THE METHODOLOGY OF MANIPULATION: STATISTICS IN PLAY

Two widely reported statistical analyses regarding Mormons in America have recently been published: the Trinity College Survey, released in December, 2011 and the Pew Report, January, 2012. Incredibly, it seems that media coverage has focused more on the later report than the former. This is especially the case in Utah and most certainly true of LDS owned Deseret News. This blog is an analysis of the methodology behind the conclusions reached in both studies.

Trinity is a well-respected private college located in Hartford Connecticut; with the exception of Yale, it is the oldest institution of higher learning in the state. Despite its Episcopal roots, years ago Trinity made it a policy not to impose any religious standards on its students. Today the liberal arts college is widely known for the diversity of its student body. Two Trinity sociology professors are the principal investigators of the report. Additionally, Rick Philips and Ryan T. Cragun (non-LDS), both affiliated with the Mormon Social Science Association, contributed to the report.

The Pew Research Center is a subsidiary of the Pew Charitable Trusts, which despite its title is actually a single entity created by an ultra-conservative family, the Pews. At one time the John Birch Society was a beneficiary of this organization. Per the IRS code, attaining non-profit status in 2004 allows the Trusts to devote a non-substantial part of its resources to lobbying the pubic sector.

The pool of respondents from each group, and therefore each reports' data differ widely. Trinity's potential respondents were any adults in the continental United States; Pew's respondents were adults who had been previously identified as active members of the Church of Jesus Christ of Latter-day Saints. Thus, the conclusions reached by the two studies are not comparable. However, the respective methodologies used to gather the collected data deserve close scrutiny. Importantly, the combined American Religious Identification Survey (ARIS) studies constitute the most comprehensive and objective statistical research of the numbers of Mormons in America available to social scientists. In contrast, the Pew study data drew from slightly more than a thousand acknowledged Mormons.

Trinity's report was based on two waves of data collected by ARIS, one in 1990 and the other in 2008. In both waves, respondents were obtained through randomized digit dialed telephone numbers. Pollsters from both waves asked respondents the open-ended question: "What is your religion, if any?" The 1998 survey polled 113,723 Americans and received 1,742 'LDS' or Mormon responses. In 2008 the poll reached 54,461 Americans, 783 of whom self-identified as Mormon. Though a Salt Lake Tribune report (updated Dec. 15, 2011) entitled “Gender Gap Widening among Mormons in Utah, but why?” claims that the Trinity study authors admitted to a margin of error of 4-5%, the Trinity College website estimates a statistical margin of error for the combined studies at less than 0.5%.

The Pew Study lists Luis Lugo, a University of Chicago PhD in political science as director, Cary Funk as researcher and Conrad Hackett as demographer. Additionally, two-thirds of Pew's 'panel of expert advisers' are LDS-affiliated, if not practicing Mormons: Mathew Bowman professor of religion at Hampden-Sydney College, Marie Cornwall, BYU sociology professor, Allison Pond of the Deseret News and Teryl Givens, professor of literature/religion at the University of Richmond. It is the Pew study which has a 95% confidence level, i. e., a 4-5% margin of error.

The Pew survey began armed with the landline and cell phone numbers of 91,590 ostensible Mormons—excluding former Mormons. Those whom Pew pollsters contacted were asked a series of questions about Mormonism. While almost half of these (45,710) were non-working numbers: faxes, business/government offices or otherwise "ineligible respondents," another 44,706 were: always busy, had no answer, engaged call-blocking/screening, categorized as "no screener completed," and two contacts had language/competency barriers. That left the Pew people with just 1174 total respondents. Of that number, 155 refused to speak with the pollsters, broke-off the conversation, were never available, or failed to return a message left by Pew. Thus, the total number of interviews upon which the Pew data based its conclusions is just 1019 members of the LDS Church (Pew Report pps. 81-84);697 interviews were considered a fresh sample, and 322 were recontacts from a prior sample. Additionally the data was admittedly skewed geographically to query more frequently from areas deemed to have more LDS residents per capita. However, there is no readily apparent source for these 91,590 numbers—only that they came from a “national sample.” Were they part of LDS Church records? If not, where did they originate? If so, given the abysmal response rate, these records seem to be extremely dated.

