AMERICAS CHANGING

RELIGIOUS IDENTITY
Findings from the 2016 American Values Atlas
AMERICAS CHANGING
RELIGIOUS IDENTITY
Findings from the 2016 American Values Atlas

Robert P. Jones and Daniel Cox

Acknowledgments
The American Values Atlas Religion Report was made possible by generous funding from The
Evelyn and Walter Haas, Jr. Fund, The Ford Foundation, The Carnegie Corporation of New York,
The Gill Foundation, The Nathan Cummings Foundation, and The Unitarian Universalist Veatch
Program at Shelter Rock. Additionally, the authors would like to thank Carolyn Davis for her
strategic support; Joanna Piacenza and Harmeet Kamboj for their editorial and communications
support; Alex Vandermaas-Peeler and Molly Fisch-Friedman for their extensive research assistance;
and Tim Duffy for his able assistance with graphics and report design and layout.

We would also like to thank Kivvit for providing communications outreach for the survey release
and Epicenter Consulting for the development of the interactive online map for the American
Values Atlas, available at http://ava.prri.org/.

Public Religion Research Institute (PRRI)

2027 Massachusetts Ave. NW, 3rd Floor
Washington D.C. 20036
www.prri.org

Released on September 6, 2017.

For more information, contact [email protected]
 Contents
7 Executive Summary

10 A Portrait of Religious Affiliation in America

18 Americas New Religious Landscape:

Three Major Trends

27 Demographic Comparisons of
Major Religious Groups

36 Religion and Politics

42 Appendix 1: Top Three Religious

Groups in Each State

43 Appendix 2: Survey Methodology

45 Appendix 3: About PRRI and the Authors

6 AMERICAS CHANGING RELIGIOUS IDENTITY
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 7

Executive Summary
The American religious landscape is undergoing a dramatic transformation. White Christians,
once the dominant religious group in the U.S., now account for fewer than half of all adults living
in the country. Today, fewer than half of all states are majority white Christian. As recently as
2007, 39 states had majority white Christian populations. These are two of the major findings
from this report, which is based on findings from PRRIs 2016 American Values Atlas, the single
largest survey of American religious and denominational identity ever conducted. This landmark
report is based on a sample of more than 101,000 Americans from all 50 states and includes
detailed information about their religious affiliation, denominational ties, political affiliation, and
other important demographic attributes.

Among the major findings:

1. White Christians now account for fewer than half of the public. Today, only 43% of Ameri-
cans identify as white and Christian, and only 30% as white and Protestant. In 1976, roughly
eight in ten (81%) Americans identified as white and identified with a Christian denomination,
and a majority (55%) were white Protestants.

2. White evangelical Protestants are in declinealong with white mainline Protestants and
white Catholics. White evangelical Protestants were once thought to be bucking a longer
trend, but over the past decade their numbers have dropped substantially. Fewer than
one in five (17%) Americans are white evangelical Protestant, but they accounted for nearly
one-quarter (23%) in 2006. Over the same period, white Catholics dropped five percentage
points from 16% to 11%, as have white mainline Protestants, from 18% to 13%.

3. Non-Christian religious groups are growing, but they still represent less than one in ten
Americans combined. Jewish Americans constitute 2% of the public while Muslims, Bud-
dhists, and Hindus each constitute only 1% of the public. All other non-Christian religions
constitute an additional 1%.

4. Americas youngest religious groups are all non-Christian. Muslims, Hindus, and Bud-
dhists are all far younger than white Christian groups. At least one-third of Muslims (42%),
Hindus (36%), and Buddhists (35%) are under the age of 30. Roughly one-third (34%) of
religiously unaffiliated Americans are also under 30. In contrast, white Christian groups are
aging. Slightly more than one in ten white Catholics (11%), white evangelical Protestants
(11%), and white mainline Protestants (14%) are under 30. Approximately six in ten white
evangelical Protestants (62%), white Catholics (62%), and white mainline Protestants (59%)
are at least 50 years old.
8 AMERICAS CHANGING RELIGIOUS IDENTITY

5. The Catholic Church is experiencing an ethnic transformation. Twenty-five years ago, near-
ly nine in ten (87%) Catholics were white, non-Hispanic, compared to 55% today. Fewer than
four in ten (36%) Catholics under the age of 30 are white, non-Hispanic; 52% are Hispanic.

6. Atheists and agnostics account for a minority of all religiously unaffiliated. Most are
secular. Atheists and agnostics account for only about one-quarter (27%) of all religiously
unaffiliated Americans. Nearly six in ten (58%) religiously unaffiliated Americans identify as
secular, someone who is not religious; 16% of religiously unaffiliated Americans nonetheless
report that they identify as a religious person.

7. There are 20 states in which no religious group comprises a greater share of residents than
the religiously unaffiliated. These states tend to be more concentrated in the Western U.S.,
although they include a couple of New England states, as well. More than four in ten (41%)
residents of Vermont and approximately one-third of Americans in Oregon (36%), Washington
(35%), Hawaii (34%), Colorado (33%), and New Hampshire (33%) are religiously unaffiliated.

8. No state is less religiously diverse than Mississippi. The state is heavily Protestant and
dominated by a single denomination: Baptist. Six in ten (60%) Protestants in Mississippi are
Baptist. No state has a greater degree of religious diversity than New York.

9. The cultural center of the Catholic Church is shifting south. The Northeast is no longer
the epicenter of American Catholicismalthough at 41% Catholic, Rhode Island remains the
most Catholic state in the country. Immigration from predominantly Catholic countries in
Latin America means new Catholic populations are settling in the Southwest. In 1972, rough-
ly seven in ten Catholics lived in either the Northeast (41%) or the Midwest (28%). Only about
one-third of Catholics lived in the South (13%) or West (18%). Today, a majority of Catholics
now reside in the South (29%) or West (25%). Currently, only about one-quarter (26%) of the
U.S. Catholic population lives in the Northeast, and 20% live in the Midwest.

10. Jews, Hindus, and Unitarian-Universalists stand out as the most educated groups in the
American religious landscape. More than one-third of Jews (34%), Hindus (38%), and Unitar-
ian-Universalists (43%) hold post-graduate degrees. Notably, Muslims are significantly more
likely than white evangelical Protestants to have at least a four-year college degree (33% vs.
25%, respectively).

11. Asian or Pacific-Islander Americans have a significantly different religious profile than
other racial or ethnic groups. There are as many Asian or Pacific-Islander Americans affili-
ated with non-Christian religions as with Christian religious groups. And one-third (34%) are
religiously unaffiliated.
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 9

12. Nearly half of LGBT Americans are religiously unaffiliated. Nearly half (46%) of Americans
who identify as lesbian, gay, bisexual, or transgender (LGBT) are religiously unaffiliated. This
is roughly twice the number of Americans overall (24%) who are religiously unaffiliated.

13. White Christians have become a minority in the Democratic Party. Fewer than one in
three (29%) Democrats today are white Christian, compared to half (50%) one decade earlier.
Only 14% of young Democrats (age 18 to 29) identify as white Christian. Forty percent identi-
fy as religiously unaffiliated.

14. White evangelical Protestants remain the dominant religious force in the GOP. More
than one-third (35%) of all Republicans identify as white evangelical Protestant, a proportion
that has remained roughly stable over the past decade. Roughly three-quarters (73%) of Re-
publicans belong to a white Christian religious group.
10 AMERICAS CHANGING RELIGIOUS IDENTITY

A Portrait of Religious Affiliation in America

The American religious landscape has undergone dramatic changes in the last decade and is
more diverse today than at any time since modern sociological measurements began.

White Christians, which once dominated the religious landscape as recently as a decade ago, now
account for fewer than half of the public. White evangelical Protestants, the single largest reli-
gious tradition, make up less than one in five (17%) Americans today. Compared to ten years ago,
significantly fewer Americans identify as white mainline Protestant (13%) or white Catholic (11%).
Mormons comprise two percent of the population.1

FIGURE 1. The American Religious Landscape in 2016

White evangelical Protestant

3
White mainline Protestant
17
White Catholic
Mormon
Orthodox Christian
24
Black Protestant
Hispanic Protestant
Jehovahs Witness
13
Other nonwhite Protestant
Hispanic Catholic

1 Other nonwhite Catholic

1
1 Jewish
1 Muslim
2 Buddhist
11
2 Hindu
Other religion
2
7 Unaffiliated
4 8
Dont know/Refused
3 1
1
Source: PRRI 2016 American Values Atlas.

1
In 2006, nearly one-quarter (23%) of the public identified as white evangelical Protestant, 18% were white
mainline Protestant, and 16% identified as white Catholic. The Pew Research Center for the People & The
Press and Pew Forum on Religion & Public Life, Religion and Public Life Survey, July 2006.
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 11

Fifteen percent of Americans are nonwhite Protestants, including black Protestants (8%), Hispanic
Protestants (4%), and Asian, mixed-race, and other race Protestants (3%). Seven percent of the
public is Hispanic Catholic.

Non-Christian religious groups constitute less than one in ten Americans. Muslims, Buddhists,
and Hindus are each roughly one percent of the population. Jewish Americans account for two
percent of the public.

