INTRODUCTION

THE FOREIGN-BORN POPULATION BY WORLD REGION OF BIRTH AND BY COUNTRY OF BIRTH,

FOR THE UNITED STATES AND STATES: 1850 TO 2010 (last revised, Feb. 18, 2023)

Presenting data from the decennial census of population for 1850 to 2000, and from the American Community Survey for 2010, for use by a wide-ranging audience of persons interested in U.S. history and state histories.     

Data on country of birth for the foreign-born population of the United States were collected in every decennial census from 1850 to 2000, and subsequently, these data have been collected in the American Community Survey (ACS). The purpose of this website is to present data on the total population by nativity (native and foreign born) and on the foreign-born population by world region of birth and by country of birth for the United States and individual states (and territories preceding statehood) from 1850 to 2010. Data shown for 2010 are five-year average data for 2008-2012 from the ACS. These data are of interest both because of the importance of immigration in American history and because there have been major differences among the states in levels and trends in both the numbers and proportions foreign-born. In addition, there have been major differences among the states in the sources of their foreign-born populations, as defined both by world region of birth and by individual countries of birth.

                The data presented here may be viewed as an update and expansion of data presented in Historical Census Statistics on the Foreign-Born Population of the United States: 1850 to 2000 (U.S. Census Bureau, Population Division, Working Paper No. 81, February 2006). More specifically, the data presented on this website update data presented in Tables 1-4 in Working Paper No. 81 and expand coverage for the United States to include data for individual states. It is recommended that users of data presented on this website refer to the report cited for a general discussion of decennial census data on the foreign-born population, including information on sources of the data, definitions, and accuracy. For graphics showing historical data on the foreign-born population of the United States, see the American Demographic History Chartbook: 1790 to 2010, Chapters 11-13.

                 It should be stressed that all census data are subject to error, as discussed in the sources noted above. In the case of data on country of birth, there are several specific concerns for the data user. The first concern is the accuracy of reporting when a country has undergone boundary changes during the lifetime of the respondent, who may not have reported country of birth based on current boundaries.

The second concern is that due to boundary changes (e.g., in Europe and the Middle East after World War I), data for specific countries frequently are not directly comparable over time. To a lesser degree, data are not always comparable for higher levels of geography. For example, the Azores were included in Africa through 1970, but were then made part of Portugal and thus included in Europe.

                 The third concern is that data on country of birth do not always reflect ethnic or cultural heritage. For example, among the foreign-born White population from Austria in 1930, 55 percent reported German as their mother tongue (language spoken in the person’s home when he or she was a child), but an additional 9 languages were reported by at least 1 percent of the foreign-born population from Austria (Demographic Chartbook, Figure 13-7). Among the foreign-born White population from Russia in 1930, 65 percent reported Yiddish as their mother tongue, 23 percent reported Russian, and 9 percent reported German. In parts of the Midwest, German was the leading mother tongue of the foreign born from Russia. 

 The fourth concern is sampling error when published data on country of birth of the foreign-born were first based on sample data (in part in 1950, and entirely since 1960). This is a particular concern in the case of data at the state level, where many of the published numbers for individual countries are very small and thus subject to high proportionate sampling error. 

                 It should be noted also that the U.S.-born children of immigrants are native (not foreign-born), and thus without continued immigration, at the national level the foreign-born population from a specific country can only decline over time due to attrition (i.e., death). Whether the foreign-born population from a specific country increases or decreases over a specific time period (apart from sampling error and other errors in the data) depends on whether immigration exceeds the total of emigration and attrition during this time period. In the case of the foreign-born population at the state level, changes over time are influenced also by interstate migration among the foreign-born.

TABLES AND GRAPHICS

This section provides a brief description of the tables and graphics included on this website.

Table A-1. This is the only A table and presents data on the foreign-born population of the United States by world region of birth and country of birth from 1850 to 2010. This four-page table attempts to make the data as comparable as possible over time and includes some notes about changes in the names of countries. As noted previously, the data user needs to be aware of issues about the  comparability over time of data on country of birth.

Tables B-1 to B-15. These 15 tables present data on the foreign-born population of the United States and states by world region of birth and country of birth for individual years from 1850 to 2010, excluding 1940 and 1950, as discussed later. These tables range in length from four pages to nine pages.

Tables C-1 to C-52. These 52 one-page tables present time-series data up to 2010 on the population by nativity and on the foreign-born population by world region of birth for the United States, the 50 states, and the District of Columbia.

