Marriage advice for chart enthusiasts
October 16, 2014 7:33 AM   Subscribe

What makes for a stable marriage?
About a decade ago, the gossip on everyone’s lips was that “1/2 of all marriages in the U.S. end in divorce.” That factoid was later disproven, but it left a lasting impression on the eligible bachelors and bachelorettes of America. In an effort to not become a part of that statistic, I started doing a little research on what makes for a stable marriage in America. [...] What struck me about this study is that it basically laid out what makes for a stable marriage in the U.S. I’ve highlighted 7 of the biggest factors below.
The Atlantic points out:
Part of the study echoes what we already know about marriage: That it's increasingly for rich people—who make a lot and can afford honeymoons.
posted by almostmanda (112 comments total) 40 users marked this as a favorite
 
Based on how much my cousin spent on her wedding this past April she should be divorced by now.

That one is by far the most interesting stat to me.
posted by phunniemee at 7:39 AM on October 16, 2014 [2 favorites]


It's interesting that the influence of wedding cost and wedding size run in opposite direction, since hosting more people at a wedding generally makes it cost more.
posted by Bulgaroktonos at 7:40 AM on October 16, 2014 [11 favorites]


Your income, the cost of your wedding, the cost of your engagement ring, the size of your wedding, and whether or not you went on a honeymoon are so related to each other that it seems ridiculous to treat them as independent factors affecting the probability that you get divorced. The fact that some of these indicators point in different directions makes it even funnier.

I do wonder if the relative likelihoods for each individual factor control for the other factors, or are just treated individually.

"It basically laid out what makes for a stable marriage in the U.S." -- it basically cries out for a "correlation is not causation" response.
posted by leopard at 7:40 AM on October 16, 2014 [15 favorites]


it basically cries out for a "correlation is not causation" response.

Correlation may not be causation, but it's highly correlated with it.

I'd be very interested to see how these statistics evolve over time; I have as sense that North America's collective expectations of what a marriage even is are changing dramatically.
posted by mhoye at 7:45 AM on October 16, 2014 [3 favorites]


So the optimal cost per guest (200 / 0) is NaN. Got it!
posted by grumpybear69 at 7:49 AM on October 16, 2014 [8 favorites]


The bit about how the longer you know each other before you're engaged, the more likely you are to stay married, is kind of a "duh" point for me. All the best marriages I know of began with really long courtships.
posted by EmpressCallipygos at 7:50 AM on October 16, 2014 [3 favorites]


it basically cries out for a "correlation is not causation" response.

This may be true for a single data point, but many correlated points are much more likely to strongly indicate causation.

There is no one single factor that guarantees a long and happy marriage (and of course you can have one of those things without the other!) but there is a constellation of factors that, if they are in place, can make the likelihood much, much greater. That some of these factors are economic does not seem terribly surprising.
posted by rtha at 7:52 AM on October 16, 2014 [1 favorite]


That it's increasingly for rich people—who make a lot and can afford honeymoons

None of this is counter-intuitive. People with better character have better relationships. And strong character (hard-work, delay of gratification, trust-worthiness, thrift) is functionally related to how much money you make.

"Can afford Honeymoons" is trying to beat the data into what it already doesn't show (i.e. that more expensive weddings are a sign of low character and more divorce). It didn't suggest expensive Honeymoons are important, only that you cared enough about your relationship to plan for one ahead of time.
posted by dgaicun at 7:53 AM on October 16, 2014 [10 favorites]


Many or most of these variables are probably highly colinear; perhaps the original study takes this into account and corrects for it (this is fairly easy to do in statistics) but the article makes no mention of whether they did or not. Without that, it pretty much boils down to things we (should) already know: get to know your partner, have a support network, be financially secure, and don't get fixated on material things.

People who go against these guidelines probably aren't doing so on account of never having heard of them--generally, I imagine, they either don't have a choice (you can't just choose to be wealthy), think that they'll be different (nobody likes to think that statistics like these apply to their relationship), or both. Or else perhaps they're getting married for a poor reason in the first place (pressure from partner/family, desire to "fix" the problems with their relationship, societal expectations, etc.)

I guess I'm just not really seeing the novelty here.
posted by Anticipation Of A New Lover's Arrival, The at 7:54 AM on October 16, 2014 [1 favorite]


I've always felt that having a happy marriage is more important than having a stable marriage, so I never understood why divorce is considered a bad thing.
posted by kyrademon at 7:56 AM on October 16, 2014 [36 favorites]


My wife and I hit "least likely" on all of them with the exception of church attendance. We've been married 31 years, make of that what you will.
posted by tommasz at 7:57 AM on October 16, 2014 [3 favorites]


I'm curious about the church thing. My mom, who goes to church every goddamn day, should have been divorced about three years into her marriage (though that would have meant I would not have been born) but due to the Catholic stance on divorce she was guilted into staying married to an emotionally-abusive, alcoholic husband. It was only after about eighteen years when some priest basically told her "fuck that noise" that she got divorced. She still feels guilty about it, though.

So, do people who frequently go to church have happier marriages than people who don't or is it just that they stay married longer when they probably shouldn't?

All signs point to me having a stable marriage after 17 years and I don't go to church. My wife goes to a pretty liberal church, but mostly just for the singing.
posted by bondcliff at 7:59 AM on October 16, 2014 [18 favorites]


The particularly scary part here is that the average cost of a wedding in the U.S. is well over $30,000,

Unlike the rising cost of a university education, this doesn't scare me because people who don't have 30k to spend on a wedding just won't spend $30k on a wedding. They'll have a small ceremony or elope. Throwing a big wedding might be for rich folks, but just getting married still only costs you as much as the fee at the courthouse, if that's all you want to spend.

It still bothers some people when I tell them I never got an engagement ring. Or wanted one. Most of the time we don't wear our wedding rings, we could have saved several hundred by just not buying those either. Most of the stuff you "have" to do when you get married is pure hokum. It can be fun hokum, but it's still hokum.
posted by emjaybee at 8:02 AM on October 16, 2014 [18 favorites]


This may be true for a single data point, but many correlated points are much more likely to strongly indicate causation.

So we look at 5 cities and the number of burglaries in a city is correlated with the number of churches. Could just be correlation, not causation.

Now we look at every city in the world and still see the same correlation. Much more likely to be causation now, right, and not another factor like "size" that causes both variables?
posted by leopard at 8:05 AM on October 16, 2014 [6 favorites]


As other people have mentioned, so many of these variables are highly correlated with each other and with income and education. I think it's more plausible to say that a couple's income and education is a bigger determinant of whether someone gets divorced rather than how many people were at their wedding per se.
posted by alidarbac at 8:07 AM on October 16, 2014 [1 favorite]


The charts appear to be based on table 3 of the paper, which is basically the results of a regression model applied to data from recently-married people.
posted by leopard at 8:08 AM on October 16, 2014


people who don't have 30k to spend on a wedding just won't spend $30k on a wedding.

hahahahahahahahahahaahahahahahahahahahahhahahahahahahahahahahahaha
posted by phunniemee at 8:10 AM on October 16, 2014 [63 favorites]


People with better character have better relationships. And strong character (hard-work, delay of gratification, trust-worthiness, thrift) is functionally related to how much money you make.

I don't even
posted by rtha at 8:14 AM on October 16, 2014 [152 favorites]


As other people have mentioned, so many of these variables are highly correlated with each other and with income and education.

