I have a degree with Mathematics, so it goes without saying that I enjoy the company of numbers. I love combing through statistics, looking for interesting correlations. It is this interest which led me recently to look at religious identification numbers from the US Census Bureau.
Comparing religious affiliation numbers between churches is not easy. Every church has there own method of reporting membership numbers. The LDS church reports everyone who has ever been baptized. Other churches, especially those with no formal membership concept, may report the number of people who attend each week. So, how does one actually compare membership numbers between sects? The best way is by asking people what religion they belong to, regardless of whether or not that church actually considers them a member. Since religion is an issue of belief, it seems reasonable to compare based upon belief, rather than membership.
The US Census Bureau has performed a study to determine US church membership based upon this identification, and the numbers are very intriguing (link). The data compares identification in 1990 with 2001. It only deals with adult population (which I consider a good idea when discussing religion).
Those who identified themselves as Mormon in 1990 totaled 2.48 million. In 2001 the number has grown to 2.78 million, giving a 12% gain over the 11 years. That seems like a strong growth number. But one must look at another factor when considering these numbers. The adult population, in the same time period, has grown from 175 million to 207 million, a growth of 18.5%. Therefore, if the Mormon church was not converting anyone, simply accounting for growth based upon the growth of population, the church should have had 2.94 million members in 2001. There is a shortfall of about 160,000 members or a growth of -6%. In other words, convert growth for the Mormon church in the United States is actually negative, the number of people leaving has exceeded the number joining.
This is not an isolated fact. A lot of Christian congregations have been seeing negative convert growth. While Christianity (in whole) has seen total growth of 5.29%, their convert growth has dropped to -13%. But this drop in membership was not felt equally among all groups. Evangelism and non-congregational Christianity seems to have been cannibalizing on the other Christian sects, as their numbers have increased. The study does note though that, since no options were provided, loosely-affiliated groups (like non-denominational and evangelical) can fluctuate a lot.
The religions that saw real growth were the non-Christian religions. Religions such as Islam, Buddhism, and Hindu saw huge growth (32% real, 14% convert). Some of that may be due to immigration, but I have my doubts. If someone wants to research that, I'd love to hear the results.
The biggest winner though was the atheists, humanists, and agnostics. Those who did not specify a religion climbed from 8% of the population to 14%. That means they had a convert growth of 87%. That's pretty astounding.
I'm not too surprised by these numbers though. Christianity has been seeing massive growth over the past 50 to 60 years. We, as a nation, are far more religious than we even were during the time of the Founding Fathers. Christianity, as a percentage of population, topped out around 90%. There weren't many people left to convert. The only way to go was down.
I also input all of the numbers into a spreadsheet on Google Docs if anyone wants to look at it (link).
Tragedy and Kitsch
5 hours ago
13 comments:
I'm doing my part.
"Is the LDS Church Shrinking?" No, according to your spreadsheet it had 12.06% growth from 1990 to 2001. This means that more people joined the church than left it. It is, however, interesting to see the growth of various religions as compared to population growth. Thanks for the links.
On a side note, I think it would be awesome to make available various census data with an interface such as this one developed at the UofU.
Kalon,
You are right. If we examine growth in a strictly nominal sense, then yes. The church is growing, as are the majority of churches on the list.
I do not think that nominal growth is an appropriate measurement for a population that is derived from a partition of a larger population, which is experiencing growth (or decay). In these cases, it is much more appropriate to view growth as market share growth.
Market share growth is generally the implied growth metric in the case of populations matching the above criteria. Imagine I claim that IE usage has decreased. Would you expect that I am claiming that IE has experienced nominal decay, or that its market share has decreased? Clearly, market share is the implied metric and is the significant one. The total number of people using the internet is a constantly increasing number (from population growth and internet adoption growth), so nominal growth is of little importance. It becomes completely reasonable to say that IE user base is shrinking when nominally it's actually increasing.
Kalon,
One clarification as well. You stated, "This means that more people joined the church than left it." In my original post I had stated, "...the number of people leaving has exceeded the number joining." We are both correct, because we are using the word "join" differently.
You have chosen to use the word in its literal sense. Technically when an 8-year-old is baptized, he has joined the church. In this sense, we see that more people have joined the church than left it.
I was using the word "join" in a slightly different fashion, which I should have made more explicit. I was referring to convert growth. I do not consider a person such as myself to have "joined" the church. I was raised LDS. When I think of people "joining" the church, I think of converts, those who take the missionary discussions. In this sense, there are definitely fewer people converting than there are leaving.
To make sure this is clear, I will explain I bit better. If there were zero people who had left the church and zero converts, we would expect the church membership to grow at the same rate as population growth. Well, almost. There are a couple of other issues. Immigration growth might reduce the membership growth. But, on the other hand, Mormon families are larger (on average) than non-Mormon families. So, regardless, we would expect to see a membership growth close to the population growth rate as a whole. In Kalon's semantics, this growth would fall under "joining," in mine it would not.
Now let's throw in these two other factors: conversion and resignation. Any significant difference in comparison to the population on whole should be accounted for by these two factors. Conversion is what I referred to above as "joining." Since Mormon growth has not kept up with population growth, conversion must be less than resignation.
Please let me know if there is a major fallacy in my argument.
"Imagine I claim that IE usage has decreased. Would you expect that I am claiming that IE has experienced nominal decay, or that its market share has decreased?" I would actually have assumed nominal decay as I did here. I think most articles I have read about browser statistics are clear to say that it is the share that they are talking about.
Anyway thanks for the clarification. I think your argument is pretty thorough given your premises. One issue is that your data is US based while the church is global. Thus even though the church's market share is shrinking in the US it cannot imply the same on the global/overall level. And hence I think that ignoring immigration and emigration is pretty significant.
