Why is
Some unintended conclusions
from the apologetic LDS press
Introduction
The last decade has seen an
increase in the frequency and acceptance of statistical studies claiming a positive
correlation between religious activity and life expectancy as well as other
measures of good health. A recent
contribution to this literature is an article by Ray M. Merrill, published in Demographic Research, Vol. 10, article
3,
Merrill’s paper presents
statistical data showing that
Merrill attributes this difference in life expectancy to a
comparatively
high level of church activity, with religiously active people often displaying
better physical health, better social support, and healthier lifestyle
behaviors, each of which promote longer life. Religious activity may also have
an independent protective effect against mortality.
Even if we assume that Dr. Merrill’s statistical data is correct, and follows proper protocol, his conclusion is misleading and inconsistent with other data. This is true regarding his claims of superior LDS health in general, as well as his broad claims with respect to correlation between life expectancy and religious activity.
The principal flaw in Merrill’s
analysis is his failure to compare the life expectancy of non-smoking Utah
Mormons with that of Americans and Canadians in general. That comparison shows that non-smoking
Mormons have a life expectancy that’s virtually identical with non-smokers in
general. The surprise is not that
non-smoking Mormons have a surprisingly long life expectancy (which they
don’t), but rather that non-smoking non-Mormon Utahans have comparatively short
life expectancies. While Merrill ponders
the question “why do Mormons live so long?” the real issue raised by the results
of his study is “What makes
This short paper discusses
Merrill’s claims in detail, taking a more critical look at what’s actually
occurring in
Caution about LDS-sponsored statistics
The science of statistics often involves inductively drawing conclusions about a population based on metrics taken from a sample (sub group) of the population. The mathematics are often protracted, and the proper protocols beyond the reach of non-specialists, making it easy for charlatans, con artists, and apologists to misuse the science as a tool for advocating causes that span the spectrum from religion to politics.
There
is both precedent and motive for the
And now, my son … I would that ye should keep the commandments of God, that ye may prosper in the land …Therefore, as they were unfaithful they did not prosper nor progress in their journey, but were driven back, and incurred the displeasure of God upon them; and therefore they were smitten with famine and sore afflictions, to stir them up in remembrance of their duty. [Mosiah 1: 7,17]
These verses from Mosiah are representative of the way the Book of Mormon repeats the message that righteous living results in prosperity, while failure to live righteously results in god’s condemnation, including natural disasters and poverty. Among Mormons these verses incite religious determination to see the church and its members viewed as blessed and prosperous; anything else calls into question, in their minds, their personal righteousness and the truthfulness of their religion.
One of the best examples of abuse of statistics by Mormon apologists is a study by Richard T. Wootton. FAIR (an apologetic arm of the Church of Jesus Christ of Latter-day Saints) claims, based on Wootton’s studies, that “For 80 years every study has shown that in relation to population Utah was number one in spawning scientists…”[2] Wootton’s analysis, fatally flawed by unscientific sampling techniques, results in conclusions that are thoroughly discredited by statistics from the National Science Foundation.[3]
Few things spell “prosper” within Mormon culture more than the acquisition of converts. Mormons believe their church is destined to “roll forth” and “fill the whole earth.” It’s not surprising, therefore, to find examples of exaggeration in LDS statistics regarding membership in the Church.
As an
active LDS apologist, Merrill is certainly subject to significant social and personal
pressure to produce statistics that paint the LDS community in a “prosperous”
light. It should come as no surprise, therefore, that some of his earlier
statistical studies regarding Mormonism are highly suspect. For example, according to Merrill:
About 70% of the 2.3 million population in the state [of
Official LDS membership
statistics claim an even higher percentage.
According to the 2004 Church Almanac,
Viewed another
way, the numbers from Merrill and the LDS almanac are approximately equivalent
to the Mormon Church claiming, as members, all the Utahans who admit to being
Mormon as well as all those claiming to be members of the Roman Catholic
Church, Baptists, Methodists, Lutherans, Presbyterians, Anglican, and Muslims.[5] In
light of the poll results from USA Today, it’s hard to imagine greater hubris
than the
Separately, Merrill also claims that
Of the 70% of the state’s population who is LDS in
The
"Attendance at sacrament meeting varies
substantially. Asia and Latin America have weekly attendance rates of about 25
percent, Europe averages about 35 percent" (Encyclopedia of Mormonism,
edited by Daniel H. Ludlow, 1992, 4:1527.)