No religious group is larger than those who are unaffiliated from religion. Nearly one in four
(24%) Americans are now religiously unaffiliated.

The religious landscape in the U.S. is highly stratified by generation. Nearly two-thirds of seniors
(age 65 or older) identify as white and Christian: White evangelical Protestant (26%), white main-
line Protestant (19%), or white Catholic (16%). Conversely, only about one-quarter of young adults
(age 18-29) belong to a white Christian tradition, including white evangelical Protestant (8%),
white mainline Protestant (8%), or white Catholic (6%). Young adults are more than three times as
likely as seniors to identify as religiously unaffiliated (38% vs. 12%, respectively).

FIGURE 2. A Generational Shift in Religious Identity

100
Dont know/Refused
90 12
18 Unaffiliated
26
80 Other religion
38 5
6 4 Other world religions
70
6 7 Jewish
6
60
9 Other Christian
10 16
50 7 Hispanic Catholic
5
15 Hispanic Protestant
40 9 8
6 Black Protestant
19
30 9
7 15 Mormon

20 6 11 White Catholic

8 26 White mainline Protestant

10 21
14
8 White evangelical Protestant
0
18-29 30-49 50-64 65+

Source: PRRI 2016 American Values Atlas.

12 AMERICAS CHANGING RELIGIOUS IDENTITY

The Geography of Religion in America

There is substantial variation in religious identity between the four regions of the United States.
Collectively, white evangelical Protestants are twice as large in the South (22%) and Midwest (20%)
as they are in the Northeast (8%). Twelve percent of residents in the West are white evangeli-
cal Protestant. In contrast, Catholics represent a much larger share of Northeastern residents.
Roughly three in ten (29%) residents of the Northeast identify as Catholic, compared to about one
in five Westerners (21%), Midwesterners (19%), and Southerners (17%). The religiously unaffiliated
represent a larger share of the population in the West (29%) than in the Northeast (24%), Midwest
(23%), or South (20%).

Notably, there is significant racial and ethnic variation among Catholics by region. White Catholics
outnumber Hispanic Catholics by a wide margin in the Northeast (20% vs. 7%, respectively) and
Midwest (15% vs. 3%, respectively). However, in the South, residents are about as likely to identify
as white Catholic (8%) as they are to identify as Hispanic Catholic (7%); and in the West, Hispanic
Catholics (12%) outnumber white Catholics (7%).

There are substantial differences in the religious profiles of the 50 states, although they follow re-
gional patterns. There are 20 states in which no religious group comprises a greater share of resi-
dents than the religiously unaffiliated. These states tend to be more concentrated in the Western
U.S., although they include a couple of New England states, as well. More than four in ten (41%)
residents of Vermont and approximately one-third of Americans in Oregon (36%), Washington
(35%), Hawaii (34%), Colorado (33%), and New Hampshire (33%) are religiously unaffiliated.

In 11 states, many of which are clustered in the Northeast, no religious group outnumbers
Catholics. No state has a higher concentration of Catholics than Rhode Island (41%), followed by
Massachusetts (34%), New Jersey (34%), New York (31%), and Connecticut (31%).

In 13 states, no religious group comprises a larger share of residents than white evangelical Prot-
estants. Unsurprisingly, most of these states can be found in the South. At least one-third of the
residents in the following states are white evangelical Protestant: Arkansas (37%), West Virginia
(36%), Tennessee (36%), Alabama (35%), and Kentucky (33%). Additionally, Mormons (51%) are the
largest religious group in Utah, one of the most religiously homogeneous states in the country.

White mainline Protestants are more numerous in North Dakota than any other state, making up
30% of the states residents.

Black Protestants are most dominant in Mississippi, where they make up one-quarter (25%) of
the population, but they are still only the second largest religious group in the state behind white
evangelical Protestants (31%).
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 13

Religious Diversity
Despite the incredible variety of religious expression and identity in the U.S., there are importance
differences in the degree of religious diversity found in each state. To measure the religious diver-
sity in the country, this report uses an index developed to measure variations in the concentration
of global religious populations.2 The index is calculated so that a score of one signifies complete
diversityevery religious group is of equal sizewhile a score of zero indicates a complete lack of
diversity and one religious group comprises the entire population of a given state.

The least religiously diverse states in the U.S. are all located in the South. Mississippi is the least
diverse state in the U.S. (0.45), followed by Alabama (0.48), Arkansas (0.49), South Carolina (0.53),
and Tennessee (0.53). Conversely, the most religiously diverse states are primarily located in the
Northeastern U.S., with New York (0.83), New Jersey (0.81), Connecticut (0.81), Massachusetts
(0.81), and California (0.8) registering the greatest degree of religious diversity.

FIGURE 3. Religious Diversity by State

NH
WA VT .781
.745 .773 ME
MT ND .744
.688 .647
OR MN
.732 SD .729
ID WI NY MA .806
.672 .628 .752 .827
WY RI .771
MI
.637
.731 PA CT .809
IA
NE .679 .759 NJ .814
NV OH
.717 IL IN
.77 UT .699 DE .778
.773 .651
.555 CO WV
VA MD .77
.768 KS MO KY .6
.68
CA .699 .613 .609
.8 NC
.572
AZ OK TN .533
NM AR SC
.768 .588
.767 .489 .528
GA
MS AL .608
TX .454 .477
.705 LA
.695
AK FL
.739 .754

HI
.44 .83
.783

Less Diverse More Diverse

Source: PRRI 2016 American Values Atlas

2
The religious diversity for each state was calculated using the Religious Diversity Index, a variant of the
Herfindahl-Hirschman Index, one of the most commonly used measures of diversity among human and bi-
ological populations. See Charles R. Laine. June 22, 1995. The Herfindahl-Hirschman Index: A Concentration
Measure Taking the Consumers Point of View. Antitrust Bulletin.
14 AMERICAS CHANGING RELIGIOUS IDENTITY

Protestant Denominations
Among Protestants in the U.S., Baptists are the largest denominational family. Roughly one-
third (32%) of all Protestants identify with some Baptist denomination, at least three times the
number who identify with the next largest denominational familiesMethodist (10%), Pente-
costal (10%), and Lutheran (8%). One in twenty identify with the Church of Christ or Disciples
of Christ (5%) denominations or with the Presbyterian denomination (5%). Only three percent
of Protestants belong to an Episcopalian or Anglican denomination. Notably, nearly one in five
(17%) Protestants belong to independent Christian churches that are not affiliated with any
Protestant denomination.

TABLE 1. Americas Largest and Smallest Protestant Denominational Families by Region

National Northeast Midwest South West

Baptist 31.9% 23.2% 23.1% 43.6% 19.3%

Non-denominational 17.1 13.7 16.8 14.9 25.4

Methodist 10.1 11.8 12.0 10.2 5.9

Pentecostal 9.7 12.1 8.0 9.6 10.6

Lutheran 7.7 6.9 16.9 2.8 8.5

Church of Christ/DOC 5.1 3.8 5.7 4.7 6.6

Presbyterian 4.8 7.5 4.2 4.0 5.6

Episcopalian/Anglican 2.9 5.1 1.6 2.6 3.4

Jehovahs Witness 1.7 2.0 1.4 1.4 2.5

Congregational/UCC 1.2 3.4 1.7 0.5 1.0

Holiness 1.2 0.9 1.3 1.2 1.1

Seventh-Day Adventist 0.9 1.3 0.6 0.8 1.3

Reformed 0.6 0.7 1.1 0.2 0.6

Prot. Peace Churches 0.5 1.2 0.7 0.3 0.3

Other Prot. denomination* 4.9 6.6 4.8 3.3 7.9

Sample Size: 47,850 6,417 11,898 21,472 8,063

* Includes those who did not specify a denominational family.

Source: PRRI 2016 American Values Atlas.

FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 15

The pattern of denominational membership among Protestants varies significantly by region.

More than four in ten (44%) Protestants living in the South affiliate with a Baptist denomination,
nearly twice as many as those living in the Northeast (23%), Midwest (23%), or West (19%). Luther-
ans remain much more prevalent in the Midwest than any other part of the country. Seventeen
percent of Protestants living in the Midwest are Lutheran, roughly twice the number who identify
with this denomination in the West (9%) and the Northeast (7%). Only three percent of Southern-
ers are Lutheran. Non-denominational Protestants are more prevalent in the West (25%) than the
Midwest (17%), South (15%), or Northeast (14%).

Certain states are also unique in that they are dominated by one denominational family. A ma-
jority of Protestants in Mississippi (60%), Arkansas (56%), Louisiana (55%), and Alabama (54%) are
Baptist, while a majority of Protestants in North Dakota (56%) and nearly half of those in Minne-
sota (49%) are Lutheran.

Jewish Identity and Denominations

Overall, 1.5% of Americans identify as Jewish when responding to a question about their religious
affiliation. However, previous research has suggested questions that frame Jewish identity in
explicitly religious terms may undercount the total Jewish population because they miss those
with a cultural or familial affinity. To address this issue, the survey included a question to identify
those who had a cultural rather than a religious connection to their Jewish identity.3 An additional
0.8% of Americans identify as culturally but not religiously Jewish.