The reason it's worth looking at all of them is that a few run counter to the others. How much you spend on your wedding ought to be highly correlated with your income, but the two have opposite effects on divorce rate.
posted by straight at 8:15 AM on October 16, 2014 [1 favorite]


I wonder if that's an effect of people spending more than they can afford and the resulting money issues? My wedding cost a fair bit, but the tab was entirely picked up by her parents so it's never been an issue; the cost of the wedding has had no effect whatsoever on our relationship.
posted by Bulgaroktonos at 8:18 AM on October 16, 2014 [1 favorite]


And strong character (hard-work, delay of gratification, trust-worthiness, thrift) is functionally related to how much money you make.

Please tell me this was meant to be sarcastic.
posted by marginaliana at 8:18 AM on October 16, 2014 [33 favorites]


People with better character have better relationships. And strong character (hard-work, delay of gratification, trust-worthiness, thrift) is functionally related to how much money you make.

I don't even
posted by rtha at 8:14 AM on October 16 [3 favorites −] Favorite added! [!]


Yeah, I really don't know how to react when dimension shifters from better-managed universes post things like this on our universe's Internet...
posted by You Can't Tip a Buick at 8:19 AM on October 16, 2014 [52 favorites]


The "correlation != causation" thing seems like a bit of a derail here, because (for at least some of these variables) it's pretty clear that nobody's arguing for a direct causal relationship. So for instance, as far as I can tell nobody's actually suggesting that you can cause your relationship with an arbitrary person to improve in stability just by having a really cheap wedding with them — that if your wedding caterer just spontaneously decides not to bill you, say, your relationship will thereby get that much more stable.

The point is clearly that the cost of the wedding is a proxy for some set of harder-to-measure social, economic and interpersonal factors. So pointing out that measured variables can be proxies for unmeasured ones is kind of like, well, hello, yeah, that's what they're saying here too.
posted by nebulawindphone at 8:22 AM on October 16, 2014 [6 favorites]


Your income, the cost of your wedding, the cost of your engagement ring, the size of your wedding, and whether or not you went on a honeymoon are so related to each other that it seems ridiculous to treat them as independent factors affecting the probability that you get divorced.

Surely the cost of your wedding is positively correlated with the size of your wedding. But their effects on divorce run in opposite directions. So even though cost and size are not themselves independent wrt each other, I'm not sure how you could arrive at these data without there being two independent factors that, in combination, explain wedding cost and wedding size.
posted by a snickering nuthatch at 8:24 AM on October 16, 2014 [1 favorite]


That it's increasingly for rich people—who make a lot and can afford honeymoons

Yeah, the redefinition of "wealthy" continues - today's outrageous luxury reserved for the monied elite? Honeymoons.

Is the middle class and the comforts it affords everyday people working everyday jobs that dim a memory? We're all neck-deep in shit if it is.
posted by Slap*Happy at 8:25 AM on October 16, 2014 [22 favorites]


My wife and I score high on evey variable! She will never be rid of me! Ha ha ha ha ha! The only question that remains is where to direct the effort I used to pour into my marriage. Model trains? Stamps? Booze?
posted by vorpal bunny at 8:25 AM on October 16, 2014 [20 favorites]


And strong character (hard-work, delay of gratification, trust-worthiness, thrift) is functionally related to how much money you make.

I'm a hard worker, I am trustworthy, and I'm frugal, and I'm broke as fuck. The only reason I didn't say anything about delaying gratification is because I'm usually too broke to DO anything to gratify myself so I can't help BUT delay gratification.

Are you SURE these are all qualities that are related to how much money you make?
posted by EmpressCallipygos at 8:29 AM on October 16, 2014 [32 favorites]


Shepherd and I did pretty much everything on the cheap. We paid for our own small wedding (despite entreaties from my mother to invite nearly everyone she knew), sort of got a honeymoon in NYC (but mostly it was for an immigration interview for me), but lordy, we sure aren't rich by any stretch. Middle class, maybe, but not rich.

I call shenannygoats on this whole thing!
posted by Kitteh at 8:31 AM on October 16, 2014


I wonder if the question was worded something like - how much did you (as a couple) spend on your wedding? One could have a large wedding and still spend very little if one's parents picked up most of the bill. For example, I had a tiny wedding in the US, probably 5 people including the officiant. Then, however, my parents insisted on throwing us a huge reception in India, with ~2000 people invited. Since this was more for their benefit than ours (which they made clear from the very beginning - something like social payback for all the weddings they had attended), we went along with it. So if asked how many people attended my wedding I'd probably say >200, but if asked how much we (the couple) spent, it would be less than $200 all told.
posted by peacheater at 8:32 AM on October 16, 2014 [1 favorite]


Yeah, none of these indicate a long and happy marriage, necessarily, just 'long'. Richer couples might be more reluctant to divorce because of the huge expense involved. Regular churchgoers (as has been pointed out above) might stay in miserable marriages out of shame. Would be interested to see data on personal happiness of couple alongside this.
posted by RokkitNite at 8:38 AM on October 16, 2014 [6 favorites]


So the optimal cost per guest (200 / 0) is NaN. Got it!
posted by grumpybear69

"OK [200 people], we're getting hitched. Meet us at the corner of X and Y. Best bring a sandwich. Wedding favors? Sure, there's a cigarette butt on the sidewalk. Knock yourself out."

IT'S MY DREAM WEDDING I'M NOT EVEN KIDDING
posted by Orange Dinosaur Slide at 8:38 AM on October 16, 2014 [21 favorites]


Perhaps another important — but unsurprising — finding was that couples who attend church regularly have much stabler marriages. In fact, couples who never go to church are 2x more likely to divorce than regular churchgoers.

Seems like they buried the lede there by failing to note that inconstant churchgoers are even more divorce-prone than us heathens.
posted by murphy slaw at 8:40 AM on October 16, 2014 [3 favorites]


"OK [200 people], we're getting hitched. Meet us at the corner of X and Y. Best bring a sandwich. Wedding favors? Sure, there's a cigarette butt on the sidewalk. Knock yourself out."

IT'S MY DREAM WEDDING I'M NOT EVEN KIDDING


Pot luck wedding.

Also not kidding. We did it--it was great!
posted by flug at 8:41 AM on October 16, 2014 [2 favorites]


I'm a hard worker, I am trustworthy, and I'm frugal, and I'm broke as fuck

On average having a good work ethic, good spending habits, and being able to work towards long term goals are certainly related economic well being. That certainly leaves a lot of room for bad luck, bad choices, incompetence, differently weighted preferences (e.g. religious poverty), and idiosyncrasy. If you've had good, stable relationships then your character (as you've described it) probably explains it. If you've always been broke and had a chaotic relationship history, maybe you've misjudged your own character.
posted by dgaicun at 8:43 AM on October 16, 2014 [4 favorites]


So for instance, as far as I can tell nobody's actually suggesting that you can cause your relationship with an arbitrary person to improve in stability just by having a really cheap wedding with them
I'm as tired of the "correlation doesn't imply causation" knee-jerk phrase as anyone else. But, it's an entirely reasonable response to the Atlantic article, which says among other things, "Their findings offer some take-aways for couples who want to minimize their chances of divorce: Have a huge wedding, but make sure it's cheap."

It's a shame some interesting and counter-intuitive results are framed by the writer at the Atlantic so terribly. But, pointing that out isn't exactly a derail; more a questionable choice at the switch house.
posted by eotvos at 8:45 AM on October 16, 2014 [4 favorites]


We had several weddings (because gay marriage legal/not legal/etc.) so should I add up all the attendees, or do an average, or what?