"Canada, the South Pacific, and the
Again, we
see a tendency for Merrill to inflate favorable statistical measures relating
to the
The
Church of Jesus Christ of Latter-day Saints often encourages and helps to
propagate the notion that it is fastest-growing religion.[6] The claim is untenable, even when using
LDS-supplied figures for Mormon membership. For example, table 1 compares the growth of
the Seventh Day Adventists with that of the
|
|
LDS[7] |
Seventh Day Adventists[8] |
|
|
11,068,861 |
11,687,229 |
|
Baptisms |
355,423[9] |
931,335 |
|
Deaths |
Not reported |
44,298 |
|
Dropped |
Not reported |
153,121 |
|
Missing |
Not reported |
91,823 |
|
Other |
Not reported |
6488 |
|
|
|
|
Table 1. Comparison of membership growth for the LDS and
Seventh Day Adventists (for year 2000).
LDS claims
about growth rate are even weaker, however, because inconsistencies in their
reporting clearly show that the
For
example, using US Census Bureau figures and the USA Today poll, the number of
Americans claiming to belong to the
There
have been instances in which the
Numerically,
this is equivalent to claiming that no members of the
In
examining Merrill’s claims regarding Utah Mormons and Mormon health, we need to
proceed cautiously, realizing that there are many ways in which statistical
studies can be corrupted by pre-existing bias, and understanding that there is
a clear bias on the part of energetic Mormon apologists, such as Merrill, to
paint Mormonism as spiritually superior to other life styles. Even so, let’s be generous and accept
Merrill’s statistical study about LDS life expectancy, although we should
remain mindful of the probability it is skewed to shine an unrealistically
favorable light on his church.
Life expectancy of non-smoking
Utahans
Life expectancy is the average number of years an individual is expected to life if current mortality rates continue to apply. It depends, obviously, on a person’s age (older people can be expected, on average, to live fewer additional years than younger ones), and is commonly stated at birth. Life expectancy is thus a statistical measure with a non-ambiguous mathematical definition. Consider a particular group, after all members of the group have lived their lives and died, the average lifetime equals the life expectancy (at birth). Such measurements are impractical in reality (you have to wait for everyone to die, so you only know the answer in hind sight) so life expectancy is usually extrapolated through calculations using current mortality rates.
Because life expectancy is a
statistical measure of a group of people, it depends heavily on the group in
consideration. For example, soldiers engaged
the war in
Over the last several decades, modern science has produced a wealth of knowledge about the health hazards of cigarette smoking, and we now know that smoking, on average, reduces a person’s expected lifetime by many years. Consequently, within any given population consisting of smokers and non-smokers (all other things being equal), the life expectancy of the non-smoking group will generally exceed the life expectancy of the group that smokes.
Dr. Merrill’s data shows that Utahns that don’t smoke, and are members of the Church of Jesus Christ of Latter-day Saints, have a longer life expectancy than Utahns who don’t smoke and don’t belong to the church.[11] According to Merrill’s data, for example, a non-smoking male LDS member who is 20 years old can expect to live another 59.0 years, but a non-smoking male non-LDS Utahan can expect to live only an additional 54.0 years. So otherwise equivalent non-LDS Utahans, in this case, have a life expectancy that’s 5 years shorter than their LDS counterparts. Merrill then proceeds with the unspoken assumption that these two groups are statistically the same in all respects, except for their membership in the church, which implies that church membership is the causal reason for the difference.
Merrill finds this difference statistically significant, and it is. However, there are two possible scenarios for this difference. One of these scenarios is favorable to Mormonism and the other isn’t. As a devout Mormon apologist, Merrill obviously favors the LDS-pleasing explanation, while failing to properly consider the alternative, even though it is a better explanation for the data.
The explanation most flattering
to the
It’s not difficult to understand why Merrill focuses on this explanation. The Mormon Church teaches that it is the “only true and living church” on earth. Furthermore, the Book of Mormon, keystone of the Church of Jesus Christ of Latter-day Saints, teaches that obedience to the church brings blessings. Indeed, this is a dominant theme in the Book of Mormon, which proclaims that those who follow the teachings will prosper in the land, while those who don’t will be cut off. As an active member of the church, Merrill’s desire to find statistical evidence supporting the notion of extra blessings for members of the church is thus understandable and expected.
Though understandable, Merrill’s
failure to consider other explanations represents an intellectual lapse,
especially since the data more firmly support the second explanation which
posits that the life expectancy of non-smoking Mormons is essentially the norm
for non-smoking people everywhere, and that non-Mormons living in
This second explanation for the
statistical data is far less flattering to Mormonism, and so it’s not
surprising that Merrill largely ignores it.
