In 1950, for the first time, an attempt was made to identify persons of mixed white, Negro, and Indian ancestry living ia certain communities in the eastern United States in a special category so that they might be included in the categories "Other races," rather than being classified as white or Negro. This identification was accomplished with varying degrees of success, however. These groups are not shown separately, but they are included in the "Nonwhite" total. The communities in question are of long standing and are locally recognized by special names, such as "Siouian" or "Croatan," "Moor," and "Tunica." In previous censuses, there had been considerable variation in the classification of such persons by race.

AGE

Definition

The age classification is based on the age of the person at his last birthday as of the date of enumeration, that is, the age of the person in completed years. The enumerator was instructed to obtain the age of each person as of the date of his visit rather than as of April 1, 1950.

Assignment of Unknown Ages

When the age of a person was not reported, it was estimated on the basis of other available information such as marital status, school attendance, employment status, age of other members of the family, and type of household. Age was estimated by this procedure in the 1950 Census for 0.19 percent of the population of the United States. This method of assigning unknown ages on the basis of related information was used for the first time in the 1940 Census when estimates of age were made for 0.16 percent of the population of the United States. In previous censuses, with the exception of 1880, persons of unknown age were shown in a separate category. The summary totals for "14 years and over" and "21 years and over" for earlier censuses presented in this volume include all persons of "unknown age" since there is evidence that most of the persons for whom age was not reported were in the age classes above these limits.

Quality of Data

According to the Post-Enumeration Survey, the quality of 1950 Census statistics on the age distribution of the population is generally good. The Post-Enumeration Survey verified the 1950 Census report on age by means of a more detailed question on exact date of birth and by on-the-spot probing of discrepancies between the two reports. In addition to age reporting errors revealed in this fashion, the Post-Enumeration Survey also shows the effect of net underenumeration of the population on the age distribution.

The net age reporting error, resulting from erroneous reporting of age for persons counted in the census, is not in the same direction for all 5-year age groups. In those from 5 to 49 years, there is generally a net overstatement resulting from age misreporting, whereas in the remainder of the age span there is a net understatement. This pattern is contrary to the expectation that there would be an overstatement in the age group 65 and over due to exaggeration of age for older persons. The absolute amounts of net age misreporting are small.

The estimated net underenumeration for the entire population was 1.4 percent. (See p. 7.) Although the Post-Enumeration Survey sample does not permit making reliable estimates of differences in net underenumeration among specific age groups, it does indicate that the net underenumeration was less than the average for the ages 15 to 39.

The combination of underenumeration and age reporting errors results generally in small net undercounts in the 5-year age groups. The largest undercounts occur in the age groups under 5 years and 45 years and over. In general, the effect of the net

the median age shown in the census appears to be low by only about 0.1 year.

The findings of the Post-Enumeration Survey with respect to age do not agree completely with evidence from vital statistics, which point to a greater undercount of children under 5 years old than that estimated by the Post-Enumeration Survey. The Post-Enumeration Survey results relating to young males 15 to 34 and to persons 65 and over do not accord with prior expectation. It was believed that there would be more underenumeration of young males (because of their mobility) and a considerable amount of overreporting of age by older persons. Plans are being made to study these discrepancies.

The accuracy of the statistics for single years of age was not measured by the Post-Enumeration Survey. There appears to be a tendency in the 1950 Census data, as in earlier census data, toward the overreporting of ages ending in 0, 2, 5, and 8, that is, the frequencies for these single years of age tend to exceed those for the adjoining years. For this reason, the medians in the tables presenting statistics in single years of age are computed on the basis of 5-year age groups.

NATIVITY

Because of the declining numerical importance of the foreignborn population, nativity has not been used so extensively for cross-classifications in 1950 as in earlier censuses. Information on the nativity and parentage of the white population and country of origin of the foreign-white stock are published in the Population Special Report, "Nativity and Parentage." The distribution of the separate nonwhite races by nativity and more detailed data on the foreign-born nonwhite population are presented in “Nonwhite Population by Race."

Definitions

The statistics on nativity are based on the responses to the question, "What State (or foreign country) was he born in?" asked of all persons. In this volume, the population is classified according to place of birth into two basic groups, native and foreign born. A person born in the United States or any of its Territories, possessions, etc., is counted as native. Also included as native is the small group of persons who, although born in a foreign country or at sea, were American citizens by birth because their parents were American citizens. Since the Republic of the Philippines was established as an independent country in 1946, persons living in the United States who had been born in the Philippine Islands were classified as foreign born in the 1950 Census whereas in earlier censuses they had been classified as native. The small number of persons for whom place of birth was not reported were assumed to be native.

Quality of Data

The data on nativity have a high degree of accuracy. According to the Post-Enumeration Survey, the estimated net underenumeration rates are 1.4 percent for the native population and 0.6 percent for the foreign born. Additional error resulting from the actual misreporting of nativity for persons properly counted in the census was negligible. The gross reporting error was 0.6 percent, and the differences tended to cancel.

CITIZENSHIP Definitions

The statistics on this subject are based on replies to the question, "Is he naturalized?" asked of all persons of foreign birth. The classification of the population by citizenship embraces two major categories, citizen and alien. Citizens are subdivided into native and naturalized. It is assumed that all natives are citizens of the United States. In addition to the citizen and alien categories, there is a third group, made up of foreign-born persons from whom no report on citizenship was obtained, designated "citizenship not

Quality of Data

In the 1950 Census, the question on citizenship was answered for approximately 92 percent of the foreign-born population and no report was made for 8 percent. The Post-Enumeration Survey results indicate that errors in the answers on citizenship were in the direction of overstating the number of naturalized foreign born and understating the number of aliens.

For the 8 percent not reporting on citizenship in the census, the Post-Enumeration Survey results indicate a citizenship distribution not essentially different from that of the foreign-born population for which the information was reported. This conclusion is contrary to the assumption that the category "Citizenship not reported" comprises persons who are actually aliens, for the most part. This assumption is reflected in the combination of this category with the alien category in some of the tables on citizenship in this and earlier censuses.

COUNTRY OF BIRTH OF FOREIGN-BORN WHITE

Definitions

The statistics on this subject are based on the respondent's answer to the question, "What State (or foreign country) was he born in?" asked of all persons. In case questions arose in the interview involving changes in international boundaries, the enumerator was instructed to decide the question on the basis of the international boundaries as of April 1, 1950. In coding and editing, however, changes in boundaries which had resulted from World War II were accepted only when such changes had been officially recognized by the United States.

Comparability

In 1940 the classification of the population by country of birth was based on the political boundaries of January 1, 1937. The corresponding 1930 data are based on the political boundaries of