Skip navigation

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.  This is the 21st in a series of blog posts on the 1950 census.


Census questionnaires have changed over time.  From 1850 to 1950, six basic questions asked in each census remained the same:  name, age, gender, race, occupation, and place of birth.  Relationship to head of household was asked from 1880 to 1950, and the citizenship status of each foreign-born person was asked from 1890 to 1950.  The nature and number of additional questions has changed over time.  This post will take a look at the major differences between the 1940 census and the Form P1, 1950 Census of Population and Housing, that was the standard fill-in-the-blank census form used in most of the continental United States. 


The questions on the standard 1940 and 1950 census forms were similar.  The most significant differences between them are: 


  • In 1940, each census page had lines for 40 persons; in 1950, this was reduced to 30 lines in order to ask “sample” questions of more people and give the enumerator space to write notes and explanations if they were needed.


  • In 1940, only two persons on each form answered sample questions.  In 1950, six people on each form were asked sample questions, and the 6th person answered several additional sample questions. 


  • In 1940, everyone was asked whether they had lived in the same place, same county, or same state in 1935.  In 1950, only persons on six “sample” lines were asked what county and state (or foreign country) they had lived in “a year ago” in 1949. 


  • In 1940, everyone was asked the highest grade of school they had attended and if they had attended school since March 1, 1940.  In 1950, only persons on six “sample” lines were asked the highest grade they had attended, whether they had completed that grade, and whether they had attended school at any time since February 1, 1950.


  • In 1940, everyone age 14 or over were asked if they had worked (or been assigned to work) on public emergency work for agencies such as the Work Projects Administration (WPA), Civilian Conservation Corps (CCC), and National Youth Administration (NYA) during the week of March 24-31, 1940. Those agencies were abolished during the 1940s so the question was not asked in 1950. 


  • In 1940, everyone age 14 or over was asked the dollar amount of wages or salary income earned during calendar year 1939, and whether they had received income of more than $50 from sources other than wages or salary (yes/no).  In 1950, only persons on six “sample” lines were asked the dollar amount received in 1949 from wages or salary; working in his or her own business, profession, or farm; or from interest, dividends, veteran’s allowances, pensions, rents, or other non-wage or non-salary income.  In addition, persons on six “sample” lines were asked how much money their relatives in the same household received from the same sources. If the respondent was not comfortable answering these income questions, the enumerator gave the person a form that could be directly mailed to the Bureau of the Census.  The enumerator was to record “10,000+” for anyone who reported an amount more than $10,000.  (Average family income in 1950 was $3,300.) 


  • In 1940, persons on two “sample” lines were asked if they had a Social Security Number (SSN) and whether deductions had been made from their wages or salary in 1939 for Old-Age Insurance or Railroad Retirement. This was not asked in 1950. 


  • In 1940, persons on two “sample” lines were asked what their usual occupation was and in what industry it had been in.  The answers to these questions could be the same or different than the person’s current occupation or industry reported in the main part of the form.  In 1950, only the person on the last sample line was asked his or her occupation in their last (previous) job and the industry in which it had been. 


  • In 1940, persons on two “sample” lines were asked if they were a military veteran, wife or widow of a military veteran, or a child (under age 18) of a military veteran. In 1950, only men on six “sample” lines were asked if they had served in the military in World War II, World War I, or at any other time.

College and university students often have two “homes” - the parental home and the on-campus dormitory or off-campus apartment or other housing.  Where should they be enumerated?  This question reaches to the heart of the basic enumeration standard first established by Congress in the 1790 census act that each person was to be enumerated at their "usual place of abode."  This is the place the person lives and sleeps most of the time, which may be a totally different place than their voting residence, legal residence, and so forth, that are established by other standards.


The enumeration of college students was not a "problem" from 1790 to 1860 since the official census day was in the summer (August, 1790-1820, or June, 1830-1860) when schools were typically not in session.  By 1800, only 37 U.S. colleges or universities had been established; by 1860 there were around 380.  In addition, the number of students enrolled in institutions of higher learning was relatively small; there were about 63,000 in 1870.  By 1950, that number had grown to nearly 2.3 million.  (This article will use the term "college" to include "university.")


1850 Census


The 1850 census instructions were the first to mention college students:  "Students in colleges, academies, or schools, when absent from the families to which they belong, are to be enumerated only as members of the family in which they usually boarded and lodged on the 1st day of June."  (Emphasis added.) This instruction placed the student's usual abode at their "college" address if there on June 1, the official census day.


1870 Census


College students were next mentioned in the 1870 census instructions under "Place of Abode" directions:  "… children and youth absent for purposes of education on the 1st of June, and having their home in a family where the school or college is situated, will be enumerated at the latter place."  Thus, students continued be enumerated at their college residence location.


