Half of all marriages - oh, wait
December 2, 2014 1:23 PM Subscribe
"It is no longer true that the divorce rate is rising, or that half of all marriages end in divorce. It has not been for some time. Despite hand-wringing about the institution of marriage, marriages in this country are stronger today than they have been in a long time. The divorce rate peaked in the 1970s and early 1980s and has been declining for the three decades since."
"Ultimately, a long view is likely to show that the rapid rise in divorce during the 1970s and early 1980s was an anomaly. It occurred at the same time as a new feminist movement, which caused social and economic upheaval. Today, society has adapted, and the divorce rate has declined again."
"Ultimately, a long view is likely to show that the rapid rise in divorce during the 1970s and early 1980s was an anomaly. It occurred at the same time as a new feminist movement, which caused social and economic upheaval. Today, society has adapted, and the divorce rate has declined again."
Of course the fact that people no longer feel feel obligated to get married is probably a big part of it.
posted by Tell Me No Lies at 1:31 PM on December 2, 2014 [36 favorites]
posted by Tell Me No Lies at 1:31 PM on December 2, 2014 [36 favorites]
Furthermore, a big contributor to the rise of divorce in the 70s (at least in the US) was the adoption of "no-fault divorces." This meant that couples could now get divorced for irreconcilable differences. From Wikipedia:
"Prior to the no-fault divorce revolution, a divorce was processed through the adversarial system as a civil action, meaning that a divorce could be obtained only through a showing of fault of one (and only one) of the parties in a marriage. This was something more than not loving one another; it meant that one spouse had to plead that the other had committed adultery, abandonment, felony, or other similarly culpable acts."
This meant that prior to this no-fault revolution, you had a lot of couples who stayed in unhappy marriages (i.e., my grandparents) because divorce involved a lot of judicial interaction. Once no-fault became a thing, many of those couples who had been staying together for those reasons were able to divorce, and and many did, resulting in a huge spike in divorce rates.
posted by cheeken at 1:31 PM on December 2, 2014 [45 favorites]
"Prior to the no-fault divorce revolution, a divorce was processed through the adversarial system as a civil action, meaning that a divorce could be obtained only through a showing of fault of one (and only one) of the parties in a marriage. This was something more than not loving one another; it meant that one spouse had to plead that the other had committed adultery, abandonment, felony, or other similarly culpable acts."
This meant that prior to this no-fault revolution, you had a lot of couples who stayed in unhappy marriages (i.e., my grandparents) because divorce involved a lot of judicial interaction. Once no-fault became a thing, many of those couples who had been staying together for those reasons were able to divorce, and and many did, resulting in a huge spike in divorce rates.
posted by cheeken at 1:31 PM on December 2, 2014 [45 favorites]
“Two-thirds of divorces are initiated by women” - an interesting statistic. I would like to see the data to back that up. The article doesn't say much about this. It is also said that men make out worse after a divorce, both financially and emotionally, though I have not seen data on this. The truth is that it leaves both sides financially impoverished, at least for a while. I believe there is quite a bit of data on this.
posted by McMillan's Other Wife at 1:34 PM on December 2, 2014
posted by McMillan's Other Wife at 1:34 PM on December 2, 2014
Tell Me No Lies that's definitely in the article, albeit in a puzzlingly hedged fashion: "Some of the decline in divorce clearly stems from the fact that fewer people are getting married — and some of the biggest declines in marriage have come among groups at risk of divorce."
posted by resurrexit at 1:35 PM on December 2, 2014 [1 favorite]
posted by resurrexit at 1:35 PM on December 2, 2014 [1 favorite]
For the less educated, divorce rates are closer to those of the peak divorce years. ... "As the middle of our labor market has eroded, the ability of high school-educated Americans to build a firm economic foundation for a marriage has been greatly reduced."Reminds me of some of the comments previously today.
posted by clawsoon at 1:41 PM on December 2, 2014 [2 favorites]
The state of our union is strong.
CHEERS! THUNDERING APPLAUSE!
Except for that one thing...
posted by blue_beetle at 1:42 PM on December 2, 2014
CHEERS! THUNDERING APPLAUSE!
Except for that one thing...
posted by blue_beetle at 1:42 PM on December 2, 2014
I've worked for divorce attorneys and anecdotally, I'd think the two-thirds of divorce being initiated by women is low. I think men make out better financially in most divorces, but worse emotionally. I don't know why women are more likely to file for divorce, but I've often thought that it's because they have more emotional support from friends and family or are just more willing to reach out to friends and family for support.
posted by readery at 1:46 PM on December 2, 2014 [4 favorites]
posted by readery at 1:46 PM on December 2, 2014 [4 favorites]
great now millenials cant even ruin their lives right... screw you, nyt
posted by boo_radley at 1:46 PM on December 2, 2014 [7 favorites]
posted by boo_radley at 1:46 PM on December 2, 2014 [7 favorites]
It is no longer true that the divorce rate is rising, or that half of all marriages end in divorce. It has not been for some time.
That's because it's stopped climbing at 100%. Everyone is divorced. Your parents are divorced. You were automatically divorced at birth.
In 1997, your great-great-grandparents were divorced retroactively by a splinter faction of The Church of Jesus Christ of Latter-day Saints, who went on to vicariously divorce Brigham Young from all 55 of his wives. There were 28 pleas of "irreconcilable differences," seventeen of "incompatibility;" eight citations of "irretrievable breakdown," and two of "unreasonable behavior."
posted by Iridic at 1:49 PM on December 2, 2014 [41 favorites]
That's because it's stopped climbing at 100%. Everyone is divorced. Your parents are divorced. You were automatically divorced at birth.
In 1997, your great-great-grandparents were divorced retroactively by a splinter faction of The Church of Jesus Christ of Latter-day Saints, who went on to vicariously divorce Brigham Young from all 55 of his wives. There were 28 pleas of "irreconcilable differences," seventeen of "incompatibility;" eight citations of "irretrievable breakdown," and two of "unreasonable behavior."
posted by Iridic at 1:49 PM on December 2, 2014 [41 favorites]
Lies, damn lies, and statistics.
http://www.the-spearhead.com/wp-content/uploads/2010/01/US-Marriage-Rate-1968-2008.jpg
About half as many people are getting married in the first place. People are just assuming they're going to get divorced in the future and are forgoing the marriage. And there's no way the authors of this piece don't know that.
