Over the past 20 years, the United States has granted permanent residency status to an average of about 1 million immigrants each year. These new “green card” recipients qualify for residency in a wide variety of ways – as family members of current U.S. residents, recipients of employment visas, refugees and asylum seekers, or winners of a visa lottery – and they include people from nearly every country in the world. But their geographic origins gradually have been shifting. U.S. government statistics show that a smaller percentage come from Europe and the Americas than did so 20 years ago, and a growing share now come from Asia, sub-Saharan Africa and the Middle East-North Africa region.
With this geographic shift, it is likely that the religious makeup of legal immigrants also has been changing. The U.S. government, however, does not keep track of the religion of new permanent residents. As a result, the figures on religious affiliation in this report are estimates produced by combining government statistics on the birthplaces of new green card recipients over the period between 1992 and 2012 with the best available U.S. survey data on the religious self-identification of new immigrants from each major country of origin.
While Christians continue to make up a majority of legal immigrants to the U.S., the estimated share of new legal permanent residents who are Christian declined from 68% in 1992 to 61% in 2012. Over the same period, the estimated share of green card recipients who belong to religious minorities rose from approximately one-in-five (19%) to one-in-four (25%). This includes growing shares of Muslims (5% in 1992, 10% in 2012) and Hindus (3% in 1992, 7% in 2012). The share of Buddhists, however, is slightly smaller (7% in 1992, 6% in 2012), while the portion of legal immigrants who are religiously unaffiliated (atheist, agnostic or nothing in particular) has remained relatively stable, at about 14% per year.1
Unauthorized immigrants, by contrast, come primarily from Latin America and the Caribbean, and the overwhelming majority of them – an estimated 83% – are Christian. That share is slightly higher than the percentage of Christians in the U.S. population as a whole (estimated at just under 80% of U.S. residents of all ages, as of 2010).2
These are among the key findings of a new study by the Pew Research Center’s Forum on Religion & Public Life examining recent trends in the geographic origins and religious affiliation of immigrants to the United States. (For information on religion among migrants not just in the U.S. but globally, see the Pew Research Center’s 2012 report “Faith on the Move: The Religious Affiliation of International Migrants.”)
About the Estimates
The U.S. government does not collect data on the religious affiliation of immigrants. However, estimates can be made using information gathered by the Department of Homeland Security on the countries of origin of new green card recipients. To estimate the religious breakdown of immigrants from each country, the Pew Research Center relied primarily on the New Immigrant Survey, a nationwide survey conducted in 2003 by scholars at the RAND Corporation, Princeton University, New York University and Yale University that asked more than 8,500 recent legal immigrants about their religion, among other questions.3
The use of survey data along with country-of-origin data improves the reliability of the estimates because, in some cases, the religious makeup of migrants differs from the religious composition of the overall population in their country of birth. This study does not automatically assume, for example, that if the population of Country A is 75% Muslim, then 75% of migrants from Country A to the United States must be Muslim. On the contrary, the study uses data from the New Immigrant Survey on the religious breakdown of new U.S. green card recipients to estimate the religious affiliation of the vast majority (95%) of legal immigrants.
On the other hand, the use of a single survey conducted at one point in time (albeit roughly in the middle of the 20-year period under examination) may introduce some time-related bias. In the absence of survey data on new immigrants from other years, this study assumes that throughout the period from 1992 to 2012, the religious breakdown of legal immigrants to the U.S. from each country of origin was the same as in 2003, at the time of the survey. For example, if the New Immigrant Survey found that 60% of new green card recipients from Country B were Christian and 40% were Buddhist, then those percentages were applied to the number of new green card recipients from Country B in every year from 1992 to 2012. This means that all of the estimated change in the religious makeup of legal immigrants reported in this study is a result of shifts in their geographic origins. This study is unable to capture any changes that may have occurred in the religious mix of migrants from a particular country.
Other evidence attests to some of the broad patterns identified by the study, however. For example, Pew Research Center surveys of U.S. Muslims in 2007 and 2011 suggest that the number of Muslims living in the United States rose in that four-year period by about 300,000 adults and 100,000 children, to a total of about 2.75 million Muslims of all ages – a rate of increase that is in line with the estimates in this report and would be difficult to explain without rising immigration. Similarly, a Pew Research Center survey of Asian Americans in 2012, combined with U.S. Census data, suggests that the number of Hindus in the United States has been increasing in large part due to rising immigration over the past two decades.
To reflect some of the uncertainty inherent in the estimates, all population numbers in this report are rounded to the nearest 10,000 and percentages are rounded to whole numbers.
For more explanation of how the estimates were calculated, see the methodology.