Friday, September 14, 2007

Why the 1900 Census!?!

A 1930 U.S. Census Image

What was it about FamilySearch's decision to index the 1900 U.S. Census that concerned people so much? One Ancestry Insider reader wrote,

Could you help answer one question I have had? Of all the projects to start indexing why did [FamilySearch Indexing] choose the 1900 census? I have my own ideas but I wonder if you have heard why. Thanks. Please keep these comments confidential.

Dear K.C., I promise I'll not tell another soul. Keep reading for the answer to your question.

Did We Just Declare War?

"Is everything OK at" a volunteer missionary of the FamilySearch department asked when FamilySearch announced the 1900 Census indexing project. "In our devotional today they announced that we were now at war."

"We Have a Bible..."

Another Insider reader wrote,

I signed up to do some indexing, but then decided it was kind of ridiculous to index [the 1900 Census] for Family Search when Ancestry and HeritageQuest have already indexed those same records. How many versions do we need?

At a recent conference an attendee asked a presenter from FamilySearch, "Why don't you just buy one of the existing indexes?The presenter froze and his voice trailed off, "We thought we had...."

The Real Answer

To get the real answer, we asked Paul Nauta, manager of public affairs for FamilySearch.

"We were seeking a record for the novice members of the Church in the United States that would give them a good indexing experience," said Nauta. Plus, the 1900 index project would produce a record set in which users would have "a high likelihood of success of finding someone in their living memory."

Many aspects of the census made it a good first project. It allowed FamilySearch to test some behind-the-scenes parameters. And "censuses are particularly easy for us to set up for indexing purposes, and of all the vital record types, censuses are the most user friendly to the novice indexer."

Nauta went on to explain that the 1900 U.S. Census includes a rich set of information. "[It] provides parents' birth place, number of children born to a mother, how many were still living, month and year of birth of each person, and more." He explained that given the loss of the 1890 census, it was particularly valuable to index the 1900 census which complements the 1880 census which they've already indexed.

An 1850 U.S. Census Image from Utah Territory

What's Next?

Nauta revealed to the Insider that "FamilySearch Indexing is doing very well at this writing. We will be adding more censuses to the indexing cue shortly." Many indexers have already seen the first batches show up from the 1850 U.S. Census. FamilySearch made the choice public at the UGA Conference in Salt Lake last Wednesday.


  1. A little more information on this topic can be found in a Deseret News article, "Volunteers are computerizing 1900 Census data in record time." Pay particular attention to reader comments, which echo feelings expressed to the Insider.

    "The 1900 census as well as many other years, are already on Ancestry and Heritage Quest online. ... To me doing the 1900 census doesn't make sense when they could be doing so many other records that haven't been done yet." (B Cole)

    "HQ's 1900 census is NOT every name indexed. The objective is FREE access. ..." (W. David Samuelsen)

    "There was no need for the LDS Church to index what has already been indexed three times (the soundex cards being the first time, with HeritageQuest and Ancestry being the second and third time). That 18 months of wasted effort by the FamilySearch Indexing volunteers (of which I am one--it is my calling in my ward), could have been spent on indexing millions of names from records that no one has yet indexed, such as the Draper Manuscripts." (Amanuensis)

    "I understand that the census records that were indexed by Ancestry were done in Asian countries. The people were paid to do it. In the case of Family Search by the LDS Church, the people doing it are those who have a deep interest in genealogy and helping others find their own families, even if they aren't LDS. Also, Ancestry charges for all their census records except the 1880 census, which the LDS Church did. The records the LDS people and friends are doing will be available either for free or at minimal cost."

    "Available at minimal cost"? That hasn't been announced yet, has it? I've been waiting to blog on the topic, but didn't think a final decision had been made.

    Something I will blog about very soon is the point about commercial Asian indexers vs. volunteer indexers. I'll give you a preview. Indexing the 1850 U.S. Census I cam across the distinctly Mormon name, "Nephi." Guess how the name is indexed on Ancestry? Stay tooned [sic].

  2. My great grandfather Patrick Dwyer lived in Kansas City in 1900. He was from Ireland and must have had a heavy accent. The enumerator spelled his name as Deguire!!! Both Heritage and Ancestry have botched this mistake in their indexes. 4 mistakes including the soundex!!! After years I finally went street by street to find him!!!!

  3. When it comes to indexing census records, there are 2 things that are paramount, besides making sure it is an "every name" indexing. (1) the name has to be interpreted and indexed by those willing to study how EACH particular census-taker forms his alpha characters in writing, AND they must be familiar with the given names and surnames to a large degree (2) the name then has to be rendered in many other spellings...for example, the surname Canady can be based on someone hearing the name spoken and then deciding to spell it that way. That same record has to be entered as Kennedy etc etc....probably the single biggest breakthrough in people finding their ancestors. The earlier the date of the census, the more likely the census-takers went by the "sound" of the name, so a knowledge of phonics is very important.
    End of comment.

  4. If you believe the Mormon Church's excuse for chosing the 1900 census as an indexing project, you also believe in the tooth fairy. It is clear that and the Church are at war. cancelled free access to their databases at the Family History Library/Centers. The Church in indexing revenue-generating records of

    The only question is which came first the chicken or the egg.

  5. Many of you people complaining about the LDS Church reindexing the 1900 Census is: (1) Census records are 'public domain' records, which means you shouldn't have to pay for something that you are allowed to look at for FREE. (2) The LDS has ALWAYS made their records available for FREE to anyone that wants to use them. (3) I am missing dozens of ancestors that I know where in the U.S. before the 1900 Census was taken! At least with the LDS volunteers they will probably do a better extraction because they know the frustration of knowing and not being able to find a relative. (4) It is frustrating to being paying a LOT of MONEY for a membership that promises Worldwide records, but they are not worldwide. I'm getting OLD and need to complete my family trees before too long.

