“In reality, there are only around 4-6 million people on this planet who consider themselves LDS.”
My suggestion was to test it by making a simple extrapolation based on one of the most recent U.S. religious self-identification surveys. My simple extrapolation yielded 5 – 6 million self-identified Mormons in the U.S. alone. In response, “someone” characterized my math skills as “phony” and “bogus”; while the survey I chose (the 2007 Pew U.S. Religious Landscape Survey) was dismissed as flawed due to “oversampling.” A link to a Joanna Brooks’ article (below) was provided to prove the “oversampling” charge.
The idea of extrapolating estimated numbers from Pew and other well respected religious self-identification surveys is exactly what researchers, journalists and organizations like the U.S. Census Bureau do. For instance the U.S. census used ARIS data to make this 2008 estimate for adult populations only:
And Joanna Brooks used ARIS data to do the same thing here:
And then there is this 2004 extrapolation across all age groups (children, teens and adults) which utilized one self-identification survey (ARIS) for adults and another (the UNC National Study of Youth and Religion) for teens and children:
“Some groups (e.g., Episcopal, Congregational, Judaism) have fewer children proportionate to their total population, and some groups (e.g., Catholic, Pentecostal, Latter-day Saints) have higher proportionate numbers of children, in which case the un-modified extrapolation to the total population would yield an undercount. For example, in 2004, 2.5% of American teenagers said they were Latter-day Saints (National Study of Youth and Religion, University of North Carolina at Chapel Hill; sample size: 3,370 nationwide), a figure higher than the 1.3% of American adults who identified themselves as Latter-day Saints in 2001. If this 2.5% figure were extrapolated to the total population, it would yield a figure of 7,334,574 total Latter-day Saints (children, teens and adults) in the U.S. for 2004 - a number nearly 2 million higher than counted in official membership records. This difference in survey results between adult and teen populations indicates that this group's U.S. membership skews young. Similar differences (pushing a group's actual numbers higher than extrapolated if it skews young and lower than extrapolated if it skews old) would be expected for all groups.” (http://www.lb9.uscou...r_relGroups.pdf)
I e-mailed Erin O’Connell, Associate Director for Communications at the Pew Research Center’s Forum on Religion and Public Life. I asked if they were aware of the Joanna Brooks criticism. (Note: The “oversampling” criticism actually originated with ARIS researchers Richard D. Phillips and Ryan T. Cragun.) Erin replied almost immediately and indicated that they were in fact aware of the Phillips/Cragun criticism and believed it to be “unwarranted.” Erin suggested a short telephone conversation with a Pew researcher who could explain in more detail. I called the researcher and enjoyed a very interesting and enlightening 30-minute conversation. Following are some of his more relevant observations:
The Phillips/Cragun “oversampling” criticism (http://www.mormonsocialscience.org/) was actually directed at the 2011 Pew Religious Devotion survey titled “Mormons in America” which can be found here: (http://www.Pewforum....ve-summary.aspx). The criticism was not directed at the 2007 U.S. Religious Landscape survey (which was my source), which is found here: http://religions.Pewforum.org/. “Mormons in America” focused on measures of “religious devotion” and was specific to Mormons; while “Landscape” focused on “self-identification” and was specific to the entire U.S. population. Apparently, “someone” conflated the two.
Phillips & Cragun have yet to express their criticism of “Mormons in America” in any academic journal or appropriate forum; therefore, while Pew is aware of the criticism, it has been unable to respond officially; nevertheless, the researcher was more than happy and quite prepared to tell me over the phone why he is convinced the criticism is “unwarranted” and also “puzzling.”
For “Mormons in America”, the researcher explained that Pew did in fact utilize an “oversample” technique in certain Utah counties with higher-than-average Mormon populations; however, the data from these counties was also correspondingly “weighted” to discount for the oversampling. This is in complete accord with long-established statistical techniques utilized within the Social Sciences. The researcher also pointed out that results from “Mormons in America” (where “over-sampling” was utilized) were virtually identical to results from prior “Religious Devotion” surveys Pew has done where the oversampling/weighting technique was not employed.
With regard to the 2007 “Landscape” survey (which was the basis of my extrapolation), the researcher made clear that absolutely no “oversampling” technique had been employed. He also pointed out that Pew (as well as other surveying organizations) regularly asks about religious preferences for the demographic portion of nearly every survey—even when the purpose of the survey is not about religion. This data is “tracked” and accumulated annually: the percentage of U.S. adults self-identifying as “Mormon” has been tracking at 1.6% to 1.7% for the past few years; but, the figure tracked at 1.9% throughout all of 2011; and it has been tracking at 1.6% so far in 2012. In fact, the tracking percentages are not all that different—usually within the margins of error—from one reputable survey to the next. Pew’s adult Mormon percentage tracked at 1.6% in 2007; ARIS tracked it at 1.4%; while both Pew and Gallup tracked it at 1.9% in 2011.
Finally, the Pew researcher thought it important to point out that the ARIS approach is not above criticism. For example, ARIS reports a large percentage of U.S. adults who simply identify themselves as “Christian.” Since ARIS does not utilize a “secondary” line of questions (like Pew and Gallup do), this percentage is significantly inflated. A large number of respondents who initially identify themselves as “Christian” will subsequently identify themselves more specifically—as “Baptist” or “Mormon” for instance—if they are given the choice.
Please feel free to draw your own conclusions--I certainly have.
Edited by Okrahomer, 25 April 2012 - 05:14 PM.