However, statistical data from the
Consider the data in table 2,
which is reproduced in graphical form in figures 1 and 2. This table reproduces some of the data from
table 4 in Merrill’s article, but it also includes the life expectancies for
people living in the
Columns 2 and 6 were taken from Merrill’s article, and columns 3, 4, and 5 were taken from actuary tables created by insurance companies. Life expectancy is a critical piece of information for insurance companies, and they spend a great deal of money collecting this data and ensuring that it is correct.
Upon examining table 1, the most
immediate observation is the nearly perfect agreement between columns 2, 3, 4,
and 5. This is true for both men and
women, and it shows that non-Smoking Utah Mormons have a life expectancy that
is almost exactly inline with the life expectancy for non-smoking Americans and
Canadians in general. In other words,
there is nothing remarkable about the life expectancy of non-smoking Utah
Mormons. It’s exactly as expected, and
in statistical agreement with the life expectancy of non-smokers across
|
Life expectancy for non-smoking males |
|||||
|
Age |
|
Future Financial[13] |
Hatton[14] |
|
|
|
20 |
59.0 |
59.39 |
59.0 |
58.7 |
54.0 |
|
40 |
40.2 |
40.16 |
40.0 |
39.8 |
36.6 |
|
60 |
22.3 |
21.86 |
22.5 |
21.5 |
19.9 |
|
80 |
8.5 |
8.19 |
NA |
7.7 |
7.5 |
|
Life expectancy for non-smoking females |
|||||
|
Age |
|
Future Financial |
Hatton |
|
|
|
20 |
63.3 |
63.71 |
63.3 |
63.2 |
59.2 |
|
40 |
43.9 |
44.14 |
43.8 |
43.7 |
40.3 |
|
60 |
25.4 |
25.49 |
25.5 |
24.9 |
22.4 |
|
80 |
10.5 |
10.32 |
NA |
9.6 |
7.7 |
Table 2. Life
expectancy for men (top) and women (below) non smokers. The data for Utah LDS and non-LDS came from
Merrill’s paper and the other three columns represent life insurance actuarial
data. Note the similarity between the
life expectancy for LDS Utahans and those for the general population in the
The real anomaly here, and the
one largely ignored by Merrill, is why the life expectancy for non-Mormon,
non-smoking Utahns is so low. While a
non-smoker in
Figure 1. Graph of life
expectancy for non-smoking men (data taken from table 1). The most striking thing about this graph is
the low life expectancy for non-Mormon Utahan’s compared to non-smoking men in
general. Data from the Hatton, Future
Financial, and Canadian actuaries clearly shows that the life expectancy for
non-smoking Mormon Utahans is inline with the non-smoker life expectancy in
American and
It’s possible that
Based on Church
records for 15,500 California Mormons during 1968 to 1975 and for 55,000 Utah
Mormons during 1970 and 1975, the … life expectancy for active Mormon men at
age 35 is about 44 years, over 7 years greater than for U.S. white males.
Note the life expectancy for the
California Mormons, and compare this with the data reported by Merrill (see
table 1). Merrill doesn’t give the life
expectancy for 35-year-old Mormons, but we can interpolate his data and arrive
at a figure of about 45.2 years. Compare
this with the life expectancy data from the
Figure 2. These data mirror those in figure 1, but show
the data for women. Note the consistent
trend – showing again that
Reading again from the abstract
in the National Library of Medicine, it says “The remaining life expectancy for
active Mormon men at age 35 is about 44 years, over 7 years greater than for US
white males.” Note that this conclusion
does not discriminate between US white males who smoke and those who
don’t. An article appearing
The Continuous Mortality Investigation (CMI), part of the Actuarial
Profession, today published new mortality tables which show – for the first
time – that the difference in mortality between smokers and non-smokers is
greater than the gap between men and women at almost every age.
This is in good general agreement with the claims for extended longevity of Mormons who don’t smoke, relative to the population as a whole. So, once again, we see than the life expectancy of non-smoking Mormons is inline with the life expectancy of non-smokers in general, and that LDS life expectancy presents no anomaly requiring an explanation.
Merrill, however, seems determined that the life expectancy of non-smoking Mormons is somehow remarkable and needs an explanation. Not surprisingly, he suggests that the reason is somehow to be found in extra blessings, of some sort, arising from membership in his church. He asserts that “differences in life expectancy of 5.8 years for males and 4.6 years for females were not directly explained by deaths attributed to tobacco use” and then proceeds to postulate that Mormon health is augmented by “factors associated with religious activity, such as better physical health, better social support, healthier lifestyle behaviors, and possibly an independent protective effect against mortality from religious activity itself.” Indeed, so forceful is Merrill with this conclusion that he repeats it in his paper no less than three times.