Liars.
posted by effugas at 1:54 PM on December 2, 2014 [2 favorites]
http://www.the-spearhead.com/wp-content/uploads/2010/01/US-Marriage-Rate-1968-2008.jpg
About half as many people are getting married in the first place. People are just assuming they're going to get divorced in the future and are forgoing the marriage. And there's no way the authors of this piece don't know that.
Liars.
posted by effugas at 1:54 PM on December 2, 2014 [2 favorites]
I don't know why women are more likely to file for divorce, but I've often thought that it's because they have more emotional support from friends and family or are just more willing to reach out to friends and family for support.
I'd bet that it's more societal than that - we tend (in the US) to see men in a relationship that dissolves as having failed, so men are less willing to pull the trigger.
posted by NoxAeternum at 1:54 PM on December 2, 2014
I'd bet that it's more societal than that - we tend (in the US) to see men in a relationship that dissolves as having failed, so men are less willing to pull the trigger.
posted by NoxAeternum at 1:54 PM on December 2, 2014
Can you maybe not cite the Spearhead? It's an avowedly misogynistic forum.
posted by kewb at 1:58 PM on December 2, 2014 [19 favorites]
posted by kewb at 1:58 PM on December 2, 2014 [19 favorites]
It is (or was years ago when I got a divorce) a convention that men wanting a divorce often let the woman file for the divorce out of some notion of being gentlemanly, so as not to make the woman look bad--rather odd since a guy ending his marriage you would imagine would not give a hoot about hos the wife felt. But so it was, back then,
We are told that it now takes two incomes to raise a family or to maintain a marriage, and that might also suggest a certain reticence to divorce.
posted by Postroad at 2:03 PM on December 2, 2014 [1 favorite]
We are told that it now takes two incomes to raise a family or to maintain a marriage, and that might also suggest a certain reticence to divorce.
posted by Postroad at 2:03 PM on December 2, 2014 [1 favorite]
Effugas, what is being measured here is the divorce rate not the number of divorces. A rate is a fraction. In this case the denominator is the number of marriages. So there's no a priori reason why falling marriage rates should lead to falling divorce rates. If 1000 people got married in 1960 and 150 were divorced 10 years later,
there's no reason to assume that of 1000 people married in 2000, 15 should be divorced 10 years later. Both are equivalent divorce rates.
The rate of marriage only affects the divorce rate if the people not getting married are disproportionately the people likely to get divorced, as the authors state is the case (so they're not hiding anything with statistics). So this would suggest that the marriages not happening are precisely the inadvisable ones (if you make the assumption that marriages that end in divorce were in inadvisable -- I realize that this assumption is not fully warranted in some circumstances and is completely false in others).
posted by If only I had a penguin... at 2:04 PM on December 2, 2014 [29 favorites]
there's no reason to assume that of 1000 people married in 2000, 15 should be divorced 10 years later. Both are equivalent divorce rates.
The rate of marriage only affects the divorce rate if the people not getting married are disproportionately the people likely to get divorced, as the authors state is the case (so they're not hiding anything with statistics). So this would suggest that the marriages not happening are precisely the inadvisable ones (if you make the assumption that marriages that end in divorce were in inadvisable -- I realize that this assumption is not fully warranted in some circumstances and is completely false in others).
posted by If only I had a penguin... at 2:04 PM on December 2, 2014 [29 favorites]
effugas: the-spearhead.com"Opponents of its editor's political views regard it as an outlet for racist and neo-Nazi material, although Tyndall himself denied these accusations.... A former editor of the magazine (of which Spearhead had several, in addition to Tyndall himself), until Tyndall's split-off in 1980, was Richard Verrall, a noted Holocaust denier..." - Wikipedia
People who deride statistics are intellectually equivalent to people deriding evolution, vaccines, and other solidly-established sciences. It is the messenger that lies, not math.
posted by IAmBroom at 2:07 PM on December 2, 2014 [6 favorites]
I read a longitudinal study of divorcing couples once (sorry, citation long forgotten), and one of the findings was that usually in as a marriage breaks down both parties are seeing it break down and thinking it's not working out and as the breakdown continues, both parties are thinking this isn't working out and realizing that they're headed for divorce. They'll even talk openly about how they're headed toward divorce. But in the end, the court system is set up so that one party is the one who files. So somebody files, and in the study there wasn't necessarily any rhyme or reason to who files -- it wasn't the more aggrieved party, the one who had moved out, the one who was angriest, etc., it was just whoever happened to file first.
But then (here's the interesting part): If you asked them after the filing, the whole story of the marriages breakdown changed. Both parties would present a story of the disintegration of their marriage where the person filing was aggrieved in various ways, and wanted their freedom and just wanted to be done with and out, while the person who didn't file was painted as the dumpee, the person who was abandonned, who really felt the loss as something they hadn't chosen, etc. etc. And again, this is a change in story -- when these same people interviewed before filing, the person who was ultimately painted as the dumpee was as eager to get out, have their freedom and get divorced as the filer. The filer, in turn, was as broken-hearted over the death of their marriage as the eventual dumpee. It was just their accounts to themselves that retroactively changed.
Anyway, I read this study years ago and it may have been conducted years before that. Presumably the interpersonal dynamics of divorce have changed as the social view and context of divorce have changed, so who knows if this pattern still holds. It's interesting, though, and an interesting view on the problems of retroactive interviewing, also.
posted by If only I had a penguin... at 2:13 PM on December 2, 2014 [19 favorites]
But then (here's the interesting part): If you asked them after the filing, the whole story of the marriages breakdown changed. Both parties would present a story of the disintegration of their marriage where the person filing was aggrieved in various ways, and wanted their freedom and just wanted to be done with and out, while the person who didn't file was painted as the dumpee, the person who was abandonned, who really felt the loss as something they hadn't chosen, etc. etc. And again, this is a change in story -- when these same people interviewed before filing, the person who was ultimately painted as the dumpee was as eager to get out, have their freedom and get divorced as the filer. The filer, in turn, was as broken-hearted over the death of their marriage as the eventual dumpee. It was just their accounts to themselves that retroactively changed.
