Tracy Reinhart is a long-time researcher who remembers way back when accessing the census meant scrolling through microfilm. Long ago she discovered her Braford ancestors’ family in Cannon, Kent, Michigan was one of those split across pages in a census. Online publishers like Ancestry and FamilySearch have to identify these split families and join them back together. That’s a fairly straightforward process unless you run into the situation Tracy ran into recently.
“Part of the 1870 census for Cannon, Kent Co. Mich. was not filmed in page order,” she told me. “As a result, when a family list carries over from one page to the next, you will find wrong family associations.” She found that for Cannon, Kent, Michigan:
- Image 28 on Ancestry.com is page 28 and ends with the Henry Wolaver family.
- Image 29 on Ancestry.com is page 30. Notice page 29 was skipped. It starts with Emma Braford. Since Emma has no family or dwelling numbers, we know that she belongs to the family on the previous page. Because the pages were filmed in the wrong order, Ancestry erroneously places her in the Henry Wolaver family. This page ends with the Harry (or Harvy) Haines family.
- Image 30 on Ancestry.com is page 31, which correctly continues with Mary Haines.
- Image 31 on Ancestry.com is page 29, the skipped page. It ends with the A. B. Brayford family.
I was interested to see how FamilySearch handled this situation. Researchers with access to both Ancestry.com and FamilySearch.org universally advise using Ancestry.com for census research and the 1870 census on FamilySearch.org is a good illustration of why.
- If you search for Cannon, Kent, Michigan, you get everyone living in the entire state of Michigan!
- If you don’t know where your person lived, but you somehow find them, FamilySearch doesn’t indicate where the person was!
The only advantage I see for searching FamilySearch’s 1870 census is that in a search you can specify another family member (in the “Other Person” field). That’s not possible on Ancestry.
But I digress…
As I compared FamilySearch.org with Ancestry.com, I noticed several interesting things.
- The image order on FamilySearch.org matches Ancestry.com.
- FamilySearch didn’t erroneously combine the Wolaver and Braford families. But they also didn’t correctly join the the two parts of the Brayford/Braford family.
- While Ancestry has 31 images for Cannon, Kent, Michigan, FamilySearch has 32. Ancestry has left out one of the pages from the microfilm! I’ve seen FamilySearch do the same thing. Neither company discloses the censure. The companies deem the image to have no genealogical value so they delete it. This is a very bad practice! There is no guarantee the decision maker understands advanced methodologies that may require a knowledge of the existence of that page, its contents, or the lack thereof. (A little looking showed this particular page is facing page 31 on folio 139. It has no names on it.)
- The digital folder number (004271429) and image number (00268) for Emma Bradford on FamilySearch.org match the image URL on Ancestry.com: https://www.ancestry.com/interactive/7163/4271429_00268. That’s kind of techie, but the takeaway is that Ancestry seems to be using FamilySearch images.
- FamilySearch misindexed the name Braford on page 30 as Bradford. Ancestry did not. Ancestry doesn’t seem to be using FamilySearch’s index.
I see several lessons we should draw from this:
- If you don’t find your ancestor on one website, check others.
- Search several images forward and backward from your ancestor.
- Your ancestor’s name can be spelled differently by the same person in the same record.
- Look at and try to understand all the information on a page.
- When the day comes that we no longer have access to microfilm, there will be errors that we can no longer detect or overcome.
- Everybody makes mistakes. Ancestry. FamilySearch. Microfilm. Everybody.
”Just a heads up for something that I never expected to find on Ancestry,” Tracy said.
Thank you, Tracy. Image credit: Ancestry.com.
This is a great post, and a great breakdown of how important it is to know your source! About the time I started to rely on Elizabeth Shown Mills' "Evidence Explained" I started to recognize that I was consuming my Ancestry matches as they were presented, and not analyzing them or giving them a critical eye. With Census data, I've learned that they have brought you close to the source you need, but it's often just close. You still need to approach it like any other source, and go through it before you can fully understand it. Go to the first page, and the last, see if there are additional notes or information as a part of the "film strip". I also verify the pages on both sides of my match, to ensure they are in order, and that they match the progression of pages you'd expect for that sub-unit. I will also often look at the townships that should be before and after the township I'm reviewing to again see the progression and see how it all fits together.ReplyDelete
I just went through an unindexed copy of the N.Y. State Census of 1825 and 1835 on a different site, and I'm happy that I was in the habit of understanding the document before I sourced it. It turned out that the "title" pages that are between township didn't actually proceed it, but came at the end of every section. Additionally, on a note attached to first page of the 1825 Census it explains the townships that were missing from that census, which included the township my ancestors lived in at the time. If I hadn't read it before I started working, I would have never found what I was looking for and ultimately written them both off as not matching my ancestors. In-fact, I had one ancestor match from 1825, and the 3 matches in 1835 gave a lot of evidence to support a theory I've been working on a long time.
Good advice, know your source before you charge off to research!
A great post indeed to remind us that the census takers were not perfect, nor the scanners at any of the companies doing that job. I am continually amazed at the number of mistranscribed names, etc. I find on census records. When we go to the original it is clear as it can be, so the lesson in both cases seems to be either be sure to look at the original or be sure to visit more than one version of the census in question if there is any difference or doubt. Great reminder of the page endings too, as this can be a problem. Would that these resource companies would hire proofreaders!!! Thank you!ReplyDelete