Skip navigation

Locating people in historical U.S. census records (1790-1940) is easier than ever due to national indexing on popular commercial genealogy websites – something that did not exist two decades ago. Before 1990 it was necessary to consult indexes in books or on microfilm for specific states, and to actually physically be at an archives, library, or other facility that had census microfilm. However, there are still times when people can be elusive in census records. Here are a few reasons; please comment with experiences you’ve had.


1. The person’s surname was recorded as a phonetic variation that sounds similar to the expected spelling due to the pronunciation or accent of the speaker, such as Weston for Westurn, Hefford for Hayford, Brechtel for Prechtel, Stephenson for Stevenson, or McAlley for McCullar.


2. The person’s name is not spelled as expected because the U.S.-born enumerator had trouble understanding the person’s non-English name due to the speaker’s language and native accent, such as Chimel for Siml, a Bohemian name, or Brechtel for Prechtel, a German name. One humorous example is “Cheese Bologna” for Narciso Bailoni, an immigrant from Italy (Austria), as shown below in the 1920 census:

Image:  Household of Cheese Bologna, Ward 4, Mount Carmel Borough, Northumberland Co., Pennsylvania, ED 103, Sheet 1A, Line 31, NARA Microfilm Publication T625, Fourteenth Census of the United States, 1920, Roll 1612.


3. The person’s middle name, nickname, or initials were recorded instead of his or her first name.


4. The person’s name is hard to decipher because the enumerator’s handwriting was poor or sloppy.


5. The creator of the index on which you are relying did understand the handwriting so the person’s name is not indexed correctly.


6. The enumerator was literate and spelled common names correctly, but lacked enough education to spell unusual names correctly.


7. The person’s mother remarried and the enumerator listed the woman’s children from her prior marriage(s) by their stepfather’s surname instead of their father’s surname.


8. The person was temporarily away from his or her family at the time of the enumeration. The person’s family may not have mentioned this person to the enumerator because they assumed he or she would be reported somewhere else—but wasn’t.


9. The person is not living where expected and/or not living in the expected household. If he or she is living with nonrelatives, the person’s name and other personal information might be misreported.


10. The person’s surname was mistakenly recorded as being the same surname as the unrelated family listed above them on the census page.


11. The person’s surname and first name were reversed by the enumerator.


12. The person changed their name in adulthood, was given a new name upon adoption as a child, or otherwise was known and enumerated by a different name than expected.


13. The person gave a pseudonym to the enumerator. For examples, see Census Fun Fact #2 – Fictional Names:  Just Call Me Another Time.


14. The person was missed by the enumerator because he or she was traveling for an extended period of time, had no fixed address, or was living in frontier areas where governmental institutions were sparse.


15. The formerly enslaved person switched surnames in the post-Civil War period and was enumerated under different surnames in 1870, 1880, and 1900.


16. The person was missed by the enumerator even though someone was presumably at home. As one example, in 1930, a census enumerator in Boston found and enumerated the non-Jewish family that lived at a Jewish school, where the husband/father was the janitor. In 1940, the census enumerator walked past the school, apparently assuming that no one lived at the school, so that same family was not enumerated.


17. Before 1900, Native Americans who were living freely with their tribes in areas outside of effective Federal Government control were not enumerated. (Those who lived on Indian Reservations were enumerated. Beginning in 1860, some Native Americans living among the general population were also enumerated.)


18. The person was born after the official census day – or died before official census day.


19. The person was enumerated on the 1900, 1910, 1920, 1930, or 1940 census, but the page on which he or she was enumerated on was mistakenly not microfilmed by the Bureau of the Census before the original paper census records were destroyed. One example of a missing page is Sheet 11A, ED 3-38, Florence, Burlington County, New Jersey.  It should be on NARA Microfilm Publication T626, Fifteenth Census of the United States, 1930, Roll 1318, but is not.


20. The person was missed or not enumerated for unknown reasons.


Thus, there are a multitude of ways that a person might have been “missed” or misreported in U.S. census records. Have you found one of these situations in your own research?  Or one not mentioned above?

NARA expects to digitally release the 1950 population census schedules for researcher use on April 1, 2022, which is 72 years after the official 1950 census day of April 1, 1950. This is the seventh in a series of blog posts on the 1950 census.


The Census Bureau developed special procedures to address the problem of enumerating transients:  persons without a fixed address and those temporarily away from home.  Without a systematic approach, these persons would either be over-enumerated (counted twice) or under-enumerated (not counted at all). These were the “T-Night” canvasses (“T” for transient) on Tuesday, April 11, 1950, and Thursday, April 13, 1950.