The issue of Jewish cultural identity has become increasingly important as younger Jews are more
likely to have a cultural affinity. Among Jews under the age of 30, fewer than half (47%) identify as
religiously Jewish, while a majority (53%) identify as culturally Jewish. In sharp contrast, more than
three-quarters (78%) of Jewish seniors (age 65 or older) are religiously Jewish, while 22% identify
as culturally Jewish.

Among all Jewish Americansthose who identify as Jewish both religiously and culturallymore
identify as Reform than any other denomination. Twenty-eight percent of Jews identify as Reform,
compared to 14% who identify as Conservative and 10% who identify as Orthodox. Two percent
identify as Reconstructionist. Notably, more than one-third (37%) of Jewish Americans claim to be
just Jewish when asked about their denominational affiliation. There is, however, a considerable
disparity in denominational membership among Jews by age.

3
To identify culturally affiliated Jews, we asked all respondents who claimed no formal religious affiliation the
following question: Do you consider yourself to be Jewish for any reason? Any respondent who said yes
or half was classified as culturally Jewish.
16 AMERICAS CHANGING RELIGIOUS IDENTITY

TABLE 2. Generational Shift in Jewish Denominational Identity

Just Something Dont know/

Reform Conservative Orthodox Reconstructionist Jewish else Refused

All Jews 28% 14% 10% 2% 37% 3% 6%=100%

18-29 20 8 15 3 44 4 6=100

30-49 25 12 13 2 37 4 8=100

50-64 34 17 6 2 33 3 4=100

65+ 35 20 3 2 34 2 4=100

Source: PRRI 2016 American Values Atlas.

Only 20% of Jews under the age of 30 are Reform, compared to 15% who identify as Orthodox.
Fewer than one in ten (8%) young Jewish Americans affiliate with the Conservative movement,
and three percent identify as Reconstructionist. More than four in ten (44%) say they are just
Jewish in terms of their denominational affinity. Jewish seniors are about ten times as likely to
identify as Reform as they are to identify as Orthodox (35% vs. 3%, respectively). One in five (20%)
identify as Conservative, and two percent identify as Reconstructionist. Approximately one-third
(34%) do not claim any denominational affiliation.

Measuring Evangelical Protestant Identity

The size of the white evangelical Protestant population in the U.S. has often been misunderstood
in recent years. The primary source of confusion stems from researchers employing differing
definitions of evangelical and their published findings often failing to clearly distinguish be-
tween the group as a whole and the distinct racial and ethnic subgroups within it.4 In this report,
evangelicals are defined as those who self-identify as Protestant Christians who also identify as
evangelical or born again.5

4
The 2015 Pew Research Center report, Americas Changing Religious Landscape, defines evangelical
Protestants based on the specific denomination with which they identify. This definition departs from that
employed in most other surveys because it does not take into account racial or ethnic identity. Subsequently
it includes many Hispanic, Asian/Pacific Islander, and mixed-race evangelical Protestants, as well as black
Protestants who do not belong to historically African American denominations.

5
All respondents who identify as Christian are then asked the following question: Would you describe
yourself as a born-again or evangelical Christian, or not? Respondents who self-identify as white,
non-Hispanic, Protestant and affirmatively identify as born-again or evangelical are categorized as white
evangelical Protestants.
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 17

Self-identified evangelical Protestants account for more than one-quarter (26%) of the public, but
whites make up only a subset of this larger group. Fewer than two-thirds (64%) of all evangelical
Protestants are white. Nearly one in five (19%) are black, one in ten (10%) are Hispanic, and six
percent identify as some other race or mixed race.

Thus, among the public overall, white evangelical Protestants account for fewer than one in five
(17%) Americans. Five percent of Americans are black evangelical Protestants, two percent are
Hispanic evangelical Protestants, and other or mixed-race evangelical Protestants make up one
percent of the public.

Like all Christians in the U.S., evangelical Protestants are experiencing a substantial racial and
ethnic transformation. Young evangelical Protestants are far more racially and ethnically diverse
than previous generations. Only half (50%) of evangelical Protestants under the age of 30 are
white, compared to more than three-quarters (77%) of evangelical Protestant seniors (age 65 or
older). Twenty-two percent of young evangelical Protestants are black, 18% are Hispanic, and nine
percent identify as some other race or mixed race.
18 AMERICAS CHANGING RELIGIOUS IDENTITY

Americas New Religious Landscape:

Three Major Trends
The National Decline in White Christian Identity
Demographically, the U.S. historically has been an overwhelmingly white Christian country, with
white Protestants alone constituting a clear cultural majority.6 In 1976, roughly eight in ten (81%)
Americans identified as white and identified with a Christian denomination.7 At that time, a ma-
jority (55%) of Americans were white Protestants.

Much of the decline has occurred in the last few decades. As recently as 1996, white Christians
still made up nearly two-thirds (65%) of the public. By 2006, that number dropped to 54%, but
white Christians still constituted a majority.8 But over the last decade, the proportion of white
Christians in the U.S. has slipped below majority. Today, only 43% of Americans identify as white
and Christianand only 30% as white and Protestant.

Although white Christians have experienced substantial losses nationally, there are notable divi-
sions in the size of the white Christian population by state. Today, the states with the highest con-
centrations of white Christians are generally found in the Midwest and Appalachia, including North
Dakota (71%), South Dakota (68%), Iowa (64%), West Virginia (61%), Kentucky (61%), Utah (61%),
Minnesota (59%), Missouri (58%), and Arkansas (58%). Conversely, white Christians are least plenti-
ful in Hawaii (20%), California (24%), New York (30%), New Mexico (30%), and Maryland (32%).

In total, fewer than half (23) of all 50 states have majority white Christian populations. This rep-
resents a significant drop from 2007, when 39 states had majority white Christian populations.9

Notably, between 2007 and 2016, the proportions of white Christians have declined in every state
but one (Hawaii), though the amount of decline varies widely by state. Many of the states that
experienced the greatest losses are concentrated in the Northeast. Massachusetts, for example,
experienced a 20-point decline in the number of white Christians living in the state over the last
decade. Hawaii, the only state to experience an increase in the proportion of white Christians,
saw an increase of six percentage points.

6
Throughout this report the term white signifies respondents who identify as white or Caucasian and who
do not identify as Hispanic or Latino.
7
The General Social Survey, 1976.
8
The General Social Survey, 2006.
9
The Pew Forum on Religion & Public Life, Religious Landscape Survey, July 2007.
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 19

FIGURE 4. Diminishing White Christian Presence in U.S.

Percent White Christian by State, 2007 vs. 2016

2007
NH
WA VT 58
55 62 ME
MT ND 67
67 86
OR MN
57 SD 74
ID WI NY MA 62
67 78 69 48
WY RI 55
MI
70 *
62 PA CT 55
IA
NE 75 67 NJ 51
NV OH
71 IL IN
43 UT 64 DE 52
56 68
68 CO WV
VA MD 41
50 KS MO KY 73
55
CA 71 65 74
34 NC
57
AZ OK TN 68
NM AR SC
46 62
39 67 65
GA
MS AL 53
TX 53 60
45 LA
57
AK FL
50 48

HI 14 14% 86%

2016
NH
WA VT 53
42 44 ME
MT ND 56
57 71
OR MN
43 SD 59
ID WI NY MA 42
56 68 56 30
WY RI 50
MI
57
50 PA CT 41
IA
NE 64 55 NJ 36
NV OH
56 IL IN
33 UT 56 DE 41
42 57
61 CO WV
VA MD 32
39 KS MO KY 61
45
CA 53 58 61
24 NC
49
AZ OK TN 57
NM AR SC
39 49
30 58 51
GA
MS AL 40
TX 46 53
36 LA
45
AK FL
40 39

HI 20
14% 86%

Note: The sample size of Wyoming in 2007 includes fewer than 100 respondents.
Sources: The Pew Forum on Religion & Public Life, Religious Landscape Survey, July 2007; PRRI 2016 American Values Atlas.
20 AMERICAS CHANGING RELIGIOUS IDENTITY

The Decline of White Evangelical Protestants

Although much of the research on religious decline has focused on losses among white mainline
Protestants and white Catholicsgroups with well-documented declinesmuch of the public
discussion on religious change has missed one of the most important stories of the last decade:
The decline of white evangelical Protestants.10

White evangelicals managed to avoid the first wave of white Christian decline in the 1990s and
into the early years of the 21st century. But between 2006 and 2016, the proportion of white
evangelical Protestants has fallen six percentage points, from 23% to 17%. White Catholics expe-
rienced a similar decline in membership, dropping five percentage points from 16% in 2006 to
11% in 2016.11 White mainline Protestants also experienced significant losses during this period,
dropping from 18% in 2006 to 13% in 2016.

