The first step to take when considering whether or not there is a gender imbalance in American Christian churches is to look for reliable evidence. I think it's safe to say that, yes, there are more women attending church than men, based on personal observation, anecdotal evidence, and statistical data. How many more, and whether or not it is a problem, are the pertinent questions to answer.
David Murrow, author of Why Men Hate Going to Church, can probably be credited with bringing attention to the imbalance and feeding the illusion that the church is feminized. He presents the difference as 39% men, 61% women, taken from the U. S. Congregational Life Survey. This is a 22% gap. Yet on p. 54 of his book, he defines an imbalanced ("gender-gapped") church as one with a gap of only 12% or more. Why? Why is this the magic number, other than "to conform to the findings of the National Congregations Study"? Is an imbalance of 44% to 56% significant enough to qualify as evidence of the church's feminization? It would say it depends upon who (which demographic) makes up that imbalance and why. (At my own church, an old, traditional church, the imbalance is 9-10%. I wrote about this here; the bulk of that imbalance is elderly widows.)
According to the Pew Study, the average imbalance in church affiliation is 6% - 47% men to 53% women. The difference between Evangelical and Mainline Protestant churches is two percentage points - Mainlines slightly higher in number of women - 46% to 54%. Catholics and Mainlines are equal. The greatest gender disparity among American Christian churches is found in "Historically black Protestant" churches and Jehovah's Witnesses, where there are roughly 2 men to every 3 women (40% to 60%). Mormon ratios are slightly higher on the female side than Protestant and Catholic churches as well - 44% men to 56% women.
I realize that these statistics do not represent church attendance, yet it is significant that a person at least identifies with a particular religion. Is this evidence that men are affiliated with yet not attending church? Most likely. Does this mean that more women attend church than men? Yes. Does this mean that the church is feminized? Again, it depends upon how you define "feminized," but still I would say, "not necessarily."
According to the same study, Hindus show 39% women to 61% men - 3 males to every 2 females. Muslims, Buddhists, and "Other faiths" are roughly 54% men to 46% women - roughly the reverse of American Christians. "Unaffiliated" are 59% men, 41% women. Again, roughly 3 men to every 2 women, same as Hindus. Here we see evidence that significantly more men than women choose a non-Christian faith, or choose to be unaffiliated. Does this mean that men are less religious than women? Well, It depends on how you define "religious." Are they less Christian than women? I'll let you decide. Is Christianity more feminized...or are females more "religious," even if only in sentiment or cultural affiliation?
According to the government's website, here are U. S. Population stats:
Population:
303,824,646 (July 2008 est.)
Age structure:
0-14 years: 20.1% (male 31,257,108/female 29,889,645)
15-64 years: 67.1% (male 101,825,901/female 102,161,823)
65 years and over: 12.7% (male 16,263,255/female 22,426,914) (2008 est.)
Sex ratio:
at birth: 1.05 male(s)/female
under 15 years: 1.05 male(s)/female
15-64 years: 1 male(s)/female
65 years and over: 0.73 male(s)/female
total population: 0.97 male(s)/female (2008 est.)
Wikipedia gives similar statistics (US demographics):
Age structure: (2007 est.)
* 0-19 years: 27.4% (male 42,667,761; female 40,328,895)
* 20-64 years: 60.1% (male 89,881,041; female 90,813,578)
* 65 years and over: 12.6% (male 15,858,477; female 21,991,195)
Human sex ratios: (2007 est.)
* at birth: 1.05 males/female
* under 15 years: 1.05 males/female
* 15-64 years: 1 male/female
* 65 years and over: 0.72 male/female
* total population: 0.97 male/female
So we see that in the general U. S. population, there are slightly more females than males, even though in the category of youth, there is a slightly higher percentage of males. The reason for the overall higher percentage of females is the significantly higher number of women over age 65. This is also what we see in our churches; mine is a prime example.
The USCLS claims that this is not true. It claims that the 39%-61% difference is present in every age category, "so the fact that women live longer than men does not explain the gender difference in religious participation." But they cite no data nor source. Murrow himself notes that older churches have a greater gender imbalance, explaining that this is because they are declining -- "hemorrhaging members." Perhaps, but where are these members going? Likely to younger, non-denominational churches, which show by far the least gender gap. Who's left, then? In large part, the elderly women.
(Murrow's explanation: "...as a congregation ages, it begins to value feminine gifts such as nurturing, stability, and close-knit community...Women stay loyal because of the relationships they've developed, but the less relational men fall away." Um, perhaps the men die off, and the women stay because they're too old to go anywhere else!)
added: I need to clarify that I'm not suggesting that elderly women account for the entire gender gap in the church. But I think it's likely they account for a good part of it in many older churches.
A word on statistics:
In "Statistical Illusion" at ChristianityToday online, Bob Smietana writes, among other things, that
... there's no official, exhaustive list of U.S. congregations, and...not all congregations track attendance.
David T. Olson, author of The American Church in Crisis, gives further details and explains his own research methodology at his website.
The spiritual health of churches is multifaceted, and is obviously much more complex than an attendance trend can portray.
Olson's contention is not with gender disparity, but with the fact that church attendance is not keeping up with population growth.
(For reviews of Olson's book see Brad Boydston and Blake Merwin.)
