Question about source vs. citation (marriage, birth certificates, etc.)

I am new to genealogy and actually firmly using Gramps. I have dabbled with it years before but only recently got enough family tree history to really start investing in it.

My question is- do marriage certificates belong as sources or citations?

I only recently started adding so I would like to do this properly from the start instead of fixing plenty of entries later. I have done a “City Building” as a repository, then “County probate court” (which is housed in that building/library) as the source, then each marriage certificate (for instance) is a citation.

But now I’m thinking the proper method is probably to put each marriage certificate as its own source even though the citation will probably be empty (or just use “today” as I’ve seen in searches.)

My reasoning is- even though there’s not much to “cite” on a one page PDF the same marriage certificate can be used to verify someone’s birth (as some of my earlier ones include the birthdate of the bride and groom). Also, it seems that it really is a “source” overall. It’ll just be somewhat messy for sorting sources by group I would think (since each would be titled differently that they cannot be grouped.)

Any advice on this?

Same could be said for birth certificates because old ones have ages of parents or professions, etc. etc.

1 Like

If I have an original paper certificate in my files, I create a separate source for it.

If I get an image from FamilySearch, for example, I create a source like this:
Title: Marriage records, v. 2-A
Author: Whitley County (Indiana). Clerk of the Circuit Court
Pub. Info: Salt Lake City, Utah : Filmed by the Genealogical Society of Utah, 1990

In the repository reference for FamilySearch, I put the microfilm number in the Call Number field. I also attach a note to the repository reference with the URL of the FamilySearch catalog entry where the film is listed.

In the citation, I attach a note with the URL of the image. I also attach a copy of the image in the Media for the citation.

Others will do it differently. The important thing is to be consistent in whatever you do so that you and others can go back and find the original record.



Besides specific Repository reports, the Repository does not print with the Source and Citation. Something to be aware of.

The Repository is: Where did I go to obtain the information. The Source is: What did I look at. With the citation: What did I find.

In your hypothetical it could be:

Repository: XYZ County Court
Source: XYZ County 1950 Marriage records
Citation: information about the marriage certificate with ideally a scan of the certificate

As much of our research is done using Ancestry and FamilySeach, etc I have these archives as the repository, the database as the source. Ideally the database will hold an image that can be downloaded. If not a transcription or a link to that specific search result can be added to the citation. On these sites there will often be something like: how to cite this record which will help.

Bottom line, you want the S/C to have the information for you or someone else to find the information again. There are as many ways to set up your R/S/C as there are Gramps users.

I also think of it as:
Repository: Ancestry, FindMyPast, FamilySearch, etc.
Source: 1861 English Census. (The same sting the repositories use)
Citation: the actual translation, image, etc.

You also need to consider if you will upload the data via a gedcom to a site such as WikiTree. How you enter data in GRAMPS will effect these uploads. Do some testing before you get too far.

DaveSch, that is a very good point. I wasn’t aware Repo info didn’t print alongside the Source and citation. That is something to keep in mind as the one method won’t show how I think it should then on a print.

Davesellers, I do appreciate you pointing out the GEDCOM points. I don’t intend to switch from gramps but it is something to keep in mind (especially if someone later on down the line decides to export it to gedcom for use in a different program.)

GeorgeWilmes- I like that you backed up the idea “do it consistently regardless.” I’ve seen this advice before. I do appreciate how you create sources. I might use that as it seems pretty smart in how to lay it out.

Thanks all. Open to any more comments but I do feel a bit better about planning this out at least. Gramps is definitely more on the “heavy” end in how much you can stuff in it but I think I’m really going to like having the ability to customize it and load it up with proper source/citation, etc.

There are a couple of things to consider, and one is, that it helps to think of the source title as a distinguishing element. Having several sources with the same title but different authors doesn’t display well, so I personally use sources like Amsterdam civil registry, or Amsterdam birth records, so that I can identify each record by its date and record number.

You can also keep in mind that the citation page field allows any kind of text, so it’s perfectly OK to put the main participant’s name in there, if you want. This is quite useful for indexes, in which you found a person (or a couple) by name.

First off, welcome!

I agree with this approach. The source will be the recordset that contains the certificate and the citation will be the individual certificate.

This holds true even with physical research. If you were to go to an archive and ask them for a marriage certificate, they would hand you a large book full of other certificates (or maybe pull it out of a large book and hand you the individual document if they were loose.

I will also second the sentiment of finding what works best for you and being consistent with it. For me, finding what works best means trying something out, running some reports to see the results, and then tweaking it. Secondly, WRITE IT DOWN!!! I can’t tell you how many times I have started/stopped/restarted my research and completely forgotten which method I had decided on worked best for me. I’ve wasted a lot of time spinning my wheels trying to remember my “process”. Now, I’ve started documenting my methodologies for why I do things the way I do. This helps me remember the process I decided on when I come back to my research and it also gives guidance to anyone who will (hopefully) pick up where I left off.



Take some time to read the help pages, there is wealth of information in there. Click on the HELP button in each screen. It may change your work flow or how you document. It can be as simple as how you enter a date, range or period of time.
There also hidden features that you will only discover by reading about each field.

I totally agree with both of these answers:

After some time finding my way with Gramps, I wrote down how I used it. It actually allows you to remember what you found as functioning but also to compare what remains to be done in your base to be in agreement with these definitions and finally to have this defined base to make possible change when you find a better idea.

As for the sources and citations resulting from this normalization, a bit like @ennoborg, I use the country code, the city code, the city then the type of register, its period and the acts it presents. It gives something like this:

When I have sub information within an element of a source, I use a pipe sign to separate what falls within the element and what falls within the citation within that element. In this case here, within the source of supporting documents for the marriages of a city in Belgium, there are the supporting documents of a particular marriage and within this each document:

I use the same kind of coding country code, place code, places, type or name of repository, for repositories:

I indicate in the reference to the deposit the references of the source:

When the repository provides a notice describing the source, much like @GeorgeWilmes indicated, I attach it in a note to the repository reference:

And in the attributes of the citation where to find the corresponding record in this repository:

All this standardization makes it much easier to sort in windows, or to create filters, for example.
I would add that I appreciate Evidence Explained by Elisabeth Mills but I only appreciate it as a help to know what information is important to keep to cite a source, not to transcribe it as it is in Gramps the way it offers them in her book. It is certainly fine in the footnotes of a book but not at all suitable for computerized research by means such as Gramps’ filters for which a somewhat fixed standard which transposes the general idea that her book provides


Just want to give a big thank you to you all. I kept playing with it and think I found a method that will work for me. And kudos to all you who document it- I keep a personal wiki and already have documented how to label sources and citations (including custom ID format to help me better sort and find things).

This is seriously one of the more welcoming communities I’ve found on the web.

PLegoux - kudos on your organization. You’ve given me even more to think about. Thanks for the super detailed write-up. Should also let me brush up a bit on my French as well :slight_smile: