Another word on “Evidence-based” and “Conclusion-based” genealogy software use

Recently I was researching a client’s ancestor and I found myself thinking about the recent discussions on “evidence-based” and “conclusion-based” genealogy software use. This case perfectly illustrates my difficulties in using genealogy database software. The case has not yet been concluded, so I am unable to publish the specifics here.

In this particular case my client’s ancestor (“S. B.”) was born ca. 1764-1765, based on analysis of several sources, including pre-1850 federal census records, the 1850 federal census, a Revolutionary War pension deposition, and an obituary.

A man with the same name was identified as a son in the 1762 will of the presumed father (“M. B.”). Obviously this immediately presents a problem, but the census records are vague enough, and the other sources late enough in life, that the possibility of the date of birth being pushed forward a few years remained, so it was possible that S. B. could have been an infant when his father wrote his will.

The most important evidence came when examining the other probate records related to M. B.’s estate. S. B. served as the administrator of the estate, in 1762-1763. By checking a 1759 compilation of laws in effect in the state, I confirmed that a person appointed executor or administrator of an estate had to be at least 17 years of age to serve as such. This would mean that S. B., son of M. B., had to have been born no later than 1745. In other words, there was simply no possibility that my client’s S. B. was the son of M. B.

Writing the report in MS Word, it is very easy to quote the relevant portion of the probate law, and cite the law book. This law provides the crucial evidence to prove that M. B.’s son S. B. was born before 1745. No other record provides this information, either directly or indirectly.

In a genealogy software program, how would one:

(1) enter a “fact” or “event” for S. B. or M. B. to reflect the existence of the probate law?

(2) cite the probate law?

I am sure that there is a way, and if I relied on genealogical software in my research, I would have to figure it out. But I have a feeling that it would be a bit convoluted, whereas it is much easier to accomplish simply using a word processor with footnotes.

The next question is then–how would this be handled differently by an “evidence-based” software user and a “conclusion-based” software user?

I am open to all comments.

If you would like to cite this post: Michael Hait, “Another word on ‘Evidence-based’ and ‘Conclusion-based’ genealogy software use,” Planting the Seeds: Genealogy as a Profession blog, posted 23 March 2012 ( : accessed [access date]). [Please also feel free to include a hyperlink to the specific article if you are citing this post in an online forum.]

6 thoughts on “Another word on “Evidence-based” and “Conclusion-based” genealogy software use

  1. I’m not a fan of the “evidence based” vs “conclusion based” software user terminology that is making the rounds of late, nor do I consider myself to be in either camp. If anything, I continue to experiment with several options, and am moving more and more toward using Word documents (which I attach as needed into my database).

    That said, I don’t see it as that difficult, at least in RootsMagic and Legacy, to do what you are asking. Since the issue is the birthdate, I would attach the probate law citation to that “fact” and write a birth note detailing my reasoning. Or, since both programs allow one to create any type of fact/event, I could also create a “Probate Law” fact, summarize the law under that fact, and create a citation to the law.

  2. Michael,

    In my gen software I’ve created a couple of tags that get used *very* frequently: Legal Context and Historical Context. But my generalization, here, needs context of its own.

    Like you, I prepare a research report at the end of every block of research. That step, done in a word processor, is essential whether the report is for a client or for sharing with cousins or for my own files.

    This Research Report is my go-to place every time I need to know exactly what research I did in a particular area or set of records. It tells me exactly who I searched for, exactly the variant name-spellings I included, exactly what “known information” or assumptions served as the basis for that block of research, exactly what was searched, exactly what yielded negative results, and exactly what I found—on everybody. And, just as important, it carries my analysis of each thing I found, as well as the legal and historical context that is critical to the analysis.

    This Research Report is my “big picture.” Genealogical software, by comparison, requires us to pick apart our data and parcel it out among pigeon holes—squirreling away each little nugget into an individual file.

    Both approaches are valuable. Both work together just fine. In the case you’ve raised—how to handle laws and the discussion of laws within genealogical software—my approach is this:

    When my Research Report is done, my data entry begins. Every abstract or transcript that appears in the Research Report can be quickly cut-and-pasted into the software—whether that “research note” deals with a deed or a law. Because my software is built around “events” in individual lives, the data entry usually takes this path:

    1. Basic Tag & Sentence:
    Under Whatever Individual, I choose the appropriate “event” tag and create the obligatory “event” sentence. If there’s a law that needs to be discussed, I use my “Legal Context” tag and proceed as if it were an event. After all, the passage of the law *was* an event relevant to my ancestor.

    To use the hotly debated terms cited in your article title, I would consider this data to be *information.* It might be *evidence.* But I certainly would not call it a *conclusion.* Different documents often provide contradictory information and my database needs to keep track of it all.

