“Protestant” isn’t one thing. It’s three different things.

The tradition that held its ground — until recently

Drive through almost any American community and you’ll pass an evangelical house of worship — the historic Southern Baptist church on the town square, a Pentecostal storefront, or the hip non-denominational congregation meeting in a converted warehouse in a growing exurb. Evangelicals are everywhere, and no religious group has shaped American culture, society, and politics more over the last fifty years.

What makes someone an evangelical? The most common answer comes from historian David Bebbington, who pointed to four hallmarks: a high view of the Bible, a strong emphasis on the atoning death of Jesus on the cross, the belief that people need a personal born-again conversion, and the conviction that faith has to show up in action. That DNA was forged in the revivals of the First and Second Great Awakenings, when preachers like Jonathan Edwards and John Wesley urged colonists toward personal conversion rather than the communal, hierarchical religion of the Congregationalists and Episcopalians. The message fit the American ethos of rugged individualism, and upstart traditions like the Baptists and Methodists exploded across the young country.

But the more interesting story may be what’s happening inside evangelicalism. In 1972, this was a Baptist movement — nearly two-thirds of all evangelicals belonged to a Baptist denomination, anchored by the Southern Baptist Convention, still the largest Protestant denomination in the country. Today, Baptists are just 38% of the tradition. The group that filled the gap? Non-denominational Christians, who were a rounding error in the early 1970s and now make up 30% of all evangelicals. The future of evangelicalism isn’t the church on the town square with a denominational sign out front — it’s the one meeting in a converted warehouse with a one-word name.

Politically, evangelicals remain the anchor of the Republican coalition: about three-quarters voted for Donald Trump in 2020, closer to 80% among white evangelicals.

Here’s the bottom line: for decades, evangelicalism was the great exception to American secularization — while every other Christian tradition eroded, it held its ground. That’s no longer quite true. The tradition has clearly slipped from its early-1990s peak. But compared to what has happened to the mainline, evangelicals are still playing a very different game.

The collapse of the old establishment

Ask an average American what a “mainline Protestant” is and you’ll probably get a blank stare. That’s remarkable, because this was once the dominant expression of religion in the United States. According to religious historian James Hudnut-Beumler, 52% of all Americans aligned with a mainline denomination in 1958. Half the country. Today, most people have never even heard the term.

The mainline is anchored by the “Seven Sisters”: the United Methodist Church, the Evangelical Lutheran Church in America, the Presbyterian Church (USA), the Episcopal Church, the American Baptist Churches USA, the United Church of Christ, and the Disciples of Christ. What sets them apart from their evangelical cousins is a generally less conservative posture. Mainline Protestants are far less likely to read the Bible literally — about a quarter do, versus roughly 60% of evangelicals — every major mainline denomination has ordained women for decades, and most have become affirming of same-sex couples. For most of the 20th century, these were the churches of the educated, middle-class establishment; the pastors of the big downtown Methodist and Episcopal churches were civic leaders, not just religious ones.

Then the bottom fell out. In the mid-1980s, the mainline share of the population plunged from the high twenties to 19% in just a few years. In religious demography, where change usually happens a point or two at a time, that kind of collapse is essentially unprecedented. Since then the decline has settled into a steady drip — 12% by the mid-2000s, 10% in the late 2010s, and now down to 8.7% in the 2024 data. The tradition that once claimed half the country can no longer claim one American in ten.

The denominations’ own meticulous record-keeping tells the same story, denomination by denomination. Since 1987, every single one of the Seven Sisters has lost at least 30% of its membership. The Disciples of Christ have shed a staggering 81% of their members. The Presbyterian Church (USA) is down 65%, from about three million members to one million. And the United Methodist Church — the largest of the Sisters — has lost 56%, with the decline accelerating sharply in the last few years as thousands of congregations disaffiliated in the denomination’s split over sexuality.

One wrinkle: church attendance among the mainliners who remain has stayed remarkably stable for fifty years. That’s a compositional effect — as the casually attached drifted away, the devout stayed. But the pews are aging, young families are scarce, and the tradition’s only real path forward is conversion in an increasingly secular country. Politically, the mainline is one of the few genuinely divided traditions left: in 2020, it split 52% for Trump and 47% for Biden — a purple church in a red-and-blue religious landscape.

Evangelical theology, opposite politics

Anyone who has ever worshiped in a predominantly African-American congregation knows the experience is unlike any other corner of Protestantism — music woven through nearly every part of the service, expansive and emotional preaching that draws verbal and physical responses from the pews, and Sunday gatherings that can stretch into a third hour while a mainline service wraps up in under sixty minutes. But the reason scholars treat Black Protestants as their own tradition runs much deeper than worship style.