1880 and 1890 Censuses


The 1880 and 1890 census instructions to enumerators flipped in the opposite direction by making the assumption that college students would and should be enumerated at the parental home:  “In the case of boarders at hotels or students at schools or colleges, the enumerator can, by one or two well-directed inquiries, ascertain whether the person concerning whom the question may arise has, at the time, any other place of abode within another district at which he is likely to be reported.” (Emphasis added).


1900 Census


The 1900 census instructions in paragraph 86 appear to flip back to the "enumerate at school" standard of 1850 and 1870 by simply indicating without further elaboration that "all persons having their usual places of abode" in colleges and other institutions should be enumerated there:


86.  Name of institution. Whenever an institution, such as a prison, jail, almshouse, hospital, asylum, college, convent, or other establishment containing a resident population, is to be enumerated, the full name and title of the institution should be written on the line provided therefor at the head of the sheet, and all persons having their usual places of abode in such institution," whether officers, attendants, inmates, or persons in confinement, should then be entered consecutively on the schedules.

1910 and 1920 Censuses

The 1910 and 1920 census instructions dug into the weeds of the college student problem in several paragraphs after laying out the definition of "usual place of abode."  Considered in their entirety, each college student was normally to be enumerated at the parental home unless they regarded the student residence as "home."


45.  Usual place of abode.—The law provides that all persons shall be enumerated at their "usual place of abode" on April 15, 1910 [or January 1, 1920]. This means the place where they may be said to live or belong, or the place which is their home.

46.  As a rule the usual place of abode is the place where a person regularly sleeps….

47.  Residents absent on census day.—There will be a certain number of persons having their usual place of abode in your district who are absent at the time of the enumeration. These you must include and enumerate, obtaining the facts regarding them from their families, relatives, acquaintances, or other persons able to give this information. Thus if a member of any family in your district is temporarily away from home on a visit, or on business, or traveling for pleasure, or attending school or college, or sick in a hospital, such absent person should be enumerated and included with the other members of the family. But a son or daughter permanently located elsewhere should not be included with the family.

48.  In the great majority of cases, however, it is more than likely that the names of these absent members of the family will not be given you by the person furnishing the information, unless particular attention is called to them.  Before finishing the enumeration of a family you should in all cases, therefore, specifically ask the question as to whether there are any such absent members as above described.

49.  Classes not to be enumerated in your district.—There will be, on the other hand, a certain number of persons present and perhaps lodging and sleeping in your district at the time of the enumeration who do not have their usual place of abode there.  These you should not enumerate unless it is practically certain that they will not be enumerated anywhere else.  As a rule, therefore, you should not enumerate or include with the members of the family you are enumerating any of the following classes: …. Students or children living or boarding with this family in order to attend some school, college, or other educational institution in the locality, but not regarding the place as their home….

57.  Students at school or college.—If there is a school, college, or other educational institution in your district which has students from outside of your district, you should enumerate only those students who have their homes or regular places of abode in your district.  (Emphasis in italics; bold indicates italics in original text.)

1930 Census

The 1930 census instructions closely followed the text of the 1910 and 1920 instructions (but with different paragraph numbers).  However, the revised “Students at school or college” text now alerted enumerators to the fact that some students might have no other home than their college residence.


68. Students at school or college.—If there is a school, college, or other educational institution in your district which has students from outside of your district, you should enumerate only those students who have their regular places of abode in your district. This will include students who live with their parents, permanently and regularly, in your district, together with certain others who have no homes elsewhere. Especially in a university or professional school, there will usually be a considerable number of the older students who are not members of any family located elsewhere and who will be omitted from the census unless you enumerate them. You should make every effort to find and enumerate all such persons.

1940 Census

In planning for the 1940 census, the Census Bureau undertook a small study of the completeness of enumeration of college students in the 1930 census who attended the University of Wisconsin (at Madison).  On January 17, 1939, staff members Charles Ellison, Robert Fluno, and Janet Meditch reported the results.  They used a sample of 80 students (55 men, 25 women) whose home addresses in the 1930 university directory were Ashland, Two Rivers, or Watertown, Wisconsin, all towns of about 10,000 population.  As a very small sample, the results of the study were not conclusive.  Despite having home addresses, it was still sometimes difficult to match the student with their “home” family because the parents’ names were not known.  They found that:

  • 10 percent of the students were not enumerated anywhere.
  • Students who lived further away from Madison were more likely to be omitted from the census than those whose hometowns were nearer to Madison.
  • Male students were more likely to be omitted than female students.
  • Underclassmen were more likely to be omitted than upperclassmen.
  • One person, a student nurse, was enumerated both at home and the university hospital despite instructions to enumerators that student nurses were only to be enumerated at the hospital where they lived and trained.
  • Generally, omission of college students was more likely than duplication - except in the case of student nurses.
Statistical Page from Study of the Enumeration of College Students at the University of Wisconsin (Madison) in 1930 from "Binder 36-L - Enumeration of Students" online at
Statistical Page from Study of the Enumeration of College Students at the University of Wisconsin (Madison) in 1930


