Percentage of US Population Affiliated with Various Religions, 1913 and 2013

18-year-old Helena Muffly wrote exactly 100 years ago today: 

Sunday, June, 1913:  

What is so rare as a day in June ,

For then if ever comes perfect days,

When song of bird and hum of bees

Bring to us fair summer’s sweetest day.

Went to Sunday school this afternoon. Took my time a getting home. I heard some of the best speaking I have ever listened to this evening. A converted Jew talked about some of the customs of the Jewish people in the Reformed Church at McEwensville.


Her middle-aged granddaughter’s comments 100 years later:

Grandma’ diary entry made me wonder: What percentage of the US population were considered members of the various religions in 1913 and 2013?

I discovered that this was a much more difficult question to answer than I thought it would be.  The data on religious affiliations were collected very differently in the early 1900s than how it is gathered now—so when the data are compiled to do a comparison, it’s kind of like comparing apples to oranges.

This gets complicated, but let me try to explain what I did to create the figure above:

In the early 1900s , the US Bureau of the Census conducted a religious census every ten years.  Religious leaders were asked how many members their congregation had; whereas in recent years, various non-profit organizations have conducted surveys where they asked a sample of the population about their religious preferences.  As a result of these differences in methodology many more people were considered to have no religious affiliation a hundred years ago than now.

Calculation of 1913 Percentages

For the figure above, I used data from an article titled “U.S. has 42,043,374 Members of Church, New Census Shows” in the May 2, 1918 issue of the Minneapolis Morning Tribune. According to the article:

The term ‘members’ has  a wide variety of uses. In most Protestant bodies it is limited to communicants or confirmed members; in the Roman Catholic, Eastern, and some other churches it includes all baptized persons, while in some bodies it covers enrolled persons.

The membership for the Jewish congregations requires some explanation. Some congregations reported as members all who contribute to the treasury of the congregation and not infrequently included women and children. The more orthodox, of the other hand, reported only those males who have incorporated the institution or have bought share or membership in it, but do not recognize as members other persons who are regular attendees or even contributors.

For the figure, I used data from the 1916 Religious Census, as reported in the Minneapolis Morning Tribune article, since this was the one done closest to a hundred years ago.  To calculate the percentages I used the US population estimate for 1916 as reported by the US Bureau of the Census. I assumed that the percentage of the population who were members of various religions did not change much between 1913 and 1916 when creating the figure.

Calculation of the 2013 Percentages

For the 2013 percentages, I used data from a survey conducted by the Pew Research Center.  Phone interviews were used to survey a sample of the US population. Respondents were asked which religion they identified with.

The survey was conducted in July, 2012—and I assume that the percentages have not changed significantly since then.

Monthly Poem

On the first of each month Grandma included a poem in the diary. For more information about the poems, see a previous post:

Monthly Poem in Diary

26 thoughts on “Percentage of US Population Affiliated with Various Religions, 1913 and 2013

  1. Flip the two years around and it would be more like what I would expect. I think that many of the other from 1916 would fall under smaller protestant groups that didn’t show up as choices in the census.

    1. The 1916 Religious Census reported results for 201 denominations—so I think that the Census Bureau probably sent the survey out to most churches. There were many different denominations and ethnically-based churches scattered across the US a hundred years ago. For example, a town in Minnesota might have had a German Lutheran church, a Swedish Lutheran church, and a Norwegian Lutheran church. My understanding is that one of key issues at the time was the large number of very small churches that wouldn’t cooperate with each other—and that people hoped the census might provide some insight into the issue.

      The main problem with the data is the difference in who was surveyed across the two years, and how they asked the question. In 1916 the census survey was sent to ministers and other church leaders who were asked how many “members” their church had. The ministers took a very narrow view of who was an official member—so many people who may have identified with a denomination or church weren’t counted. Whereas, in 2012, individuals were asked which religious organizations they identified with—so people who may not have attended a church in decades, but who culturally identified with a denomination or religion, were counted.

    2. Yes before I saw the dates I had flipped the results in my mind! But your explanation clears some things up. I mean what if there were no churches in some remote areas? People may have still had services in their homes, etc… Very interesting though; data would change dramatically depending on whether the question is do you attend services regulary or do you consider yourself religious?

  2. Quite a “wordy” entry for Grandma! I know, in the past, we’ve wondered if the poems were originals of hers, but this is one I remember from literature class in school: it was one of my favorites!

    1. I wonder if she might have “customized” parts of someone else’s poem to suit how she felt about the month of June. Your comment led me to “google” the first line, “What is so rare as a day in June.” I discovered that it is part of the prelude to the first part of a very long poem by James Russell Lowell called The Vision of Sir Launfal. You can see the entire poem on the Camelot Project site at the University of Rochester

      The line “What is so rare. . ., ” is the first line of the fourth stanza.

      The words Grandma used after the first line don’t match Lowell’s poem–which suggests that she probably started with it and and then made some changes.

      Thank you for encouraging me to dig deeper into the poem.

    1. You comment reminds me of a discussion I once had with a friend. who grew up in Minnesota. I wondered why Minnesota had so many colleges affiliated with the Lutheran church compared to the number in other states–and she said, “It’s because some were founded by Norwegian Lutherans and some were founded by Swedish Lutherans.” 🙂

  3. I am so impressed with your research, your calculations and conclusions, and then the final chart which works right along with your blog comparisons from 1913 to 2013. And from your grandma’s diary, it sounds like she also had an inquiring mind!.

  4. It is interesting how tricky it is to compare the information collected. And also to note the different ways religious information was collected and is collected. I think our information on religion is collected in the regular national census.

    1. It’s interesting how different countries collect different data. According to the US Census Bureau, the US collected data on religious bodies from 1906 to 1936. It also says that information was collected on religion as part of the General Population Survey in 1957, and that they are no longer allowed to collect it on a mandatory basis. This is what the Census Bureau FAQ sheet says about why it is no longer collected:

      “The U.S. Census Bureau does not collect data on religious affiliation in its demographic surveys or decennial census. Public Law 94-521 prohibits us from asking a question on religious affiliation on a mandatory basis; in some person or household surveys, however, the U.S. Census Bureau may collect information about religious practices, on a voluntary basis.Therefore, the U.S. Census Bureau is not the source for information on religion, nor is the Census Bureau the source for information on religious affiliation.”

      1. Fascinating. It makes sense that they couldn’t ask a question like that in a mandatory form. I can’t remember how the question is worded in our census. I should though because we had a census just a few months ago 🙂

  5. Hard to hit the like button. I’m worried about our world today. Whatever the census says, we’ve changed. On Sunday mornings Americans used to go to church and everything closed down on Wednesday evening, because so many people went back for prayer meeting. When I was a kid, everyone was getting in the car to go to church on Sunday mornings. We waved at each other and commented on new hats. We also didn’t have any churches that protested at funerals or blew up abortion clinics. Religion in and of itself is not good, but the connections it makes between people and their God – that is good. Thanks for your work.

    1. You described it perfectly. Your description of going to church on Sunday morning when you were a child is very similar to what I remember. Thanks for the thoughtful comment.

  6. Those numbers were not what I expected, honestly. Interesting nonetheless. I’d love to know how the Muslim population has changed, though I don’t think there would’ve been too many Muslims in America at that time. I’m probably very wrong, but you never hear much about the Islamic religion in old American books and stuff…

Leave a Reply to pattisj Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s