Also, Empress, it amuses me greatly to think that by dgaicun's metric, your moral character must be worse than those of Donald Trump, or Bernie Madoff, or Dick Chaney. Well, at least you have something to shoot for!
posted by rtha at 8:46 AM on October 16, 2014 [19 favorites]


I'm a hard worker, I am trustworthy, and I'm frugal, and I'm broke as fuck

On average having a good work ethic, good spending habits, and being able to work towards long term goals are certainly related economic well being. That certainly leaves a lot of room for bad luck, differently weighted preferences (e.g. religious poverty), and idiosyncrasy. If you've had good, stable relationships then your character (as you've described it) probably explains it. If you've always been broke and had a chaotic relationship history, maybe you've misjudged your own character.


You seem to be combining statistics with a word that can mean whatever you want it to mean. I'm not sure why you're doing that!
posted by selfnoise at 8:47 AM on October 16, 2014 [14 favorites]


I'm curious about the church thing.

Benefits likely come as much/more from extensive shared social context, many friends and extensive who share values, etc etc etc, than they do from doctrinal beliefs per se.

There is backup on this from the fact that benefits seem to accrue regardless of the particular church attended. If this were all or mostly related to specific beliefs or doctrines, there would be massive differences in effect depending on which specific church you attend and which specific doctrines you believe in. Instead, the benefits seem to accrue to churchgoers of a broad variety of beliefs, indicating it is likely something more to do with the things that are common among many/most religions rather than those particulars of doctrine and teaching that are most of the matters of contention among them.

For those of us who are not churchgoers, I'm guessing you could replicate some or many of the benefits by having a shared social commitment of similar extent--maybe volunteerism, working together for some mutual cause, or the like. Working towards mutual shared goals in some kind of a structured environment with many friends who are also working towards those mutual shared goals.

Churchgoing fills that niche (plus a number of others, of course) for many in our society.
posted by flug at 8:49 AM on October 16, 2014 [5 favorites]


Pot luck wedding. Also not kidding. We did it--it was great!

My dream wedding is actually a similar "stone soup" kind of thing - I know enough people with enough quirky talents that I could probably ask a bunch of my friends to take care of a lot of the various trappings for free, in lieu of having to get me a gift. Even the officiant is taken care of (that guy I said I know who was a Jesuit seminarian? Yup - and I know him well enough to know that his sermon would be BALLS-OUT CRAZY AWESOME.)
posted by EmpressCallipygos at 8:54 AM on October 16, 2014 [1 favorite]


For persons under the age of 55, most of the best marriages I know have at least one party who is either divorced or was in a long term, might as well be married relationship, that went south before the strong relationship happened.
posted by crush-onastick at 8:57 AM on October 16, 2014 [2 favorites]


I got married 11 years ago after dating my partner for 7.5 years, had a 115 person wedding which cost about $9K if you count the value of donated services (photography and DJ), and went on a honeymoon. I go to church irregularly, my husband just broke a streak of not setting foot inside a church building except for weddings or funerals for 20 years for the sole reason that I needed someone to look after our kids while I sang in the service. But the single factor that has most strongly contributed to our marriage being a strong one was the two years of couples counseling we underwent before we were engaged. Wish they'd looked into that.
posted by KathrynT at 8:59 AM on October 16, 2014 [7 favorites]


The "correlation != causation" thing seems like a bit of a derail here, because (for at least some of these variables) it's pretty clear that nobody's arguing for a direct causal relationship

Some quotes from the Randal Olson piece that seem to indicate a direct causal relationship:
In other words, Bridezilla = Divorcezilla. Don’t let advertisers fool you into spending your life savings on your wedding.

Whatever you do after your marriage, don’t skimp on the honeymoon!

Crazy enough, your wedding ceremony has a huge impact on the long-term stability of your marriage.
The problem with highly correlated predictors is that they make regression coefficients difficult to interpret. If income and education are highly correlated, it's fundamentally hard to tell if a highly-educated wealthy person is unlikely to get divorced because of his income or because of his education. But that doesn't stop anyone from showing you a chart with a precise number on it.

More to the point, the relationship between wedding expense and divorce rates is almost certainly going to vary by household income, so it's silly to produce a chart that shows one single number as "the" impact of wedding expense on divorce.
posted by leopard at 8:59 AM on October 16, 2014 [6 favorites]


I'm curious about the church thing.

In addition to flug's comments on increased community involvement, virtually all of the religious people I know have a religious component to their concept of marriage. They are less likely to get married primarily for tax or health insurance reasons, since the marriage is a contract with God, with the State as secondary. They are also more likely to believe that the ideal marriage is for life, and not until it stops being convenient (this brings up it's own issues, but it does lower the divorce rate). Many Churches require some form of additional pre-martial counseling, and have significantly longer waiting periods for marriages. There's impulsively rushing off to get married if your Church needs 6 months notice. Also, many priests/ministers can and will refuse to married people they don't think are a good fit. Sure, this is not true for 100% of religions, but it is enough to have a significant effect on the data.
posted by fermezporte at 9:03 AM on October 16, 2014 [3 favorites]


But the single factor that has most strongly contributed to our marriage being a strong one was the two years of couples counseling we underwent before we were engaged. Wish they'd looked into that.

This was something we did as well but for a few months instead of two years. We did it because we started dating I lived in the States and he lived in Canada, and because I had a minimum wage job plus an old DUI on my record, I couldn't visit him so he had to visit me. Once we got engaged, the counseling became important because we weren't going to live with each other until we got married. It was incredibly helpful to us. It allowed us to learn about aspects of our personalities/goals/desires that weren't immediately obvious.
posted by Kitteh at 9:05 AM on October 16, 2014


OK [200 people], we're getting hitched. Meet us at the corner of X and Y. Best bring a sandwich.

Ceremony is long enough to eat a sandwich? Too long!

Speed it up, I got another wedding down by the park in half an hour.
posted by madajb at 9:06 AM on October 16, 2014 [7 favorites]


Interesting data, but my kingdom for an error bar. I get that it's not aimed at an audience of statisticians, but still. In fact, if he's aiming at an audience that won't understand error bars, I think he should be much more careful with the jokey "...therefore you should definitely avoid X" correlation/causation stuff. Ugh.

As a side note, from skimming the paper it looks like a lot of the data was culled from surveys on Amazon's Mechanical Turk. Is that generally regarded as a decent survey source? (Or have I badly misread that?)

Jpfed - Surely the cost of your wedding is positively correlated with the size of your wedding.

Not the weddings I've been to, or at least not strongly. One pair of friends had a lavish wedding in which the venue hire alone can't have been far south of £15k (other friends planning a wedding at the time had looked it up), and others with similar numbers of guests have been under £2k, all-in. I saw a lot of weddings when I was growing up, thanks to being involved in my local church, and formed the definite impression that the more expensive-looking weddings roughly correlated with how excitable (and prone to getting visibly upset by delays/hitches) the people involved were. Anecdote not data and all that but, correcting for income levels, I'd expect inexpensive weddings to weakly predict chilled-out couples and families.
posted by metaBugs at 9:13 AM on October 16, 2014 [4 favorites]


So even though cost and size are not themselves independent wrt each other, I'm not sure how you could arrive at these data without there being two independent factors that, in combination, explain wedding cost and wedding size.

