Overview:
Have you ever wondered why the US census doesn't include an explicit question on respondents' religion?
As analysis of the data from the 2020 UK census filters into the public domain, a question arises: Why does the US census not ask questions pertaining to religion? Only about a quarter of Americans know the answer.
In the mid-to-late 1950s, the Census Bureau asked a question on religious affiliation ahead of the 1960 census. One of the worries at the time was that such questioning would challenge the separation of church and state. The director of the Bureau at the time felt that a question on religion wouldn’t threaten the enshrined notion. A voluntary question to 431 Milwaukee households saw only three declining to answer.
In 1957, the question was broadened to 40,000 households, but concerns were raised, including from some religious groups that had experienced persecution. Though the question was not included in the 1960 census, the data from 1957 was published. There were 35,000 responses, of which two-thirds of people aged 14 and older identified as Protestant, about a quarter declared themselves Catholic, and just over 3% were Jewish. In total, 96% reported an affiliation, while 2.7% reported no religion, and 1% did not answer. Interesting data, nonetheless.
During the next two decades, there was some debate about including the question until it was decided in 1976 that there would be no religion question in the 1980 census. The Census Bureau’s director, Vincent P. Barabba, declared that “asking such a question in the decennial census, in which replies are mandatory, would appear to infringe upon the traditional separation of church and [s]state. Regardless of whether this perception is legally sound, controversy on this very sensitive issue could affect public cooperation in the census and thus jeopardize the success of the census.”
Later that year, Congress formally prohibited the census from asking a question about “religious beliefs” or “membership in a religious body,” given that this was something that citizens were legally required to complete.
The issue of church and state separation is seen as the most common criticism of including such a question, with others thinking the question would deter respondents from filling out their forms. Response rate is a crucial metric to consider.
In a paper in the Journal of American History, Kevin Schultz found the following:
To American Jews, the proposal seemed a clear “violation” of the separation between “Church and State” that, in the shadow of the Holocaust, appeared a frightening means of centralized control. To Catholics, the proposal’s success would have achieved a “long desired objective,” and so the Catholic Church in America threw all its institutional weight behind the measure. To most Protestants, the issue was less contentious, perhaps because it was all so new. In the 1957 words of the Christian Century, the nondenominational voice of the Protestant establishment, because “religious affiliation has become hardly more than a matter of sociological identification in America,” many Protestants were hard put to understand all the fuss about just “another automatized item punched on an IBM card.” But there they were, three sides embroiled in a two-year debate about whether the federal government should put a question about religion on the United States census of 1960.
This is a shame, because the data would be a fascinating insight into the religious/nonreligious landscape in the US. Instead, this has fallen to other organizations, such as the Pew Research Center, Gallup, and the NORC.
With modeling to suggest that Christianity will fall to below 50% of the US population in the coming decades, these are statistics that many demographers will be keeping a close eye on, not least in terms of the intersection with politics.