Let’s examine, briefly, these claims:
1) Better physical health and Healthier
lifestyle behaviors
According to Merrill, the LDS dietary code, known as the Word of Wisdom may be responsible for the difference in life expectancy between non-smoking Utah Mormons and non-smoking non-Mormon Utahns. This is an unlikely cause for differentiation, however, as the Word of Wisdom has little dietary value outside of its prohibition against smoking (a prohibition that was not uncommon in Joseph Smith’s day, when the Word of Wisdom was written). The Word of Wisdom is generally vague and fails to mention important food groups all together. For example, the Word of Wisdom never mentions dairy products and it only weakly alludes to vegetables, never mentioning them by name. It proposes moderate use of meat (to be used only when it’s cold) and says that grains, wheat in particular, are the staff of life. It also makes no mention of exercise or physical activity as a prerequisite for good health (though it does mention it as a consequence of following the dietary rules).
The original Word of Wisdom allowed moderate use of wine (home made) but modern Mormons eschew wine all together. In doing so they undoubtedly benefit from less inclination toward overindulgence and the many associated illnesses that accompany alcoholism – but they may also deny themselves the clear health benefits from moderate use (which, ironically, the original Word of Wisdom prescribed).[17] The same can be said for tea, which is prohibited under the Word of Wisdom, but has been shown to have high correlation with several aspects of good health.[18]
Obesity is one of
|
State |
Percent of Obese Adults |
Rank |
|
|
28.4 |
1 |
|
|
28.1 |
2 |
|
|
27.7 |
3 |
|
|
26.0 |
4 |
|
|
25.6 |
5 |
|
|
25.2 |
6 |
|
|
25.2 |
|
|
|
25.2 |
|
|
|
25.0 |
9 |
|
|
24.9 |
10 |
|
|
24.8 |
11 |
|
|
24.6 |
12 |
|
|
24.5 |
13 |
|
|
24.4 |
14 |
|
|
24.0 |
15 |
|
|
24.0 |
|
|
|
23.9 |
17 |
|
|
23.9 |
|
|
|
23.8 |
19 |
|
|
23.7 |
20 |
|
|
23.6 |
21 |
|
|
23.5 |
22 |
|
|
23.2 |
23 |
|
|
23.2 |
|
|
|
23.0 |
25 |
|
|
22.9 |
26 |
|
|
22.6 |
27 |
|
|
21.9 |
28 |
|
|
21.8 |
29 |
|
|
21.7 |
30 |
|
|
21.7 |
|
|
|
21.5 |
32 |
|
|
21.2 |
33 |
|
|
20.9 |
34 |
|
|
20.9 |
|
|
|
20.8 |
36 |
|
|
20.3 |
37 |
|
|
20.2 |
38 |
|
|
20.2 |
|
|
|
20.1 |
40 |
|
|
20.1 |
|
|
|
20.1 |
|
|
|
19.9 |
43 |
|
|
19.9 |
|
|
|
19.6 |
45 |
|
|
19.0 |
46 |
|
|
18.8 |
47 |
|
|
18.4 |
48 |
|
|
16.8 |
49 |
|
|
16.4 |
50 |
|
|
16.0 |
51 |
|
Total average |
22.8 |
|
Table 3.
State-by-state comparison of adult obesity rates (2004).
A more recent study by Merrill
paints an even more interesting picture.
This study was based on data from figures obtained in 1996, 2001, and
2003-2004 by the Utah Health Status Survey.
The results show that
Depression is another metric for
gauging general health. Merrill asserts
that Mormons have better mental health than their non-Mormon counterparts, yet
Doctors here have for years talked about the widespread use of
antidepressants in the state. But there was no hard evidence until a national
study that tracked drug prescriptions came to an unexpected conclusion: Antidepressant drugs are prescribed in Utah
more often than in any other state, at a rate nearly twice the national
average.[22]
Since
2) Better social support
Mormons are well known for taking care of each other, and they actively promote that public image. While the anecdotal stories may be inflated, it’s probably true that, overall, Mormons benefit from strong social support. Social support can be (and is) found among many other groups, too, and it’s not clear that Mormonism provides vastly superior social support, compared to what many Americans find in other churches and organizations.
While Mormon society can provide support to the faithful, it can also be the source of a great deal of stress for ex-Mormons and non-Mormons. One of the ways this can happen is through acts of discrimination (both deliberate and incidental). Consider the following story:[23]
William Carlton May thought a promotion to chief executive officer of
Totally Awesome Computers seemed logical. After all, he says, he had been
running the company's day-to-day operations for three years.