What were the results of Pew's study? With the caveat that more than half of respondents reside in the Mountain-west and a third in Utah, Pew's study concludes that Mormons were generally happy with their lives/communities, and had families where the husband supported a stay-at-home wife/mother. Two-thirds believe Mormons are not accepted in mainstream society, but also believe that they are becoming more accepted. Most are Republican and are unhappy with the federal government. Mormons were divided on whether the U.S. was ready for a Mormon president, men feeling more positive than the women surveyed. 88% of Pew respondents were Caucasian. All stressed the importance of family. Almost all described living a devout lifestyle and characterized their theology as Christian. The amount of formal education and religious orthodoxy showed a positive correlation. (Note: Utah has several so-called universities, most with LDS institutes of higher learning attached, as well as LDS-owned Brigham Young University. However, historically at least, Utah high school graduates are more likely to enroll in a local/Utah university than be accepted or even seek acceptance out of state. Yet it is only the University of Utah that has any world standing, according to the Academic Ranking of World Universities).

More than 90% of Pew respondents accept the Book of Mormon as the literal word of God, written by ancient American prophets and Joseph Smith as a prophet. But even in this study, 23% of the one-thousand plus queried could not say that they believed “whole-heartedly” in all the teachings of the LDS Church. [Emphasis added] Other conclusions addressed LDS income levels, both inside and outside Utah vis a vis the remainder of the U.S.

ARIS/Trinity was able to reach a much broader database than Pew by a huge factor. As such, the margin of error is much less than in the Pew Study. At the beginning of 1990 the LDS Church claimed 1.3 million Utah Mormons or 77.2% of the people in the state of Utah. At the end of 2008 official LDS numbers posted 1.8 million members in Utah or 68% of Utah's population. During this 18 year time period the church also claims to have increased its numbers from 4,175,000 to 5,974,041 and likewise its percentage of the U.S. Population from 1.7% to 2%.

Trinity's statistics tell a much different story. According to the Trinity Study, the U.S. Mormons in 1990 numbered 2,487,000 and 3,158,000 in 2008, making the percentage of adult LDS a constant at just 1.4% of the population. Children were not queried; however, the authors explain that this would not significantly alter their findings because “...research shows that Mormon birthrates are declining...” Trinity also reports that the discrepancy can be accounted for by “...large numbers of ex-Mormons who continue to be counted by the organization.” According to this report, though Mormonism appears to attract a large number of converts, it also has nearly as large a number of apostates (just as many 'out-switchers' as 'in-switchers') and thus the over-all per capita number of LDS is not gaining, but is “treading water.” Additionally, LDS “market share” in Utah fell from 69% in 1990 to 57% in 2008.

Other interesting conclusions drawn by ARIS/Trinity are that Utah Mormons are the only U. S. religious group that maintain a majority of the population in any state. Apostasy rates are rising among young men in Utah, leading to a gender imbalance with women outnumbering men 3 to 2. Utah Mormons have a statistically larger household than Mormons elsewhere by .5 persons and Mormons are twice as likely to be Republicans as are other Americans. Like the Pew study, this study also confirmed a rising income rate among Utah Mormons.

These studies raise questions for me. Does the LDS Church count baptisms for the dead in their annual statistics of new converts? If one looks at the LDS website the number of baptism per year (excluding children born of record) are counted given. Could this number which is added to the number of children born into LDS families include those who Mormons have baptized posthumously. Over the past decade there have been articles of LDS missionaries visiting graveyards, for example, East Granby Center Cemetery in Connecticut, to record data from tombstones. For years it has been well-known that Church representatives have researched obituaries, real estate, tax and other public records to forward on to the Church genealogy department. A Holocaust victims group’s disapproval of Mormons baptizing Jews was well-publicized. More recently, an April 5, 2008 letter from the Vatican Congregation for Clergy, asks episcopal conferences to direct all bishops to keep the LDS from microfilming and digitizing information contained in parish registers“...so as not to cooperate with the erroneous practices of the Church of Jesus Christ of Latter-day Saints."

Could this be the true source of Mormonism's “growing numbers and the reason for the discrepancy between the Trinity/ARIS numbers and those claimed in annual general conferences by the LDS Church?” This is pure speculation on my part, but it would be interesting to know the answers to whether baptisms for the dead are included in the LDS annual statistical increase of new convert baptisms, which historically have not included children of record born in that year.

The Trinity survey claims that social scientists are more apt to rely on self-reported numbers than other ways of verifying numbers which could be affected by public relations interests. One thing is clear: Mormons are told and taught that they are more numerous and widespread than good research reveals. Trinity further posits: “Organizations, however, sometimes have an incentive for overstating their membership.” Could it be that the LDS Church has such an incentive?