If you know of any good religious identification studies for the rest of the world, let me know. There are a number of developments that do not imply good growth elsewhere. The Mormon church just recently consolidated three areas in Europe under one Area President. Perhaps there is something else going on, but its unlikely that the church would be consolidating if it were seeing massive growth.
Also, looking at church growth as a percentage of membership has been on a steady decline for the past 10 years. The total number of missions, stakes, wards, etc has leveled off quite a bit. The number of converts per missionary have decreased at the same time that converts/missionary has also decreased. Pretty much every metric you look at tells the same story, and it ain't pretty.
If you're interested in LDS demographics, http://cumorah.com is a great site to check out. They have quite a bit of information. If you love numbers like myself, you'll find it very interesting.
I would like to contest one of your premises. Namely that immigration growth isn't significant. I found this census document discussing immigration differences from the year 1990 to 2000. It seems to indicate that quite a bit of population growth indeed has come through immigration. In fact from 1990 to 2000 the US saw a population growth around 32 million. The foreign born population increased about 11 million. Thus about 1/3 of the population growth came from immigration during that time period.
Since so much of population growth came from immigration it make sense that we would see the religious shares in the US gradually move toward religious shares on a global scale. Hence religions such as Islam, Buddhism, and Hindu increasing in share and religions such as the LDS decreasing in share.
You bring up a good point, and thanks for finding that immigration report. I had briefly searched for it, but had trouble finding it. There's a lot of good data in the report. You also make a good argument that immigration should not be ignored; I was unaware it was as high as it was. There are a few points I would highlight, regarding this new data.
If you look at page 5 of the study, you will see something that should not come as a surprise to anyone. 51.7% of US immigrants are from Latin America. Latins are overwelmingly Christian, so this is even worse news for the overall Christian numbers. Even with the heavy influx of Catholics, Catholicism is in a major decline.
Mexico also has roughly a .5% Mormon population. So we should see a fair number of Mormons in this group. In fact, to account for the influence of immigration, all we need to find our 123,000 Mormon immigrants between 1990-2000. I doubt there were that many, but I'm sure there were at least half that many.
So, clearly immigration tempers the numbers for Mormons (while making them worse for Catholics). But, even if you take the numbers and remove 10 million people from the adult population in 2001, you will see that Mormon convert growth is still negative (but not by much). So, even if not a single immigrant were Mormon (which I know to be false, by my own acquaintances), then the church would still be experiencing negative growth (but much less negative growth).
On looking at the numbers it seems that the growth of non-Christian religions may be partially connected to immigration. I doubt that it is the full story though. In running numbers, trying to account for the growth solely through immigration leaves these non-Christian religions with a deficit. But it is clearly a major factor. Anecdotally, I know of a number of people who have begun to move away from Christianity towards Eastern religions.
I found another interesting tidbit that concerns the original article. I spoke briefly concerning the difficulty of comparing church affiliation when using the numbers that the individual sects provide. One can see this evidenced when looking at the vast discrepancy between the numbers provided by the Mormon church itself and the actual real-world data.
It would not be appropriate to compare, too closely, the nominal values reported by the church and those in the above study, because the above study deals only with adult population. It is reasonable to consider growth numbers, as those should remain fairly consistent. During the 1990-2001 period, the church reported an average yearly gain of 1.67%. The actual average yearly gain was 1.04%. The church's numbers claim that 1.89% of Americans are Mormon, whereas the actual number is 1.34%. That results in a 60% overstatement of growth and a 41% overstatement of market share.
Obviously I am not accusing the church of falsifying their numbers. The church has made it quiet clear what their numbers are, the total number of people who have ever been baptized (regardless of attendance). I just think it's a lousy metric to report, evidenced by its vast discrepancy from real-world data.
"But, even if you take the numbers and remove 10 million people from the adult population in 2001..." Well I don't think it is quite as easy as that. The report specifically states on page two that native population increase was 9.3 percent. This kind of means that any increase over 9.3% are in fact conversions. The reason that is only kind of is because of what you already mentioned about some portion of the immigration population already being a certain denomination. However even accounting for .5% of immigrations being Mormons this is still under the actual growth exhibited by the LDS church.
This means we have a pretty big discrepancy between these two studies. This study is saying that the population increased by a factor of 13% overall. The original study I presented showed growth of 18%.
There are a few possible ways of accounting for this. One is the fact that my study is from 1990-2001 rather than 1990-2000. That's one more year. That's definitely not enough to account for a 5% deficit, but it's a start.
I'm not really sure how to account for the remainder of the discrepancy. Perhaps they simply contacted more people or something.
The rest of the discrepancy might be accountable in the consideration that my original study only looks at adult population, whereas your study shows aggregate growth across all ages.
By your same logic, we could remove growth rate of immigrants (3.7%) from the total growth rate (18.5%) to get a cutoff of 14.8%. Do you see the problem with using 9.3% as the cutoff?
To account for the discrepancy, the only reasonable thing to do is apply the same immigrant distribution to the religious identification numbers. That would give (13-9.3)/13 = 28%. 28%*(207,980-175,440) = 9,261. In other words, we would expect to see a growth of 9.26 million adult immigrants in the religious identification study. This is even less than the 10 million I allowed in my previous comment (I intentionally low-balled my original number).
Great catch about the discrepancy of the numbers. I agree that the extra year is probably the biggest contributor to the difference. Another contributor might be that the foreign-born population of one year contribute to the native born population of the next year. I guess it's a definite testament to the difficulty of reconciling numbers.
Post a Comment