Determining Jurisdiction for Citation of Online Imaged Vital Records

I often find it difficult to determine the jurisdiction level (i.e. county or state) to use when creating a citation for online imaged documents, particularly vital records.

For example, at EE 9.33 County-level Certificates: Online Databases & Images (County Level) (EE, 1st ed., p. 459), example #3 for child "Infant Day."  The jurisdiction as lead element in the citation is Shelby County, Tennessee.  Looking at this record online {1}, it is a typical death certificate stating at the top "State of Tennessee, State Board of Health, Bureau of Vital Statistics, Certificate of Death."  Place of Death is listed as Shelby County.

Does the fact that this record is actually at the Shelby County website, gives credence to using that place as the jurisdiction?  What if this same imaged certificate was found in an collection for Tennessee Death Records?  My tendency would be to specify the jurisdiction as Tennessee, and include the agency per the QuickCheck Model on p. 430.



Submitted byEEon Sat, 10/17/2020 - 09:28

Jeff, yours is the most common dilemma researchers face when they use vital records imaged online. We’ve had a number of discussions on this in the forum across the years. This time, let’s try to dissect the issue in a different way.

EE 9.33 "County-level Certificates" (which you cite) lays out the essential elements for citing a vital record. Those essentials are

  1. Name of creator (agency and jurisdiction)
  2. Collection (if one exists)
  3. Specific entry—by year, number (if there is one) and name of person
  4. Repository where the register is now housed
  5. Repository location (city, state).

An important takeaway here is this: The creator and the repository are not always the same.

EE 9.1. “Basic Issues” (things relevant specific to the kind of records in this chapter)  tells us this:

"The basic issues that determine the structure of your citations in this record category will be

  • whether the record was created at the local or state level;
  • whether the original record you are citing is a “loose” document from an agency file, an official copy recorded in a clerk’s record book, or an image copy consulted in filmed or digitized form;
  • whether the derivative record you are citing is an official certificate issued by an agency, or whether it is a database entry or index; and
  • whether you are citing an imaged copy [online] ..."

With regard to the first bullet, the fact that you are obtaining this at the website of the Shelby County Recorder, not a state-level website, does tell you that you are viewing the local copy.

With regard to the last bullet, let's examine the underpinnings of 9.33's example for an online image: example 3 at page 459.

  1. When citing images gathered online, we use layers.
  • In one layer, we cite the image that we see.
  • In another layer, we cite the website provider, the URL and/or path that we take to get to the image number, and the image number.
  • In the last layer, we cite whatever source-of-the-source data the provider gives us (if any).
  1. In the image layer, we cite only the details we can discern from the image itself or its context (surrounding entries, the start of the roll of imaged microfilm, etc.)
  • We do not include anything we do not see (i.e., things we cannot verify for ourselves).
  • We do not include the source-of-the-source data the provider gives us. (We cannot personally attest the accuracy of that data.)
  • Our citation of the online-imaged document will not contain all the information we would include if we obtained that record from the original repository. That’s just logic, because we did not obtain this from that repository.

Re the QuickCheck Model on p. 430: That is for citing a record you have received directly from the agency from which you requested it. In other words: you know, from your own experience, that this exact record exists at this exact place you are citing. With online images, however, we don't know that. The provider's source-of-the-source data could (and sometimes does) err.


Submitted byJeffH13on Sun, 10/18/2020 - 21:52

Thanks for clarifying, especially the usage of the QuickCheck Model on p. 430.  It seems to me that FHL microfilms (and the online images of them) give good source data in the first few frames of the film.


Submitted byEEon Mon, 10/19/2020 - 10:35

Yes, Jeff. The initial images of each microfilm (or each "item" on a film) produced by GSU/FHL provide us with information essential to understanding our source and identifying it correctly. Ditto for film produced by the National Archives. Across the past decade of online images, FamilySearch has also labored to give us sound citations for many of their offerings.