But owner Dell Schanze kept saying the then-vice
president would be perfect for the job, except he wasn't Mormon, May contends. Schanze allegedly told May, "You cannot make decisions
the way I want you to because you are not LDS."
That allegation could be part of a weeklong trial slated to begin Monday on
May's wrongful-firing lawsuit against Superdell's
Computers Inc., a Sandy-based company that does business as Totally Awesome
Computers (TAC).
May's attorneys plan to present evidence intended to show Schanze
was a volatile boss who discriminated against minorities and people who were
not members of The Church of Jesus Christ of Latter-day Saints.
Anecdotal stories of similar acts
of discrete discrimination abound. All
too often, in
LDS opposition to atheists,[24]
women’s rights,[25] and
gays[26]
is well known. The
Additionally, the Church has
fought for legislation giving religious organizations the right to discriminate
based on religious thought and action even when the job at hand (a janitorial
position, for example) has no religious/spiritual component.[29] And, it was only relatively recently that the
A further indicator of stress among non-Mormons in a Mormon-dominated state is offered by yet another LDS apologist:[33]
For example, the suicide rate among less active LDS church members aged
25 to 29 was seven times higher than among their active church peers.
Nationally, the suicide rate among 20- to 34-year old males was 2.5 to 3 times
higher than among active LDS church members of the same age. Suicide risk was
also 3 to 6 times higher among nonmembers in comparison to active members of
the LDS church.
According to
At one suicide every 28 hours,
So, Utah’s suicide rate for young
men is the highest in the nation, yet LDS apologists claim the rate among
active LDS men is only about 1/3rd the national average. Assuming
the apologists are correct, then
As with the attenuated life expectancy of non-smoking non-Mormons from Utah, this statistical data offers further support for the hypothesis that the LDS-dominated culture of Utah is dangerous for the health of non-Mormons, and young non-Mormon men in particular. The data isn’t explained by longer life among Mormons due to “better social support.” On the contrary, the data suggest that there are significant risks to the health of non-Mormons due to social/political agendas pursued by the church and its members.
3) Independent protective effect against mortality
from religious activity itself
According to Merrill, Mormons “possibly” benefit from “an independent protective effect against mortality from religious activity itself” and he cites work done by Wilkins 2003; Ross and Mirowsky 2002; Klein et al. 2002; McCullough et al. 2000; Ellison and Levin 1998.
These studies showing a presumed correlation between religious involvement and life expectancy are coming under increasing scrutiny. For example, in an article titled The Epidemiology of Religion (For "Reality Check" in Skeptical Briefs, Vol 11, no. 3, September, 2001) Vic Stenger writes:[35]
Epidemiological
studies, in general, are notoriously difficult to interpret reliably because of
so-called "confounding factors." A given study may indicate a
correlation between an illness and some factor, but this does not necessarily
prove that the factor is the cause (or cure) of the illness. In an amusing
example given by psychiatrist Richard P. Sloan, a study might find that lung
cancer is more prevalent among people who carry around matches in their
pockets. It would not follow that matches cause lung cancer.
Vic Stenger points out that
A review of the literature from a more critical perspective has been provided by R. P. Sloan, E. Bagiella, and T. Powell in the British medical journal Lancet. They concluded that the published work lacks consistency and is not based on sufficiently large samples of data. Linda Gundersen has also reviewed the subject in Annals of Internal Medicine and finds that the conclusions of several of the studies are doubtful.
Stenger points out that most of
the epidemiological studies regarding the purported health effects of religion
have been based on church attendance. A major problem with this metric is that
people (and institutions like the Mormon Church, as we’ve already seen) tend to
embellish their religious involvement.
This serves to seriously skew the research and lends considerable
uncertainty to the studies made. In an
excellent article titled Church, Lies, and Polling Data, Andrew Walsh
says:[36]
In 1993, Chaves, Kirk Hadaway
of the
Scholars whose work has relied on the polling data
were particularly unhappy with the challenge. The Rev. Andrew Greeley-Catholic
priest, novelist, columnist, and sociologist-was incensed by the study,
denouncing it in a Religion News Service column as "a sloppy piece of
work" that naively extrapolated regional findings into national ones. Hadaway, Chaves, and Marler
responded that they weren’t attacking the validity of poll data, merely
pointing out its sharp limitations. "Americans misreport how often they
vote, how much they give to charity, and how frequently they use illegal drugs.
People are not entirely accurate in their self-reports about other areas as
well," Hadaway wrote in the magazine Christian
Century. "Males exaggerate their number of sexual partners, university
workers are not very honest about reporting how many photocopies they make.