FIGURE 5. The Decline of White Christians, 2006 - 2016

White evangelical Protestant White mainline Protestant White Catholic

25 23.0
22.5
21.4 20.8 21.3
20.8
19.4 19.9
20
17.8 18.3
17.5 17.5 17.3
17.0 16.8
15.5 15.5
15 14.9 14.2 13.7
16.0 16.1 13.3 12.8
15.6
14.7 14.2 14.4 13.9
10
13.0
12.3 12.1
11.4
5

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

Sources: The Pew Research Center for the People & The Press and Pew Forum on Religion & Public Life, Religion and
Public Life Survey, 2006-2009; PRRI 2010 American Values Survey; PRRI/Brookings 2011 Pluralism Survey; PRRI 2012
American Values Survey; PRRI 2013-2016 American Values Atlas..

10
For a longer discussion of this trend and its implications, see Robert P. Jones, The End of White Christian Amer-
ica (Simon & Schuster, 2016).

11
The Pew Research Center for the People & The Press and Pew Forum on Religion & Public Life, Religion and
Public Life Survey, July 2006.
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 21

The Aging of White Christians

White Christians are not only declining, they are aging. Only slightly more than one in ten white
evangelical Protestants (11%), white Catholics (11%), and white mainline Protestants (14%) are
under the age of 30. Approximately six in ten white evangelical Protestants (62%), white Catholics
(62%), and white mainline Protestants (59%) are at least 50 years old.

No religious group has older members than white evangelical Protestants and white Catholics.
The median age of white evangelical Protestants and white Catholics is 55 years old, slightly high-
er than white mainline Protestants at 54 years old. Unitarian-Universalists are also much older
than members of other religious groups: The median age is 54 years.

Four decades earlier, the median age of these white Christian groups was much lower. In 1976,
the median age of white Catholics was just 40 years old, while the median age of white Protes-
tants was 45 years old.12

In contrast, about one in three Hispanic Protestants (32%), religiously unaffiliated Americans
(34%), Buddhists (35%), and Hindus (36%) are under the age of 30. More than four in ten (42%)
Muslim Americans are under the age of 30. The median ages of Hindus (32 years), Muslims (32
years), Buddhists (36 years), religiously unaffiliated Americans (37 years), and Hispanic Protes-
tants (37 years) are below 40.

The Mormon Exception

Although Mormons are a predominantly white Christian religious tradition, there is little evidence
to suggest that they are experiencing similar declines.13 Currently, 1.9% of the public identifies as
Mormon, a number identical to findings from a 2011 study of Mormons in the U.S.14

Mormons are also much younger than other white Christian religious traditions. Nearly one-quar-
ter (23%) of Mormons are under the age of 30. Fewer than half (41%) are age 50 or older.

12
General Social Survey, 1976. It was not possible to identify evangelical and mainline Protestants in the 1976
General Social Survey.
13
Eighty-three percent of Mormons are white, non-Hispanic.
14
Mormons in America: Certain in Their Beliefs, Uncertain of Their Place in Society, Pew Research Center, Jan. 12, 2012.
22 AMERICAS CHANGING RELIGIOUS IDENTITY

FIGURE 6. Religious Affiliation by Age

18-29 30-49 50-64 65+

Muslim 42 42 12 4

Hindu 36 51 10 3

Buddhist 35 38 19 8

Unaffiliated 34 37 19 10

Hispanic
32 44 17 8
Protestant

Hispanic Catholic 25 44 21 10

Orthodox
23 37 26 15
Christian

Mormon 23 36 23 18

Black Protestant 19 34 29 18

Jehovahs Witness 19 34 28 19

Jewish 18 30 25 27

Unitarian/
14 29 28 27
Universalist

White mainline
14 28 30 29
Protestant

White evangelical
11 28 32 30
Protestant

White Catholic 11 27 34 28
0 10 20 30 40 50 60 70 80 90 100

Source: PRRI 2016 American Values Atlas.

FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 23

The Growth of Nonwhite Christians

Increasing Racial and Ethnic Diversity Among Protestants

Today, three in ten (30%) Americans are white Protestant. This is a considerable decline over the
past couple of decades; in 1991, white Protestants represented half (50%) the public.15 Not only
have white Protestants experienced a substantial decline as a proportion of the general population,
they also represent a shrinking proportion of all Protestants. In 1991, 83% of all Protestants were
white, compared to two-thirds (67%) today. One-third (33%) of all Protestants are now nonwhite.

However, the degree of racial and ethnic diversity among Protestants varies considerably between
denominational families. More than nine in ten Lutherans (92%) and roughly eight in ten Meth-
odists (83%), Presbyterians (83%), and Episcopalians (80%) are white, non-Hispanic. In contrast,
fewer than six in ten (58%) Baptists are white, and a sizeable share of members are black (30%)
or Hispanic (5%). Similarly, only half (50%) of Pentecostals are white, while one-quarter (25%) are
Hispanic, and 17% are black. Protestants who belong to non-denominational Protestant churches
are also somewhat diverse: Two-thirds (67%) are white, 13% are black, and 10% are Hispanic.

The Ethnic Transformation of the U.S. Catholic Church

For most of the past 25 years, American Catholics have been overwhelmingly white. In 1991,
more than eight in ten (87%) U.S. Catholics were white, non-Hispanic.16 Today, 55% of Catholics
identify as white, non-Hispanic, and more than one-third (36%) are Hispanic. Notably, much of the
shift in the ethnic composition of the American Catholic community occurred over just the last
couple of decades.

Generational differences in the ethnic and racial make-up of American Catholics also suggest that
a substantial cultural shift is underway. Fewer than four in ten (36%) Catholics under the age of 30
are white, non-Hispanic, compared to a majority (52%) who are Hispanic. In contrast, more than
three-quarters (76%) of Catholic seniors (age 65 or older) are white, while only 17% are Hispanic.

Demographic differences between Hispanic and white Catholics also suggest that the Hispanic
Catholic community is poised to make further gains. First, Hispanic Catholics are nearly twice as
likely as white Catholics to be parents of children under the age of 18 (30% vs. 16%, respectively).
The discrepancy in parental status is driven largely by age differences between Hispanic Catholics
and white Catholics. Second, Hispanic Catholics are more likely to have larger families: Close to
three in ten (28%) Hispanic Catholic parents have at least three children under the age of 18 living
at home, while 21% of white Catholic families report the same.

15
Pew Research Center for the People & the Press, Political Values Survey, 1991.
16
Pew Research Center for the People & the Press, Political Values Survey, 1991.
24 AMERICAS CHANGING RELIGIOUS IDENTITY

The Shifting Cultural Center of the Catholic Church

Historically, the U.S. Catholic Church was centered in the Northeast, where European immigrants
from predominantly Catholic countries, such as Italy and Ireland, first settled. Even today, Catholics
represent a disproportionate size of the population in many states in the region, including Rhode
Island (41%), Massachusetts (34%), New Jersey (34%), Connecticut (31%), and New York (31%).

However, due to recent patterns of immigration from predominantly Catholic countries in Latin
America, particularly Mexico, the Catholic population is experiencing growth across the South-
west. In 1972, roughly seven in ten Catholics lived in either the Northeast (41%) or the Midwest
(28%).17 Only about one-third of Catholics lived in the South (13%) or West (18%). Today, a major-
ity of Catholics now reside in the South (29%) or West (25%). Currently, only about one-quarter
(26%) of the U.S. Catholic population lives in the Northeast.

Regional age differences among Catholics also indicate that Catholicisms center of gravity is
shifting. American Catholics residing in the Northeast and Midwest are substantially older than
Catholics in the Southwestern United States. Fewer than half of Catholics in the Northeast (47%)
and Midwest (44%) are less than 50 years of age. In contrast, a majority (58%) of Catholics living in
the Western U.S. are under the age of 50.

Rise of the Religiously Unaffiliated

The religiously unaffiliatedthose who identify as atheist, agnostic, or nothing in particu-
larnow account for nearly one-quarter (24%) Americans. Since the early 1990s, this group
has roughly tripled in size.

Religious identity is highly stratified by age, with younger Americans (age 18-29) most likely to be
religiously unaffiliated and seniors (age 65 or older) least likely to identify this way. However, the
relationship between age and unaffiliated status is not completely linear. While there is consider-
able variation at the youngest and oldest ends of the age spectrum, there is less variation in the
rate of unaffiliated identity among Americans in their 40s (23%), 50s (18%), and 60s (16%).

Although unaffiliated Americans tend to be younger than religiously affiliated Americans on av-
erage, the group collectively is older today than it was a generation ago. Today, about one-third
(34%) of unaffiliated Americans are under the age of 30, while nearly three in ten (29%) are at
least 50 years old. In the 1970s, half (50%) of all unaffiliated Americans were under 30 years old,
and only 17% were age 50 or older. The median age of someone who was unaffiliated during that
decade was also seven years younger than it is today: 29 vs. 36, respectively.

17
General Social Survey, 1972.
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 25

FIGURE 7. Growth of the Religiously Unaffiliated, 1976-2016

Percent of population

25 24

20
20
17
14 14
15
12
9
10 8
7 7 7

0
1976 1980 1984 1988 1992 1996 2000 2004 2008 2012 2016

Sources: General Social Survey, 1976-2012; PRRI 2013-2016 American Values Atlas.