Image:  1940 Census--"Enumeration, One Day was Devoted to the Enumeration of Trailer Camps and Other Places Inhabited by Transients" (National Archives Identifier 6200777).


Tuesday, April 11, 1950, was the date for “an intensive drive to cover in a single night the occupants of certain places usually devoted to transients” such as hotels, YMCAs, and tourist courts or camps (campgrounds).  YMCA (Young Men’s Christian Association) facilities provided safe and affordable lodging to young men moving to cities from rural areas, with more than 100,000 rooms nationally by 1940.  Enumerators were to be stationed at these facilities from 5 p.m. to 11 p.m. on Tuesday evening and again from 7 a.m. to 12 noon the next day.


T-Night procedure featured the use of Individual Census Report Forms (ICRs) that were to be completed by each person themselves, in contrast to the standard procedure of the enumerator recording information on the population schedule during an interview with a household member.  ICRs were used to avoid unnecessary invasion of privacy or interference with personal and business activities that would result if people were interviewed in hallways or their rooms or were delayed from checking out of hotels and similar places.


The area Crew Leader or someone he designated to be the “enumerator in charge” supervised the T-Night enumeration.  The T-Night supervisor consulted hotel and YMCA management to determine the number of ICRs needed, room number designations, and which guests were expected to check out the next day.  Enumerators assigned to T-Night worked from the hotel lobby. They could ask guests passing through the lobby if they had completed their ICRs, telephone guests on the house phone to request they pick up or return forms, review forms as they were returned, and seek clarifications when necessary.

American Hotel Association letter to member hotels, December 29, 1949, about the 1950 censusLetter from T. Harry Gowman, President, American Hotel Association, December 29, 1949, to member hotels, concerning the forthcoming 1950 census. From Supplement to the Enumerator's Reference Manual, 1950 Census, page 14.


T-Night ended on Wednesday, April 12, 1950, at 12 noon, but follow up efforts to “clean up” the enumeration of a hotel could continue until Friday, April 14, 1950, at 5 p.m.  If a hotel guest left without completing an ICR, the enumerator would fill out an ICR based on the hotel register information, including name, address, race, and gender, and indicate “REG” (for “register”) on the ICR.


Due to the self-enumeration procedure, the Bureau warned enumerators not to “relax and wait for results.”  Instead, it was that the enumerators “take affirmative action throughout the procedure, to insure maximum results. The larger the volume of returns from transients, the better will be the quality of the census.  Although we wish maximum returns, these must be obtained with due regard to the common courtesies extended to guests and management....”


Thursday, April 13, was the date for a similar effort at missions and flophouses.  T-Night enumerators assigned to these facilities were to “station themselves at the main entrance or the lobby of the place” and were to directly interview guests and any resident staff and employees.  The information collected was recorded on the regular Form P-1 population schedule.


Adapted from Supplement to the Enumerator’s Reference Manual, 1950 Census of the United States, pages 10-14, and Crew Leader’s Manual, 1950 Census of the United States, pages 26-37.

NARA expects to digitally release the 1950 population census schedules for researcher use on April 1, 2022, which is 72 years after the official 1950 census day of April 1, 1950. This is the sixth in a series of blog posts on the 1950 census.


The Bureau of the Census developed “field enumeration procedures” to make sure the 1950 census would be as complete a count as possible. Systematic coverage of each Enumeration District (ED) was essential.


Each enumerator was ordinarily assigned one ED and given a portfolio containing all needed supplies. It included a written description of the ED as well as a map showing its boundaries.  An enumerators whose ED adjoined an unincorporated place would receive either two maps or one map and an aerial photograph.  One of these would show the complete ED and the other would show on a larger scale where the ED adjoined an unincorporated area.  An unincorporated area was a community having no legal boundaries; in other words, a named geographic place without a governing body separate from the surrounding area.


In cities and towns having blocks (areas bounded on all sides by streets), the enumerator was to canvas one block or square at a time, and one just one side of the road at a time: no crisscrossing the street!  If the blocks were numbered on the ED map, the enumerator was to work in the order indicated by those numbers. The enumerator was to begin each block at a corner and proceed clockwise around the block until returning back to the beginning point. If the block included courts, alleys, dead-end streets, or passageways, and so forth, the enumerator was to check those as well, such as shown in the image below.

1950 census - how to enumerate a complete city block

Image:  Example of how to enumerate a complete city block.


Enumerators were admonished to be careful to visit all dwelling units of apartment buildings, including those in the basement.  The crew leaders would furnish a list of all apartment numbers in buildings with 20 or more units, but the enumerator would have to make his or own list for smaller buildings.


