How to Start
Links to 11 great sites
Grandfathers in Queries
Beginning Internet Skills
Off the web
People after 1900
The LDS 1880
Googling your Ancestors
Main Genealogy Page
Essays on Genealogy
I once spent a pleasant afternoon talking to my grandmother about family stories. She liked to have something to do with her hands, so she made a pan of cornbread, a pot of stew and an apple pie while we talked. She measured a few things, but not many. When I asked her how she could cook without a recipe she shrugged and told me "After a while you get to know what to do." She started helping her mother cook when she was four years old, on a farm in Kansas; at seventy, she was pretty good at it.
In November 2002 the LDS church (our friends, the Mormons) put transcriptions of the 1880 United States Census, 1881 Canadian Census and 1881 UK Census on-line at www.familysearch.org
The three censuses are the fruit of 17 years labor by countless volunteers. The search engine attached to the database is a work of genius. The help screens attached to the database, however, lack examples. Since it came out, I've spent a couple of hours a week on their site, looking for everyone I have in my database born between 1805 and 1880. Some people are easier to find than others. This is a list of things to try if you don't find your ancestor the first try. I don't claim to be a genius at searching, but I have spent a lot of hours with the data. After a while you get to know what to do.
The 21 May 2003 Roots Web Review carried a short introduction and a link to this page. Several RWR readers sent me some additional tips. I added them to Section 2
The form gives you spaces for Census, first name, last name, Head of Household first name and last name, Birthplace country, birth year, birth year range. When you select the 1880 US census, it allows you to specify race and state. If you pick a state, it opens a field for county, then town.
The first try, for me, is a wife living with a husband. I'm going to use Joseph Kerlin (1828 - 1885), who married Sarah Elizabeth Bergen (1836 - 1924) as my example throughout this piece. They lived in Johnson County, Indiana and had five children. Here was my first try:
The first thing you'll notice - married women are under their married name, since that is how they are on the census. When I have two windows open, one set to Roots Web World Connect and one to the 1880, I sometimes slip and find myself using the woman's maiden name on the 1880.
I found them on the first try. Joseph was born, married and buried in Johnson County. He was a slam dunk. Your results may vary, as they say. Mine do. If I don't find people on the first try, I try again. I'm still looking for Charles Lafayette Boyd, a traveling minister. I've tried 40 or 50 times to find him. Some examples follow, based on my experience. Which one to try next depends on how common the first name is, how common the last name is, how old the people were, how populous the county was, and how many facts I have about the people. If the search above hadn't worked, I would have tried:
Look close - I upped the age range for Sarah to + or - 5. People lied about their age, enumerators got it wrong, the birth date you got from great aunt Matilda was wrong, or the person who posted the data on the Internet typed 1854 when he meant 1845. If the surname is rare, my first try is usually just the husband and wife's names, with no birth year. I won't illustrate that one. You can try for the husband alone, in case his wife is going by "Polly":
I won't illustrate looking for Sarah alone, either. A lot of men went by just their initials, so:
Texans, in my experience, are the worst for using initials instead of full names, but it happened all over. Men did it more than women did. Next we'll try for all the Kerlins in Johnson County:
You would not do this if their name was Smith and they lived in Indianapolis, or if you did not know where they were living in 1880. I sometimes look for any one with the surname in the county, once I find the couple, to see if I can find the siblings. If you know the county, and they just don't come up under their surname, try the first names alone:
If their names were Samantha and Eltweed, you could try the whole country, especially if you were confident of Samantha's birth year. Not Sarah and Joseph however. I just tried Sarah and Joseph, with Sarah's birth year 1836, and got 200+ hits for 44-something Sarahs.
The search engine behind the database is amazing, but not perfect. It will do a soundex search unless you check the "Exact Spelling" box. So, if you are looking for a William, you'll get Wm. and Will as well. If you are looking for surname Pack, you'll get a least Pax as well. I've had problems with double consonants in the less-common surnames. For instance, I had to look through all of the Barbaras married to all of the Williams in Montgomery County, Pennsylvania to find my William Casselberry; he was hiding under the name Casselbery, with one "r".
I won't illustrate the rest of the suggestions, nor will I number them. Play with it; after a while you'll get a feel for what works and what doesn't.
First name, last name and birth year (+/- 2) for all the children who would have been under 16 at the time. If the surname was common, add state or state and county. If not, search the whole nation. I found a John and Mary who were on the census as "J" and "M", thanks to their daughter Viola. Children are especially helpful if they have rare names and the parents have common ones. Children tend to have reasonably accurate birth years, too. There is a lot of difference between 5 and 10. There isn't a lot of difference between 55 and 60.