Actual attendance at museums, symphonies and operas does not match survey
results. We should not expect religious behavior to be immune to such
misreporting.”
After the first wave of scholarly discussion, Hadaway et al. returned to
We’ve already seen evidence that the
In addition to the problem of self reporting religious activity, Stenger points out that:
none of these reports have the statistical
significance that merit their being regarded as definitive. Furthermore, the studies have been unable to
adequately establish a clean causal connection between religion and health. As
Sloan and his coauthors explain, when confounding variables are considered,
these can explain most if not all of the effects observed. For example, a 1972
study that is often cited as evidence for a positive association between church
attendance and health was later found, by its own authors, to be due to failure
to control for people with reduced mobility. People in poorer health were
simply less likely to go to church.
Additional points raised by Stenger include:
· If there is a correlation between positive thinking and health, should we conclude that positive thinking causes good health or that good health causes positive thinking?
· Far fewer prison inmates are atheists in proportion to their representation in the general population, and far more are Christians. Should we conclude that Christians are more likely to be criminals, or is it possible that some criminals that claim a religious conversion do so because Parole boards tend to look favorably on convicts that become “born again?”
· Filtering by the media gives a false sense of what the evidence says, on balance. For example, studies exist which indicate a negative influence of religion on health, although you have to dig to find them since they are rarely reported in the media or in uncritical reviews. For example, Koenig and his collaborators selected only favorable studies in their review and ignored even some of their own work that gave conclusions opposite to the ones they preferred to see.
The significance of this last point should not be underestimated. News reporters have a vested interest in telling people news that sells, and most people want to believe that their religious involvement gives them some sort of benefit. For example, in the March 2002 issue of Skeptical Inquirer, Kevin Christopher reports:
A tiny
Associated Press story, dated December 12, 2001, and written with barely enough
copy to fill two column-inches of space in any newspaper, announced the findings
of the Mayo Clinic's just-published clinical study on the healing powers of
intercessory prayer. In a nutshell, the Mayo researchers could not find any
evidence that such prayer affected the health of the study subjects. In
fairness to AP, a story about what scientists say they did not find doesn't
exactly burst into flames on any editor's desk (and, at least, the AP reported
something). AP reported given the heavy ink poured forth at the media's altar
of Medicine and Spirituality, one would have thought such findings--especially
Mayo Clinic findings--might merit more attention from health and religion
editors.
The study in
question, conducted by Dr. Jennifer Aviles and colleagues, and led by Dr.
Stephen L. Kopecky, is titled "Intercessory
Prayer and Cardiovascular Disease Progression in a Coronary Care Unit
Population: A Randomized Controlled Trial." Published in the December 2001
issue of the Mayo Clinic Proceedings, Aviles et al. found no significant effect
on medical outcomes after hospitalization in a coronary care unit."
Merrill, of course, makes no mention of such studies.
When comparing two groups and trying to understand the extent to which a single factor is responsible for something like enhanced life expectancy, it’s necessary to use observations in which samples from the two groups are exactly the same, except for the factor under investigation. For example, religious people should not be compared with the general population. Instead, they should be compared with groups that have similar healthy habits, and whose only difference is in their religious practices.
Sometimes it’s not feasible to identify two groups that
are exactly the same, except for a single factor. We can still explore the causes of the factor
if we look at the statistical correlation of that factor with many different
groups. By using may different groups,
the slight differences between them (outside the factor of interest) can be
expected (sometimes) to average out, leaving the factor of interest as a trend
among the data.
Keeping these issues clearly in our mind, let’s consider
what we should find if, as Merrill claims, there is “a robust association
between religious involvement and lower mortality.”
If Merrill’s claim is true, then we should see a strong correlation between life expectancy and populations that have a high degree of religious involvement. To test Merrill’s assertion we can plot the life expectancy for different states and countries as a function of some measure of their religious involvement. Keep in mind, however, the earlier studies that show people exaggerate their religious involvement. One group that we expect would not exaggerate, however, is those who claim no religious affiliation. Thus, we can plot the life expectancy of the 50 states against the percentage of people in each state that claims no religious affiliation.
Fortunately, these data are readily available. A recent poll conducted by USA Today lists the number of non-church goers in each of the 50 continental states. Combining these data with life expectancy data for each state, we should (if Merrill is correct) see a negative correlation between the prevalence of non-believers in a given state and the state’s life expectancy. In fact, however, the correlation is positive, and in dramatic disagreement with Merrill’s assertion.
