Search billions of records on


USGenWeb Free Census Project Help, A basic overview of the census & its history

1. Federal vs. state and territorial.
2. How the census changed over the years; population schedules vs. several other kinds of schedules (forms) etc.
3. The layout of the census schedules

The US Federal Census has been taken every ten years since 1790. In addition, a few off-years had federal enumerations in certain localities.

In the first few enumerations, only very basic information was recorded, but as each census year came and went more and more information was recorded.

Our project deals primarily with Federal Census enumerations, not those done by states, territories, or other localities.

Even in the Federal census, in most years several kinds of census forms were filled out. In addition to population schedules, some years also had mortality schedules, slave schedules, military schedules, Indian schedules, and several more.

We are concerned primarily with population schedules.

Every year from 1790 to 1920 is currently available, and 1930 will be available on April 1, 2002.

However, a large number of the early enumerations have unfortunately been lost, and nearly 100% of the 1890 enumeration is lost.

Our webpages attempt to show exactly what is still surviving and available for transcription assignment. (This is a work in progress and will be for the next few months.)

The census has changed quite a bit over two centuries.

We can roughly divide it up into three periods in our work.
1. From 1790 to 1840 the schedules only showed the name of the head of each household, and a tally of the number of persons in the household in broad age groups.
2. From 1850 to 1870 the schedules show the name of every single person counted, and additional information not found earlier.
3. In 1880 the modern census began, and from this date to the latest ones available we have enumerations done by professional census workers, using special districts within the regular state and county areas.

You may be familiar with Soundex and Miracode index records for the 1880 and later enumerations. We don't deal with those, our project is to transcribe the actual census itself. With modern search engines and computer techniques there is little need for an index, compared to the pre computer age.

Some people don't understand the meaning of "transcribe" or "transcription." It means literally to create an exact duplicate, a mirror image.

We are all aware that the census record is swimming in errors, especially misspelled and erroneous names. Our job is not to correct or amend, not to improve upon the census. It is simply our job to exactly reproduce it. There are exceptions to every rule, of course, but the exceptions in this case are few.

We want whatever the original record says, "right" or "wrong." In fact, we want it so exactly that we get fussy over things like periods after initials. If there is a period after an initial or abbreviation in the original, then we want one too. If there is not one there then we absolutely don't want one. If there is bad grammar, we want it exactly as it is in the census.

There are exceptions. For instance, if the old double "s", looking like an "f" is used, for instance Maf or Mafs, convert to a double "s", as in Mass. Also, on certain lines we require an entry, even if not on the original these would be for family number, dwelling (or house) number, surname, line number, and the like. In addition, any ditto marks or the ditto abbreviation "do" are to be converted into the word or words being referenced. And if an occupation or birthplace is horribly misspelled and then dittoed, we horribly misspell it each time.

In each census year the categories, columns, questions, etc. would change, sometimes a little, sometimes a lot. In addition, things like the page numbering system would change.

Pages and sheets.

Depending on the year and locality, pages could carry anywhere from zero to five different numbering schemes. This is a big problem we have been having.

Despite some instructions that can be found in various places, the following is the official "right" way that we want to do it.

Whenever the original enumerator's page numbers are given, we list those as page number and any and all others as sheet number. The original numbers nearly always will start at page one for the district, county, or counties assigned to that enumerator, and count up from there to the end of his assignment.

These are almost never printed or stamped.

We need to understand the form of the censuses, especially the early ones.

Each enumerator rode around on his horse for weeks and months with a ledger book, recording his information. What you see on microfilm is NOT that book, or even the writing of that enumerator. The work was collected from each enumerator and rewritten by a court clerk (before 1880) into the old ledger books that have been filmed.

Even that is not quite correct.

For several census years in the mid 1800's, the original rewritten book was then recopied by hand, and in one year believe it or not this second recopy was then recopied again (now a fourth generation copy). This last generation copy in each case is the federal copy to go to Washington, D.C., and therefore the copy that has been microfilmed.

So our "original" is not so original, but it is what we have to work with.

Many years later, sometimes after the original books had fallen apart and been rebound (sometimes out of order), numbers were put on them. These are the sheet numbers, sometimes printed or stamped, sometimes handwritten.

How do we know which handwritten numbering system is the one we should use? That is a very tough question, because every year and locality is different. We want to label as page number whatever is closest to being the enumerator's original numbering, if any. In many cases there is no original numbering shown.

Each transcriber needs to become an "expert" on the enumeration that they are performing by familiarizing themselves with what it looks like, with special attention to little quirks and idiosyncracies.

In all cases your State Census Coordinator is your best resource. Be sure and stay in contact at least on a monthly basis.


When transcribing the census, MAKE A BACKUP COPY!

Census Project since 1997

A Gift of the Past for the Future! Started in February, 1997, The USGenWeb Free Census Project is an all-volunteer project to transcribe census records in a standard format in order to make them available to genealogical researchers on the Internet.

Free CSS Template by