Anyway, I read this study years ago and it may have been conducted years before that. Presumably the interpersonal dynamics of divorce have changed as the social view and context of divorce have changed, so who knows if this pattern still holds. It's interesting, though, and an interesting view on the problems of retroactive interviewing, also.
posted by If only I had a penguin... at 2:13 PM on December 2, 2014 [19 favorites]
Any discussion that claims men fair worse than women in divorce needs to establish that this statement is not influenced by childcare. There is simply no equitable way to split childcare costs when one parent has primary custody; lost time from work and personal life taking care of children's various needs isn't recompensable.
Additionally, in many states a noncustodial parent's childcare payments are limited (in Pennsylvania, they max out at 50% of income). Obviously the upper limit on a custodial parent's expenses is above 100%, as medical costs can bankrupt US citizens, and others are forced into massive credit debt.
posted by IAmBroom at 2:14 PM on December 2, 2014 [3 favorites]
Additionally, in many states a noncustodial parent's childcare payments are limited (in Pennsylvania, they max out at 50% of income). Obviously the upper limit on a custodial parent's expenses is above 100%, as medical costs can bankrupt US citizens, and others are forced into massive credit debt.
posted by IAmBroom at 2:14 PM on December 2, 2014 [3 favorites]
In my 5:04 comment, that should be there's no reason to assume that 15 shouldn't be divorced 10 years later.
posted by If only I had a penguin... at 2:14 PM on December 2, 2014
posted by If only I had a penguin... at 2:14 PM on December 2, 2014
“Two-thirds of divorces are initiated by women” - an interesting statistic.
posted by McMillan's Other Wife at 1:34 PM on December 2 [+] [!]
obligatory eponysterical.
posted by We put our faith in Blast Hardcheese at 2:15 PM on December 2, 2014 [5 favorites]
posted by McMillan's Other Wife at 1:34 PM on December 2 [+] [!]
obligatory eponysterical.
posted by We put our faith in Blast Hardcheese at 2:15 PM on December 2, 2014 [5 favorites]
kewb,
Penguin,
The only reason the divorce rate is lower is because people aren't getting married in the first place. A broken up (or avoided) relationship just doesn't count as a divorce; the alternative is not marriage! The article is trying to argue "it's safe to get married, the high divorce rate is a myth". If they were dryly stating facts rather than making a case they'd declare the sharply reduced marriage number. Instead it's implied just as many people are getting married now, but they're marrying for love rather than economic benefit. Or something.
Bottom line, there's no way that graph at the top is remotely honest. Those sets are of very different sizes!
posted by effugas at 2:16 PM on December 2, 2014 [1 favorite]
The National Marriage Project at the University of Virginia publishes an annual report titled The State Of Our Unions which includes data on US marriage rates since 1960. From 1970 through 2008, the US marriage rate has declined from 76.5 to 37.4 marriages per 1,000 unmarried women.There. Now nobody needs to conflate the truth, with people they don't like.
Penguin,
The only reason the divorce rate is lower is because people aren't getting married in the first place. A broken up (or avoided) relationship just doesn't count as a divorce; the alternative is not marriage! The article is trying to argue "it's safe to get married, the high divorce rate is a myth". If they were dryly stating facts rather than making a case they'd declare the sharply reduced marriage number. Instead it's implied just as many people are getting married now, but they're marrying for love rather than economic benefit. Or something.
Bottom line, there's no way that graph at the top is remotely honest. Those sets are of very different sizes!
posted by effugas at 2:16 PM on December 2, 2014 [1 favorite]
I imagine all the marriages that fell apart in the 70s and 80s were ones that shouldn't have happened in the first place and when it became possible and practical for an unhappy spouse to get the hell out, they did.
The likely candidate for the "unhappy spouse" being the woman, since men hadn't been encouraged or forced to depend on marriage for a place in the world.
So, yay disruptive feminism!
(At least, that was probably part of it.)
posted by edheil at 2:21 PM on December 2, 2014 [10 favorites]
The likely candidate for the "unhappy spouse" being the woman, since men hadn't been encouraged or forced to depend on marriage for a place in the world.
So, yay disruptive feminism!
(At least, that was probably part of it.)
posted by edheil at 2:21 PM on December 2, 2014 [10 favorites]
There's nothing dishonest about the graph - it's plotting divorce rates, not total numbers of divorces. One could argue that divorce rate is not a meaningful statistic and that we should be looking at the number of divorces instead, but that doesn't make the graph dishonest - it just says that of the people who got married, a certain number got divorced. The reasons for that could be many -- and one is almost certainly that people that shouldn't have got married (and would have gotten married in say the 1960s) didn't. I don't think the article is trying to argue that it's now "safe to get married" and that we should throw caution to the winds when selecting potential spouses.
posted by peacheater at 2:23 PM on December 2, 2014 [3 favorites]
posted by peacheater at 2:23 PM on December 2, 2014 [3 favorites]
So there's no a priori reason why falling marriage rates should lead to falling divorce rates.
The 1,000 person sample isn't a random sample of the population, it's a sample of people who have been married at least once. If the population as a whole get's married less and the people who choose not to get married today are people who would have been more likely to get divorced in the past, then your 1,000 person sample is less likely to divorce.
I suppose another way to say it is that since there is less pressure to ever get married today the people most likely to divorce don't ever get married, there is some self-selection bias in the population of people who have been married.
posted by VTX at 2:23 PM on December 2, 2014 [5 favorites]
The 1,000 person sample isn't a random sample of the population, it's a sample of people who have been married at least once. If the population as a whole get's married less and the people who choose not to get married today are people who would have been more likely to get divorced in the past, then your 1,000 person sample is less likely to divorce.
I suppose another way to say it is that since there is less pressure to ever get married today the people most likely to divorce don't ever get married, there is some self-selection bias in the population of people who have been married.
posted by VTX at 2:23 PM on December 2, 2014 [5 favorites]
The article is trying to argue "it's safe to get married, the high divorce rate is a myth".