All of the Kerlins born in 1828, looking for Joseph, then all of them born in 1836, looking for Sarah. Set the year range to "Exact" for the first try, then (+/-) 2, 5, or even 10, depending on how many hits you get.
All of the Sarah's in Johnson County, Indiana, followed by all of the Josephs.
Try a middle name. I found Arthur Monfort Morgan in Sedgewick County, Kansas as "Manfort Morgan" by looking for all of the Morgans born in 1819, (+/- 2) living in Kansas.
If you don't know where your ancestors were in 1880, and their names are too common for a nation-wide search, the three obvious places to try are the states where they were born, married and died. The not-so-obvious places are the states where their siblings lived in 1880; that failing, the states where the siblings married and died, or the state where their parents died. Families often moved west as a group, and parents often moved in with their children for the last years.
If your ancestor has a rare first name, or a rare surname, or was over 65 at the time, he will be easier to find than if his name was John Smith, born in 1855 and living in Chicago. Don't expect the age, spelling, birth state, or parents' birth state to be accurate all of the time. People lie, make mistakes and forget things.
The transcription isn't perfect either; the people who did it were human. If you find someone who is close but not plumb, it pays to check them against the original, either on microfilm or the Ancestry census images. I found William Jefferson Belcher, who went by "WJ" in Texas. The transcriber had him as "MJ". I knew it was him because three out of four of his children matched what I had and his father-in-law was living right next door. On the Ancestry images, it looked like a coin toss; heads it's an "M", tails a "W". Until you've looked at it, it is hard to believe an "M" and a "W" can be almost identical in a 19th-century hand.
Remember too, the census itself wasn't 100% accurate. Enumerators varied in their skill, penmanship, diligence and sobriety. The 1880 transcription is a boon, but, as always, cross-check and verify
See also Those Elusive Edes, a search through the Ancestry.com census images for all years for some people named Ede, which may have some lessons you can use.
Kate Johnson of Salt Lake City, Utah writes:
[Ed. note: the FHL in Salt Lake is three stories tall and a Mecca for genealogists. However, many LDS churches have Family History Centers, which have the CDs Kate mentions. They are like little branches of the FHL. The Mormons let anyone, of any faith, use their FHCs for free.]
The Census database on CDs has better (or maybe just different) search capabilities than the on-line version. For example, if you know that a family should have had two brothers with particular names, you can set up a Boolean search to look for the Peterson families with sons named Peter and Fred. On the other hand, the CD search doesn't accommodate variations in spelling as does the web version.
The CD databases are more complete, and have some things that aren't on the web; for example, Scottish Parish Records, which one would have to pay for on the internet. They also have the 1881 Scottish census.
The web site tells you were they are located. Click on "Family History Library System", then "Family History Centers". You should call and make sure they have the CDs.[Ed. Note: You can also look in the yellow pages, under "Churches - Church of Jesus Christ of Latter Day Saints". The large churches will have several numbers, including one for "Family History Center."
G. David Thayer of Salem, Oregon writes:
Here is an example. If you search for the surname Delaplain, you get names like:
Both "Delaplaine" and "Diliplane" have the same Soundex code: D414. Entering either name as a surname in the RootsWeb SSDI with Soundex turned on yields exactly the same list of names (as indeed it should). From the examples above one might think that they use the first two characters of the surname and then apply some sort of Soundex scheme for the rest of it. I thought so too, for a while, but then I recall discovering some examples where that didn't work, either.
Barbara Samuelson writes:
I was looking for a Bartley family in a county where I knew they were in 1880. I entered the Bartley surname, but they didn't pop up on the "supposedly-Soundex" list. So I took one of the unrelated names that did appear on the list and clicked to their census information. By clicking on the "Household" link in the individual view, you get the entire household of the individual. On the household view you can click to either the Next Household or the Previous Household.
You can click backward or forward through the whole county if you want to, and see all the individuals listed. I stumbled across my Bartley family listed as "Bartlet." They should have shown up on the search, but apparently the database search is not perfect in that respect. This way, you can also find individuals who are the victims of transcriber errors where the census was either mis-read or was mis-typed.
Just as in the microfilm census, with this method you can find out a lot of great information about neighboring families, and perhaps even stumble across some lines that you had no idea were in the area.