FIGURE 8. Percent Unaffiliated by Age

45

38
40

35 29

30

23
25
18
20 16

15 12

10 8

5
18 29 30 39 40 49 50 59 60 69 70 79 80+
0
18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80+

Source: PRRI 2016 American Values Atlas.

26 AMERICAS CHANGING RELIGIOUS IDENTITY

Atheists, Agnostics, Seculars, and the Religiously Unattached

There are notable differences among the unaffiliated in terms of their religious identity. Only
about one-quarter of the unaffiliated identify as atheist (14%) or agnostic (13%), while fewer
than one in five (16%) identifies as a religious
person. The majority (58%) of Americans who FIGURE 9. Unaffiliated Americans:
are not religiously affiliated do not identify as a Mostly Secular
religious person.18

Younger unaffiliated Americans are only 16 14 Atheist

somewhat more likely than unaffiliated se-

Agnostic
niors to identify as atheist or agnostic (31% vs. 13

24%, respectively), while unaffiliated seniors Secular

person
are twice as likely as younger unaffiliated
Americans to identify as a religious person Religious
58 person
(20% vs. 10%, respectively). A majority of unaf-
filiated young adults and seniors say they are
secular (59% vs. 56%, respectively). Source: PRRI 2016 American Values Atlas.

18
This includes seven percent of the unaffiliated who identify as spiritual but not religious.
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 27

Demographic Comparisons of
Major Religious Groups
The Enduring Gender Divide
Women continue to make up the majority of most religious groupsincluding most Christian groups.
A majority of black Protestants (58%), white evangelical Protestants (56%), Hispanic Catholics (55%),
white mainline Protestants (54%), white Catholics (53%), and Hispanic Protestants (53%) are women.

Among Mormons and Jewish Americans, the gender split mirrors that of the general public48%
male, 52% femalewhile the gender divide among Buddhists and Orthodox Christians is exactly
even (50% male, 50% female).

Men make up a majority of religiously unaffiliated Americans (55%), Muslims (59%), and Hindus
(66%). Fewer than half of unaffiliated Americans (45%), Muslims (41%), and Hindus (34%) are
women. Atheists and agnostics are more heavily male-dominated than the unaffiliated overall:
Men account for 64% of atheists and 63% of agnostics.

The gender imbalance is also quite prominent among Unitarian Universalists, among whom near-
ly two-thirds (64%) are women. Only 36% are men.

Educational Attainment
Levels of educational attainment vary widely between religious groups, even among white Chris-
tians. Unitarian-Universalists and Hindus are, on average, the most educated, while Hispanic
Catholics are the least formally educated.

Close to half (48%) of white evangelical Protestants have a high school education or less, compared
to fewer than four in ten white mainline Protestants (38%), white Catholics (37%), and Mormons
(34%). Roughly one-third of Mormons (32%) and more than one-third of white mainline Protestants
(36%) and white Catholics (39%) have a four-year college degree, compared to only one-quarter
(25%) of white evangelical Protestants.

Among Protestants, there are substantial variations by denominational family. Presbyterians and
Episcopalians have the highest levels of educational attainment. Roughly half of Presbyterians (48%)
and Episcopalians (52%) have at least a four-year college degree. In contrast, significantly fewer
Methodists (36%), Lutherans (34%), Baptists (19%), and Pentecostals (15%) graduated from a four-
year college or university. A majority (55%) of Baptists and six in ten (60%) Pentecostals have no
more than a high school education.
28 AMERICAS CHANGING RELIGIOUS IDENTITY

FIGURE 10. Gender by Religious Affiliation

Male Female

Hindu 66 34

Muslim 59 41

Unaffiliated 55 45

Orthodox
50 50
Christian

Buddhist 50 50

Mormon 48 52

Jewish 48 52

Hispanic
47 53
Protestant

White Catholic 47 53

White mainline
46 54
Protestant

Hispanic Catholic 45 55

White evangelical
44 56
Protestant

Black Protestant 42 58

Jehovahs Witness 37 63

Unitarian/
36 64
Universalist
0 10 20 30 40 50 60 70 80 90 100

Source: PRRI 2016 American Values Atlas.

FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 29

FIGURE 11. Education by Religious Affiliation

High school or less Some college College graduate Post-graduate

Hispanic Catholic 70 16 8 4

Jehovahs Witness 68 23 6 3

Hispanic
65 20 10 4
Protestant

Black Protestant 52 26 12 10

White evangelical
48 27 15 10
Protestant

Unaffiliated 44 25 17 14

Muslim 44 22 17 16

White mainline
38 25 19 17
Protestant

Buddhist 37 26 18 19

White Catholic 37 24 22 17

Mormon 34 33 19 13

Orthodox
28 20 25 26
Christian

Jewish 22 18 27 34

Hindu 19 11 31 38

Unitarian/
16 20 22 43
Universalist
0 10 20 30 40 50 60 70 80 90 100

Source: PRRI 2016 American Values Atlas.

30 AMERICAS CHANGING RELIGIOUS IDENTITY

Nearly half (49%) of Catholics have a high school education or less, while more than one in five
(21%) have some college education, and 29% report having at least a four-year college degree.
However, levels of educational attainment vary considerably by ethnicity. Seven in ten (70%) His-
panic Catholics have a high school education or less, compared to 37% of white Catholics. White
Catholics are about three times as likely as Hispanic Catholics to have at least a four-year college
degree (39% vs. 12%, respectively).

Nonwhite Christian religious groups have considerably lower levels of education. A majority of
black Protestants (52%), Hispanic Protestants (65%), and Hispanic Catholics (70%) report having a
high school education or less. Fewer than one-quarter of black Protestants (22%), Hispanic Prot-
estants (14%), and Hispanic Catholics (12%) have a college degree.

Among non-Christian religious groups, Muslims have the lowest levels of educational attainment.
More than four in ten (44%) Muslims have a high school degree or less, compared to 37% of Bud-
dhists, and about one in five Jews (22%) and Hindus (19%). Jewish and Hindu Americans are highly
educated; strong majorities have a four-year college degree (61% vs. 69%, respectively), including
more than one-third of each group who have post-graduate degrees (34% vs. 38% respectively).

However, among Jewish Americans, there are considerable differences in educational background
by denomination. Orthodox Jews are far more likely than Reform Jews to have no more than a
high school education (44% vs. 14%, respectively).

No religious group has a higher proportion of members with post-graduate degrees than Unitar-
ian-Universalists. Nearly two-thirds (65%) have a college education, including more than four in
ten (43%) who have an advanced degree.

Religiously unaffiliated Americans are less educated than many religious groups. More than four
in ten (44%) have a high school education or less. One-quarter (25%) have some college expe-
rience, such as an associates degree, and roughly three in ten (31%) have a four-year college
education. However, self-identified atheists and agnostics have significantly greater educational
experience. About one-third of atheists (34%) and agnostics (31%) have no more than a high
school education, while 42% of each group have at least a four-year college degree.

Economic Status
Household income

Religious groups with the lowest levels of household incomes tend be Christians from racial and
ethnic minority groups. More than four in ten black Protestants (43%), Hispanic Protestants (45%),
Jehovahs Witnesses (45%), and Hispanic Catholics (49%) report household earnings of less than
$30,000 per year.
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 31

White Christian groups report higher levels of income. Fewer than one-third of white evangelical
Protestants (28%), Mormons (26%), white mainline Protestants (22%), and white Catholics (19%)
have household incomes of less than $30,000 per year.

Among non-Christian groups, Muslims report the lowest levels of household income. Nearly four
in ten (38%) Muslims have household incomes of less than $30,000 per year, compared to fewer
than one-third of Buddhists (31%), Hindus (20%), and Jews (16%). One-quarter (25%) of Hindus
and 30% of Jewish Americans have annual household incomes in excess of $100,000. Unitari-
ans-Universalists are also fairly well off financially. Only 18% report living in households making
less than $30,000 annually, and 22% report having incomes exceeding $100,000.

The income levels of religiously unaffiliated Americans closely mirror Americans overall.

Home Ownership

Patterns of homeownership vary considerably among religious groups. Roughly three-quarters of

white Catholics (78%), white evangelical Protestants (75%), and white mainline Protestants (75%)
report they own their home. About two-thirds (66%) of Mormons say they own rather than rent.

Nonwhite Christians are far less likely to report owning their home. Fewer than half of Jehovahs
Witnesses (48%), black Protestants (46%), Hispanic Catholics (38%), and Hispanic Protestants
(38%) are homeowners.

Among non-Christian groups, Jewish Americans (69%) are much more likely than Buddhists
(54%), Muslims (36%), or Hindus (31%) to own their home. More than seven in ten (72%)
Unitarian-Universalists own their home. A majority (54%) of religiously unaffiliated Americans
are homeowners.

Health Insurance Status

A slim majority (51%) of white Catholics and close to half of Mormons (48%) and white mainline
Protestants (46%) report they receive health insurance through their employer. More than four
in ten (43%) white evangelical Protestants say they have employer-based health insurance. Three
in ten (30%) white evangelical Protestants have government-sponsored health care such as
Medicare or Medicaid, while fewer white mainline Protestants (27%), white Catholics (25%), and
Mormons (22%) report the same.