Figure 3 shows a scatter plot in which the horizontal axis represents the percent of a state’s population that claimed to belong to “no religion” (taken from the USA Today poll) and the vertical axis represents the life expectancy for the given state. Given Merrill’s assertion about a robust religious association resulting in a lower mortality rate, we would expect the correlation in figure 3 to be negative. That is, the more non believers in a given state, the lower the state’s life expectancy. Instead, we see the opposite trend.
Figure 3. Scatter plot (red dots) showing the life expectancy (in years, at birth) for the 48 lower states as a function of the percent of the population that does not profess a religion.[37] The blue line is a least-squares linear fit to the scatter plot, and shows that there is a slight positive correlation between increased life expectancy and the fraction of the population composed of individuals who don’t profess membership in any religion.
The straight line in figure 3 is a linear regression of the scatter plot. That is, it represents the best (statistical) linear approximation to the data. We can test the statistical significance of the linear regression in figure 3 by calculating the uncertainty of the slope (see appendix 2). This calculation shows that the slope equals 0.1322 and the standard deviation of the slope is 0.0446. This means it’s highly unlikely that the correlation is negative, and only turned up positive because of a statistical fluke.
We can look at a similar plot involving the life expectancy of different nations, plotted against the national rates of non-participation in religion (see figure 4). As with figure 3, life expectancy correlates positively with reduced religious activity. Of course in all these data there are many factors involved. More industrialized countries with greater wealth tend to be less religious, and poorer countries tend to be more religious. These factors should not, however, be necessarily decoupled from the analysis because religion is such a wide-spread phenomenon, and because it affects every aspect of public policy. It is, in a word, woven throughout the social fabric. For example, religious activity could adversely affect mortality if religiously devout and/or fundamentalist people were less liberal and less supportive of national health-care systems.
Again we draw the conclusion that religious activity is negatively correlated with life expectancy. This conclusion is diametrically opposed to Merrill’s assertion that there is “a robust association between religious involvement and lower mortality.” Clearly, if there is any benefit from religious activity at all, it is a very small effect (not robust at all) and unable to compensate for whatever extenuating factors religious apologists might invoke for the positive correlation between lack of religious activity and enhanced life expectancy.[38]
Figure 4. Scatter plot (red dots) showing the life expectancy (in years, at birth) for 52 countries as a function of the percent of the population that does not attend church at least once each week.[39] The blue line is a least-squares linear fit to the scatter plot, and shows that there is a slight positive correlation between increased life expectancy in a country and the fraction of the population composed of individuals who don’t attend church at least once each week. The data in this figure compare well with the data in figure 3, and both indicate a positive correlation between increased life expectancy and decreased involvement in organized religion.
The data plotted in figures 3 and 4 should not be misconstrued as proving a correlation between less religious activity and life expectancy. The data are, however, inconsistent with Merrill’s assertion that there is a strong correlation between increased church activity and increased life expectancy.
Conclusions
Merrill’s analysis showing that