Tables D-1 to D-52. These 52 one-page tables present time-series data up to 2010 on the ten leading countries of birth of the foreign-born population for the United States, the 50 states, and the District of Columbia. The rankings in these tables for data for 1960 to 2010 are based on the point estimates of data collected on a sample basis and thus are subject to sampling variability.

One limitation of these tables is that the published detail on country of birth, especially for non-European countries, was quite limited for censuses before 1960. Thus, a country may have been in the top ten countries of birth for a state, but data for that country were not published. For the most part, this does appear to be a serious problem, with one notable exception: missing data for Florida on the foreign-born from Cuba. Data on the foreign-born from Cuba were first published in the 1870 census, then again in the 1880 census, when Cuba was the leading country of birth of the foreign-born for Florida.  For the 1890 census, no data were published separately for Cuba, which was included in the West Indies.  Data showing Cuba were published for states for 1900, but in 1910, 1920, and 1930, data on the foreign-born for Cuba were not published at the state level. The lack of data for Cuba seriously distorts the presentation of data for Florida in Table D-11.

For use in Table D-11 for Florida, the foreign-born population from Cuba has been estimated for 1890, 1910, 1920, and 1930. The procedure for doing this used published data for 1880 and 1900, and also for 1950, when data on the foreign-born from Cuba and other West Indies were published for White for each state and for Nonwhite for some states, including Florida.  (See later discussion on data on the foreign-born population for 1940 and 1950.)   The procedure used linear interpolation over time between 1880 and 1900 and between 1900 and 1950 to create preliminary estimates for Cuba and other West Indies. These preliminary estimates were then controlled pro-rata to published totals for the West Indies.

Tables E-All Countries and Table E-Afghanistan to Table E-Yugoslavia. These 124 one-page tables present time series data (from 1850 to 2010 or some part of this span) on the ten leading states of residence for the foreign-born population from all countries and for individual countries of birth. Data are shown for each country with at least one year in the 1850-2010 span in which there were 10,000 or more foreign-born from that country and for which the data are available at the state level. As noted earlier for the D tables, for 1960 to 2010, the rankings are based on sample data.

In addition to tables for countries, these tables include a Table E for Atlantic Islands. The reason is that this area was a prominent area of birth for some states (most particularly Massachusetts, Rhode Island, and California) and for some cities in those states (e.g., Fall River, Massachusetts). It was not until the 1920 census that any data were published separately for the Azores, when the Azores accounted for 77 percent (33,995 of 44,340) of the foreign-born population from the Atlantic Islands.

Figures 1 to 52. These 52 one-page graphics show selected data from the C and D tables for the United States, the 50 states, and the District of Columbia. The graphics include numbers in thousands and short abbreviations for country names.  The corresponding C and D tables given in the headnote for each figure, provide more detail.

DATA ISSUES BY YEAR

                 This section provides a brief discussion of data issues in the 1850 to 2010 period, as reflected in the B tables. These tables use the country names as shown in the respective censuses (e.g., British America for 1850 to 1880 and Canada subsequently). Unfortunately, there are several cases where data on the foreign-born population by country of birth were published with more detail for the United States than for individual states. These cases are identified with an asterisk in Table A-1. 

1850 and 1860. Data presented in Table B-1 (1850) and Table B-2 (1860) reflect data published in census reports for those censuses. For reasons that are not clear and with a methodology that was not stated, the numbers for the total foreign-born population (but not the numbers by country of birth) by state for 1850 and 1860 were revised slightly for publication in the 1870 census, and these revised totals for 1850 and 1860 were carried forward in historical tables in subsequent censuses and in Working Paper No. 81, Table 14. In some states, the revised totals for the foreign-born population were lower than the original totals for the foreign-born with country of birth specified. The two sets of numbers on the total foreign-born population by state in 1850 and 1860 can be compared in Table B-1 and Table B-2, respectively, to see the size of the changes.

1870 to 1900. These data are presented in Table B-3 (1870), Table B-4 (1880), Table B-5 (1890), and Table B-6 (1900). There do not appear to be any major data issues, although it is unfortunate that data for Cuba, which were shown starting in 1870, were not shown separately in 1890, when Cuba was combined with the rest of the West Indies in data shown for the United States and states.