Nonwhite Christians are much less likely to have employer-based health insurance than white
Christians, but there are notable differences in patterns of coverage among these groups. About
one-third of black Protestants (35%), Hispanic Protestants (31%), Jehovahs Witnesses (31%),
and Hispanic Catholics (29%) report having health insurance through their employer. Jehovahs
32 AMERICAS CHANGING RELIGIOUS IDENTITY

Witnesses (33%) and black Protestants (30%) are more likely to receive health insurance through
government programs than Hispanic Protestants (21%) and Hispanic Catholics (23%). But nearly
one-third of Hispanic Protestants (29%) and Hispanic Catholics (32%) report they have no health
insurance coverage at all.

No group is more likely to receive health insurance through their employer than Hindus. More
than six in ten (62%) Hindus have employer-sponsored health care coverage. Fewer than half
(47%) of Jewish Americans, four in ten (40%) Buddhists, and nearly three in ten (29%) Muslims re-
ceive health insurance through an employer. Hindus are also far less likely to receive health care
through government programs, such as Medicare or Medicaid, with only six percent reporting
they receive this type of coverage. In contrast, at least one in five Buddhists (21%), Muslims (25%),
and Jews (25%) have government-sponsored health insurance. Nearly one in five (19%) Muslims
report having no health insurance coverage.

The health care insurance status of religiously unaffiliated Americans closely resembles that of
Americans overall. Four in ten (43%) receive it through an employer, while 19% report receiving it
through the government. Fifteen percent have no health insurance.

Approximately four in ten (43%) Unitarian-Universalists also receive their health insurance
through an employer, while close to one-third (29%) have government-sponsored coverage. Only
four percent of Unitarian-Universalists report having no health insurance at all.

Race and Ethnicity

The patterns of religious identity among major racial and ethnic groups are quite distinct. Nearly
seven in ten (68%) white Americans are Christian and nearly half (47%) are Protestant. Twen-
ty-seven percent of whites are evangelical Protestant, while 20% are mainline Protestant. An addi-
tional 18% of whites are Catholic. One-quarter (25%) of all whites are religiously unaffiliated.

Black Americans are also predominantly Christian. Three-quarters (75%) of black Americans are
Christian, although a much greater proportion are Protestant (67%), and only six percent iden-
tify as Catholic. Roughly one in five (19%) black Americans are religiously unaffiliated. Hispanic
Americans are also predominantly Christian, although they are about twice as likely to identify as
Catholic (48%) than Protestant (25%). One in five (20%) Hispanics are unaffiliated.

Unlike other racial and ethnic groups, fewer than one in three (30%) Asian and Pacific Islander
(API) Americans are Christian, with roughly equal numbers identifying as Protestant (16%) and
Catholic (13%). More than one-third (34%) of API Americans are religiously unaffiliated, with
significant numbers identifying as members of non-Christian traditions, such as Hindu (15%),
Buddhist (7%), or Muslim (7%).
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 33

FIGURE 12. Religious Profiles of White, Black, Hispanic, and Asian Americans

100

90
19 20 Other religion
25
80 Unaffiliated
34
Hindu
70 6
4
Buddhist
60
Muslim
18
48 15
50 Jewish

Other Christian
40 7
Catholic
7
30 67
47 Protestant
13
20
25
10
16

0
White, non- Black, non- Hispanic Asian or Pacific
Hispanic Hispanic Islander

Sources: PRRI 2016 American Values Atlas.

Marital Status and Family Size

Divisions in marital status among American religious communities largely occur along racial
lines. At least six in ten white mainline Protestants (60%), white Catholics (61%), white evangelical
Protestants (64%), and Mormons (68%) are married. In contrast, members of nonwhite religious
traditions are much less likely to be married. Only 37% of black Protestants, 45% of Hispanic Prot-
estants, and half (50%) of Hispanic Catholics report being married.

Notably, among non-Christian groups, Hindus (66%) are far more likely than Muslims (49%) and
Buddhists (35%) to be married. More than six in ten (62%) Jewish Americans also report being mar-
ried, although there are some differences by denomination. Nearly three-quarters (74%) of Ortho-
dox Jews are married, compared to about six in ten who belong to Conservative (64%) and Reform
(59%) denominations. Roughly four in ten (42%) religiously unaffiliated Americans are married.

White Christians stand out in that they do not have young children living at home. Nearly
three-quarters of white evangelical Protestants (73%), white mainline Protestants (73%), and
34 AMERICAS CHANGING RELIGIOUS IDENTITY

white Catholics (75%) do not have children under the age of 18 living at home. Roughly seven
in ten Jews (71%) and Buddhists (72%), and approximately two-thirds of religiously unaffiliated
Americans (68%), black Protestants (67%), and Hindus (67%), also report that they have no chil-
dren living in their household. In contrast, only about half (50%) of Hispanic Catholics and Hispan-
ic Protestants (52%) have no children in their households.

Among Americans who report living with children in their household, Mormons have by far the
largest families. More than four in ten (42%) Mormon parents say they have at least three chil-
dren in their household under the age of 18. Only 28% of Mormons living with children report
having only one child. By comparison, fewer than one-quarter of white evangelical Protestants
(24%), white mainline Protestants (23%), and white Catholics (21%) with children say they have at
least three kids in their household.

Nearly three in ten (29%) Jewish parents report having at least three children in their household,
although family size varies dramatically by denomination. More than six in ten (62%) Orthodox
Jewish parents say they have at least three children living in their household, compared to 17%
of Jewish parents who identify as
Reform who say the same.
FIGURE 13. The Religious Profile of LGBT Americans
Hindus have some of the smallest
100 3 3 Dont know/Refused
families. A majority (55%) of Hindus
with children say they have only 90
Unaffiliated

one child in their household, while 24 Other religion

only four percent say they have 80
Other world
three or more. Muslim families are 46 religions
70
larger, with more than one-quarter 6 Jewish
(26%) of those with children in their 60 7 Other Christian
household reporting three or more. 4
50 4 Hispanic Catholic
8 4
The Religious Identity 40
Hispanic Protestant
6
of LGBT Americans 11 Black Protestant
5
30
LGBT Americans are far less likely Mormon
13 6
to identify with any religious group 20 White Catholic
6
than the public as a whole. Near-
White mainline
10 8
ly half (46%) of Americans who 17 Protestant

identify as lesbian, gay, bisexual, or 6

0 White evangelical
transgender (LGBT) are religiously All Americans LGBT Americans Protestant

unaffiliated. Fewer identify as Chris-

Sources: PRRI 2016 American Values Atlas.
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 35

tian. Only six percent of LGBT Americans are white evangelical Protestant, while similar numbers
identify as white mainline Protestant (8%) and white Catholic (6%). Fewer than one in ten identify
as black Protestant (6%), Hispanic Catholic (5%), or Hispanic Protestant (3%). LGBT Americans are
somewhat overrepresented among non-Christian religions. Six percent identify as Jewish (2%),
Buddhist (2%), Muslim (1%), or Hindu (1%).

There are stark generational differences among LGBT Americans in their religious identity. A
majority (56%) of LGBT young adults (age 18 to 29) are religiously unaffiliated, compared to
one-quarter (25%) of LGBT seniors (age 65 or older). There are no significant gender differences
in the religious profile of LGBT people.

Notably, Buddhists and Unitarian-Universalists have a much higher proportion of LGBT mem-
bers than other religious traditions. One in seven Buddhists (14%) and Unitarian-Universalists
(14%) identify as LGBT.
36 AMERICAS CHANGING RELIGIOUS IDENTITY

Religion and Politics

Political Affiliation
No religious group is more closely tied to the Republican Party than white evangelical Protestants.
Nearly half (49%) of white evangelical Protestants identify as Republican, about one-third (31%)
are independent, and just 14% are Democratic. Mormons also lean heavily Republican, with more
than four in ten (44%) identifying with the GOP, compared to 12% who are Democrats. White
mainline Protestants (34% Republican, 26% Democrat) and white Catholics (34% Republican, 26%
Democrat) also lean more toward the Republican Party.

Notably, older white Christians are generally more likely to be affiliated with the Democratic
Party than those who are younger. White evangelical Protestant seniors are about twice as likely
as those under the age of 30 to identify as Democrat (18% vs. 10%, respectively); they are about
equally as likely to identify as Republican (49% vs. 47%, respectively). Similarly, while nearly one-
third (31%) of white Catholic seniors are Democrats, only about one in five (21%) white Catholic
young adults identify as the same. The generational gap is slightly smaller among white mainline
Protestants: Twenty-eight percent of seniors, compared to 21% of young adults, are affiliated with
the Democratic Party.

Nonwhite Christians tend to lean more toward the Democratic Party, although there is significant
diversity in the partisan preferences of these groups. No religious group more strongly identifies
with the Democratic Party than black Protestants. More than two-thirds (68%) of black Protes-
tants identify as Democrats, compared to only four percent who identify as Republican. Hispanic
Catholics are more than four times as likely to identify as Democrat than they are to identify as
Republican (40% vs. 9%, respectively). Hispanic Protestants also lean Democratic in their political
preferences, although by a more modest margin (30% vs. 17%, respectively). More than four in
ten (41%) Hispanic Protestants are independent.