...er... is it? Because I thought it was very simply saying "hey, when people tell you half of all marriages end in divorce, that's not actually true anymore."
Isn't it a good thing if fewer people get married but those marriages are better?
posted by showbiz_liz at 2:24 PM on December 2, 2014 [11 favorites]
...er... is it? Because I thought it was very simply saying "hey, when people tell you half of all marriages end in divorce, that's not actually true anymore."
Isn't it a good thing if fewer people get married but those marriages are better?
posted by showbiz_liz at 2:24 PM on December 2, 2014 [11 favorites]
effugas, one could also argue that if you feel comfortable getting married in the current climate and have both married and unmarried peers, you are likely to not get divorced. Your assumption is that divorce occurs due to marriages happening, not marriages happening under the wrong circumstances.
If people are living in a culture that creates more ongoing marriages and the people considering marriage have parity with their married peers, then there's no reason their marriage wouldn't follow the trend. You are correct in saying any arbitrary couple couldn't have a successful marriage, but that isn't the point.
posted by mikeh at 2:26 PM on December 2, 2014 [1 favorite]
If people are living in a culture that creates more ongoing marriages and the people considering marriage have parity with their married peers, then there's no reason their marriage wouldn't follow the trend. You are correct in saying any arbitrary couple couldn't have a successful marriage, but that isn't the point.
posted by mikeh at 2:26 PM on December 2, 2014 [1 favorite]
Well, I'd quote the New York Times, but apparently they block that. It is true there's a paragraph where they say maybe marriages aren't starting in the first place. So there's that.
posted by effugas at 2:28 PM on December 2, 2014
posted by effugas at 2:28 PM on December 2, 2014
I imagine all the marriages that fell apart in the 70s and 80s were ones that shouldn't have happened in the first place and when it became possible and practical for an unhappy spouse to get the hell out, they did.
That would be my parents. I've never known two people with less in common than those two but they somehow managed to stay married for twenty-five years up until a few years after they left the Catholic church.
posted by octothorpe at 2:28 PM on December 2, 2014
That would be my parents. I've never known two people with less in common than those two but they somehow managed to stay married for twenty-five years up until a few years after they left the Catholic church.
posted by octothorpe at 2:28 PM on December 2, 2014
This seems to reinforce my belief that the divorce rate spiked because America shifted from marriage as an economic institution to one primarily rooted in happiness. It also explains why lower income folks have a higher divorce rate.
I'm not sure why we should read the article as saying that marriage is somehow safer. Instead it speaks to the affluence that women have gained in the US that they can delay marriage. Both parties who get married reap the benefit of prolonged matching costs.
posted by politikitty at 2:31 PM on December 2, 2014 [4 favorites]
I'm not sure why we should read the article as saying that marriage is somehow safer. Instead it speaks to the affluence that women have gained in the US that they can delay marriage. Both parties who get married reap the benefit of prolonged matching costs.
posted by politikitty at 2:31 PM on December 2, 2014 [4 favorites]
I know this is counter to conventional wisdom, but I (a married straight male for 21 years) actually have a pretty positive view of divorce. I say, THANK DOG for divorce. I think it is one of the best social institutions ever and that seeing it as a failure or a sign of moral decay is shortsighted and idiotic.
People change, and sometimes people don't change. Divorce means we don't have to remain in a relationship that is dysfunctional or violent or unfulfilling or just simply no longer works. Divorce means we can move and grow.
I concede that sometimes it may mean people bail on a relationship that might with work be the "best thing ever", and that it can be disruptive to children, that not all divorce is a reason for celebration. But imo the benefits vastly outweigh the negatives.
posted by edgeways at 2:32 PM on December 2, 2014 [16 favorites]
People change, and sometimes people don't change. Divorce means we don't have to remain in a relationship that is dysfunctional or violent or unfulfilling or just simply no longer works. Divorce means we can move and grow.
I concede that sometimes it may mean people bail on a relationship that might with work be the "best thing ever", and that it can be disruptive to children, that not all divorce is a reason for celebration. But imo the benefits vastly outweigh the negatives.
posted by edgeways at 2:32 PM on December 2, 2014 [16 favorites]
But can we blame Teh Gays ? I was told we could blame Teh Gays.
posted by Pogo_Fuzzybutt at 2:34 PM on December 2, 2014 [12 favorites]
posted by Pogo_Fuzzybutt at 2:34 PM on December 2, 2014 [12 favorites]
Effugas, you're missing the point of what a rate is. It's the number of divorces divided by the number of marriages. A person who doesn't get married isn't counted in the top OR the bottom of that fraction.
Para 1: So image 100 couples, 50 of them get divorced. Now image that two randomly selected of those couples never got married. On average, you'd expect that two randomly selected couples would be one who otherwise would have gotten divorced and who would not have gotten divorced. So those two couples don't get married. Now you have 49/98 getting divorced. Still 50% see? Cause that's how rates work. When people don't get married, they never get divorced, but they never end up in the denominator either.
Para 2: On the other hand, if the two couples who never got married in the first place were somehow disproportionately likely to be the couples who would have ended up divorced, then you would have a divorce rate of 48/98 which is a divorce rate lower than 50%. But it's not the declining marriage rate that did it, it's the fact that the people not getting married are the people likely to get divorced.
Para 3: If that pattern holds (the people who would get divorced don't get married) then you could have rising marriage rates and still falling divorce rates. So say in addition to those two couples who didn't marry above, 10 new couples think about getting married (5 of them would divorce). Of the 5 who would divorce, two of them decide not to marry. Now you have a divorce rate of 51/103. A lower divorce rate despite a rising marriage rate.
So in the previous three paragraphs, we examined three possibilities: Para 1: Falling Marriage rate only. Para 2: Falling Marriage Rate and Falling Rate of Inadvisable Marriages. Para 3: Falling rate of inadvisable marriage rate only. Only paragraphs 2 and 3 resulted in falling divorce rates. What do those two paragraphs have in common? Falling rate of inadvisable marriages. Paragraphs one and two have falling marriage rate in common, but obviously that cannot on its own cause a falling divorce rate as shown by the math in Para 1. And paragraph 3 showed how a falling rate of inadvisable marriages could cause a decline in divorce rates even with rising marriage rates.
