Monday, March 30, 2009

Breaking Down Data: Sociopolitical Views by Religious Affiliation

In discussions about the shape of America’s religious landscape, the most common assertion is that the United States is a Christian nation with Christian values. A typical phrasing of this claim often sounds something like the quote from Bishop E.W. Jackson, Sr. in this Wall Street Journal article, which reads, “To deny that the country is fundamentally Christian in its culture and its heritage is just not true.” Yet in that same article, Richard Land, president of the Southern Baptist Ethics and Religious Liberty Commission, points out that America is a pluralistic nation; indeed, America may boast the widest spread of religious belief within its population of any country in the world. But throughout this decade, and especially so in election years, political headlines focusing on tendencies of the greater electorate have spotlighted Christians, and Evangelicals -who constitute one-fifth of the voting population- in particular. What would be interesting to study, then, given the multitudes of religious groups in the United States and the tendency to default that the nation’s voice is inherently Christian, and more specifically Protestant, is this: which religion can say that its members share opinions closest to that of the “Average American”? Using data from last year’s U.S. Religious Landscape Survey conducted by the Pew Forum on Religion & Public Life, we can take an in-depth look at the social and political views of Americans broken down by their religious affiliation. Furthermore, by comparing the responses of each religious group to the total population’s response to questions posed in the survey, we can establish some basis as to which religious affiliation’s social views most closely mirror the national average; that is, the viewpoints espoused by an average American.

In its Social and Political Views (PDF) section, the Pew survey categorizes the respondents to its questions into fifteen separate religious or non-religious affiliations: Evangelical Protestant, Mainline Protestant, historically black Protestant, Catholic, Mormon, Jehovah’s Witness, Orthodox Christian, Jewish, Muslim, Buddhist, Hindu, Atheist, Agnostic, Secular Unaffiliated, and Religious Unaffiliated. These fifteen groups effectively make up the entire American population, showing the wide swath of belief systems that exist within the country. Those being surveyed were asked a series of questions intended to gauge how they felt about certain tenets of American society, such as how threatened they feel their values are by Hollywood, as well as their views on hotly debated issues like the acceptance of homosexuality and the legality of abortion. The Hollywood prompt read as such: “I often feel that my values are threatened by Hollywood and the entertainment industry,” to which the respondents had the option to agree, disagree, or do neither. For this specific prompt, 42% of the total population agreed that Hollywood was threatening, while 56% disagreed and 3% did not know or refused to answer. By comparison to the national average, 67% of Mormon survey participants agreed that the entertainment industry threatened their values, twenty-five points above the national average and the highest percentage of “Agrees” to this question; Atheists had the lowest percentage of “Agree” responses, with 19%.

By showing the discrepancies between national average responses to the Pew survey questions and responses of each group of believers or non-believers, I wanted to come up with an answer to which affiliation was the "most American" in their opinions. I will put it this way: if America had to send a member of one cohort to talk to an extraterrestrial who was curious about the social and political values of our country, which religion would be best suited? My method was to choose a number of prompts compiled in the survey that were well suited to providing good data for studying this question; I would then list the national average percentage response to one option of each prompt (for example, what percentage of respondents replied with “Agree” to the aforementioned Hollywood question), take the absolute value of the difference in percentage points between the national average and the responses of each group, and then find the average difference across the eleven prompts that I picked. Whichever group had the lowest average difference can be said to be closest to reflecting average American sociopolitical viewpoints, while those subsets with the highest average differences are groups that are far outside the mainstream of American sociological thought. The questions that I chose were ones that stuck out as representative of the most important political topics; these questions also had a sizable difference between the highest percent of positive responses and the lowest: I often feel that my values are threatened by Hollywood and the entertainment industry; I am satisfied with the way the political system is working in this country; I describe my political views as conservative; I describe my political views as moderate; I describe my political views as liberal; I worry the government is getting too involved in the issue of morality; I think abortion should be legal in most/all cases; Homosexuality is a way of life that should be accepted by society; Evolution is the best explanation for the origins of human life on earth; If given a choice, I would rather have a smaller government providing fewer services than a bigger government providing more services; The best way to ensure peace is through military strength (INSTEAD OF: Good diplomacy is the best way to ensure peace).

These questions, the "total population" percentage responses, and the percentage responses of each affiliation were entered into an Excel file, which can be downloaded here. Once I had input all the data, I proceeded to calculate a “difference average,” or the average discrepancy between the total population response and the response of each affiliation. As stated before, I took the absolute value of the differences in percentages for each prompt, with the exception of prompts 3, 4, and 5, the “political views” questions: for these, I divided each difference by three before including it in the sum to be averaged, because there were three possible responses to one basic question, being “Describe your political views.” Once all the religious affiliations had their own total sum of discrepancies, their sum was divided by 11 to create the "difference average." To recap, the lower the difference average, the closer the religious affiliation is on the whole to the sociopolitical views of an average American. The averages were a bit surprising:


Difference Avg.