1910. Data for Poland starting in 1860 and for Bohemia starting in 1870 were shown separately, even though they were not independent countries. In 1910 census reports, persons born in Poland were distributed primarily to Austria, Germany, and Russia, based on their reported mother tongues. And persons born in Bohemia were included with Austria. The 1920 census of course reflected changes to the map of Europe after World War I, including the creation of the countries of Poland and Czechoslovakia.  Table B-7 (1910) includes estimates for Poland in 1910, as published in 1920 census reports, with corresponding estimates for Austria, Germany, and Russia in 1910. In addition, Table B-7 includes estimates for Bohemia in 1910, with corresponding changes to the estimates for Austria. The result is a reasonably consistent set of numbers for Poland for 1900 to 1920 and a reasonably consistent set of numbers for Bohemia for 1900 and 1910. The new nation of Czechoslovakia was larger than Bohemia, and thus data for Bohemia in 1910 and Czechoslovakia in 1920 are not directly comparable.

1920 and 1930. These data are presented in Table B-8 (1920) and Table B-9 (1930). There do not appear to be any major data issues.

1940 and 1950. With the exception of 1850 and 1860, data on the foreign-born by country of birth have been collected for the total population (i.e., including all races).  In 1850 and 1860, the data were collected and published for the free population.  It is assumed here, as it was in 1870 census and subsequent census reports, that the entire slave population was native. This appears to introduce a small amount of error, suggested by the fact that only 0.2 percent of the Black population in 1870 was foreign born.

For the censuses of 1870 through 1930, data on the foreign-born population by country of birth were published for the total population. For the census of 1940 and 1950, the focus in publications was on the White population.  For 1940, the number of Nonwhite foreign born was published for states, with no country of birth detail. For 1950, data on the Nonwhite foreign-born population with some world region and country of birth detail was published for several states with the largest Nonwhite foreign-born populations.

Published data on the foreign-born country of birth by White and Nonwhite for 1930 and 1960 have been used in combination with the published data described for 1940 and 1950 to develop estimates of the Nonwhite foreign-born population by country of birth for 1940 and 1950, with these estimates then being added to the published data for the White foreign-born population to produce estimates for the total foreign-born population, which are included in the tables on this website. A short paper on the development of these estimates is available as an appendix on this website and as a table.

These estimates for 1940 and 1950 were prepared only for the United States, not for individual states; however, the methodology for preparing estimates for the United States could be extended to develop estimates for individual states.  It would be a much bigger project to develop estimates for all states, which would then need to be adjusted to conform to national totals as well as to available state totals. 

It would appear that a good alternative to developing estimates for unpublished data for 1940 and 1950 would be to use the IPUMS (Integrated Public Use Microdata Series) 100-percent files for the 1940 census and, when it becomes available, for the 1950 census.  However, it turns out that there are major problems with using the IPUMS files. There are some substantial differences between published numbers and IPUMS numbers.  For example, the published number of Nonwhite foreign born in the 1940 census is 175,758 compared with 215,017 (22 percent larger) in the IPUMS file. There are several countries for which number of White foreign born differs substantially between the published number and the IPUMS number, perhaps reflecting editing procedures used in preparing the published data.   One general problem with the IPUMS data, based on my use of it for small-area historical research, is that it includes individuals on the enumerator schedules whom the enumerators crossed out (e.g., because the enumerator found out after listing the individuals that they should have been enumerated elsewhere based on their usual place of residence). These individuals were not included in published population totals, but they were included in the IPUMS files (perhaps to maximize information available for genealogical research). If there were no published data on the foreign-born population by country of birth for 1940, then the choice would be to use the IPUMS data. But the fact that over 98 percent of the foreign-born population was classified by country of birth in published data for 1940 (all of the White foreign born) and for 1950 (all of the White foreign born and the majority of the Nonwhite foreign born) suggests that it is best to use the published data on the foreign born population by country of birth and to develop estimates for the less than 2 percent for whom data were not published in census reports.

1960 and 1970. These data are presented in Table B-10 (1960) and Table B-11 (1970).  There do not appear to be any major data issues. As noted earlier, the 1960 census was the first in which all data on the nativity of the population and on the country of birth of the foreign-born population were collected on a sample basis. In order to provide a general indication of the size of the sampling error, some rough approximations are shown here for 1960 (when the sample size of 25 percent was higher than in subsequent years) of estimated numbers and (in parentheses) their standard errors: 50 (15), 100(20), 250 (30), 500 (40), 1,000 (50), 5,000 (110), 10,000 (160), 25,000 (250), and 50,000 (350).

Some rough approximations of standard errors of percentages (expressed in percentage points) for 1960 follow.  These standard errors depend on both the base of the percentage and the estimated percentage itself.

                                     Base of Percentage