Saturday, May 9, 2009

Using the Minnesota State Census Indexes

One of my current projects involves gathering information about the widows and children who fled from the Dakota Indians during the 1862 war. I've been curious about how they rebuilt their lives -- did they return to the Minnesota River valley? if not, where did they go? how did they support themselves? and so on. Because some of the widows remarried quickly, this can be slow going. But it was curious that, while I could find families in the 1857, 1860, and 1870 censuses, I could not find them in the 1865. Why would that family 'disappear' for that census, when there is other evidence the family group remained in Minnesota?

As I gain more experience with the Minnesota State Census Indexes (both at Ancestry.com and on the Minnesota Historical Society website), I'm learning that effective use may involve more than the usual options. To understand why some different techniques are needed, it was useful to me to discover how the index was constructed. The indexes for all the State Censuses at Ancestry.com were outsourced and created in a very short period of time, from microfilm copies of the census records. MHS's search works a little differently, but is based on the same index database. Fortunately, MHS can make corrections more easily than Ancestry.com, and updates to some records have already taken place. There remain other problems, including those resulting from illegible microfilm and unfamiliarity with our ethnic surnames.

One of the larger unresolved problems exists with the 1865 Index, where many records were entered into the index database with no surname at all. If the census taker wrote the family surname for the head of household and only assumed but didn't enter that surname for the rest of the family, the database index has only the given name. Therefore, just entering the surname with a given name for the spouse and children will result in no record found.

For example, let's say the family includes John Doe (head), wife Mary, and children James and Ellen. Searching for Mary Doe on the index will not find her or the children with John. To find her, the search should be based on her given name, the specific census and the most specific location you know. For sure, include the county, and if at all possible the name of the township or village where the family lived. You may be able to find the specific location from the U.S. census for 1860 and 1870. Your result will come back with a blank surname, but you'll have the opportunity to match up each potential match with the right family.

It's unclear when, if ever, this problem will be completely resolved. Don't give up! When you find one of these in your family, you can make a WOTR (Write On The Record) entry to add the surname, which will eventually make it into the MHS index (see my research notes at ).

Good luck in all your searches.