25 February 2012

Citing What It Does Not Say

This is part of one of the online trees for a relative of mine. I've eliminated the name and the precise date of birth because I'm more interested in how Ancestry.com handles citing sources than who this specific individual is whose birth is being cited. I've also seen problems like this countless times so I don't want to appear to be picking on this entry only.

I've see all six "sources" used in this instance. The Ancestry.com tree indicates that all six are sources for the fact that this person was born on a certain date in a certain place. The problem is that all six sources do not say that. Tying these sources to a precise date and place of birth is indicating that they are more accurate than they really are.


  • The 1900 census provides a month and year of birth along with the state. 
  • The 1920 and 1930 census provides an age and a place of birth. The age does not necessarily suggest a specific date--which this citation seems to indicate that it does. 
  • The death index does provide the date and place of birth.
  • The U. S. National Home for the Disabled Volunteer Soldiers provides an approximate year of birth and a state of birth. 
  • The World War I draft card provides a county and state of birth along with a date of birth.
Why not just link them all to the precise date and place? After all, that's a heck of a lot easier. It may be, but it's a heck of a lot less accurate.

There's only 1 of the six records that provide the specific location and date--so really only one should be listed as a source with that precise information. 

Indicating a record says something it does not is confusing and in this case it looks like there are more sources with that level of detail than there are.

The date and place for an event when tied to a source for that should only be as precise as that source indicates.

Sure it takes a little more work, but it makes our work more accurate and makes analysis easier. 

  • Month and year in Illinois should be tied to 1900 census.
  • Approximate year and location of Illinois should be tied to the 1930 census and Soldier's Home information.
  • Specific date and specific place should be tied to the death index.
  • The approximate year and county and  state should be tied to the draft card.

4 comments:

debbehagner said...

Elizabeth Shown Mills like to cite every piece of data that is on the document. I like to cite the document. I don't care for being overkill with the citation. Birth Certificate cite to the Birth tag or fact. But Elizabeth Shown Mills want to take the document and pick it apart. For example the Death Cetificate would have birthday and birth place (she would cite that) as welll as death date and place. Every piece of info on the death certificate would have a citation.

Andrew Hatchett said...

What you say is true-however, just how does one do what you propose should be done when working with the Ancestry Member Trees application?

Don't just show us the problem... show us the solution; and by that I mean you should take your proposed solution and actually construct a one person Ancestry Member Tree showing us just what to do each step of the way.

Judy G. Russell said...

Ancestry's repeated habit of suggesting AS FACT what barely qualifies as theory is exceedingly annoying. But hey... just click on the leaf. You'll feel better...

Bubba said...

Easiest way is to create alternate dates/places of birth and link the source citation to the birth date/place that corresponds to the amount of detail the source provides.

We'll work on a longer post about this.