The Black Church developed as a distinct institution because it had to. Even after the Civil War, African-Americans were denied full access to American society — and many white Protestant congregations excluded them too. So the Black community built its own denominations, beginning with the African Methodist Episcopal Church, founded by Richard Allen in Philadelphia in 1816. And because segregation shut Black Americans out of the social clubs, civic halls, and political venues open to whites, the local church consolidated all of it under one roof: worship, socializing, economic life, and political organizing. No white Protestant tradition has ever played that kind of all-encompassing role.

The tradition’s footprint in the population has been shrinking, though. Black Protestants were about 9% of American adults in the early 1970s and stayed remarkably stable — around 7–8% — for decades. But over the last twenty years the share has slid steadily, down to 4.4% in the 2024 data.

The largest denominations in the tradition are the National Missionary Baptist Convention at about 2.4 million members, the National Baptist Convention at 1.56 million, the AME Church at just over a million, and the Church of God in Christ in the Pentecostal wing. Geographically, the center of gravity is the Deep South — 8% of Mississippians belong to one of the major Black Protestant denominations. And the boundaries are fuzzier than the categories suggest: 17% of Southern Baptists are African-American, which means there are over two million Black members of America’s flagship evangelical denomination.

Here’s what makes the category necessary. On theology, Black Protestants look a lot like evangelicals — rates of biblical literalism are similar, and many Black pastors preach sermons that would fit comfortably in an evangelical pulpit. But at the ballot box, the two traditions could not be further apart.

That thirteen-to-one margin is why RELTRAD carves out a third category rather than folding Black Protestants in with evangelicals. It’s also a standing reminder for anyone studying American religion: race may not always be the variable you’re looking at, but it always looms in the background.

Look up a denomination’s RELTRAD classification

Some of these assignments are intuitive. Plenty aren’t — the Evangelical Lutheran Church in America is mainline despite its name, the United Church of Christ and the Churches of Christ sit in different traditions entirely, and two denominations with nearly identical Baptist names can land in different families. Type a denomination below to see where the standard scheme puts it.

Classifications follow the RELTRAD scheme (Steensland et al. 2000) as implemented in the General Social Survey. One caveat for survey nerds: for ambiguous answers — “just Baptist,” “just Methodist,” or non-denominational — the coding also uses the respondent’s race and church attendance to make the call, so individual survey respondents can be sorted differently than their denomination’s home tradition.

A short annotated bibliography of RELTRAD

• 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 79(1): 291–318.

  The founding document. This is the paper that introduced the RELTRAD classification scheme and made the case — historical, theological, and empirical — for treating evangelical, mainline, and Black Protestantism as three distinct traditions rather than points on a liberal-conservative spectrum. If you read one thing on this list, read this.

• Stetzer, Ed, and Ryan P. Burge. 2016. “Reltrad Coding Problems and a New Repository.” Politics and Religion 9(1): 187–190.

  It turns out much of the RELTRAD syntax circulating among researchers contained coding errors that misclassified thousands of respondents. This short note documents the problems and establishes an open, corrected code repository so that everyone’s “evangelical” means the same thing.

• Burge, Ryan P., and Andrew R. Lewis. 2018. “Measuring Evangelicals: Practical Considerations for Social Scientists.” Politics and Religion 11(4): 745–759.

  What happens when you count evangelicals by their denomination versus just asking “are you a born-again or evangelical Christian?” This paper walks through how much the two approaches overlap, where they diverge — including self-identified evangelicals who never attend church — and which measure to use for which research question.

• Shelton, Jason E., and Ryon J. Cobb. 2017. “Black Reltrad: Measuring Religious Diversity and Commonality Among African Americans.” Journal for the Scientific Study of Religion 56(4): 737–764.

  RELTRAD treats Black Protestants as a single family; this paper opens the box and builds a classification of the diversity inside the Black Church — Baptist, Methodist, Pentecostal, and beyond. Essential reading if the Black Protestant section above left you wanting more.

• Lehman, Derek, and Darren E. Sherkat. 2018. “Measuring Religious Identification in the United States.” Journal for the Scientific Study of Religion 57(4): 779–794.

  The loyal opposition. Lehman and Sherkat argue that RELTRAD’s reliance on race and its handling of ambiguous answers introduce systematic problems, and they propose an alternative classification. Read it alongside the Steensland piece to understand the live debates in how scholars sort American believers.

• The RELTRAD code repository — github.com/ryanburge/reltrad

  Working R syntax for building the RELTRAD variable from raw General Social Survey data — the corrected code from the Stetzer and Burge paper above. Fork it and you can reproduce every trend line on this page.

• The Association of Religion Data Archives — thearda.com

  The largest free archive of religion survey data in the world, with hundreds of downloadable datasets, denominational membership statistics, and RELTRAD-coded variables built into many of its files.

• The General Social Survey Data Explorer — gssdataexplorer.norc.org

  The GSS is the backbone of every trend line on this page — the longest-running high-quality measure of American religious affiliation, fielded since 1972. The Data Explorer lets you build your own tables in the browser, no R required (though R is more fun).