In other words, falling marriage rates do not a priori mean falling divorce rates. Please see your grade 8 teacher if you have any questions.
posted by If only I had a penguin... at 2:40 PM on December 2, 2014
Para 1: So image 100 couples, 50 of them get divorced. Now image that two randomly selected of those couples never got married. On average, you'd expect that two randomly selected couples would be one who otherwise would have gotten divorced and who would not have gotten divorced. So those two couples don't get married. Now you have 49/98 getting divorced. Still 50% see? Cause that's how rates work. When people don't get married, they never get divorced, but they never end up in the denominator either.
Para 2: On the other hand, if the two couples who never got married in the first place were somehow disproportionately likely to be the couples who would have ended up divorced, then you would have a divorce rate of 48/98 which is a divorce rate lower than 50%. But it's not the declining marriage rate that did it, it's the fact that the people not getting married are the people likely to get divorced.
Para 3: If that pattern holds (the people who would get divorced don't get married) then you could have rising marriage rates and still falling divorce rates. So say in addition to those two couples who didn't marry above, 10 new couples think about getting married (5 of them would divorce). Of the 5 who would divorce, two of them decide not to marry. Now you have a divorce rate of 51/103. A lower divorce rate despite a rising marriage rate.
So in the previous three paragraphs, we examined three possibilities: Para 1: Falling Marriage rate only. Para 2: Falling Marriage Rate and Falling Rate of Inadvisable Marriages. Para 3: Falling rate of inadvisable marriage rate only. Only paragraphs 2 and 3 resulted in falling divorce rates. What do those two paragraphs have in common? Falling rate of inadvisable marriages. Paragraphs one and two have falling marriage rate in common, but obviously that cannot on its own cause a falling divorce rate as shown by the math in Para 1. And paragraph 3 showed how a falling rate of inadvisable marriages could cause a decline in divorce rates even with rising marriage rates.
In other words, falling marriage rates do not a priori mean falling divorce rates. Please see your grade 8 teacher if you have any questions.
posted by If only I had a penguin... at 2:40 PM on December 2, 2014
Isn't the falling divorce rate just another data point demonstrating how the (western) world is getting better? Life is better for everyone regardless of what the gloomy millennials may say. Crime is down, divorce is down, our kids are smarter, we are not as racist, nor homophobic, life expectancy is lengthening, murder is falling. Everything is better.
posted by Keith Talent at 2:42 PM on December 2, 2014 [5 favorites]
posted by Keith Talent at 2:42 PM on December 2, 2014 [5 favorites]
"In other words, falling marriage rates do not a priori mean falling divorce rates. Please see your grade 8 teacher if you have any questions."
Maybe your 10th grade teacher will cover sampling bias for you.
posted by klangklangston at 2:46 PM on December 2, 2014 [11 favorites]
Maybe your 10th grade teacher will cover sampling bias for you.
posted by klangklangston at 2:46 PM on December 2, 2014 [11 favorites]
> "I say, THANK DOG for divorce."
Agreed. I think all the decades of handwringing over divorce rates have made a lot of people forget that divorce is actually an important social justice issue.
posted by kyrademon at 2:47 PM on December 2, 2014 [13 favorites]
Agreed. I think all the decades of handwringing over divorce rates have made a lot of people forget that divorce is actually an important social justice issue.
posted by kyrademon at 2:47 PM on December 2, 2014 [13 favorites]
oops, on consultation with my grade 8 teacher, I see that the rate in Para 3 is 51/106.
posted by If only I had a penguin... at 2:50 PM on December 2, 2014
posted by If only I had a penguin... at 2:50 PM on December 2, 2014
Klang: Serious question: please explain. I wasn't thinking about sampling at all, but about population rates.
Do you mean that the people most likely to get divorced are the ones who don't end up in the "married sample"? That was my point. That that's what causes the declining divorce rates, not the falling marriage rates themselves. To put it in the sampling language: It's not because your sample is smaller (a sample smaller but still large enough to be reliable should produce the same estimated parameter), it's because some kinds of people are more likely to drop out of your sample than others. A larger sample with the same pattern in people dropping out of the sample would also find lower divorce rates.
posted by If only I had a penguin... at 2:54 PM on December 2, 2014 [1 favorite]
Do you mean that the people most likely to get divorced are the ones who don't end up in the "married sample"? That was my point. That that's what causes the declining divorce rates, not the falling marriage rates themselves. To put it in the sampling language: It's not because your sample is smaller (a sample smaller but still large enough to be reliable should produce the same estimated parameter), it's because some kinds of people are more likely to drop out of your sample than others. A larger sample with the same pattern in people dropping out of the sample would also find lower divorce rates.
posted by If only I had a penguin... at 2:54 PM on December 2, 2014 [1 favorite]
And again, this is a change in story -- when these same people interviewed before filing, the person who was ultimately painted as the dumpee was as eager to get out, have their freedom and get divorced as the filer.
It's not surprising that people's stories change, and I suspect post divorce stories may be more accurate. At the time it's difficulty to be clear on anything.
posted by Tell Me No Lies at 3:02 PM on December 2, 2014 [1 favorite]
It's not surprising that people's stories change, and I suspect post divorce stories may be more accurate. At the time it's difficulty to be clear on anything.
posted by Tell Me No Lies at 3:02 PM on December 2, 2014 [1 favorite]
People are just assuming they're going to get divorced in the future and are forgoing the marriage.
This seems like a pessimistic way of saying that people no longer feel pressured into marriage in the first place.
Think of the number of couples since 1968 who moved in together and realized it wasn't going to work out, then broke up and moved on. Think of the gay people who didn't feel the need to be closeted (though of course now some of them are starting to be represented in the married column because they are happily married to their partners, not because of the need for a beard). Think of the people who felt like it was perfectly normal not to be married by 25 or even 30, and so avoided "settling" just because it's what you do. Think of the people who were not raised to think that you should just marry whoever you're dating at 22, because that's how life works.