Due to the inconclusive nature of this small study, January 26, 1939, Morris H. Hansen, Chief of the Statistical Research Division, suggested that the Census Bureau undertake a more extensive nationwide sampling of college student enumeration in the 1930 census in order to revise and improve instructions to enumerators.  However, a more extended study was likely not undertaken since the 1940 census instructions to enumerators concerning college students were very similar to 1930 but with minor additional complexity:

306.  Persons to be counted as members of the household include the following: …. Members of the household attending schools or colleges located in other districts, except student nurses away from home and students in the Naval Academy at Annapolis, or in the Military Academy at West Point, or in any other training school or institution operated by the War or the Navy Departments or the United States Coast Guard. (Emphasis added.)

1950 Census

Nine years later, in advance of planning for the 1950 census, Mr. Hansen decided to again raise the college student issue.  On January 7, 1948, Mr. Hansen forwarded the 1939 study to Census Bureau Director James Clyde Capt and other senior officials.  The agency’s Technical Advisory Committee on General Population Statistics reviewed the issue and the procedure subsequently proposed was that "a student who is away from home attending college will be considered and enumerated as a resident of the enumeration district in which he lives while attending college and not of the district in which his home is located." The legality of this procedural change was reviewed and approved by the Acting Attorney General, whose letter to the Secretary of Commerce on March 25, 1950, quoted the Acting Census Director:

"One of the problems facing the Bureau of the Census at each enumeration of the population is the handling of persons who have concurrent residence in two places…. It has been necessary to devise rules for enumerating such persons in a uniform manner.

For census purposes persons who appear to have two concurrent residences have generally been considered to be usual residents of the households in which they live in the area where they work.  To illustrate, there are people who work in one community but maintain a home elsewhere at which they are present for weekends or less frequently.  It has been customary census practice to count such persons as usual residents of the community in which they sleep more than half the week. Inmates of prisons, tuberculosis sanitariums and other institutions in many cases have homes from which they came and to which they will return.  These persons have traditionally been counted as usual residents of the institution, which is the place where they lived and sleep most of the time.  A similar treatment has long been given to members of the armed forces who are enumerated as inhabitants of the community in which they are stationed.

The principal exception to this rule in the past has been the college student living away from home.  To be consistent, college students should be enumerated as residents of the community where they spend most of their time.  Students living in college communities for as much as 9 months of the year should be enumerated as residents of those communities….”

The magnitude of this problem is not large.  In the Current Population Survey for October, 1949, we found that there were about 2.3 million college students in the United States, and that about 1.3 million of these would have been enumerated as residents of the college community even under 1940 procedures.  The 1950 rule therefore involves a change for only a part of the total college student population…."

Thus, in 1950, the Census Bureau finally and permanently determined that enumerating college students at their college residences where they normally lived and slept was better policy that was more consistent with the standard historical definition of "usual place of abode" or "usual place of residence." It would (or should) certainly have been less confusing to enumerators and the public.  Enumerating college students at their college residence is still the rule today - as parents of students may remember if they read and followed the directions for the recent 2020 census!

Notes for readers:

(1) Statistics on institutions of higher learning and the number of their students are from Thomas D. Snyder, ed., 120 Years of American Education:  A Statistical Portrait (Washington, DC:  U.S. Department of Education, 1993) and Digest of Education Statistics (2013), "Table 303.10:  Total fall enrollment in degree-granting postsecondary institutions, by attendance status, sex of student, and control of institution:  Selected years, 1947 through 2023."


(2) For discussion of who were residents of a place, see the prior blog post:  1950 Census:  "Usual Place of Residence."


(3) The University of Wisconsin study results and other memorandums on the subject of enumerating college students are from "Binder 36-L - Enumeration of Students" (National Archives Identifier 205683298).

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.  This is the 20th in a series of blog posts on the 1950 census.