In towns or villages without distinct blocks, the enumerator was to canvass the population street by street. In some cases, it would be desirable to interrupt the canvass of a principal street in order to enumerate persons living on a side street, but the enumerator was to use his or her judgment on how best to do it. While conducting the census, the enumerator was to draw lines on the ED map to indicate which sides of the road he or she had completed as an aid to keeping track of work completed and work yet undone, as shown below.

1950 Census Field Procedures Example of How to Enumerate When There Are Not Distinct Blocks

Image:  Example of how to mark ED map with locations already enumerated.


In rural areas, the enumerator was to begin the enumeration on a major road that intersected the boundary lines of the ED, then enumerate the remainder in a logical manner. The Bureau instructed: “Follow each path and lane” and “Ask who the neighbors are and where they live.”  The neighbors should be asked to point out the locations of houses on the map. In rural areas, the Bureau noted, “there are many places which cannot be seen from any road, and many places of 3 or more acres on which no one lives.”  While canvassing the ED, the rural enumerator was to mark the map with each house’s dwelling number as indicated on the population schedule.


All enumerators, whether urban or rural, were to correct the map(s) they received by adding roads not shown on the map, crossing out abandoned or nonexistent roads, correcting erroneous street names, and marking driveways leading to structures not easily visible from the road.  Two examples of these map annotations are shown below.

1950 Census - Example of Map Corrections to be Made by Enumerator on His or Her Map

Image:  Example of how mark corrections on ED map.


The maps marked up by enumerators were temporary records intended to help them do their work in a logical manner and demonstrate to their Crew Leaders (supervisors) and to the Bureau of the Census that a complete enumeration had been conducted. The permanent ED maps—created by the Bureau’s Geography Division for planning purposes prior to the enumeration—can be found online in the National Archives Catalog in “Enumeration District and Related Maps, 1880-1950” (National Archives Identifier 821491).


These were some of the basic field enumeration procedures followed by the 1950 census enumerators.  In the coming weeks, we’ll look at a few more.


Adapted from Urban and Rural Enumerator’s Reference Manual, 1950 Census of the United States, pages 2-6.

Government forms usually look boring.  White paper.  Black ink.  Dull.


The Census Bureau chose green or blue ink for many 1950 census forms.  Color, but still just one color.


However, the Bureau chose two colors for a very important form of the 1950 census:  Form 17Fld-1, Portfolio Control Label.  It’s artistic – as far as forms can be – and almost screams, “I’m important.”  This single oversize sheet measured about 10 1/4 inches by 17 1/2 inches.  The printing was all blue except for a flag-like red-white-blue stripe that was usually on the right 1/3 of the form.  The words “1950 Census of the United States” were printed in blue ink in the white stripe.

1950 Census, Form 17Fld-1, Portfolio Control Label

Image from “[Folder 14] Form 17Fld-1, Portfolio Control Label, 17th Decennial Census, 1949-1950” (NAID 195980248) in series “Narrative Histories, Committee Minutes, and Procedural Manuals Primarily Relating to the 17th Decennial Census” (NAID 5634057), Record Group 29, Records of the Bureau of the Census.


(Of course, there are always nuances.  When the 1950 census becomes available, researchers will also notice that some Portfolio Control Labels have the red-white-blue stripe on the left side, and some have no stripe as all. It's possible the stripe was cut off of some forms.  No, I don't know why at this time.)


This form was likely pasted on the outside of large envelope.  When the enumerator was done querying the people and farms in the enumeration district, he or she placed completed population schedules, agricultural schedules, maps, and other administrative materials inside this envelope and returned them to the Crew Leader.  If the enumerator made the serious mistake of accidentally losing the portfolio, the Portfolio Control Label served as a “postage due” mailing label to the appropriate Census District Office.


This label also served as an operational checklist for Field Office verification of completed work and for Washington Office verification of coding and key punch activities for statistical compilations.


When the population schedules were microfilmed by the Bureau of the Census in 1952, the Portfolio Control Label was normally the first page of the records for each enumeration district.  Sadly, its beautiful two-color scheme is lost in the black-and-white monochrome microfilm. Fortunately, the full color example shown above survives in the administrative records of the Bureau of the Census.


1950 census, microfilmed portfolio control label

Image:  Portfolio Control Label for Enumeration District 65-78, Long Beach, Los Angeles County, California, NARA Microfilm Publication T628, Seventeenth Census of the United States, 1950.  This image contains no Title 13 restricted data.