Black Protestant




Secular Unaffiliated







Jehovah's Witness


The graph at right shows the spectrum of American sociological opinion by affiliation of belief. The x-axis is chopped up into three sections, by means of how far away each affiliation was from the mean Difference Average in terms of standard deviations: the average Difference Average was 10.82, with a standard deviation of 5.45. The first group can be said to make up mainstream religious thought; the second group ranges from just outside the mainstream to "slightly less than radical"; the third group is on the fringes of American sociological opinion-holding.

According to this method, the group whose viewpoints most reflect that of the American population are Catholics, an affiliation that makes up 23.9% of the American population, according to Pew, yet has only elected one of their own as President (so, when ET comes with some tough questions, an average Catholic seems to be best suited to give him or her answers). Even more surprising is the fact that Evangelicals appear to have more radical social beliefs in America than both Muslims and Hindus, which makes more sense upon inspection of the data: on three separate questions –those regarding abortion, homosexuality, and evolution– Evangelicals’ responses were at least twenty-four percentage points away from the national average. The groups displaying the most radical viewpoints are not necessarily a shock, as we never really see too many atheists, agnostics, or Jehovah’s Witnesses on our mainstream news networks, or being elected to office for that matter.

What these results tell me is that there is a well-defined mainstream of religious thought in America, and it is indeed Christian. However, it is comprised of moderate Christian voices: these are people who are skeptical of Hollywood and the political system, who are moderately conservative in their politics, and who for the most part accept the ways of life of others, as reflected in their overall above-average acceptance of homosexuality and abortion rights. It is this mainstream, along with members of historically black Protestant churches who fall barely outside its current, that comprise a bit more than half of the United States population, with Evangelicals present for another 26.3%. It is a bit unnerving that Evangelicals have the potential to hold a large sway on the political process with what might be called “moderately radical” sociopolitical viewpoints; however, as the last election may have shown, when the mainstream is flowing for the most part in one direction, that will be where the end results are found.


  1. William, your post on the Pew Forum’s survey and its very interesting findings brought forth some well thought-out questions about the voice of America. I have forever held that view that America, as you mentioned, is a pluralistic nation consisting of many religions. Your thesis arguing that the voice of America, at least portrayed in elections and as viewed by the rest of the world, largely seems to be Evangelical Protestant is a side that I had never really considered. Therefore, your thoughtful reflection on it and methodical analysis of the data really demonstrated a well-researched argument supported by empirical evidence. I do like your use of rhetoric, with terms like “while” and “however.” This really showed that you anticipated and were ready to counter any issue presented by the reader. Furthermore, the format of this recent post is different compared to your other posts, and definitely added something interesting for the reader. Having read many scientific articles, I felt that this format was very professional and commend your use of it.

    With any post there is always room for some improvement. While I do think that images in a blog post should be large, I would argue that the opening image may be too large and hence, take away from your introduction. I did appreciate your links to Excel documents that you created as well as graphs that were self-generated. I would have explained to the reader what the difference average meant in simpler terms (e.g. ~3 indicated a score similar to the average views of Americans and ~19 was a score far more different than the average American). Although this can be easily seen on your Excel sheet, for the few users who do not have the ability to open the document, that information could have more conveniently been placed on your post. I found it very interesting that Atheist and Agnostic views were so far from mainstream views. I think mentioning this in your conclusion, along with the similarity of perceptions between average Americans and the Muslim community, may have been enlightening to a reader, who may not have realized how similar the two are. Overall, I think this was a very enlightening post and your presentation of it showed your expertise and was appreciated.

  2. I am VERY impressed with the fact that you took data from different surveys to answer your own question and did not rely on someone else's curiosities for this post. Along with the fact that you decided to design your own analysis of such date to find your answer, this post definitely legitimizes you as a scholar.

    If you chose to expand this I would like to hear why you think Catholics, a group usually thought to be ultra conservative and almost "extreme," ended up being the most representative of the average American. Also, an analysis of a group's over arching doctrine and how close its followers usually adhere to said doctrine would be highly beneficial if you chose to expand it. For instance, the Catholic realm is extremely large and diverse. It would be interesting to find a survey of how much of the Catholic population actually adheres to the standards of the catholic church. This might explain your findings better.

    As far as the logistical and editorial critique of this post, I would say it is sometimes hard to follow the explanation of the data. Simplifying your methodology might help (And free up more space for your conclusion as well). Though I could sit down and take a couple more minutes to go through it and figure it out, most curious readers will not. I appreciated the thoroughness, but, as I was saying, your time could be spent better else where.

    Overall, Bravo!

  3. Another additional thought:

    Evangelicals and protestants can be, and often are, the same people. A qualification of this case probably needs to be made or else people may see it as an oversight or confusion in definition. Admitting possible limitations in your study, or the study you used for data, boosts credibility in the long run.


Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.