A Question on Religion in the US Census

Zahid Bukhari

The United States Census Bureau produces quality data about almost all aspects of the nation’s people and the economy. One can find in its annual reports personal data on class, race, ethnicity, gender, age, education, profession, occupation, income level, marital status, place of birth and origin of immigrant country. Besides the details of households’ facilities, the census reports also include data on Americans’ agricultural and industrial activities and resources. However, it is quite astonishing that the information on the religion of the American people is totally absent from the national census. These socio-economic variables are especially important in understanding the human behavior. While we question most of these variables, the “R” question is conspicuously absent. Several other developed countries including England, Canada and Australia, on the other hand, include the religion question in their censuses.

The impact of religion and religious practices on the socio-political behavior of an individual person is increasingly becoming a significant topic to be studied. In recent decades, religion has emerged as one of the most important factors in the national as well as international arenas. Historically, social scientists have tried to deny or ignore its importance and instead asserted the waning impact of religion in developed societies. For them religion would have had no role in the future of a modern human being. However, all the claims of disappearing importance of the religion factor from human behavior proved the opposite. The last two presidential campaigns showed the decisive role played by certain religious groups and moral value voters on the outcome of the elections’ results and ultimately on the direction of the national policies. The relationship of religion-based institutions with family related issues, community development projects and social service initiatives has also grown tremendously in recent years in our society.

The first U.S. census was taken in 1790 and it counted 3.9 million inhabitants. Initially, questions on other topics, including religion, were not part of the census. Gradually, the census included several other topics in its questionnaire. The Census Bureau took censuses of religious bodies from 1850 to 1936 with only a brief interruption. They compiled the statistics on membership number of churches and synagogues by requesting the religious bodies themselves. The Bureau cancelled the program in 1946 and one of the reasons was the opposition of Christian Scientists, a group doctrinally opposed to enumeration.

In 1956, the U.S. Census Bureau announced that it was considering a question on religion in the census of 1960. That announcement initiated a vigorous debate in the national media. Different religious groups and other civil organizations voiced opposition to the issue. As a result, Bureau scraped the idea of including a question on religion in the forthcoming census of 1960. They argued that a census question of religious affiliation invades privacy and violates the 'separation of church and state' provision in the First Amendment.

In 1976, the 94th Congress amended Title 13 of United States Code (that part of U.S. legal code that pertains to the census), allowing for the collection of a mid-decade census of population. However, the amendment specifically prohibited asking religious-orientated questions by stating “that a person may not be compelled to disclose information regarding his religious beliefs or membership in a religious body.” The result is that we have data on all aspects of human life other than religion.

Ironically, the Bureau did conduct a test of a question on religion in March 1957 using the Current Population Survey, a Bureau-directed monthly survey of 35,000 households. The question was: “What is your religion? Baptist, Lutheran, etc.” In fact, 96% of respondents complied, that is, they reported a religious preference. According to that survey the Americans of 14 years and above stated their religion as following:

White Protestant 57.4% 
Black Protestant 8.8 % 
Catholic 25.7 % 
Jewish 3.4 % 
Other religion/not reported 2.2 % 
No religion 2.7% 
Total 100%

The US population was 170 million in early 1957 and now, fifty years later, it reached 300 million. During this period, the American religious landscape has also changed significantly in four ways: 1) the adherents of religions other than Christianity are increasing every year; 2) the number of mixed-religion households is on rise (due to inter-faith marriages); 3) among religions, the composition of various denominations or sub-groups is changing and they are also becoming more important due to their socio-political activities; and 4) the number of Americans unaffiliated with religion is also growing.

Various research institutions and survey organizations have been doing surveys about the religious preferences and the impact of religion on American society. They have been producing valuable data and useful reports. However, in the absence of any official data, various ethno-religious groups have made their own estimates and this practice is also to counter estimates from other practioners. The exercise has sometimes confused the religious picture of the American population. For example, what are the numbers of the evangelical Christians in the population? A recent Baylor University study put the percentage at 33.6, roughly 100 million people. At the same time, a study by the Bliss Institute of University of Akron put the percentage at 26.3, roughly 79 million people – a difference of almost 21 million followers. The same is the case with the Catholic population numbers. The range of the estimates is between 22% and 26%, a difference of about 12 million people. The Latino population has surpassed the African-American population in the United States. How many of them are Catholics or Protestants? One may estimate on the basis of survey results but receiving the exact answer is difficult to achieve.

The same issue applies to the Jewish population. In 2000-2001, a $6 million National Jewish Population Survey, sponsored by the United Jewish Communities, announced that the Jewish number dropped by 400,000 within a decade from 5.8 million to 5.2 million. The Jewish scholars raised several questions about the methodology of the survey in a colloquium held at Brandeis University after the results were released. Some of the Jewish scholars recommended strongly that the religion question should be asked in the decennial census.

The number of American Muslims is another interesting episode. On February 21, 1989, the New York Times published a story and put the American Muslim number at 6 million. After almost twelve years, Tom Smith, a professor at the University of Chicago’s, in a sponsored study for the American Jewish Committee, estimated that the total Muslim population in the USA was 1.9 million in 2001. According to Smith the “Muslim population is commonly overestimates by a factor of three or four ….” Other scholars, however, strongly disagreed with his estimate and put the numbers in the range of 6 – 8 million. Even Hollywood took part in the debate when, in 2005, one of the characters in the film Syriana claimed there were 10 million Muslims in America.

Another troublesome question for the scholars and researchers is whether Americans are becoming more unaffiliated to any religion or non-religion over time. Their numbers are on the rise but the extent of their growth is unknown. The percentage of this category has been estimated in the range of 10% to 14%. The social and political behaviors of this category need more analysis. However, the absence of any reliable data does create a serious problem.

We should also keep in mind that with the census data on religion, all of the questions would not be resolved automatically. However, the students, scholars, media pundits and policy makers would have some reliable numbers to report and offer informed critical analysis of Americans’ behavior in the social, economic and political fields. The religion data would certainly enhance the capacity of the religious communities for planning their future activities, establishing the places of worship, developing certain businesses, and implementing community development programs all around the country

What we need is a comprehensive religion census in the USA. That census would provide the data about religious preference, unaffiliated or non-religion, religious bodies, places of worship, faith based institutions including educational, health and social services. The process could be started by including a question on religion in the forthcoming census of 2010. Congress should act now and allow the Census Bureau to obtain this information about the American population.

Zahid Bukhari was the former Project Director of the American Muslim Studies Program at the Prince Alwaleed Bin Talal Center for Muslim-Christian Understanding.