Jpfed: correct, but it could just be that wedding expense is irrelevant if you control for income and wedding size, except for a handful of highly financially irresponsible people who are especially likely to get divorced. That would generate a wedding cost effect that only applies to a small subset of the population.
posted by leopard at 9:16 AM on October 16, 2014 [1 favorite]


That certainly leaves a lot of room for bad luck, bad choices, incompetence, differently weighted preferences (e.g. religious poverty), and idiosyncrasy

What's great is that we can figure out the contours of the universe dimension-shifting Internet posters come from by looking at what they leave out when they make statements like this. I am guessing, from what our slider has left out, that dgaicun is visiting from one of the universes that has a jubilee every seven years or so, wherein debts are forgiven and asset distribution is radically flattened.

When you slide back over there, dgaicun, you can tell people from your universe campfire stories about the scary universe you visited, the one where *spooky ghost noise* the chief predictor of individual wealth is family wealth, and where *more spooky ghost noises* class mobility is a.... MYTH!

have a confederate, dressed like an old-timey fat cat from our universe, jump out of the tent when you say "MYTH!" It'll scare the bejeezus out of everyone.
posted by You Can't Tip a Buick at 9:17 AM on October 16, 2014 [44 favorites]


What metabug said. Lavish weddings are expensive. Large weddings, not necessarily.
posted by Lesser Shrew at 9:19 AM on October 16, 2014


Not sure why I decided to skim the article (the one with the stats and research) or even post, but here goes.

There is a limitation that the authors of the actual study (ie, Francis and Mialon); the study population took an mechanical Turk survey for .5 to .75 cents. The majority of their sample was not wealthy (as a couple, approximately 73% if the study population earned less than 75 K). So as a starting my guess that this population would not have 20 K ready to go for weddings, etc, but who knows.

If I were to use this data to make a predictive model, I wouldn't want to just see a healthy sample size, which they do have with an n of 3000 or more, but are the variables still significant after a multivariate analysis and are they statistically significant (not just the little graph of converted hazard ratios that they are using for the Atlantic piece). The paper does have this if anyone is interested (Appendix Table 1) and things like cost of wedding did not actually become significant unless it was $0 to $1000 OR $20000 and above.

The one variable that was never mentioned in the Atlantic piece, which was signficant via the multivariate analysis (and increased the likelihood of a stable marriage) was having children. No idea why that fell of on the Atlantic piece, either.

On preview, someone else noticed the mechanical Turk survey. I wondered the same thing when I read their research because there are some medical journals that would not publish it based on that bias. I just ran a pubmed search and google scholar search to see if this is published in a peer reviewed journal; it does not appear to be, and putting it on the Atlantic, well, in the sciences you would blow up your chance for publishing anything now, but maybe it is different for economists.
posted by Wolfster at 9:24 AM on October 16, 2014 [12 favorites]


phunniemee: Based on how much my cousin spent on her wedding this past April she should be divorced by now.

That one is by far the most interesting stat to me.
That. Is not. The way statistics work.
posted by IAmBroom at 9:31 AM on October 16, 2014 [3 favorites]


None of this is counter-intuitive. People with better character have better relationships. And strong character (hard-work, delay of gratification, trust-worthiness, thrift) is functionally related to how much money you make.

dgaicun are you sure you are not the lovechild of Malcom Gladwell and Rush Limbaugh?
posted by incolorinred at 9:35 AM on October 16, 2014 [2 favorites]


grumpybear69: So the optimal cost per guest (200 / 0) is NaN. Got it!
"Guests" may be an inexact variable for "people present". The optimal wedding would therefore be between two people who were never present, with no one officiating, which does make sense: a clerical or computer error in the government wedding registry would probably last a very long time.

It seems that the ideal cost for entering such data into the registry is therefore $200. Seems high, sure, but isn't a long-lasting marriage worth it?
posted by IAmBroom at 9:35 AM on October 16, 2014 [1 favorite]


Unlike the rising cost of a university education, this doesn't scare me because people who don't have 30k to spend on a wedding just won't spend $30k on a wedding.

In a world where there are no credit cards, perhaps.
posted by Jahaza at 9:36 AM on October 16, 2014 [4 favorites]


I'm curious about the church thing. My mom, who goes to church every goddamn day, should have been divorced about three years into her marriage (though that would have meant I would not have been born) but due to the Catholic stance on divorce she was guilted into staying married to an emotionally-abusive, alcoholic husband. It was only after about eighteen years when some priest basically told her "fuck that noise" that she got divorced.

Yeah, and this might also factor into say, wedding cost - i.e. if you're in a culture that has a more traditional view on marriage, you may have a more traditional (and cheaper) wedding.

Not to say that I think marriage shouldn't be a commitment, but: I tend to see divorce rates as a sign of the prevalence of women's rights in a particular society.
posted by capricorn at 9:36 AM on October 16, 2014 [5 favorites]


Since there is a lot of talk about the outputs and not much about the inputs, I read (well, read and skimmed) the original study. There are a few interesting take-aways:
The paper is about "duration", not "stability" or "happiness"
Respondents' average marriage duration 6.2 years
Respondents were all from Amazon Mechanical Turk with US IP addresses (50-75 cents for 5 minutes)
Responses from participants over 60 years old were discarded
68% of respondents had never been divorced
Whites seem to be slightly over-represented in the sample, Hispanics slightly under-represented.
Incomes from $25K-$75K over-represented, >$125K under-represented
High-school education very under-represented, college very over-represented

My biggest question is about how they considered the durations of two-thirds of the responses, those that had never divorced? It seems easy to include data about someone that has been married for 30 years, but what to do with the still-married newlywed? Does the one-year-and-still-married person get counted the same as the one-year-then-divorced person? As far as I can tell the study doesn't say.

The study seems designed to do only one thing, explore the correlation between wedding spending marriage duration. From their conclusion: The wedding industry has consistently sought to link wedding spending with long-lasting marriages. ... Overall, our findings provide little evidence to support the validity of the wedding industry’s general message that connects expensive weddings with positive marital outcomes.

(On preview they do talk about the mTurk sample: Samples of mTurk workers have been found to be more representative of the US population than in-person convenience samples, standard internet samples, and typical college samples ... Moreover, the internal and external validity of experiments conducted with mTurk have been shown to be comparable to that of laboratory and field experiments)
posted by achrise at 9:37 AM on October 16, 2014 [2 favorites]


According to table 2 on the paper, the wedding expenses go in different directions in the bivariate and multivariate models and have little statistical significance beyond the 0-1000 range. The fact that that they go in different directions from one case to another is pretty substantial evidence for multicollinearity, no?

To put that in layman's terms: ignore the "cost of the wedding" graph.

Good God I'm procrastinating so hard I just interpreted an econometric model.
posted by Ndwright at 9:38 AM on October 16, 2014 [14 favorites]


It's interesting that the influence of wedding cost and wedding size run in opposite direction, since hosting more people at a wedding generally makes it cost more.

My favorite wedding ever was the one for my friend Wendy, whose father was a fairly poor pastor at a small, primarily black congregation. She wore her nicest Sunday dress; her husband wore his best suit. Pastor-dad performed the ceremony. I'd estimate about 150 in attendance. The wedding march was played on a cassette tape on a boom box at the front of the church. Sister Peaches, a talented women from the church, sang a special song. The bishop offered a special prayer. Afterward, there was a potluck reception in the little civic center of their little town. Everyone brought some food. Someone brought a karaoke machine. Lots of friends took turns offering a song and good wishes. There was intermittent dancing and singing for a couple of hours. Nothing fancy; nothing ostentatious. If they spent more than $100 on everything, I'd be shocked. And it was absolutely the most joyous wedding I've ever been to. I wish we had more $100 weddings.
posted by Pater Aletheias at 9:39 AM on October 16, 2014 [12 favorites]


We're getting married next week, so it's probably a little late to listen to any of this advice. We'll just have to do whatever the hell we want and keep communicating with each other about our needs. Oh well!
posted by backseatpilot at 9:40 AM on October 16, 2014 [10 favorites]


Hmm. Checking my own experiences, the only things I did wrong were (a) not invite enough guests (only about 30 friends and family -- mind you, very small families on both sides) and (b) not going to Church (because hello? We're both atheists).

