As I understand it, the Genealogical Proof Standard (GPS) is the backbone of Genealogy today, and yet it has not been mentioned in a question asked here for over three years, and only in 20 questions and answers total.

At the same time the number of questions per day asked here continues to usually be around or slightly more than 1 per day, and rarely has it touched 2 per day, so I think we need some new ideas for how to increase that.

For a long time I have been wanting to become better at routinely applying the GPS to my own amateur research, and have been wondering how I can use the time I spend here to do that.

Although I will talk about the GPS, @ColeValleyGirl has pointed out that the GPS is not the only way to perform rigorous genealogical research so anything I say about the GPS can be read as applying equally to UK Society of Genealogists' formulation of Standards and Good Practice.

I think a lot of our users start researching their family history by wondering who their ancestors are, then start to quickly flesh out a family tree using hints from Ancestry.com etc, and come up with some elaborate and romantic hypotheses for where they came from. The one thing that is usually lacking in these trees, and lacking in a big way, is evidence to support the many hypotheses that each tree depends on if it is to be considered more than a shadow of reality.

It is here that I think we can help those users and ourselves to become more rigorous in our genealogical approach by endorsing the asking of a new "style" of question.

This new "style" is in no way intended to become the norm for questions here - because that would certainly alienate many of our potential users - it is intended as just another style of question and might offer a path to a form of Peer Review.

As an example of the Q&A style that I am thinking could lead to conclusions that might meet the Genealogical Proof Standard, or at least to the level of something that might be considered an early/advanced draft of something that could, I have now posted this self-answered question, and invite other answers to it:

Proving parents of Thomas Morgan of Tre-bryn, Pencoed, Coychurch, Glamorgan, Wales about 1808 were Thomas and Ann? .

Does this idea have any merit? Would asking some questions in this way help/hinder our efforts to build a site with more and better Q&As to attract the best genealogists and family historians that we can to our site?

Having recently viewed (at @JanMurphy's suggestion) a webinar by Warren Bittner called "Proof Arguments - How to write them and why they matter", I am thinking that a question could perhaps be structured along the lines of:

This is my proof argument tying Thomas Morgan to his birth family written "Bittner-style", have I made my case sufficiently?

<insert proof argument here>

with "Yes" being indicated by votes on the question and/or answers to that effect, with other answers being "No, because ..."

3 Answers 3


I'm all in favour of improving the quality of genealogical research, but believe strongly this approach won't achieve that objective, nor the objective of improving the question volume around here.

You run the risks of:

  1. alienating beginners, if they perceive such a stringent approach is the norm here;
  2. alienating those who don't follow the US-based BGG GPS (the UK Society of Genealogists has a different formulation of Standards and Good Practice which overlaps with but is not the same as the GPS)

I think that providing subtle and well-targeted guidance/editing for individual questions matched to the apparent experience-level of the OP is a much better way to go... which doesn't preclude a very experienced OP providing the level of rigour you're proposing. In some ways, it's veering towards the 'peer review' we were discussing here 3 years ago: https://genealogy.meta.stackexchange.com/a/1417/104

  • Yes - it is definitely along the lines of your Peer Review post. I've updated my question to try and make clear that alienation of any potential users of our site is something that I know we must always be very careful about. The "format" I am thinking about is not intended to supplant any of the ways we ask and answer questions at the moment but to offer ideas on how an extra, more rigorous, set of questions could be added to our site.
    – PolyGeo Mod
    Mar 2, 2016 at 22:07

I support this idea as long as the questions and proofs are difficult and interesting.

If we are not careful, I think we risk ending up with lots of questions about random ancestors that are not really useful to anyone except the OP.

What I would propose is that if one posts such a proof, it utilises a unique or interesting resource, or newly published database, or is a very difficult problem you've solved, or contains something else that could inspire someone else's research. We don't need lots of questions about people browsing exhaustively through BMD indexes and censuses, because (hopefully) most people on this site know how to do that already. What might be useful is a proof using an underutilised resource, such as (thinking of your recent post, @PolyGeo) settlement examinations.

For your example, getting the birth certificate of Grace Martin Wills would probably be 99% of the proof to the question given. I could be wrong, but I don't really see what more to the proof there would be unless you suspect the birth certificate either intentionally or accidentally contains false information. Please correct me if I'm wrong, but for this reason I don't think that is a particularly good example of a proof Q&A that would be useful to the community.

I also don't think the structure of having people vote up short answers like "I agree with your proof" is particularly useful. I think it would be better structured if in your question you can explain the dilemma or problem you are trying to solve, and then self-answer with the proof as an answer. The question itself need not be long but ought to include the relevant background information. People with substantial suggestions (more than just a comment on the proof) could add that as another answer, or could be encouraged to expand and rework the proof as a separate answer. Upvotes of the proof answer(s) are implicitly community endorsements of that proof, and after a reasonable time of voting the OP could decide whether to accept their own proof or the suggestions/proof contributed by someone else.

I do not purposefully and routinely use the Genealogical Proof Standard, although in the course of my research I end up (consciously or not) covering all the components. As long as the relevant parts of GPS are all there I don't think there is a need for the proof to be structured too much as a "template", because that tends to get boring to read. We don't want boring questions and answers because that certainly will not encourage people to participate. As you mentioned, I think it would be fine to include citations throughout, or at the bottom of the post, rather than listing them right in the middle as Step 2.

Questions in this format could perhaps be given a new tag such as .