  6. Dear Marie Katonak,

    Thanks for your comment. Let me respond to your points one by one.

    (1) "Public domain means you shouldn't have to pay for something that you are allowed to look at for FREE." Sorry Marie, but someone always has to pay. There are a variety of public domain records which may be viewed for FREE at NARA. Surely you believe you should have to pay the airlines to fly to see these documents. Or you can have the documents sent to you, for which NARA will charge you. That's OK, isn't it? BTW, when you fly out to NARA, check their schedule first. Congress is cutting their budget so they've had to cut back on their library hours and increase their copying fees. Marie, someone always pays.

    (2) Don't put too much stock in the LDS Church always providing these records for free. Many top level managers at FamilySearch have been pushing hard for over a year, urging the LDS Church to allow them to charge some users for some records. Call it "cost recapture."

    (3) I think you make an excellent point here. A commercial firm must cap their quality at some level. A group of dedicated volunteers sponsored by a Church that wants to record every soul is far more likely to incur the extra costs to push quality to a higher level. How ironic, then, is the large number of indexers who admit to checking when they can't read a name.

    (4) I don't know that Ancestry "promises" that they have more records than they actually have. There are many ways of checking what records Ancestry has before you subscribe. Cancelling is easy. Click the Help link in the upper-right corner. If cancelling isn't one of the questions listed, type in "How do I cancel my subscription?" The article will instruct you to go to your My Account page and click on the Cancel Subscription link. It's that easy. Price complaints by people who won't cancel their subscriptions are getting as OLD as you are.

    -- The Insider

  7. Dear Readers,

    You can find many more comments about this issue over at Dick Eastman's place.

    The Ancestry Insider would like to thank him for referencing this article.

    -- The Insider

  8. Dear Tooth-Fairy Cynic,

    The Ancestry Insider remanded himself, leaving us, his staff, to respond to your comment. What could that mean? Thank you for your comment.

    -- The Ancestry Insider Staff

  9. Dear Mr./Mrs. Canady/Kennedy,

    Thanks for your comments. Good suggestions for indexers interpreting handwriting and researchers searching indexes.

    -- The Insider

  10. Dear Patrick Dwyer descendant,

    When you suspect a name has been misindexed, it's time to use Ancestry's Exact Matches searching. An Exact Matches search of the 1900 Census for everyone born in Ireland and living in Kansas City, Missouri returns 3,509 matches. Adding some other piece of info can cut this to a reasonable number. I added a first name of Patrick, which cut the matches to 188. Browsing through these matches showed just a few starting with D, which made it simple to find your Patrick Dwyer was indexed as Patrick Deguise and enumerated as Patrick Deguior. He lived at 536 Harrison St. with his wife Margarette and children Mary, Richard, Thos., Maud and William.

    Another nice feature of is that you can add corrections when you find indexing mistakes. Corrections are displayed immediately and are rolled into the index after one or more months.

    -- The Insider

  11. I have no problem with FamilySearch indexing the 1900 U.S. census records. The more the merrier! While I commend Ancestry for allowing the user to add corrections to its database, there are still plenty of misinterpreted names left untouched. As a genealogy librarian, if patrons were not successful with Ancestry or Heritage, I would steer them back to the microfilm and/or the home-grown works (if they existed) like census books produced by local genealogical societies or individuals. The latter tend to be much more accurate because the volunteers or individuals are usually more familiar with the names they are indexing. I see FamilySearch's plunge into indexing the 1900 census as a win for the searcher. BTW, what right of it is mine to tell FamilySearch it shouldn't be doing this? Its their time, money, and product.

  12. NOT AT WAR

    I specifically asked David Rencher who said, "Absolutely NOT. We want Ancestry to succeed. We simply wish they would share more."

    Happy Dae.

  13. Hopefully the top level managers at FamilySearch will take the Ancestry Insider's wise words to heart and remember that nothing is free, not even the time and effort put in by volunteer indexers.

    Consider Amazon has built a multi-billion dollar business at least in part by getting people to submit unpaid reviews and product ratings. Consider how many people would be willing to submit such reviews without pay if users of had to pay to view them. If you want people to volunteer to transcribe records for you, it seems unwise to then charge them to view the fruits of their labor. Consider the potential for to put up a volunteer indexing system. How many people will be willing to index for knowing that the fruits of their labor would be sold for profit by Ancestry?

    FamilySearch has extremely ambitious goals, which it likely cannot afford to achieve on a paid labor basis, so it needs to think carefully about the trade off in getting paid for the contributions it is making versus the contributions the volunteer indexers are making. FamilySearch might decide to rely on church member's dedication. It might try to come up with some scheme whereby those who contribute get access, or improved access, but how will quality be effected if their transcribers are motivated by desire to get access? It might be workable with sufficient quality-monitoring mechanisms. Then again, Ancestry might be able to work out a means of trading access for indexing labor.

    Right now FamilySearch appears to have a competitive model for achieving its goals by attracting labor instead of money. Unless they are overwhelmed by "freeloaders" they will do well think long and hard before trying to get labor and money out of their users, because providing access is already quite cheap and getting cheaper by the day diminishing the importance of the freeloader, while the cost of labor is not following a similar trajectory.

    It's a pretty basic decision about whether they want to compete with Ancestry as a second provider of paid access or instead compete on the basis of leveraging volunteer labor. Trying to mix the two will be a very fine line to try to walk.

    Or at least that's how I see it.


Note: Only a member of this blog may post a comment.