Got married on our tenth anniversary (to the day), did it cheap-ish because we paid for it ourselves, had a modest honeymoon (again, paid for it ourselves), etcetera. That was over eleven years ago ...
posted by cstross at 9:40 AM on October 16, 2014 [3 favorites]


I tend to see divorce rates as a sign of the prevalence of women's rights in a particular society.

Hah, yes. I mean, I want people to be happy and it's a shame when marriages end, but I've seen too many people argue along the lines that "feminism (/irreligion) increases divorce rates!", like that's a bad thing. I tell myself that it's a weird blind spot in their reasoning, rather than an active desire to have more people trapped in miserable situations.
posted by metaBugs at 9:41 AM on October 16, 2014 [14 favorites]


dgaicun: And strong character (hard-work, delay of gratification, trust-worthiness, thrift) is functionally related to how much money you make.
And you probably vote Republican.

How much money you make is most influenced by how much money your parents made, which is essentially unrelated to strong character.
posted by IAmBroom at 9:42 AM on October 16, 2014 [5 favorites]


People with better character have better relationships. And strong character (hard-work, delay of gratification, trust-worthiness, thrift) is functionally related to how much money you make.

I don't even


I've seen this kind of attitude expressed among people in the South most of my life. The idea that most of the wealthy, and in particular, the old money wealthy, are morally superior due to their good breeding--that they are "the quality people"--the attitude goes even farther than the comment above in ascribing the "better character" to the quality of the family lineage rather than personal accomplishment, so it's sort of a "blood will out" mindset.

What seems most probable to me from the research findings is that having more wealth makes it easier to weather rough times, and since there's a lot of data showing that money concerns are often among the chief causes of conflict in marriages, it only makes sense that our system is primed to make it easier for the wealthy to appear more virtuous if we assume their better odds of success in marriage are driven by innate character traits rather than being a product of the material and spiritual differences in life circumstance that money can provide.
posted by saulgoodman at 9:45 AM on October 16, 2014 [17 favorites]


My biggest question is about how they considered the durations of two-thirds of the responses, those that had never divorced?

This is the Cox proportional hazards model mentioned in the paper. Marriages have a baseline "failure rate" which represent how likely a couple is to get divorced at each moment in time. Then this failure rate gets adjusted up and down for the predictors in the model like income, wedding cost, etc (the adjustment factor is the same at each moment in time).

So you can use both dissolved marriages and ongoing marriages to fit this sort of model, because the lack of a dissolution event provides some information about what the failure rate is.
posted by leopard at 9:47 AM on October 16, 2014 [3 favorites]


Yeah - people who have more money are less likely to have money problems that create so much stress that it breaks the relationship. This seems like a no-brainer interpretation of that set of data, and makes more sense than "oh, the rich are just magically better because of reasons."
posted by rmd1023 at 9:51 AM on October 16, 2014 [4 favorites]


Unlike the rising cost of a university education, this doesn't scare me because people who don't have 30k to spend on a wedding just won't spend $30k on a wedding.

In a world where there are no credit cards, perhaps.


By "don't have" I mean "don't have access to" whether via debt or family. Not everyone has 30k worth of available credit or family money. Thankfully, that still doesn't prevent you from getting married if you want to. I have known more than one couple that got married, in fact, because one of them needed insurance but couldn't get it, and there are the tax benefits, etc. that come with the piece of paper. A big traditional wedding is a luxury, and as more Americans drop below the "comfortable" income line, that will become even more true. Marriage itself remains a relative bargain, at least until it goes south and you have to hire lawyers.
posted by emjaybee at 9:51 AM on October 16, 2014 [1 favorite]


The stat I'd like to see is the relationship of destination weddings to duration, perhaps measured in miles from where the couple plans to actually live.
posted by Obscure Reference at 9:53 AM on October 16, 2014 [6 favorites]


Weddings are a scam. you can spend an exorbitant amount on a small number of guests or a large number - Our wedding cost ~$4,000 for 80 guests, we catered with indian food for 800 bucks; it was more than enough food and was better than any wedding food ever. My wife found a great dress for 8 bucks, we rented a cheap venue and spent the rest on booze, it was awesome.

My interpretation of this would be that if you value the company of your friends and family and want to celebrate your relationship you don't need to spend a lot; but if you spend a great deal you likely value something else: tradition, opulence, appearance, etc. and your relationship is secondary.
posted by Colby_Longhorn at 9:53 AM on October 16, 2014 [3 favorites]


In fairness to dgaicun, they did stress that the relationship exists "on average"; It doesn't follow from what they're saying that any given rich person is morally superior, or vice versa. I still don't agree with their point (wealth/class mobility is way too low for that to be true, unless you want to get into some really disturbing territory), but let's not attack them for something that they didn't say. (IANAMod, obviously)
posted by metaBugs at 9:55 AM on October 16, 2014


Colby_Longhorn: I think suggesting that someone who wants some of the frills and pomp and circumstance means someone thinks their relationship is secondary is kind of a radical interpretation.
posted by rmd1023 at 10:04 AM on October 16, 2014 [4 favorites]


In additional fairness to dgaicum--if we could, and I imagine it is a significant "if", successfully operationalize the components in his/her statements I would bet it is significantly accurate. The tendency on this board to partially demonize persons who are financially successful ( not wealthy/filthy rich/inherited wealthy) or have successful and stable careers is sometimes troubling for me. There is quite often an immediate polarization of good guys = financially struggling versus bad guys=financially successful. Coming from a financially stable back ground, valuing delayed gratification, behaving with integrity, being frugal, having high impulse control etc probably does increase the probability of long term marriage. These do not have to be viewed as judgmental--just potentially observable/measurable phenomena. of course they can be "explained away" by a variety of factors--good luck, right family, right time/place--but that does not make them less predictive.
posted by rmhsinc at 10:13 AM on October 16, 2014 [6 favorites]


30,000 on average is a down payment for a house or the pay off amount for your student loans - it's not some "frills" at your fancy dress party. IMO it does imply some twisted priorities, but I'm a fucking cheapskate.
posted by Colby_Longhorn at 10:14 AM on October 16, 2014 [2 favorites]


we catered with indian food for 800 bucks; it was more than enough food and was better than any wedding food ever

I wish I could have done this, but my stepmother once came with my father and I for one of our lunch excursions to a local Indian restaurant and she complained that the butter chicken was too spicy. Butter chicken. It's like the mildest dish, I can't even

Our wedding turned out pretty great anyway, we had it in a nice restaurant, kept it to immediate family and close friends. You got a choice of beef tenderloin or halibut, both were actually done incredibly well (I find for big catered events the food sometimes comes out a bit less than ideal, but because of the size of our wedding it was basically like just ordering it at a nice restaurant, cuz that's what it was). It was pricey for having 25-30 guests, but reasonable in general terms for the cost of a wedding (we apparently spent like a third of the American average). We also went on our honeymoon like 3 years later when we could actually afford it instead of going into more debt.
posted by Hoopo at 10:15 AM on October 16, 2014