Would you have applied to be a census enumerator in 1950 after reading the brochure, "A Message to Job Applicants"?  Let's review its contents:


Requirements and desired skills for applicants for 1950 census enumerator:

  • U.S. citizenship
  • High school education or equivalent
  • Understands written and spoken instructions
  • Writes neatly and legibly
  • Does simple arithmetic quickly and accurately
  • Talks easily with people and gains their cooperation and confidence
  • Has good judgment, patience, and understanding
  • Has the "combined qualities of a bookkeeper, salesman, and diplomat"


If selected for the job, the applicant will receive training in:

  • Interviewing people - including practice interview sessions
  • Asking census questions
  • Making entries on the forms
  • Reading a map


Applicants who successfully complete the training will: 

  • Act as a representative of the Federal Government
  • Collect, record, and transmit confidential information to the Federal Government
  • Take an oath not to reveal any of the information he (or she) obtains
  • Be assigned a specific area called an Enumeration District (E.D.)
  • Receive a map of the E.D. and enough questionnaire forms to do the job
  • Visit every place in the E.D. where someone might live
  • Make a record of everyone found in the E.D.
  • Make a record of the places people live in (housing)
  • Make a record of the crops and livestock (agriculture) in rural areas
  • Record the information received accurately and neatly on the forms provided according to instructions
  • Make more than one visit to a household if needed


Hours and Pay

  • An enumerator will receive a "daily wage" for each day of training time - to be received about 2 weeks after training is completed. Training would begin on Monday or Tuesday, March 27 or 28, 1950.
  • An enumerator typically works 8 hours a day, but not necessarily "9 to 5" in order to reach people in the E.D. during the evening or on weekends. Enumeration would begin on April 1, 1950. A city assignment could be completed in about two weeks.  Assignments in country (rural) areas might take an extra week or two.
  • An enumerator who finished early and did a satisfactory job might be given additional assignments.
  • An enumerator will receive an amount for each person, farm, or other unit of enumeration. A good worker can average $7.50 to $8.50 or more for a full day - to be received about three weeks after completing an E.D.


Accurate and complete work by 140,000 census enumerators working under 8,000 crew leaders out of 450 census "district" offices and 14 census "area" offices is critical for:

  • Apportionment of Representatives to Congress (and state and local legislative bodies)
  • Administration of federal, state, and local government programs
  • Planning of government and business projects
  • Solving housing and agricultural problems


Ready to apply?  Visit your Census District Office today!


Adapted from Binder 68-D, "Taking the 1950 Census - A Message to Job Applicants" (National Archives Identifier 206240454).


Taking the 1950 Census - A Message to Job Applicants

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.  This is the 19th in a series of blog posts on the 1950 census.


This is our eighth detailed look at the Form P1, Population and Housing Schedule, that was used in most of the United States.  This post will focus on questions 35 to 38 that were asked of the person on the last “sample” line of each form who were 14 years of age and over.  You could call these people the "3 1/3 Percenters" - the select 1 out of every 30 persons who answered six additional census questions.  


Item 34 was not a question, but a direction to the enumerator:   "To enumerator:  If person worked last year (1 or more weeks in item 30):  Is there any entry in items 20a, 20b, and 20c?  If Yes, skip to item 36.  If no, make entries in items 35a, 35b, and 35c."


Items 35a, 35b, and 35c relied upon the same directions as items 20a, 20b, and 20c that were previously discussed here, but referred to the person's previous job instead of their current job. 


  • Item 35a,  "What kind of work did this person do in his last (previous) job?"  This question relied upon the same rules and instructions as Item 20a.


  • Item 35b,  "What kind of business or industry did he work in (in previous job)?" This question relied upon the same rules and instructions as Item 20b.


  • Item 35c,  Class of worker (in previous job): Private employer (P), government (G), in his or her own business (O), or without pay on family farm or business (NP).  This question relied upon the same rules and instructions as Item 20c.


Item 36,  If ever married (Mar, Wd, D, or Sep in item 12):  Has this person been married more than once?  The possible answers are Yes, No, or blank if the person has never been married.


Item 37,  How many years since this person was (last) married, widowed, divorced, or separated?  To make answering this question easier, the enumerator was told to look at Items 12 and 36 before asking the question.  Thus:


  • For persons married only once ("No" in item 36) who were coded "Mar" in item 12, the enumerator was to ask, "How many years since he (or she) was married?"
  • For persons married more than once ("Yes" in item 36) who were coded "Mar" in item 12, the enumerator was to ask, "How many years since he (or she) was last married?"
  • For persons coded "Nev" in item 12, this item was to be left blank.
  • Only whole years (not partial years) were to be counted, except that the box "Less than 1 year" was to be checked if appropriate.


Item 38, If female and ever married (Mar, Wd, D, or Sep in item 12):  How many children has she ever borne, not counting stillbirths?  The enumerator was to count all children ever born alive to this woman, including children by any previous marriage, children now deceased, and children not living in the household. Stillbirths were those in which the infant never breathed.  Logically, neither the woman's adopted children or stepchildren were to be counted.  If the respondent asked whether adopted children should be counted, the enumerator was to tell the woman that they should not be, but the enumerator was not to inquire whether any of the children had been adopted.  


Adapted from Form P1, 1950 Census of Population and Housing, and Urban and Rural Enumerator's Reference Manual, 1950 Census of the United States, pages 59-60