Importantly, there are substantial generational differences in partisan identity among black
Protestants and Hispanic Catholics. Black Protestant seniors are far more likely to identify as
Democratic than young black Protestants (79% vs. 58%, respectively). A majority (56%) of Hispanic
Catholic seniors identify as Democratic, while only about one-third (35%) of young Hispanic Cath-
olics identify as the same.

The pattern of partisan attachment is also lopsided in favor of Democratic identity among non-
Christian religious groups. Nearly half of Jewish (47%) and Muslim Americans (46%) identify as
Democratic, while far fewer identify as Republican (20% vs. 6%, respectively). Buddhists (37%
Democrat, 9% Republican) and Hindus (36% Democrat, 9% Republican) also show far more
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 37

FIGURE 14. Party Affiliation by Religious Affiliation

Republican Independent Democrat Other/Dont Know

White evangelical
49 31 14 6
Protestant

Mormon 44 38 12 5

White Catholic 34 35 26 5

White mainline
34 35 26 6
Protestant

Orthodox
26 43 26 5
Christian

Jewish 20 28 47 5

Hispanic
17 41 30 11
Protestant

Unaffiliated 11 47 33 8

Buddhist 9 47 37 7

Hindu 9 40 36 15

Hispanic Catholic 9 38 40 13

Muslim 6 39 46 10

Jehovahs Witness 4 32 13 51

Black Protestant 4 23 68 5

Unitarian/
3 33 57 7
Universalist
0 10 20 30 40 50 60 70 80 90 100

Source: PRRI 2016 American Values Atlas.

38 AMERICAS CHANGING RELIGIOUS IDENTITY

affinity for the Democratic Party. However, four in ten (40%) Hindus and nearly half (47%) of
Buddhists are independent.

Among Jewish Americans, partisan attachments vary considerably by denominational affiliation.

Roughly six in ten (59%) Reform Jews and 47% of Conservative Jews identify as Democrats, com-
pared to only 23% of Orthodox Jews. Nearly four in ten (38%) Orthodox Jews identify as Republi-
can, and 29% identify as independent.

Religiously unaffiliated Americans are also more likely to identify as independent (47%) than they
are to identify as Democratic (33%) or Republican (11%) combined. Among religiously unaffiliated
subgroups, atheists are more Democratic leaning, with more than four in ten (44%) identifying as
Democrat.

No religious group is less likely to identify as Republican than Unitarian-Universalists. Only three
percent of Unitarian-Universalists are Republican. A majority (57%) identify as Democrat, and
one-third (33%) are independent. Also notable are the unique profile of Jehovahs Witnesses.
Reflecting their historical theological commitment to political neutrality, more than half (51%) of
Jehovahs Witnesses either refused to answer the question about partisan identity or responded
by saying they didnt know.

Shifting Religious Coalitions of Democratic and Republican Parties

The religious coalitions of the Democratic and Republican Parties are drifting further apart. In
2006, more than eight in ten (81%) Republicans identified as white Christian, including white
evangelical Protestant (37%), white mainline Protestant (22%), and white Catholic (20%). Half
(50%) of the members of the Democratic Party also identified as white Christian, including white
evangelical Protestant (17%), white mainline Protestant (16%), or white Catholic (16%). Today,
roughly three-quarters (73%) of the Republican Party is white Christian, but fewer than one-third
(29%) of the Democratic Party identifies this way.

Still, both political parties are being impacted by the larger tectonic changes in the religious land-
scape. These changes can be seen clearly in the generational profiles within each party.

More than eight in ten (84%) Republican seniors (age 65 or older) identify as white Christian,
including 42% who are white evangelical Protestant. Significantly fewer (57%) young Republicans
(age 18-29) are white Christian, and only about one-quarter (24%) are white evangelical Protes-
tant. Young Republicans are also three times more likely than seniors to be religiously unaffiliated
(19% vs. 6%, respectively).

The Democratic coalition is also shedding white Christian members, who are fast becoming a
minor constituency within the party. Among Democratic seniors, close to half identify as white
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 39

FIGURE 15. Widening Religious Gap Between Republicans and Democrats, 2006 vs. 2016

2006 2016
100 100 Unaffiliated
4
9 11
5 Other world
90 90
3 3 26 religions
3 5
6
80 80 3 Jewish
3
8
4 3 Other Christian
70 20 3 70
6 Hispanic Catholic
16
60 18 60
10 Hispanic Protestant

50 50 4 Black Protestant
22
18
Mormon
40 16 40
17
White Catholic
30 30
White mainline
16 10 Protestant
20 37 20
35 White evangelical
11
Protestant
10 10
17
8
0 0
Republican Democrat Republican Democrat

Sources: The Pew Research Center for the People & The Press and Pew Forum on Religion & Public Life, Religion and
Public Life Survey, 2006; PRRI 2016 American Values Atlas

Christian, including white evangelical Protestants (15%), white mainline Protestants (17%), or
white Catholics (15%). Only 15% of Democratic seniors are religiously unaffiliated. By contrast,
white Christians make up only 14% of young Democrats, including white evangelical Protestants
(3%), white mainline Protestants (6%), or white Catholics (4%). Among young Democrats, black
Protestants (14%) and Hispanic Catholics (10%) rival the number of white Christians in the group.
A whopping 40% of young Democrats are religiously unaffiliated.
40 AMERICAS CHANGING RELIGIOUS IDENTITY

FIGURE 16. Religious Identity by Age Among Republicans

100 Other religion

10 6
90 13 4 Unaffiliated
19
5
Other world religions
80 5
17 Jewish
6 4
70
4 4 Other Christian
6 18
60 Hispanic Catholic
4 15
22
50 Hispanic Protestant
13 18
17 Black Protestant
40
Mormon
30 16
White Catholic
42
20 37
32 White mainline Protestant
24
10 White evangelical Protestant

0
18-29 30-49 50-64 65+

Sources: PRRI 2016 American Values Atlas.

FIGURE 17. Religious Identity by Age Among Democrats

100 Other religion

15
90 20 Unaffiliated
29 Other world religions
80 40 4
5 Jewish
70 6 6
4 Other Christian
9
60 6 17 Hispanic Catholic
6
50 12 Hispanic Protestant
7 20
5 Black Protestant
40 10 15
Mormon
30 5 17 12
17 White Catholic
20 14
7 13 White mainline Protestant

10 4 9 White evangelical Protestant

6 15
10
5
0
18-29 30-49 50-64 65+

Sources: PRRI 2016 American Values Atlas.

FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 41

Political Ideology
White evangelical Protestants and Mormons are the most conservative religious groups in the
country. More than six in ten (62%) white evangelical Protestants and a majority (57%) of Mor-
mons identify as politically conservative. Liberals make up only 12% and 15%, respectively, of
these religious traditions. White mainline Protestants also lean more conservative than liberal
(38% vs. 25%, respectively), while white Catholics are about twice as likely to identify as conserva-
tive than liberal (42% vs. 22%, respectively).

There is wide variation in the political ideology of nonwhite Christian groups. Despite their strong
Democratic leanings, black Protestants are about as likely to be conservative (30%) as they are to
be liberal (29%). Similarly, Hispanic Protestants are about equally as likely to identify as conser-
vative (33%) and liberal (29%). Hispanic Protestants are much more likely to be conservative than
they are to be liberal (41% vs. 23%, respectively).

Non-Christian communities are more liberal leaning. Close to half of Jews (45%), Buddhists (48%),
and Hindus (45%) identify as liberal; significantly fewer Jews (27%), Buddhists (17%), and Hindus
(16%) identify as conservative. Muslims are also more likely to identify as liberal (38%) than they
are to identify as conservative (20%).

Religiously unaffiliated Americans also lean liberal in their politics. More than four in ten (43%)
identify as liberal, while 21% are conservative.