God the money thing is so obvious to me. If you have money that's one less thing to fight about. Subtle and even maybe drastically different attitudes about money can be minimized if there is plenty of money to go around. People with money generally have more stable lives, single or married. Money also gives you flexibility to address issues in your marriage. Whether that's hiring a maid, getting a babysitter when you need a break or going to marriage counseling.
posted by whoaali at 10:23 AM on October 16, 2014 [11 favorites]


Huh, I guess our divorce bucked the odds.
posted by desjardins at 10:27 AM on October 16, 2014 [3 favorites]


From the first link:
In the research paper, the authors suggest that the financial burden incurred by lavish, expensive weddings leads to financial stress for the couple, which ultimately tears the marriage apart. They found that women, in particular, are vulnerable to divorce after expensive marriages: women in couples who spent $20,000 or more on their wedding are 3.5x more likely to end up divorced than their counterparts who spent less than half that.
I can't get the paper to open, but can someone explain if the study authors also somehow limit the effects of expensive weddings to women only? Because I'm guessing that there weren't just a lot of expensive lesbian weddings that led to this conclusion, which leads me to believe men were involved in the divorces, too.
posted by jaguar at 10:44 AM on October 16, 2014 [4 favorites]


flug: "Benefits likely come as much/more from extensive shared social context, many friends and extensive who share values, etc etc etc, than they do from doctrinal beliefs per se."

Recalling from my seminary years, studies suggest it's some combination of shared values (which are a strong predictor of marriage duration and happiness); community support (having friends, also a strong predictor); membership in a community that explicitly values and normalizes marriage (weakish predictor but makes people somewhat more likely to stick it out, for good or for bad); and access to supportive services -- which includes everything from required premarital counseling to get married in the church, to pastoral counseling for individuals or couples in crisis, to "date night" where the teen club watches little kids so parents can go out on a date, to couples retreats. You can BUY all of those things if you have enough money, but membership in a religious congregation often provides access to a variety of "services" designed to support couples and families at low or no cost.

(The lesson of this for my future-pastor classmates was, take your premarital and marriage counseling duties seriously and do continued education in those areas, and look hard at what gaps in your community need filling for married people who DON'T have a ton of money to burn as you plan church social and religious activities.)
posted by Eyebrows McGee at 10:46 AM on October 16, 2014 [4 favorites]


I can't get the paper to open, but can someone explain if the study authors also somehow limit the effects of expensive weddings to women only? Because I'm guessing that there weren't just a lot of expensive lesbian weddings that led to this conclusion, which leads me to believe men were involved in the divorces, too.

Lesbians were not included in this study (by definition, for this particular study, they only included people married to the opposite gender).

Also, remember that they surveyed the study population via mechanical turk and this does not mean you have both halves of a couple, just men and women who took the survey.

In their multivariate analysis, they did break it down by gender,and it was only significant for all subjects, and females with marriage costs over 20 K.

There were other discrepant results by gender according to their data; for example, a partners looks as having a decreased likelihood of stability/duration/ of marriage was statistically significant for all subjects, males, but not females.

I do think that it would be an interesting thing to try to address via other published articles and the field, but you know what? Their paper does have a discussion section. I wonder if this is someone's project for a masters degree or something?
posted by Wolfster at 11:10 AM on October 16, 2014


@ emjaybee - I didn't have (and didn't want) an engagement ring either. The way engagement rings were viewed/shown/made a big deal of in the community in which we lived at the time played a role in that, but finances were a factor too. I didn't want to spend so much on something that was basically a short-term, me-only status symbol. Instead, we have hand-crafted wedding rings that we designed together. They're a lot more us. (Posted because I've met very few other women who don't have and didn't want engagement rings.)
posted by Flipping_Hades_Terwilliger at 11:11 AM on October 16, 2014 [3 favorites]


There were other discrepant results by gender according to their data; for example, a partners looks as having a decreased likelihood of stability/duration/ of marriage was statistically significant for all subjects, males, but not females.

OK, but a guy can value his female partner's looks without it being reciprocal. A wife cannot divorce her husband without the husband also getting divorced.
posted by jaguar at 11:18 AM on October 16, 2014 [1 favorite]


The tendency on this board to partially demonize persons who are financially successful ( not wealthy/filthy rich/inherited wealthy) or have successful and stable careers is sometimes troubling for me.

I have a good education from a prestigious college, a decent job with excellent benefits, and a stable, long-term relationship where we own a home in one of the most expensive cities in the U.S.

Being of "good moral character" maaaaaybe contributed a teeny bit to all of this; some of it was hard work; mostly, it was luck. Luck being born when I was, where I was; luck going to college at a time when I graduated with a debt of about $10,000; luck because a whole ton of people I've never met and never will meet worked for decades to change laws and culture so that someone like me - a woman, a lesbian, a person of color - could have a margin of error of sufficient depth and width that even when I tripped, I didn't end up in the abyss. Acknowledging all the shoulders I'm standing on, and asking other people to do the same, is not exactly demonization.
posted by rtha at 11:21 AM on October 16, 2014 [35 favorites]


I think it's also kind of telling how merely voicing the idea that "character" (which itself needs some unpacking) is not correlated with income was treated as an attack on rich people's character, rather than what it was, a defense of less wealthy people's character.
posted by en forme de poire at 11:28 AM on October 16, 2014 [14 favorites]


In additional fairness to dgaicum--if we could, and I imagine it is a significant "if", successfully operationalize the components in his/her statements I would bet it is significantly accurate. The tendency on this board to partially demonize persons who are financially successful ( not wealthy/filthy rich/inherited wealthy) or have successful and stable careers is sometimes troubling for me.

I don't think anyone is "demonizing the rich" when they point out to dgaicum that assuming "a person who has a lot of money is someone who worked hard" may be putting the cart before the horse.

Yes, it is true that financial success might indicate that a person be possessing of other good qualities. But it is not true to assume that it is the biggest indication that a person possesses those qualities. It's yet another "correlation does not equal causation" point.
posted by EmpressCallipygos at 11:29 AM on October 16, 2014 [5 favorites]


I was going to say that dgaicun was only talking about correlation, but then I re-read his comment and nope, s/he is actually arguing for a mechanistic relationship: "And strong character (hard-work, delay of gratification, trust-worthiness, thrift) is functionally related to how much money you make."
posted by en forme de poire at 11:37 AM on October 16, 2014 [1 favorite]


Back to the actual article, I'm really curious about the reasoning behind these - for example, the wedding size. Are they trying to show that if your family all showed up, then you have more support system and probably can weather bumps better?

(also, I may need to be excused to plan a belated honeymoon)

In seriousness, though, the honeymoon thing may also relate to who has the financial security/time off to plan a honeymoon. Better jobs decrease stress.
posted by corb at 11:48 AM on October 16, 2014 [2 favorites]


rtha-you are absolutely correct. Humility and graciousness seem to me to be the best way to accept the good fortune which some of us have experienced. And I take no issue with what you said--but on this board there is a generous amount of sneering, castigating and belittling those who have achieved financial success . And it is not limited to the ultra-wealthy--phrases such as rich assholes, pricks, rich white men, entitled capitalists, capitalist pigs/fuckers (or some such) appear regularly without in any sense discriminating between those who use their good fortune with stewardship, responsibility and generosity or with indulgence and exploitation. I get as tired of railings against the fortunate as I do about judgements and demeaning comments about the less fortunate (hardly ever on this board). I also assumed a generous interpretation of dgaicun statements and not a literal interpretation. If you think this board, judging by language/adjectives used, is not negatively biased towards the financially successful upper middle, professional and lower upper class then we are not reading/hearing the same thing. My point, quite succinctly, is that in most cases (rich/middle/poor) it is usually 60% luck, 20% good/bad decisions, 20% effort/or lack of. Percentages are negotiable.
posted by rmhsinc at 12:08 PM on October 16, 2014 [7 favorites]


On average having a good work ethic, good spending habits, and being able to work towards long term goals are certainly related economic well being

But maybe not to good proofreading and grammar?

