If you are really interested in what went on during the early census check out this site http://homepages.rootsweb.com/~haas/learningcenter/ and access census records. Here is a sample of some of the information that they offer: To make matters worse not every census taker used good quality ink or pens and to make the ink last longer many census takers added water. Many of the microfilmed census records available are also copies of the original census records. We'll talk more about that when I discuss the individual census year. From the 1790 Census through the 1820 Census the Census Takers were required to furnish their own paper, rule the paper to fit the format the government mandated and to bind the schedules themselves. Therefore not ever census schedule was made of high quality paper. Since census takers were paid by the number of names they recorded sometimes the cost and problems of taking the census far exceeded their payment. Here's an example: "Sir: I beg to report that I have been dogbit, goose-pecked, cowkicked, briar-sratched, shot at, and called every 'fowel' that can be tho't of. have worked 12 days and made $2. I have had enough and I beg to resign my position as a census taker for Crittenden Township." So wrote Roger Waite to a marshal of census enumerators for the State of Vermont on August 24, 1790-the year of the first national census in the United States. (2) And, last but not least by any means, is the handwriting and the spelling of the census takers. Keep in mind that the census taker recorded names as they SOUNDED. Since many of the people interviewed were illiterate they could not spell their names, even if they had been asked. Accuracy of the Information on the Census The accuracy of the census records must be taken with a grain of salt. Census takers were often school teachers who were unemployed during the summer or a farmer who wanted to supplement his income, in other words they were just regular people. Some were conscientious and took their job seriously while others were just out to put as many names down as possible to collect their pay (much like employees are today). Prior to the 1850 census each census taker was responsible for being sure that all names in a given area where recorded. And, they mapped out their own route. A census taker (called an enumerator) was to begin at a given point, maybe with his next door neighbor and go from house to house, farm to farm, one after the other asking his questions. But, if on a given day the census taker needed to pick up a pig at so & so's house he might well pick up the pig and then stop at the houses along the way to gather his census information. In towns a census taker might go up one road hitting every house on the right side and then down the left side or he might cross the street in a zig zag pattern. That's why it's so important to get several names before and after your own family. With the 1850 census the census taker was given a map and instructions on the order in which to take the census. By the 1850 census you can pretty well rely on the people listed to each other actually lived next to each other. In far flung counties a census taker covered a lot of ground to reach each household. If he reached a farm and no one was home who could responsibly answer the questions, the census taker would question children, servants, old people who's memory wasn't always reliable, neighbors or write down what he knew personally and guess at the rest or leave it blank. Thus the information on a given individual can vary from year to year according to the person who was furnishing the information. The census takers didn't consult any type of original records, all information was passed verbally and from memory. In families that had 10 or 12 or even more children, it wasn't unusual for one to be overlooked or for birth dates to get mixed up between children. When asked what year they immigrated people would say "oh about xxx" without consulting any paperwork. What they answered was what was written down. Many of the people interviewed were people who's native language wasn't English. Many didn't completely understand the questions asked and if they did answered in broken English or English heavily accented. The Census Taker wrote down what he heard which led to some really bizarre names being listed. Say your ancestor's name out loud and have someone else write it down (preferably someone who doesn't KNOW how to spell it) and see what spellings you come up with. For example: My Haas line is pronounced Hass and I almost never find them in the census records as Haas. And remember that the pronunciation of your ancestors name today might not be what it was "back then". An example: My father-in-law pronounced my mother-in-law's maiden name as "Ambers" although her name was actually Ambrose. Had I not heard him say the name I'd never in a million years checked the spelling Ambers. Some families are listed more than once in a census. Because of the time it took to complete some of the censuses, especially in rural areas, families often moved and ended up getting counted twice, not always with the same information. For the same reason some families were missed entirely. In some states census records were "padded" to up the population count so that an area could qualify for statehood. Watch for duplicate entries of a family with slightly different information or names left out or added. And the reverse was true, states that wanted to decrease their taxes often under reported on the census and whole areas are left out. Census takers were instructed to count the population on a given date, for example June 1st, and even if it was August before the count was actually made for a given family the information was to be accurate as of June 1st. If someone was alive on June 1st and deceased by the time the census taker got to the house that person would be listed as living. If a baby was born in July (for example) it would not be listed since it wasn't alive on June 1st. This can be extremely confusing and frustrating since most census schedules list the actual date the census information was written down. It has caused many family researcher untold hours of unnecessary research. In the following explanation of each census watch closely for the date the census was to be recorded as of. Along the same lines some of the later schedules that ask for a birth year will record a child as 2/12 meaning the child is two months old. Most researchers look at the date the information was recorded and quickly come up with a birth date - but remember the child would be 2 months old AT THE CENSUS ENUMERATION DATE - not the date the information was recorded.