Instead, we have the freedom to divorce, and also the social space to be unmarried as adults. I can't see why that could possibly be a bad thing.
posted by Sara C. at 3:05 PM on December 2, 2014 [12 favorites]
This seems like a pessimistic way of saying that people no longer feel pressured into marriage in the first place.
Think of the number of couples since 1968 who moved in together and realized it wasn't going to work out, then broke up and moved on. Think of the gay people who didn't feel the need to be closeted (though of course now some of them are starting to be represented in the married column because they are happily married to their partners, not because of the need for a beard). Think of the people who felt like it was perfectly normal not to be married by 25 or even 30, and so avoided "settling" just because it's what you do. Think of the people who were not raised to think that you should just marry whoever you're dating at 22, because that's how life works.
Instead, we have the freedom to divorce, and also the social space to be unmarried as adults. I can't see why that could possibly be a bad thing.
posted by Sara C. at 3:05 PM on December 2, 2014 [12 favorites]
We should be happy with the divorce rate whatever it is. Nobody should be trapped in a shitty marriage so that society can pat itself on the back with irrelevant statistics.
posted by Talez at 3:07 PM on December 2, 2014 [17 favorites]
posted by Talez at 3:07 PM on December 2, 2014 [17 favorites]
Suppose you've got 100 marriages in 1970, and a 50% divorce rate. That leaves you with 50 marriages.
Suppose you've got 100 marriages in 2008, and a 15% divorce rate. That leaves you with 85 marriages.
Except you don't have 100 marriages in 2008. You have 50 marriages in 2008, and a 15% divorce rate. That leaves you with 43 marriages.
We can argue about whether 43 is significantly lower than 50 but it's not like everybody got all relaxed and chill and there's like 85 marriages now so go right ahead and get hitched. That is not reality. Placing all those populations on an equally scaled graphs really does imply that it is, and clucking that somebody should really have paid more attention in 8th grade doesn't make it any less of a lie.
posted by effugas at 3:15 PM on December 2, 2014 [2 favorites]
Suppose you've got 100 marriages in 2008, and a 15% divorce rate. That leaves you with 85 marriages.
Except you don't have 100 marriages in 2008. You have 50 marriages in 2008, and a 15% divorce rate. That leaves you with 43 marriages.
We can argue about whether 43 is significantly lower than 50 but it's not like everybody got all relaxed and chill and there's like 85 marriages now so go right ahead and get hitched. That is not reality. Placing all those populations on an equally scaled graphs really does imply that it is, and clucking that somebody should really have paid more attention in 8th grade doesn't make it any less of a lie.
posted by effugas at 3:15 PM on December 2, 2014 [2 favorites]
“Two-thirds of divorces are initiated by women” - an interesting statistic.
Just based on personal observation, a lot of men are very passive-aggressive about breaking up, including breaking up a marriage. They just fuck the relationship up until she breaks it off or files, so he doesn't have to be the bad guy.
posted by George_Spiggott at 3:17 PM on December 2, 2014 [15 favorites]
Just based on personal observation, a lot of men are very passive-aggressive about breaking up, including breaking up a marriage. They just fuck the relationship up until she breaks it off or files, so he doesn't have to be the bad guy.
posted by George_Spiggott at 3:17 PM on December 2, 2014 [15 favorites]
My former wife and I were a little annoyed that the courts required one of us to initiate the divorce. It was a mutual decision and we (metaphorically) walked into the room together to declare it. Still, since we had to choose I let her be the one to file it. I guess it did seem a bit more gentlemanly to me. We had no idea we were contributing to the statistical disparity.
posted by meinvt at 3:24 PM on December 2, 2014 [1 favorite]
posted by meinvt at 3:24 PM on December 2, 2014 [1 favorite]
People who deride statistics are intellectually equivalent to people deriding evolution, vaccines, and other solidly-established sciences.
Okay, I'll bite.
To quote Wikipedia, "Statistics is the study of the collection, analysis, interpretation, presentation, and organization of data."
Statistics does all of those pretty well except for interpretation; that makes sense as interpretation requires a deep understanding of the data being analyzed. Interpretation also informs what data you want to collect (analysis is just math).
Where statistics gets its dodgy reputation from is the interpretation phase. It's not strictly fair as it is not the statisticians who are going hog wild, but interpretation is where lying with statistics starts and trust in them stops. It is an area ripe for abuse and rife with abuse, and a good reason for any intelligent person to look cynically at whatever statistics they are presented with.
There's plenty of hard evidence that statistics are frequently twisted and turned and plain old lied about. To view them with suspicion is far more reasonable then assuming a picture being painted is true because at one time it was associated with data sets and mathematical analysis.
posted by Tell Me No Lies at 3:27 PM on December 2, 2014
Okay, I'll bite.
To quote Wikipedia, "Statistics is the study of the collection, analysis, interpretation, presentation, and organization of data."
Statistics does all of those pretty well except for interpretation; that makes sense as interpretation requires a deep understanding of the data being analyzed. Interpretation also informs what data you want to collect (analysis is just math).
Where statistics gets its dodgy reputation from is the interpretation phase. It's not strictly fair as it is not the statisticians who are going hog wild, but interpretation is where lying with statistics starts and trust in them stops. It is an area ripe for abuse and rife with abuse, and a good reason for any intelligent person to look cynically at whatever statistics they are presented with.
There's plenty of hard evidence that statistics are frequently twisted and turned and plain old lied about. To view them with suspicion is far more reasonable then assuming a picture being painted is true because at one time it was associated with data sets and mathematical analysis.
posted by Tell Me No Lies at 3:27 PM on December 2, 2014
We should be happy with the divorce rate whatever it is. Nobody should be trapped in a shitty marriage so that society can pat itself on the back with irrelevant statistics.
Amen.
posted by Tell Me No Lies at 3:28 PM on December 2, 2014 [2 favorites]
Amen.
posted by Tell Me No Lies at 3:28 PM on December 2, 2014 [2 favorites]
I wonder if the filing thing is also a reflection of household labour work - in that in most relationships I know, women file paperwork. They may not make the ultimate decision, but they're the ones sending stuff off to the insurance company, to the school, to the government (even for family and tax purposes) so it just follows they'd be doing the paperwork for divorce.
posted by geek anachronism at 3:30 PM on December 2, 2014 [6 favorites]
posted by geek anachronism at 3:30 PM on December 2, 2014 [6 favorites]
Effugas, so your thinking we should be trying to maximize the absolute number of intact marriages. OK, if that's your goal, these graphs can't help you figure out if that's happening. But I don't think most other people see that (maximum number of intact marriages) as a goal we should he pursuing, so nobody else is using these stats to evaluate reality against that goal. The graphs aren'y misleading. They're just not speaking or intending to speak to the absolute number of intact marriages. Nobody but you is attempting to interpret them as speaking to that.
I don't think maxi.using the number of intact marriages is a worthy goal, but I can imagine reasons why one might think it is. Reasonable people can disagree. If you would like to know how we're doing on that goal, head over to census.gov, find the marital status question, multiply the proportion by the US population and divide by two. That's a decent measure of the number of intact marriages. I'm sure its declined and still declining. It would probably be interesting to read some thoughtful analyses of the implications/consequences of this, but it's a whole other topic from the one covered in the article. Graphs aren't misleading just because they don't answer the question you're interested in, unless they claim to answer that question. This one doesn't.
posted by If only I had a penguin... at 3:30 PM on December 2, 2014 [10 favorites]
I don't think maxi.using the number of intact marriages is a worthy goal, but I can imagine reasons why one might think it is. Reasonable people can disagree. If you would like to know how we're doing on that goal, head over to census.gov, find the marital status question, multiply the proportion by the US population and divide by two. That's a decent measure of the number of intact marriages. I'm sure its declined and still declining. It would probably be interesting to read some thoughtful analyses of the implications/consequences of this, but it's a whole other topic from the one covered in the article. Graphs aren't misleading just because they don't answer the question you're interested in, unless they claim to answer that question. This one doesn't.
posted by If only I had a penguin... at 3:30 PM on December 2, 2014 [10 favorites]
To me, the most interesting bit was not the nitpicking about the exact statistics, but about the class implications of how marriage rates intersect with our increasingly grossly unequal society:
The marriage trends aren’t entirely happy ones. They also happen to be a force behind rising economic and social inequality, because the decline in divorce is concentrated among people with college degrees. For the less educated, divorce rates are closer to those of the peak divorce years. ...
Overall, the marriage trends resemble those in many other areas of American life. For people on the wealthier side of the class divide, life is better than it used to be in many ways. For people on the other side, the situation is much more complicated.
The people dropping out of the marriage game are the same people who are being royally hosed economically, and those are connected. And at the upper/middle end of things, I am sure there are a lot of two-earner marriages where a divorce would mean a precipitous decline in living standards for both people -- many, if not most, middle class families don't have deep reserves and couldn't easily sustain two households on their current income.
There's a lot more to the economics of this than this short article touches on, though the actual research they are summarizing hopefully looks at it deeper.
posted by Dip Flash at 3:33 PM on December 2, 2014 [4 favorites]
The marriage trends aren’t entirely happy ones. They also happen to be a force behind rising economic and social inequality, because the decline in divorce is concentrated among people with college degrees. For the less educated, divorce rates are closer to those of the peak divorce years. ...
Overall, the marriage trends resemble those in many other areas of American life. For people on the wealthier side of the class divide, life is better than it used to be in many ways. For people on the other side, the situation is much more complicated.
The people dropping out of the marriage game are the same people who are being royally hosed economically, and those are connected. And at the upper/middle end of things, I am sure there are a lot of two-earner marriages where a divorce would mean a precipitous decline in living standards for both people -- many, if not most, middle class families don't have deep reserves and couldn't easily sustain two households on their current income.
There's a lot more to the economics of this than this short article touches on, though the actual research they are summarizing hopefully looks at it deeper.
posted by Dip Flash at 3:33 PM on December 2, 2014 [4 favorites]
it meant that one spouse had to plead that the other had committed adultery, abandonment, felony, or other similarly culpable acts." no idea why I remember this but apparently the cause cited in Cher's divorce from Sonny Bono was "involuntary servitude."
posted by The Whelk at 3:36 PM on December 2, 2014 [1 favorite]
posted by The Whelk at 3:36 PM on December 2, 2014 [1 favorite]
I always figure women initiate divorce because they end up doing most of the emotional work in a marriage and keeping a relationship together and just end up realizing it's not worth it to feel neglected and disconnected from a partner who doesn't/won't participate.
posted by discopolo at 4:07 PM on December 2, 2014 [5 favorites]
posted by discopolo at 4:07 PM on December 2, 2014 [5 favorites]
Maybe your 10th grade teacher will cover sampling bias for you.
I'm not seeing what sampling bias has to do with this, either. The parameter we're trying to estimate about this population (divorce rate) pertains to the population of married people. It doesn't affect our estimate of the parameter just that this population has decreased, although it probably affects our level of confidence in that estimate by increasing the standard error. To argue for sampling bias, you'd have to talk about a specific problem in the study design that led to bias in the selection criteria - just changing the size of the population doesn't do it unless you're doing something weird like pooling results or trying to do a meta-analysis.
Anyway, seconding If I only had a penguin... that people are trying to read more of a narrative about people getting married than is stated in the article or the data. Effugas has done a good job of demonstrating why the divorce rate doesn't tell us anything about the number of married people, but nobody has actually laid a finger on the thesis or analysis of this article, which is solely concerned with divorce rates, marriage duration, and some speculation about why the divorce rate has decreased. This article is about divorce and the population of married people, and specifically about demonstrating that our cultural narrative about "rising divorce rates" is false.
If you want to construct a counterargument that the divorce rate is a red herring if we're concerned with the state of marriage and that we should include marriage rates and the size of the married population as well as divorce rates, that is great and I would completely agree with you. If you want to argue that the smaller group of people who are getting married these days tend to have more stable relationships because they're getting married by choice instead of by economic necessity, I'd agree with you there too. But arguing that a graph of divorce rates which was intended specifically to disprove the claim "the divorce rate is rising" is somehow making claims about the size of the married population is specious.
posted by dialetheia at 4:41 PM on December 2, 2014 [3 favorites]
I'm not seeing what sampling bias has to do with this, either. The parameter we're trying to estimate about this population (divorce rate) pertains to the population of married people. It doesn't affect our estimate of the parameter just that this population has decreased, although it probably affects our level of confidence in that estimate by increasing the standard error. To argue for sampling bias, you'd have to talk about a specific problem in the study design that led to bias in the selection criteria - just changing the size of the population doesn't do it unless you're doing something weird like pooling results or trying to do a meta-analysis.
