August 11, 2014

Puerto Rican Population Declines on Island, Grows on U.S. Mainland

Appendix C: Data Sources

The data in this report come mainly from five U.S. Census Bureau sources: the American Community Survey for 2006-2012; the Puerto Rican Community Survey for 2006-2012; the decennial census for 1980-2010; the Current Population Survey for 2007-2013 and annual population estimates for the island of Puerto Rico. For all sources except the population estimates, data was accessed via the Integrated Public Use Microdata Series (IPUMS).15

The ACS is the largest household survey in the United States, with an initial sample of about 3.7 million selected addresses among the 50 states and District of Columbia. The PRCS, which is similar in design and methodology, includes an initial sample of 37,000 selected addresses in Puerto Rico. The survey covers the topics previously covered in the long form of the decennial census. It is designed to provide estimates of the size and characteristics of the resident population, which includes people living in households and group quarters. The ACS and PRCS are the source of most information about characteristics of the Puerto Rican origin population.

The specific data sources for this report are the 1% sample of the 2006-2012 ACS and 2006-2012 PRCS Integrated Public Use Microdata Series (IPUMS) and the 5% sample of the 1980, 1990 and 2000 censuses IPUMS provided by the University of Minnesota. The IPUMS assigns uniform codes, to the extent possible, to data collected by the decennial census and the ACS from 1850 to 2012.

In the interest of greater accuracy, the Pew Research Center developed new survey estimates for the 2006-2009 ACS and PRCS that are consistent with the results of both the 2000 and 2010 censuses. These new survey estimates are based on the Census Bureau’s revised annual intercensal population estimates, i.e., those that are consistent with both the preceding and subsequent censuses.

Characteristics of Puerto Ricans in the U.S. and on the island are based only on those who self-identify as Hispanics of Puerto Rican origin. Not everyone born on the island of Puerto Rico or currently living on the island of Puerto Rico is considered Puerto Rican in this analysis, as not all of them are of Puerto Rican origin. In most cases, individual or household characteristics refer to those at the time of the survey. However, earnings, household income and poverty status are based on the respondent’s income characteristics in the 12 months prior to the survey.

Due to differences in the way in which the IPUMS and Census Bureau adjust income data and assign poverty status, data provided about these variables might differ from data on these variables that are provided by the Census Bureau. Dollar figures in the report were converted to 2012 dollars using the research series of the consumer price index (CPI-U-RS) for the mainland and the consumer price index downloaded from the Instituto de Estadísticas de Puerto Rico for the island. In addition, in table 1 of Appendix A, personal earnings and household income figures for people living on the U.S. mainland are inflated by 15% in order to account for the higher cost of living on the island of Puerto Rico.

The 1980, 1990 and 2000 decennial censuses are the source of population data about Hispanics of Puerto Rican origin for those years.

In addition, characteristics of recent migrants in two time periods were generated using data from both the decennial census and American Community Survey. For recent migrants from Puerto Rico living in the U.S. in 2012, the 2012 ACS was used to analyze people who self-identified as Puerto Rican, were born on the island of Puerto Rico and first came to live in the U.S. between 2000 and 2012. For recent migrants from Puerto Rico living in the U.S. in 2000, the 2000 decennial census was used to analyze people who self-identified as Puerto Rican, were born on the island of Puerto Rico and first came to live in the U.S. between 1988 and 2000.

Finally, the analysis of migration patterns from the island to the mainland and across states and regions within the mainland uses a pooled sample of the 2006-2012 ACS in order to obtain a large enough sample size. The IPUMS variable MIGPLAC1 was used to identify moves—because the ACS records residence in the year prior to the survey, and the analysis combines six years of data, the data likely include multiple moves by some people. As a result, the report refers to “moves” rather than “movers.” Furthermore, while MIGPLAC1 includes moves to different households within a state, the analysis did not include these intra-state moves.

Current Population Survey

The Current Population Survey (CPS) is a monthly survey of about 55,000 households conducted jointly by the Bureau of Labor Statistics (BLS) and the Census Bureau. The CPS was the source of information about the reasons that island-born Hispanics of Puerto Rican origin moved to the mainland United States. To achieve an adequate sample, surveys of 2007-2013 were combined.

Because of data limitations, the analysis could not isolate only those island-born Puerto Rican origin migrants who moved from the island to the mainland, so the sample probably includes a small number who moved from another U.S. territory or foreign country. According to the 2012 American Community Survey, 92.7% of island-born Puerto Ricans who migrated to the U.S. from abroad in the 12 months prior to the survey had come from Puerto Rico.

The category “job-related reasons” includes new job or job transfer; to look for work or lost job; easier commute; and other job-related reasons. The “household/family” reasons include change in marital status; to establish own household; and other family reasons. The “housing” category includes wanted better neighborhood; for cheaper housing; and other housing-related reasons. The residual “other reasons” includes a general other-reasons category, as well as attend/leave college and health reasons.

Other data

Population estimates for the island are the source of the 2013 Puerto Rico population total, for some estimates of migration from the island, and for population change in island municipios. The estimates are generated annually, along with revised and updated estimates for prior years. They are based on government data on births, deaths, immigration and migration, rather than an actual count. The bureau’s annual estimates include a breakdown of the components of population change, including net migration. The bureau also produced estimates for 1980-1990 and for 1990-2000.

Maps

The maps of the 50 states and D.C. showing the 2010 and 2000 county-level Puerto Rican populations are based on U.S. Decennial Census Summary File data, as well as 2006-2010 5-year American Community Survey data, downloaded from the U.S. Census Bureau’s American Fact Finder. These maps show the distribution of Hispanics of Puerto Rican origin, as well as approximations of the distributions of island-born and mainland-born Hispanics of Puerto Rican origin. Approximations were used for the island born because of limitations in the population groups available in the ACS data, which group together all people born on the island of Puerto Rico, including some Hispanics who are not of Puerto Rican origin and some non-Hispanics. According to Pew Research analysis of the 2012 American Community Survey, 93% of people born in Puerto Rico who are currently residing in the U.S. are Hispanics of Puerto Rican origin, 3% are non-Hispanic, and the remaining 4% are Hispanics of other origin groups. Thus, the island-born approximations used in these maps could overstate the true number of island-born Hispanics of Puerto Rican origin.

Another limitation of the data used to produce the maps is that the overall Puerto Rican origin population living stateside includes a small number of people born in another country (equaling 2.2% of the total Puerto Rican origin population in the U.S., according to analysis of the 2012 American Community Survey). To produce an approximation of the county-level mainland-born Puerto Rican population, the population of all people born on the island of Puerto Rico was subtracted from the total Puerto Rican origin population in each county. As a result, because some people who are not Hispanics of Puerto Rican origin are included in the island-born population, this could lead to a slight underestimate of the mainland-born Puerto Rican county populations (as an overly large number was subtracted from the base of all Puerto Ricans). Second, because some people who were not born on either the island or the U.S. mainland are included in the total Puerto Rican population, this could lead to a slight overestimate of the mainland-born Puerto Rican population.

For 2010, data are also approximations for both the island-born and mainland-born due to the differences in data sources—2010 SF-1 data is used to calculate the entire Puerto Rican origin population, while 2006-2010 ACS data is used to calculate the island-born and mainland-born components of this population. For 2000, decennial census Summary File data is used for all three populations.

  1. Steven Ruggles, J. Trent Alexander, Katie Genadek, Ronald Goeken, Matthew B. Schroeder, and Matthew Sobek. Integrated Public Use Microdata Series: Version 5.0 (Machine-readable database). Minneapolis: University of Minnesota, 2010.