I'd really like to see a citation for this too. I'd say I am (very luckily) someone who has experienced economic well being and a wonderful marriage (despite no churchgoing and relationship of less than a year before it) but I know plenty of people who don't have economic well being who have a better work ethic, spending habits, and ability to work toward long term goals than I do.

The only thing more poisonous in U.S. society these days than our growing income gap/loss of economic opportunity is the self-congratulation of the haves that they somehow deserve their good fortune more.
posted by bearwife at 12:16 PM on October 16, 2014 [6 favorites]


Rmhsinc, this is the first comment I made in response to dgaicun's statement. Please point out to me where you see me "castigate" dgaicun, or the rich.
posted by EmpressCallipygos at 12:27 PM on October 16, 2014


Empresscallipygos--I don't believe I mentioned you or anyone in particular. If you disagree with the general observation fine but this was not about any specific person. No need to take this personally.
posted by rmhsinc at 12:38 PM on October 16, 2014


No need to take this personally.

Irony is a harsh mistress.
posted by poffin boffin at 12:41 PM on October 16, 2014 [6 favorites]


Irony is a harsh mistress. Seriously, I have no idea what this means or its purpose. Seriously.
posted by rmhsinc at 12:45 PM on October 16, 2014 [1 favorite]


The impact of gender on the results is a good measure of model uncertainty. In reality every divorce impacts one man and one woman (there are no same-sex couples in the data) so any gender effect is because the men in the sample happened to have different experiences than the women in the sample.
posted by leopard at 12:50 PM on October 16, 2014 [1 favorite]


I agree that it's probably a measure of model uncertainty more than anything else, but you could also imagine other explanations. For example, if women on mTurk are poorer than men, they might be more harmed by expensive weddings. Or, since women are often more involved in wedding planning than men, they might have a better idea of the real costs, while the men are just making a wild guess, or they are calculating the costs differently (subtracting parental support or wedding gifts, for example).
posted by fermezporte at 1:34 PM on October 16, 2014 [1 favorite]


Ok, but still: expensive weddings are more likely to end in divorce for the women in the sample than for men in the sample, but one can't (as the first linked piece did) generalize that out to expensive weddings are more likely to result in divorce for women but not men.
posted by jaguar at 1:55 PM on October 16, 2014 [2 favorites]


I think it's also kind of telling how merely voicing the idea that "character" (which itself needs some unpacking) is not correlated with income was treated as an attack on rich people's character, rather than what it was, a defense of less wealthy people's character.

Remember, it ain't a class war until the poor fight back.
posted by bgal81 at 1:55 PM on October 16, 2014 [3 favorites]


Seriously, I have no idea what this means or its purpose.

You were telling me not to take what you said personally, when you were yourself seemed to be taking people's critique of dgaicun's statement personally. This is irony.
posted by EmpressCallipygos at 2:25 PM on October 16, 2014 [1 favorite]


None of this is counter-intuitive. People with better character have better relationships. And strong character (hard-work, delay of gratification, trust-worthiness, thrift) is functionally related to how much money you make.

People who are healthy tend to be more able to work hard. Lack of physical and/or mental health does tend to generate poverty and divorce. The things that create poverty and low income - crappy wages, crappy hours, crappy working conditions - contribute to divorce. Many poor and low income people work hard, are thrifty, trust-worthy, and have little choice but to delay gratification, or may grab a bit when and as they can, because opportunities are limited. The character and values of the wealthy have some rather ugly underpinnings. marriage: ... it's increasingly for rich people I think like everything else, if you have money, many troubles are smoothed.

Googling character is interesting. Traits for (Moral) Character include: Empathy, Sincerity, Integrity (includes honesty), Resilience, Respect, Humility, Competence, Fairness, Wisdom, Courage. I think we live in a culture that values money and power, and devalues people.
posted by theora55 at 3:44 PM on October 16, 2014 [4 favorites]


but on this board there is a generous amount of sneering, castigating and belittling those who have achieved financial success . And it is not limited to the ultra-wealthy--phrases such as rich assholes, pricks, rich white men, entitled capitalists, capitalist pigs/fuckers (or some such) appear regularly without in any sense discriminating between those who use their good fortune with stewardship, responsibility and generosity or with indulgence and exploitation.

Nobody's done this in this thread, so I don't even know what this complaint is doing *here*. Despite my (certainly by global standards) wealth, I've never felt personally targeted in discussions where disparaging language has been used like that. Like, there's been no castigating of rich people in this thread that I can see, only moralistic sniffing about people of good character who obviously can't be poor. I mean eeewwww.
posted by rtha at 3:52 PM on October 16, 2014 [5 favorites]


empress--as you noted "I seemed" to react personally which I didn't. You did react on a personal level to what I said.
posted by rmhsinc at 4:15 PM on October 16, 2014


There is not really a contradiction between "How many people attend the wedding" and "How much you spent on the wedding". The way I'm interpreting this is that when you control for wedding size, increased spending correlates with a greater chance of divorce. And similarly, when controlling for spending, decreased number of guests correlates with a greater chance of divorce.

That being said the first link is full of unsupported conclusions and other offensive statements, particularly the "Bridezilla = Divorcezilla." If your study (dealing with only opposite-sex marriage) shows women have more expensive weddings than men you have an unrepresentative sample and/or one sex is misreporting.
posted by mountmccabe at 4:33 PM on October 16, 2014


If you've always been broke and had a chaotic relationship history, maybe you've misjudged your own character.

So EVERY person who has persistent economic hardship and problems in their interpersonal life is deficient in character? And grown people out there, for example- through absolutely no fault of their own- born into poverty stricken communities, broken homes, no social supports, etc. and exhibit the descriptions you have outlined above must be deficient as people? Wow.

We are certainly getting a good look at some people's "character" in here.
posted by incolorinred at 7:40 PM on October 16, 2014 [6 favorites]


What we see above is that dating 3 or more years before getting engaged leads to a much more stable marriage. This finding probably comes as no surprise, but it should stand as a warning to those who are eager to get married right away. Don’t jump into marriage before you really get to know someone.

Well.... my wife and I have known each other since we were 16, started 'dating' last Augustish (2013), got engaged in December, and got married this August (2014). We're 37. I sure hope we know each other by now. :) I wonder what we do to the model...

(I know this isn't how statistics work; I teach data analysis. :) I'm just easily amused and used to breaking boxes/forms/models/etc. I'd love to see some data on length of knowing each other total, instead of the assumption that people jump straight into dating...)
posted by joycehealy at 7:45 PM on October 16, 2014


Fact: studying stuff like this has basically been the last 11 years of my life. I can cite sources for everything I'm about to type, but short-term here are a few things to know that matter (I've linked to statistically significant studies, US Census Data and other science-driven, peer-reviewed goodness, but there are tons of books and other things I'm too tired to bother with this late):

Average age at first marriage for USians is now 27 (women) or 28 (men) - the highest it's ever been in recorded history. Also, 1 in 4 young adults today may never marry. TONS more cool info at the Pew Research Center's site here.

That oft-cited 50% divorce rate is a relic from the 1980s and has actually dropped down to 40% since 1990 and is typically averaged using income, background, ethnicity and two different age cohorts (52% before age 20, 34% if the couple marries aged 20-23).