  • 1
    I think your answer nails several of the weaknesses I was feeling with my proposal. Not using this "format" for questions that a birth certificate would quickly resolve makes a lot of sense. As you seem to have realised, this question is the one that got me re-thinking along these lines. I'll wait to see if there are other answers before accepting, and then try a "GSP as an answer" on one of my questions. I was thinking of coining a genealogical-proof tag too:-)
    – PolyGeo Mod
    Feb 28, 2016 at 0:18
  • 1
    @PolyGeo Just noticed we already have a proof tag, though I must admit I'm not sure exactly what it's used for
    – Harry V. Mod
    Feb 28, 2016 at 1:32
  • Perhaps genealogical-proof is the only proof tag that we need on this site but before testing what others think on such a tag merge, am I on the same page with you that application of the genealogical-proof is the indication that the author (along with later editors) believes and wants a particular Q&A to be rigorous [GPS level] about any evidence and conclusions presented in it?
    – PolyGeo Mod
    Feb 28, 2016 at 5:50
  • 1
    @PolyGeo Yes, I think that tag should be reserved for questions for which the main goal is formulating a rigorous proof. If there are existing unanswered questions maybe they could be reworked to be in this format but I don't think it makes sense to modify answered questions to make them into a genealogical-proof question – it would be better to ask as a new question.
    – Harry V. Mod
    Feb 28, 2016 at 12:46
  • I've just posted an example GPS Q&A as genealogy.stackexchange.com/questions/10419/… It is a relatively simple example so that it can act as something that a large percentage of our users can easily absorb the details of and see through to the style of Q&A it portrays.
    – PolyGeo Mod
    Mar 5, 2016 at 10:51
  • @PolyGeo I think that question shows how this maybe isn't a great format for this site, because now that you've updated the proof (much improved now!) it makes a lot of what Jan and I said redundant. Small refinements are great but major reworkings of an answer are generally avoided (after all, people's votes don't change if you drastically change an answer). Do we now go back and edit our critique, or what? Certainly I think it is good to have self-answer Q&As pertaining to tough research problems but I'm not sure framing it as a Genealogical Proof Standard 'exercise' is the best way.
    – Harry V. Mod
    Mar 7, 2016 at 18:27
  • I think it has been a useful exercise to work through this one as a test case, and I'm in two minds as to the value of such questions based on this one example. Something to bear in mind is that I think SE general advice is that overhauling answers is encouraged, but revisions made to questions should bear existing answers in mind so as not to make them redundant. That is why I tried to only improve the details of information items in the question and placed any new evidence in my answer instead.
    – PolyGeo Mod
    Mar 7, 2016 at 21:55
  • W.r.t. your and @JanMurphy's answers I think it is your choice whether to revise your answers. Certainly, I think they remain largely applicable to both the question asked and in light of new evidence they triggered me finding in my answer. They should still accrue net upvotes (and I would not expect to see any downvotes on either). If people like my proof now perhaps I'll see a few more upvotes over time. My feeling is that we should not write off this format just yet, but certainly that first test is not an ideal example.
    – PolyGeo Mod
    Mar 7, 2016 at 22:00
  • 1
    @PolyGeo Sorry I didn't mean to come across as critical of your question/answer, and I agree we shouldn't write off the format yet. Just a bit of refinement needed. I'll add a note to my answer to the effect that the proof is now much improved, but will leave the other comments as they were.
    – Harry V. Mod
    Mar 8, 2016 at 1:03
  • No problem - everything you said was constructive - I think my Q&A became unwieldy because I did not think hard enough early enough about what parts of my content should go in the question versus my answer. I feel like I learned a lot about my Morgans and that part of Glamorgan through the injection of your and @JanMurphy's expertise.
    – PolyGeo Mod
    Mar 8, 2016 at 1:47

I think that outlining all five elements of the BCG's Genealogical Proof Standard is a huge ask for a single question.

To recap, the elements are:

  • reasonably exhaustive research;
  • complete, accurate citations to the source or sources of each information item;
  • tests—through processes of analysis and correlation—of all sources, information items, and evidence;
  • resolution of conflicts among evidence items; and
  • a soundly reasoned, coherently written conclusion.

Existing questions on the site are likely to address one of these elements. The most obvious example is one of my own examples where I asked: How can I ensure that I am making a reasonably exhaustive search? but with a little digging we could find examples for all the other elements.

As a thought-experiment, imagine if there were tags for each element:

I'm not saying that we need to run out and make these tags right this moment. But these are important concepts that we need to think about -- something that is also true of the principles of good practice put forth by the Society of Genealogists. I don't see the principles of the SoG and the BCG's GPS as competing standards -- both are helpful guides of the issues we should be thinking about.

In a future, more mature version of the site, can we imagine a scenario where it might make sense to have questions focused on these topics, or any of the other elements in other standards? Again, I stress that I'm not proposing we rush out and make tags for each of these principles -- the psuedo-tags were a suggestion to look at the problem in a fresh way.

Rather than trying to cram all five elements of the GPS in one question, it might be more helpful to focus on which one we're having difficulty with.

As far as proof statements go, a reminder: a proof statement doesn't have to be long -- sometimes a paragraph is sufficient. For simple cases, a clear, succinct statement of how you arrived at your conclusion will be sufficient.

In general, I wonder if some of us (myself included) aren't getting a bit stuck on the "let's find this record!" approach and not looking at other parts of the research process.

enter image description here

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .