A census can have as much as a 25% error rate. It depends greatly on the part of the country you're looking at. The farther south, the more errors for the most part. The average person, in the South, normally went thru 2nd or 3rd grade or just long enough to learn enough arithmatic to run a farm. Reading was way down on the list of priorities. The worst census is the 1870, other than the loss of the 1890. That was the first one after reconstruction and the population of the South had doubled, nore than that in some areas. One of the main purposes of the census is to determine how many seats in the House each state will be alotted. The Union wanted to insure that the South wouldn't be in control of the House of Representatives so they had the enumerators deliberately skip families in the states that had been a part of the Confederacy. In some counties with a large population of both whites and blacks, as many as 1 in 5 families were missed, on purpose. A census should never been taken as anything more than a starting point. If the year of birth turns out to be within 5 years one way or the other it is considered perfect. I learned a lot of this at the "Heritage Quest Road Show" when it came here a couple of years ago. If you ever get the chance to go to one you won't be disappointed. Just in case some of you don't know what Heritage Quest is, it's one of the best of the Genealogical publications. Their seminar (Road Show) can't be beat. Charlotte Rodney Hall wrote: > > How common are errors in the census? I have a family with the father Daniel > having 3 different ages up to 5 years apart. Wife's name is the same but > her ages vary even more most of the children are the same except one child > was named Emil and one census has a Samuel whose age is 2 years different. > I don't know wether to consider them as the same family or not. > > Rodney Hall > Subknave@msn.com