown is a source from which no information as to its authority can be determined. Negative represents a source that is known to be error prone.

Claims that are both primary and original are generally considered to be unbiased, though not necessarily without error. Secondary sources may suffer memory bias due to a time lapse occurring between the event and the recording as well as from copy errors. Derivative sources may also suffer from these as well as author bias, intentional or otherwise. All sources may contain lies.

From experience we know that certain types of records are more accurate and have less bias than others. For instance, birth certificates ( original, primary ) almost never have incorrect birth dates. Baptism records ( original, secondary ), even those recorded within just a few months of birth, may contain incorrect birth dates caused by memory lapses of the mother, as astounding as that seems. Death certificates and tombstones ( original, secondary ), which usually occur many years after birth and the informant is generally not the mother, often contain incorrect birth dates. Census records ( original, primary ) are notorious for having inaccurate ages due to the fact that the informant was more than often not the head of household, and who, especially early on, had little formal education resulting in date subtraction errors or simply guesswork. Had the census authorities required birth years instead of ages, they would probably be much more accurate. Census records, however, are very accurate for documenting parental relationships, so there you go.

Categorizing sources can be very tricky and is sometimes more art than it is science. Gigatrees therefore allows you to not only categorize your sources, but to also override the Reliability Assessments derived from those categories for individual claims when necessary.

For many sources, multiple source categories might apply. When determining the appropriate source category to use, you should choose the least authoritative category that applies to the claims being extracted from it. For instance, if you have a source that includes scans of original documents, along with some memoirs and derivative accounts, but you are only using the images of the original documents from that source, they you can rightfully choose original as your source category. If however you include claims from the other areas of the source, then you should choose a different source category. A serious genealogist might, in this case, split their single source up into several source entries, each with their own category.