If you marry after 30, your chances of divorce drop significantly (numbers are from 2004, but it's about 8%). (Figure 12, page 20)

Dr. John Gottman's "four horsemen" predict divorce with 94%-97% accuracy rate.

I could go on and on but cited enough links for y'all to dig into, if you're truly curious.

Dr. Helen Fisher, Lucy L. Brown, Arthur Aron and Bianca Acevedo published research showing that fMRI brain scans of happily married couples were correlative with initial findings on the same test subjects after 21 years. So what's the secret to a happy marriage?

Go get yourselves scanned at the Kinsey Institute to find out if you and your partner's brains light up with the love pattern like those awesome scans I linked to above, and if so, repeat annually. Apparently head-over-heels can last at LEAST 21 years, and I know Dr. Fisher plans on continuing her research until she (or all her test subjects) are no longer physically able to do the work. Her TED talks are pretty entertaining, too.
posted by Unicorn on the cob at 8:58 PM on October 16, 2014 [17 favorites]


If my answer seems random, I just wanted to throw some actual relationship science in this thread vs. just personal experiences (I love hearing people's stories, but they are often contradictory or unrelatable to others, for some reason) or the polarizing language from the Atlantic piece.

I watch people fall into assumptions and weird patterns about dating/attraction/partner expectations that are totally unrealistic based on bullshit blog fodder and Hollywood scripting, fairy tales, and watching their parents/grandparents/friends/community fall apart during divorce.

Tons of shitty books with shitty advice come out every year and become best-sellers. People have opinions, and some are great at selling the idea of happiness with some bullshit like "visualizing it!!!" a la The Secret.

But on the flipside, there's a formula to it if you look hard enough - you may not fit the formula, but understanding even a fraction of it might help make the difference between finding the partner who fits your specific relationship needs (and vice versa) and constantly repeating the same mistakes or just guessing blindly and failing.

Relationship science is still a young and misunderstood field because it's not exactly neuroscience, it's not exactly anthropology and it sure as hell isn't just sociology or psychology. Yet, it's seen by some as quackery; a macro view of the population's behavior, statistics, scientific studies, census data, economic and household trends, divorce and birth rates, etc. is a constantly moving animal.

Money, though? Yeah, it matters, but armed with enough actual knowledge on what works and what doesn't, you can hopefully avoid letting money become A Thing that destroys something as awe-inspiring as true love for a lifetime.
posted by Unicorn on the cob at 9:18 PM on October 16, 2014


Cost of a wedding and how many people attend the wedding are related almost linearly, only if the location is fixed. Which for a particular person it probably is, but not for all weddings.

In other words, it's entirely possible that a 10-person wedding in some really expensive venue (restaurant in NYC, say) to cost more than a 200-person wedding somewhere else (fire hall two hours outside Omaha).

So it could be that big, inexpensive weddings are for some reason more "successful" (for whatever definition of 'successful' you want to use), while small, expensive weddings are not. There could be a third variable in play, which is basically cost-per-person, that if we had access to it might show a better correlation. We don't really know.

Anecdotally, I could kinda see how this might work; the small, expensive wedding might force a lot of unpleasant choices ("who gets invited?") that cause stress, which is avoided in the large/cheap scenario. Or it could be an urban/rural thing, driven by other reasons.
posted by Kadin2048 at 12:05 AM on October 17, 2014


So I've had a look at that paper. As other's have noted, this is a self selected sample so one should take any results from it with an absolutely gigantic pinch of salt! They also cut out respondents older than 60 for no apparent reason? I would imagine happily married people at 70 might be quite important!

The paper is slightly confusingly written, while they mention it is a multivariate model, they seem to be implying that, for each factor, they included gender and then the factor of interest, and nothing else?

I hope not. As others have mentioned, some of these factors will be correlated, so they should all be put into the model at the same time. If they did do that, they don't seem to have attempted any kind of model selection (dropping factors which aren't statistically associated with the outcome), and they haven't included interactions (two factors working together, like wealth and cost of weddings), which is something they probably should have done.

Here's something that should worry one about this study. Gender was apparently predictive of divorce. I suppose this could be noticing that women tend to get remarried more often, although I'm not sure the arithmetic makes sense there.
posted by Cannon Fodder at 1:28 AM on October 17, 2014


The particularly scary part here is that the average cost of a wedding in the U.S. is well over $30,000

This statistic irks me to no end. I just want to scream that Average is not the best measure for this kind of data!!!!! Median is what you want. The average cost of a wedding is highly skewed by those who spend ungodly amounts of money. I looked at the paper linked above and they have a breakdown of wedding costs, grouped by amount (see the bottom of page 12). Fully 50% of weddings are <=$5,000, which would put the median just around $5,000. People can take a breath and relax now.

I am going to be emailing The Knot about this and hopefully they won't put out that godawful statistic next year and actually use the median.
posted by LizBoBiz at 6:06 AM on October 17, 2014 [6 favorites]


"Multivariate model" means all the variables are put in at the same time. They indeed should have thrown out useless variables and looked at interaction effects. The gender effect is strong evidence of a sampling bias and it was especially egregious for the "popularizing" pieces to treat it as an independent predictor of divorce (if you spend a lot of money on a wedding, the bride is somehow more likely to get divorced than her husband, Bridezilla=Divorcezilla hur hur).
posted by leopard at 6:08 AM on October 17, 2014 [4 favorites]


Leopard: I assume so but technically a multivariate model is merely a model with more than one independent variable in it, the way the results section is written confused me on whether they were simply controlling for gender each time.
posted by Cannon Fodder at 6:27 AM on October 17, 2014


They stratify the multivariate results by gender, but it's clear that the multivariate model includes all the controls (including gender). The bivariate results are the ones that don't have any controls.
posted by leopard at 8:03 AM on October 17, 2014


I had my own wedding and one of my best friends' in mind while reading this. My friend had about 250 people at her wedding. She had it on a family members' farm, her mom and bridal party (myself included) made almost all the food, etc. It was probably pretty cheap for the number of people.

In comparison, I had about 40 people at my wedding, but we had to rent a venue and get a caterer etc.

The difference was not that we wanted to be more "fancy", it was that our wedding was not LOCAL. We had recently moved and anyway, our friends and family were spread out all over the country. Most of them had to travel. We had no local connections to a free venue or any space to make/store homemade food. My friend had deep local connections, friend and family, to the place she got married.

I suspect this has something to do with the seemingly-inverse trends between money spent and # of attendees. If you managed to invite more people AND spend less money, then you likely had a deep local network of resources helping you with free or cheap stuff.

(Yes, I still get a little vexed that all the "budget wedding" sites go crazy about "have it in someone's yard! Have a potluck!" OK, how are all your friends in hotel rooms going to make/bring a dish?? Not all of us have family members with a nice big yard..)

(Of course I know that having a deep local network is likely ALSO a good predictor of a lasting marriage... but it would be more appropriate to contextualize that than to act like the frivolity of spending more money on your wedding causes your future divorce)
posted by nakedmolerats at 8:32 AM on October 17, 2014 [4 favorites]


A tonne of people showed up for our potluck wedding. Looks like we're together for good! Also, we only had a mini-moon until about six months later so I'm going to split the difference on that one. :)

Interestingly, the chart on combined household income reference point is $0-25k. So basically, living anywhere above the poverty line does a marriage good. Noted, thanks.
posted by flyingfox at 3:30 PM on October 17, 2014


« Older The Mother of All Supercuts   |   Why chemicals are like blenders, not margarita... Newer »


This thread has been archived and is closed to new comments