The percentage of U.S. Catholics who consider themselves “strong” members of the Roman Catholic Church has never been lower than it was in 2012, according to a Pew Research Center analysis of new data from the General Social Survey (GSS). About a quarter (27%) of American Catholics called themselves “strong” Catholics last year, down more than 15 points since the mid-1980s and among the lowest levels seen in the 38 years since strength of religious identity was first measured in the GSS, a long-running national survey carried out by the independent research organization NORC at the University of Chicago.

The decline among U.S. Catholics is even starker when they are compared with Protestants, whose strength of religious identification has been rising in recent years. About half (54%) of American Protestants – double the Catholic share (27%) – described their particular religious identity as strong last year, among the highest levels since the GSS began asking the question in 1974.


The General Social Survey has surveyed nationally representative samples of American adults every year or two since 1972. Among other questions, respondents are asked: “What is your religious preference?” Those who name a religious affiliation are then asked: “Would you call yourself a strong [RELIGIOUS PREFERENCE] or a not very strong [RELIGIOUS PREFERENCE]?”1 A person who identifies as “Catholic,” for example, is asked, “Would you call yourself a strong Catholic or a not very strong Catholic?” while someone who identifies as Methodist is asked, “Would you call yourself a strong Methodist or a not very strong Methodist?” and someone who identifies as a Southern Baptist is asked, “Would you call yourself a strong Southern Baptist or a not very strong Southern Baptist?”2

Church Attendance Falling Among Catholics

Strength of religious identity is associated in the GSS with higher levels of religious commitment, such as more frequent attendance at worship services. In general, “strong” Catholics report going to Mass more often than do Catholics as a whole, and “strong” Protestants say they attend church more often than do Protestants overall.

However, over the past four decades, self-reported church attendance has declined among “strong” Catholics as well as among Catholics overall. The share of all Catholics who say they attend Mass at least once a week has dropped from 47% in 1974 to 24% in 2012; among “strong” Catholics, it has fallen more than 30 points, from 85% in 1974 to 53% last year.

Among Protestants as a whole, self-reported church attendance has been fairly stable, although the share of those who attend at least once a week was somewhat higher in 2012 (38%) than in 1974 (29%). Self-reported church attendance among “strong” Protestants has fluctuated over the years, but the share of frequent attenders was not significantly different in 2012 (60%) than in 1974 (55%).


In 1974, Catholics were more likely than Protestants to report attending religious services at least once a week (47% vs. 29%). By 2012, the situation had reversed: Protestants overall were more likely than Catholics to say they attend church weekly or more often (38% vs. 24%). Similarly, in 1974 “strong” Catholics reported going to church more frequently than did “strong” Protestants (85% vs. 55%), but in recent years “strong” Protestants have reported attending church about as often as “strong” Catholics do (60% vs. 53% in 2012).

Widening Gap

From 1974 until the late 1980s, U.S. Catholics were roughly comparable to Protestants on the question of the strength of their religious identity. Since the mid-1990s, however, Protestants have consistently had higher strength of religious identity than Catholics, and the gap has widened in recent years.

In addition, the percentage of U.S. Catholics who say they are “not very strong” Catholics has been increasing, while the percentage of Protestants who say they are “not very strong” members of their denomination has declined over time. In 1990, 45% of both Catholics and Protestants said they were “not very strong” members of their faith. By 2012, the share of “not very strong” U.S. Catholics had increased to 62%, while the share of “not very strong” Protestants had dropped to 36%.


Comparing Protestant Traditions and Catholics

Within the overall Protestant category, respondents can be separated into three major religious traditions: black Protestants, mainline Protestants and evangelical Protestants. For example, members of the Southern Baptist Church and the Assemblies of God belong to the evangelical Protestant tradition, while members of the United Methodist Church and the Presbyterian Church (USA) are part of the mainline Protestant tradition, and members of the National Baptist Convention and the Church of God in Christ are in the black Protestant tradition.