non-smoking Utah Mormons live longer than non-smoking non-Mormon Utahans is
probably correct. His conclusions,
however, are badly flawed. Without
looking at the life expectancy of non-Mormon non-Smokers nation wide, Merrill
assumed that the question needing an answer is “why do Mormons live so
long.” In reality, though, non-smoking
Mormons have a life expectancy that is virtually identical with that of
non-smoking Americans and Canadians in general.
The question needing an answer is actually “why do non-smoking
non-Mormons from
The answer to this question is
wide open, but one obvious area for future study centers on the roll played by
Mormons in creating a stressful environment for their non-Mormon
neighbors. As a theocratic state (in
reality, if not by design)
Ex-Mormons are especially
susceptible to stress caused by the Mormon Church. Stories from ex-Mormons contain a common
theme of exclusion, bigotry, and persecution at the hands of the
Appendix 1
|
State |
|
% LDS -- |
#LDS -- |
% LDS -- self reported |
#LDS -- self reported |
Difference |
|
|
4,530,182 |
1 |
45,302 |
0.6 |
27,181 |
(18,121) |
|
|
5,743,834 |
6 |
344,630 |
6.4 |
367,605 |
22,975 |
|
|
2,752,629 |
0.3 |
8,258 |
0.8 |
22,021 |
13,763 |
|
|
35,893,799 |
1 |
358,938 |
2.2 |
789,664 |
430,726 |
|
|
4,601,403 |
2 |
92,028 |
2.7 |
124,238 |
32,210 |
|
|
3,503,604 |
2 |
70,072 |
0.5 |
17,518 |
(52,554) |
|
|
830,364 |
0.3 |
2,491 |
0.5 |
4,152 |
1,661 |
|
DC |
553,523 |
0.3 |
1,661 |
|
- |
(1,661) |
|
|
17,397,161 |
0.3 |
52,191 |
0.7 |
121,780 |
69,589 |
|
|
8,829,383 |
1 |
88,294 |
0.7 |
61,806 |
(26,488) |
|
|
1,393,262 |
14 |
195,057 |
27.4 |
381,754 |
186,697 |
|
|
12,713,634 |
0.3 |
38,141 |
0.4 |
50,855 |
12,714 |
|
|
6,237,569 |
0.3 |
18,713 |
0.5 |
31,188 |
12,475 |
|
|
2,954,451 |
0.3 |
8,863 |
0.6 |
17,727 |
8,863 |
|
|
2,735,502 |
1 |
27,355 |
1 |
27,355 |
- |
|
|
4,145,922 |
0.3 |
12,438 |
0.6 |
24,876 |
12,438 |
|
|
4,515,770 |
0.3 |
13,547 |
0.5 |
22,579 |
9,032 |
|
|
1,317,253 |
0.3 |
3,952 |
0.7 |
9,221 |
5,269 |
|
|
5,558,058 |
3 |
166,742 |
0.5 |
27,790 |
(138,951) |
|
|
6,416,505 |
0.3 |
19,250 |
0.2 |
12,833 |
(6,417) |
|
|
10,112,620 |
0.3 |
30,338 |
0.4 |
40,450 |
10,113 |
|
|
5,100,958 |
0.3 |
15,303 |
0.5 |
25,505 |
10,202 |
|
|
2,902,966 |
0.3 |
8,709 |
0.6 |
17,418 |
8,709 |
|
|
5,754,618 |
1 |
57,546 |
0.9 |
51,792 |
(5,755) |
|
|
926,865 |
1 |
9,269 |
4.5 |
41,709 |
32,440 |
|
|
1,747,214 |
1 |
17,472 |
1 |
17,472 |
- |
|
|
2,334,771 |
9 |
210,129 |
8 |
186,782 |
(23,348) |
|
|
1,299,500 |
0.3 |
3,899 |
0.6 |
7,797 |
3,899 |
|
|
8,698,879 |
1 |
86,989 |
0.3 |
26,097 |
(60,892) |
|
|
1,903,289 |
3 |
57,099 |
3.3 |
62,809 |
5,710 |
|
|
19,227,088 |
0.3 |
57,681 |
0.3 |
57,681 |
- |
|
|
8,541,221 |
1 |
85,412 |
0.7 |
59,789 |
(25,624) |
|
|
634,366 |
1 |
6,344 |
0.8 |
5,075 |
(1,269) |
|
|
11,459,011 |
1 |
114,590 |
0.4 |
45,836 |
(68,754) |
|
|
3,523,553 |
0.3 |
10,571 |
1 |
35,236 |
24,665 |
|
|
3,594,586 |
4 |
143,783 |
4 |
143,783 |
- |
|
|
12,406,292 |
0.3 |
37,219 |
0.3 |
37,219 |
- |
|
|
1,080,632 |
0.3 |
3,242 |
0.3 |
3,242 |
- |
|
|
4,198,068 |
0.3 |
12,594 |
0.7 |
29,386 |
16,792 |
|
|
770,883 |
0.3 |
2,313 |
1 |
7,709 |
5,396 |
|
|
5,900,962 |
0.3 |
17,703 |
0.6 |
35,406 |
17,703 |
|
|
22,490,022 |
0.3 |
67,470 |
1 |
224,900 |
157,430 |
|
|
2,389,039 |
57 |
1,361,752 |
73.3 |
1,751,166 |
389,413 |
|
|
621,394 |
0.3 |
1,864 |
0.6 |
3,728 |
1,864 |
|
|
7,459,827 |
0.3 |
22,379 |
0.9 |
67,138 |
44,759 |
|
|
6,203,788 |
3 |
186,114 |
4 |
248,152 |
62,038 |
|
|
1,815,354 |
0.3 |
5,446 |
0.7 |
12,707 |
7,261 |
|
|
5,509,026 |
0.3 |
16,527 |
0.4 |
22,036 |
5,509 |
|
|
506,529 |
7 |
35,457 |
11.3 |
57,238 |
21,781 |
|
|
655,435 |
4 |
26,217 |
4 |
26,217 |
- |
|
|
1,262,840 |
4.7 |
59,353 |
4.7 |
59,353 |
- |
|
Total |
293,655,404 |
|
4,338,706 |
|
5,467,398 |
1,128,692 |
Appendix 2
Positive correlation between no religious affiliation and life
expectancy
Appendix 3
Positive correlation between less church activity and life expectancy
