My favorite big family memes

It’s difficult to study large families in the US because most surveys have what is called “right-censored data,” where they cap it at, say 4 and above, so we can’t look specifically at, say, eight or above families. However, I found one dataset, the Cooperative Election Study, that does actually have single-child number of children under 18 values. So I did some quick numbers to see how common the proverbial big Catholic/Latter-day Saint/Orthodox Jewish/Evangelical family is. 

Of course, where exactly you put the “big family” line is always to some extent arbitrary. As longtime readers know I have eight children, and with every additional child we had the same moment of “three kids is a lot, but four kids is a looot,” and then “four kids is a lot, but five kids is really a lot,” and so on. It’s a Sorites Paradox situation. So I’m putting the “big family” line here somewhat arbitrarily at 6 children plus under the age of 18. As seen below, while about 1-2% of Protestants and Catholics have big families, it’s 6% of Latter-day Saints and 3% of Muslims. So we literally have about 3-6x the big, young families of the mainstream US traditions. For what it’s worth, I only know of four other people in my (Millennial) generation that have our same family size of eight or more—and they all work for the Church or Church-affiliated schools. 

% “Big Family”

Protestant	1.6
Roman Catholic	1.3
Mormon	6.2
Eastern or Greek Orthodox	0.8
Jewish	1.2
Muslim	2.8
Buddhist	0.2
Hindu	0
Atheist	0.4
Agnostic	0.5
Nothing in particular	1.5
Something else	2.2

What this means is that, our relatively small numbers notwithstanding, we are actually a relatively large proportion of the big families in the US, about 1 in 20 big, young families are Latter-day Saints, while one in three are still Protestants, and about one in five or so are Catholics.   

Religious Affiliation of 6+ Families in the US (%)

Protestant	36.6
Roman Catholic	17.4
Mormon	4.6
Eastern or Greek Orthodox	0.3
Jewish	2.3
Muslim	1
Buddhist	0.1
Hindu	0
Atheist	2.3
Agnostic	2.7
Nothing in particular	20.6
Something else	12.1

Unfortunately, the Catholic option doesn’t split it out for, say, Traditional Latin Mass versus Novus Ordo Catholics, but they do have Jewish religious tradition, so breaking that down even more we see that about 7% of Orthodox have large, young families, while for all other Jewish traditions it’s 1-2%. Also, it’s worth noting that this is probably not including a lot of the even larger families of the Haredi community, since they don’t like taking surveys, so it’s probably around 7% of Modern Orthodox. However, there are only 8 Orthodox with large families in the sample, so at this point we’re straining our ability to derive too much from the numbers involved. 

When we split it out by religiosity, about 2% of people who say that religion is “very important” have large families, while for everybody else it’s about 1%. This means that about half of all large families come from people who say religion is “very important” in their lives, while 1/10 say “not too important,” and 14% say “not at all important.” 

So the secular big family, while a minority, is a significant minority. This actually tracks my own non-representative experience. I am part of several large-family Facebook groups that share tips on how to manage having one child in the emergency room while your house is leaking while your car is breaking down while…. (one calls to mind the Jim Gaffigan quote, “having a large family is like you’re drowning…and somebody hands you a baby.”) And while there are definitely a lot of religious people, there actually are also some not particularly religious women (and it is usually women, I’m one of the very few men in those groups) that have just decided to have a large family. 

Methodological Minutiae

As far as I can tell the CES only has the “children under 18” number, so we are not including all of the empty nesters who had large families at home. This gives us 701 “big, young families” in 2024. At these relatively small cell sizes I’m ignoring weights, plus it’s worth noting that for really small percentages like these some of the people selecting that they have a lot of children are people who aren’t filling out the survey right (see base rate fallacy), so I suspect that these numbers are somewhat inflated, but they should suffice for comparative purposes.