    2. Narrative:
    If I want my genealogical/biographical narrative to include more details about this (as I typically do), I’ll enter that narrative in whatever screen the program provides for entering narrative that will show up in the “life stories.” However, I also recognize that what I write here is a *summary.* It’s an overview that reflects my *interpretation* of what happened. I may include a brief abstract or a quotation from the document, to make a critical point; but my wording is still shaped by my assumptions as to what transpired, and whatever I write here remains subject to change. Often, the narrative that goes here is a cut-and-paste of analyses done within my Research Report, where the freeform style of a word processor allows flexibility to organize material in all kinds of ways and better develop my thoughts.

    To use again those debatable terms: This narrative summarizes the “evidence” from that document. It often compares it to contradictory “evidence” in other documents, and it presents my very temporary “conclusion.”

    3. Source Citation:
    After creating the “event” and its discussion, I open my software’s source screen (freeform) and paste in the appropriate citation already created in the Research Report.

    4. Raw Evidence:
    My software has two screens attached to the Source Citation screen for an “event.” One is Research Notes. The other is Comments.
    – In that Research Notes screen, I do another cut-and-paste from my Research Report. I paste in the full abstract or transcript, with every nuanced detail that the document offers.
    – In that Comment screen, I paste in whatever observations I made about that document in my Research Report.
    In other words, the “Research Note” presents the raw evidence. The attached “Comment” presents my thoughts about that document, which may or may not rise to the status of a “conclusion.”

    5. Associates:
    At this point, if the document involved other individuals, I can do a simple copy-and-paste of all the above into the file for each associate. My software allows copying the Source Citation, Research Notes, Comments all in one step.

    6. Composite Research Notes:
    My software also provides a person-based “Research Notes Report” that will automatically consolidate all my raw evidence for that person–and my comments upon that evidence–in chronological sequence, with full documentation automatically attached to each abstract, transcript, or comment.

    This is my other, major go-to document. Here, I can analyze everything I have on a person, every exact detail, exactly the way the document presented it. This is where I can compare the exact detail from one document to the exact detail in all other documents found for that person. This is where I see patterns between details and associates. And, because it’s in a timeline format, it’s easy to see holes in my research.

    Within this framework, I would have great difficulty in trying to say that software should be “evidence-based” or “conclusion-based.” To me, the two are inseparable and circular, like a chicken that created an egg that created a chicken that created an egg.

    IMO, the underlying cause of the debate is a fundamental problem with expectations from software: that is, that all “reporting” can or should be within a relational database that requires us to isolate bits of data and pigeonhole it into predetermined screens. I love my genealogical software, but it’s not the only tool in my tool kit!


  3. This is tricky. The fact here that is being disproved is the parenthood of M.B., with the birthdate a piece of evidence. In RootsMagic, while one could put “bef. 1745” as a birthdate entry, cite the will and probate and mark it as disproven, that doesn’t tell the whole story. What is really being disproven is the idea that M.B. was the father. To handle this in the same fashion you would need to add M.B. to the tree and mark his relationship as disproven. RootsMagic does not let you mark relationships as disproven, though, and even if it did it is unclear if you would want to clutter the tree with dead ends.

    Myself I tend to use a combination of genealogy software and Evernote. If I’m collecting facts from documents for relationships that are I will use RootsMagic. If, however, I’m working on a brick wall and am gathering lots of speculative possibilities, I use Evernote. That way I can put each document in it’s own file and link to them from one or more master notes, formatted in whatever way fits the investigation I’m doing.

    So, for example, recently I was researching the family of Sheldon Nichols of Ontario. I was going off of census records, published histories, and other Internet records. I did this in RootsMagic in a new tree. It worked great.

    I am also, though, trying to track down his father David Nichols. This is much more difficult because I don’t have a birth date or place of birth. For this I’m collecting information on a bunch of David Nichols. A tree structure does not help here, so instead I save the documents I find in Evernote and and link to them from a spreadsheet and several other notes where I can arrange the various candidates in ways that are more useful.

  4. Is there any reason you cannot use software to create a tree for the WRONG person this would enable synchronizing data with proper person.
    On the research tree you can put a note in name section (Rootsmagic) or in general notes for Family Tree Maker this person is not son of XyZ person.
    Source for this fact can be the separate ‘Rebuttal Tree’.
    If you wish to create Tree with father in it as optional father then in Rootsmagic in proof section in right side window you can select disputed from drop down menu.

    I presently use software as it helps me to organise my research trail and document as I go. At the conclusion or at budget limit I will give client I report done in Word detailing information given by client, information sought by client and then my findings both positive and negative. Client then decides if they need more or it is time to settle the account.
    I offer the family tree(s) done and Gedcom and always provide copies of all records relevant to the research in digital and hard copy (if requested)
    How we do the research and present it really is a matter of knowing our client and asking how they might like the information given to them.
    I have clients who are happy to get the information in note form via E-mail, others want a tree, others want to just be able to understand what it is I am saying and how it fits with what they already know.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s