No religious group is as politically progressive as Unitarian-Universalists. Seven in ten (70%) Uni-

tarian-Universalists identify as liberal; only six percent are conservative.
42 AMERICAS CHANGING RELIGIOUS IDENTITY

Appendix 1: Top Three Religious

Groups in Each State
Unaffiliated White evangelical Protestant White mainline Protestant Mormon
Catholic Black Protestant Asian-Pacific Islander Protestant

White evangelical Protestant

National 24 20 17
Arkansas 37 16 15

Unaffiliated West Virginia 36 24 17

Vermont 41 22 16 Tennessee 36 18 15
Oregon 36 18 14 Alabama 35 15 15
Washington 35 18 14 Kentucky 33 21 16
Hawaii 34 24 10 Mississippi 31 25 14
New Hampshire 33 27 17 Oklahoma 30 21 14
Colorado 33 20 15 South Dakota 29 24 17
Maine 31 19 18 Missouri 28 20 19
Alaska 31 16 12 North Carolina 28 19 14
Montana 30 22 20 South Carolina 27 19 16
Nevada 29 21 10 Indiana 25 23 20
California 28 27 9 Georgia 22 20 19
Idaho 27 20 15
Catholic
Wyoming 26 21 17
Rhode Island 41 23 11
Nebraska 26 21 19
Massachusetts 34 28 10
Arizona 26 23 14
New Jersey 34 19 11
Maryland 26 18 17
Connecticut 31 27 12
Michigan 25 18 18
New York 31 25 9
Kansas 23 21 21
New Mexico 30 25 12
Virginia 23 20 15
Louisiana 26 18 17
Florida 23 21 15
Illinois 25 23 13

Tied Pennsylvania 24 23 17
Wisconsin 25 25 20 Delaware 24 21 13
Ohio 23 23 18 Texas 23 19 18
Minnesota 22 22 22
White mainline Protestant
Mormon North Dakota 30 22 20
Utah 51 23 7 Iowa 25 22 21
0 10 20 30 40 50 60 70 80 90 0 10 20 30 40 50 60 70 80 90

Sources: PRRI 2016 American Values Atlas.

FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 43

Appendix 2: Survey Methodology

The 2016 American Values Atlas (AVA) is a project of PRRI. Results for all demographic, religious
affiliation, and political affiliation questions were based on 101,438 bilingual telephone inter-
views (including 60,355 cell phone interviews) conducted between January 6, 2016 and January
10, 2017 by professional interviewers under the direction of SSRS. The AVA was made possible
by generous grants from The Evelyn and Walter Haas, Jr. Fund, The Ford Foundation, The Carn-
egie Corporation of New York, The Gill Foundation, The Nathan Cummings Foundation, and The
Unitarian Universalist Veatch Program at Shelter Rock.

Throughout 2016, at least 1,000 interviews were completed each week, with about 600 inter-
views conducted among respondents on their cell phones. Each week, interviewing occurred
over a five-day period, from Wednesday through Sunday or from Thursday through Monday.
The selection of respondents within households was accomplished by randomly requesting to
speak with the youngest adult male or female currently living in the household.

Data collection was based on stratified, single-stage, random-digit-dialing (RDD) of landline

telephone households and randomly generated cell phone numbers. The sample was designed
to represent the total U.S. adult population from all 50 states, including Hawaii and Alaska. The
landline and cell phone samples were provided by Marketing Systems Group.

The weighting was accomplished in two separate stages. The first stage of weighting corrects
for different probabilities of selection associated with the number of adults in each household
and each respondents telephone usage patterns. In the second stage, sample demographics
were balanced to match target population parameters for gender, age, education, race and His-
panic ethnicity, region (U.S. Census definitions), population density, and telephone usage. The
population density parameter was derived from 2010 Census data. The telephone usage pa-
rameter came from an analysis of the July-December 2015 National Health Interview Survey. All
other weighting parameters were derived from an analysis of the U.S. Census Bureaus March
2016 Current Population Survey.

The sample weighting was accomplished using iterative proportional fitting (IFP), a process that
simultaneously balances the distributions of all variables. Weights are trimmed so that they
do not exceed 4.0 or fall below 0.25 to prevent individual interviews from having too much
influence on the final results. The use of these weights in statistical analysis ensures that the
demographic characteristics of the sample closely approximate the demographic characteris-
tics of the target populations.
44 AMERICAS CHANGING RELIGIOUS IDENTITY

The margin of error for the total sample is +/- 0.4 percentage points at the 95% level of confi-
dence. The margin of error for the issue sample is +/- 0.6 percentage points at the 95% level of
confidence. The design effect is 1.4. In addition to sampling error, surveys may also be subject
to error or bias due to question wording, context, and order effects.

State Sample Sizes

State Total Sample State Total Sample

United States 101,438 United States 101,438
Alabama 1,485 Montana 524
Alaska 573 Nebraska 747
Arizona 2,042 Nevada 977
Arkansas 1,008 New Hampshire 432
California 9,640 New Jersey 2,779
Colorado 1,657 New Mexico 726
Connecticut 1,073 New York 7,072
Delaware 302 North Carolina 3,544
Florida 6,076 North Dakota 331
Georgia 2,928 Ohio 4,074
Hawaii 438 Oklahoma 1,154
Idaho 609 Oregon 1,648
Illinois 3,587 Pennsylvania 4,610
Indiana 2,288 Rhode Island 369
Iowa 1,325 South Carolina 1,636
Kansas 1,091 South Dakota 367
Kentucky 1,463 Tennessee 2,139
Louisiana 1,410 Texas 6,956
Maine 594 Utah 1,056
Maryland 1,727 Vermont 313
Massachusetts 1,952 Virginia 2,862
Michigan 2,997 Washington 2,264
Minnesota 2,060 West Virginia 784
Mississippi 833 Wisconsin 2,213
Missouri 2,171 Wyoming 244
FINDINGS FROM THE 2016 AMERICAN VALUES ATLAS 45

Appendix 3: About PRRI and the Authors

PRRI
PRRI is a nonprofit, nonpartisan organization dedicated to research at the intersection of religion,
values, and public life.

Our mission is to help journalists, opinion leaders, scholars, clergy, and the general public better
understand debates on public policy issues and the role of religion and values in American public
life by conducting high quality public opinion surveys and qualitative research.

PRRI is a member of the American Association for Public Opinion Research (AAPOR), the Ameri-
can Political Science Association (APSA), and the American Academy of Religion (AAR), and follows
the highest research standards of independence and academic excellence.

We are also a member organization of the National Council on Public Polls, an association of polling
organizations established in 1969, which sets the highest professional standards for public opinion
researchers. PRRI is also a supporting organization of the Transparency Initiative at AAPOR, an
initiative to place the value of openness at the center of the public opinion research profession.

As a nonpartisan, independent research organization, PRRI does not take positions on, nor do we
advocate for, particular policies. Research supported by our funders reflects PRRIs commitment
to independent inquiry and academic rigor. Research findings and conclusions are never altered
to accommodate other interests, including those of funders, other organizations, or government
bodies and officials.

History

Since PRRIs founding in 2009, our research has become a standard source of trusted
information among journalists, scholars, policy makers, clergy, and the general public. PRRI
research has been cited in thousands of media stories and academic publications, and plays a
leading role in deepening public understanding of the changing religious landscape and its role
in shaping American politics.

For a full list of recent projects, see our research page: http://www.prri.org/research/

PRRI also maintains a lively online presence on Facebook (http://www.facebook.com/prripoll) and

Twitter (http://www.twitter.com/prripoll).
46 AMERICAS CHANGING RELIGIOUS IDENTITY

About the Authors

Robert P. Jones, CEO

Dr. Robert P. Jones is the CEO of PRRI and a leading scholar and commentator on religion, values,
and public life. He is the author of The End of White Christian America and two other books, and
numerous peer-review articles on religion and public policy. Dr. Jones writes a column for The
Atlantic online on politics and culture and appears regularly in a Faith by the Numbers segment
on Interfaith Voices, the nations leading religion news magazine on public radio. He is frequently
featured in major national media such as MSNBC, CNN, NPR, The New York Times, The Washington
Post, and others.

Dr. Jones serves as the Co-Chair of the national steering committee for the Religion and Politics
Section at the American Academy of Religion and is a member of the editorial board for Politics
and Religion, a journal published by Cambridge University Press for the American Political Sci-
ence Association. He is also an active member of the Society for the Scientific Study of Religion, the
Society of Christian Ethics, and the American Association of Public Opinion Research. He holds a
Ph.D. in Religion from Emory University, where he specialized in sociology of religion, politics, and
religious ethics. He also holds an M.Div. from Southwestern Baptist Theological Seminary and a B.S.
in Mathematics and Computing Science from Mississippi College. In 2013, Dr. Jones was selected by
Emory Universitys Graduate Division of Religion as Distinguished Alumnus of the Year. In 2016, Dr.
Jones was selected by Mississippi Colleges Mathematics Department as Alumnus of the Year.

Before founding PRRI, Dr. Jones worked as a consultant and senior research fellow at several think
tanks in Washington, D.C., and was assistant professor of religious studies at Missouri State University.

Daniel Cox, Director of Research

Dr. Daniel Cox is the Research Director of PRRI, specializing in survey research, youth politics, and
religion. He has coauthored several academic book chapters and journal articles on topics relat-
ing to religious polarization, anti-Muslim attitudes in the U.S., religious tolerance of atheists, and
the origins of free-market ideology among evangelical Protestants. He regularly writes for major
news outlets like FiveThirtyEight, Yahoo! News, and U.S. News & World Report. Dr. Cox holds
an M.A. and a Ph.D. in American Government from Georgetown University, as well as a B.A. in
Political Science from Union College. Prior to joining PRRI, he served as Research Associate at the
Pew Forum on Religion & Public Life, where he worked as part of the core research team. He is an
active member of the American Association of Public Opinion Research (AAPOR) and the Ameri-
can Political Science Association (APSA).
PRRI
2027 Massachusetts Ave NW, 3rd Floor
Washington D.C. 20036
www.publicreligion.org