1940 U.S. Census Index is Now Complete
August 09, 2012
When the 1940 U.S. census was released to the public on April 2, 2012 it was anticipated that the 3.8 million images containing about 131 million names would not be indexed until much later this year. Last week while I attended the BYU Genealogy Conference, both Ancestry and FamilySearch announced the completion of the index. Time for another genealogy happy dance!
While the entire index is now available at Ancestry, the last of the index will be published in the next couple of weeks at FamilySearch. Both are free. FamilySearch had more than 150,000 volunteer indexers to complete the task. Since both organizations employed different methods of indexing the records, the question has to be asked - which is more accurate? Randy Seaver in his article, "1940 U.S. Census Comparisons - Summary and Conclusions" described his study of the two indexes. In most cases, of his small sample, FamilySearch had better accuracy. In fact, he concluded that FamilySearch had a 12% error rate compared to Ancestry's 23% error rate. I guess I'd prefer a 0% error rate, but there are numerous variables involved in indexing historical records such as these. Thanks to all the volunteers who made this possible so quickly!
1940 Census webinars
To learn more about the 1940 census, view our two webinars:
- Navigating the New Census Tools in Legacy Family Tree by Geoff Rasmussen
- Navigating the 1940 U.S. census by Thomas MacEntee
Posted by Geoff Rasmussen in News & Views | Permalink
|
Comments (4)
It should be noted that while Ancestry's error rate appears to be higher, that they also allow posts of alternate indexing which are searchable. FamilySearch has been pondering the correctability issue for over a year and to date has no method to do so. Errors quite often result in records becoming "invisible" so the genealogy community needs to be more sensitive to this issue.
Posted by: Leon Morse | August 10, 2012 at 08:20 PM
Leon Morse is right about the correctability issue with FS. FS would have been much more accurate had they trained and tested their arbitrators. This was a major complain of indexers who spent countless hours insuring what they indexed was accurate, even utilizing other public records like Find-A-Grave to find a family, only to have arbitrators change it to an incorrect listing. One one sheet I did, I had five families with a total of 24 names that I searched out when the handwriting was atrocious. The arbitrator chose to opt for what was in the authorized lists---nothing even close to what was even decipherable---and effectively lose 24 people to future researchers. The same problem continues now with the ship lists and arbitrators who don't read or understand the directions. SL seems to turn a deaf ear and in some cases send condescending letters when you complain after the umpteenth time. So much so that after almost 80k in names with 98 percent accuracy, I have chosen to give up on FS as it is too frustrating.
Posted by: sukiyhtaky | August 14, 2012 at 10:09 AM
While I agree with the frustrations, one work around for the missing data via a search engine is to view each image one at a time and decipher the writing yourself. Yes this can be time consuming but it still beats cranking through reels of microfilm. Also, if the family has been in the area for awhile you can often find them by either searching or scrolling through images for a known neighbor. A hearty thank you to all we labored to insure accuracy even if the end product does not reflect your hard work!
Posted by: Frank | August 14, 2012 at 03:23 PM
anyone who has taken the time to view an orginal will agree mistakes can happen. I am just greatful for those who do the work, and want to thank all for it.
Posted by: Cindy Tanksley | August 15, 2012 at 07:21 AM