Of these major religious traditions, black Protestants and evangelical Protestants were most likely to report having a strong religious identity in 2012. Mainline Protestants are the least likely of the three Protestant traditions to report a strong identity with their faith. In 2012, 69% of black Protestants affirmed a strong religious identity, as did 57% of evangelical Protestants, 35% of mainline Protestants and 27% of Catholics.


‘Strong’ Protestants Stable As A Share Of The General Public, ‘Strong’ Catholics Declining

Viewed as a share of all U.S. adults, “strong” Protestants have remained remarkably stable over time. In the 2012 GSS, 27% of American adults identified as Protestants with a strong religious identity – exactly the same percentage as in 1974.3

Over the same period, however, the share of the population that identifies as Protestant but not “strong” Protestant declined substantially, from 36% in 1974 to 23% in 2012.4  In other words, the increasing concentration of “strong” Protestants seems to be due, in part, to attrition in the ranks of Protestants who do not claim a strong religious identity.


“Strong” Catholics, by contrast, have been declining both as a share of all Catholics and as a share of the U.S. public. The proportion of all American adults who identify as “strong” Catholics has fallen in the GSS from 12% in 1974 to about 7% today. The share of the public that identifies as Catholic but not “strong” Catholic, on the other hand, has risen slightly, from 14% in 1974 to nearly 18% in 2012.5



The General Social Survey is a project of the independent research organization NORC at the University of Chicago, with principal funding from the National Science Foundation. All results are produced using a composite weight that is the product of three weight variables included in the 1972 to 2012 GSS cumulative dataset: WTSSALL (which adjusts for the number of adults in the household), OVERSAMP (which adjusts for oversamples of African Americans in 1982 and 1987), and FORMWT (which adjusts for problems with the randomization of various questionnaire forms in the 1978-1985 waves of the GSS).

Respondents with a religious affiliation who did not answer the religious strength measure are excluded from the analysis of religious strength. This amounted to less than 1% of the people who took the survey in 2012, but it was higher in some other years. See Smith, Tom W. July 1993. “Little Things Matter: A Sampler of How Differences in Questionnaire Format Can Affect Survey Responses.” National Opinion Research Center, University of Chicago, GSS Methodological Report No. 78 (PDF).

The coding of denominational affiliations into Protestant religious traditions uses RELTRAD syntax for Stata (PDF). See Steensland, Brian, Jerry Z. Park, Mark D. Regnerus, Lynn D. Robinson, W. Bradford Wilcox, and Robert D. Woodberry. 2000. “The Measure of American Religion: Toward Improving the State of the Art.” Social Forces, volume 79, issue 1, pages 291-318. This approach to classifying Protestants into mainline Protestant, evangelical Protestant and black Protestant traditions has been widely used by social scientists in the analysis of GSS religion data. Among respondents who identify as some type of Protestant or only as Christian in the GSS RELIG variable, 94% are classified into a religious tradition by this syntax.


1 The question about strength of religious identity has been a core GSS question since 1974. It was not included in the 1972 and 1973 waves of the GSS. (return to text)

2 In recent years, increasing shares of GSS respondents identify themselves only as “Christian” (6% in 2012) and do not provide a more specific religious identity. Since Roman Catholic and Eastern Orthodox Christians usually identify themselves as Catholic and Orthodox, respectively, respondents who identify only as “Christian” are classified in this analysis as Protestant. If respondents who identified only as “Christian” are excluded from the analysis, the share of Protestants reporting strong identity follows the same general pattern, peaking at 52% in 2012. (return to text)

3 Respondents with a religious affiliation who did not answer the strength of religious identity measure are excluded from these calculations. (return to text)

4 Protestants who did not identify as strong include those who said they were “not very strong” Protestants and those who volunteered they were “somewhat strong” Protestants. (return to text)

5 Catholics who did not identify as strong include those who said they were “not very strong” Catholics and those who volunteered they were “somewhat strong” Catholics. (return to text)

Photo Credit: © Anderson Ross/Blend Images/Corbis