Anyway, seconding If I only had a penguin... that people are trying to read more of a narrative about people getting married than is stated in the article or the data. Effugas has done a good job of demonstrating why the divorce rate doesn't tell us anything about the number of married people, but nobody has actually laid a finger on the thesis or analysis of this article, which is solely concerned with divorce rates, marriage duration, and some speculation about why the divorce rate has decreased. This article is about divorce and the population of married people, and specifically about demonstrating that our cultural narrative about "rising divorce rates" is false.
If you want to construct a counterargument that the divorce rate is a red herring if we're concerned with the state of marriage and that we should include marriage rates and the size of the married population as well as divorce rates, that is great and I would completely agree with you. If you want to argue that the smaller group of people who are getting married these days tend to have more stable relationships because they're getting married by choice instead of by economic necessity, I'd agree with you there too. But arguing that a graph of divorce rates which was intended specifically to disprove the claim "the divorce rate is rising" is somehow making claims about the size of the married population is specious.
posted by dialetheia at 4:41 PM on December 2, 2014 [3 favorites]
So it turns out no one is lying with statistics, people just have no fucking clue how to understand them.
posted by MisantropicPainforest at 4:52 PM on December 2, 2014 [5 favorites]
posted by MisantropicPainforest at 4:52 PM on December 2, 2014 [5 favorites]
I'm not seeing what sampling bias has to do with this, either.
The "Married in the 70s" sample probably contains a decent number of people who didn't want to have sex before marriage, and therefore got married.
The "Married in the 90s" sample probably doesn't contain nearly as many people who got married out of a desire to have sex.
It doesn't seem crazy to me to think that those 2 scenarios have different outcomes. It does seem like it would be a more apples-to-apples comparison if the 90s group included couples who lived together for some time, >2 years, for example.
posted by selfmedicating at 4:56 PM on December 2, 2014 [1 favorite]
The "Married in the 70s" sample probably contains a decent number of people who didn't want to have sex before marriage, and therefore got married.
The "Married in the 90s" sample probably doesn't contain nearly as many people who got married out of a desire to have sex.
It doesn't seem crazy to me to think that those 2 scenarios have different outcomes. It does seem like it would be a more apples-to-apples comparison if the 90s group included couples who lived together for some time, >2 years, for example.
posted by selfmedicating at 4:56 PM on December 2, 2014 [1 favorite]
Changes in the divorce rate are interesting insofar as they can be used to describe changes in marriage culture and predict the likely result of the choice to marry. The rate can be used to make inferences about social changes, but as the sample becomes more self-selecting and less representative of the general population, its predictive value declines.
posted by Svejk at 4:59 PM on December 2, 2014
posted by Svejk at 4:59 PM on December 2, 2014
That has nothing to do with sampling bias. Like at all.
posted by MisantropicPainforest at 4:59 PM on December 2, 2014
posted by MisantropicPainforest at 4:59 PM on December 2, 2014
effugas, I think you have the right way of looking at it, but are concentrating on the wrong end.
Now, I am assuming you're seeing marriage as a good. Which, in most cases, it likely is -- now so more than ever, as people are sticking with it. Not getting married is not essentially bad. It can mean people decided not to stick with a dysfunctional relationship (emotional or physical abuse, lack of long-term compatibility) or that they approached their situation more responsibly without the social pressures of marriage.
I know people will immediately point to children and property, the two social pillars of marriage (as government is love-blind), but would you rather have friends who had a tumultuous marriage, decided to have a kid to see if it'd work, then got divorced among disagreement, or two people who had a loving relationship and had a kid without getting married -- because they wanted to have a child -- with the acknowledgment that they might not be together forever, with discussion pre-child of what them splitting would mean for the kid.
Obviously there are many cases in-between, but the latter would be unthinkable in traditional marriage-based society. And I've seen a lot of the former, with friends and acquaintances who felt it was their social responsibility to get married, have kids, and stick with marriage.
posted by mikeh at 5:22 PM on December 2, 2014 [1 favorite]
Suppose you've got 100 marriages in 1970, and a 50% divorce rate. That leaves you with 50 marriages.Look at it this way -- rather than fifty marriages that people decided didn't work, you have seven. That doesn't mean that there aren't kids, it doesn't mean the relationships were perfect, it only means that fewer people entered a legal and social contract that they later decided was not right.
Suppose you've got 100 marriages in 2008, and a 15% divorce rate. That leaves you with 85 marriages.
Except you don't have 100 marriages in 2008. You have 50 marriages in 2008, and a 15% divorce rate. That leaves you with 43 marriages.
Now, I am assuming you're seeing marriage as a good. Which, in most cases, it likely is -- now so more than ever, as people are sticking with it. Not getting married is not essentially bad. It can mean people decided not to stick with a dysfunctional relationship (emotional or physical abuse, lack of long-term compatibility) or that they approached their situation more responsibly without the social pressures of marriage.
I know people will immediately point to children and property, the two social pillars of marriage (as government is love-blind), but would you rather have friends who had a tumultuous marriage, decided to have a kid to see if it'd work, then got divorced among disagreement, or two people who had a loving relationship and had a kid without getting married -- because they wanted to have a child -- with the acknowledgment that they might not be together forever, with discussion pre-child of what them splitting would mean for the kid.
Obviously there are many cases in-between, but the latter would be unthinkable in traditional marriage-based society. And I've seen a lot of the former, with friends and acquaintances who felt it was their social responsibility to get married, have kids, and stick with marriage.
posted by mikeh at 5:22 PM on December 2, 2014 [1 favorite]
Guys. It's an advocacy piece, responding to people's fears of divorce by saying, heh, it's better now, look how much safer it is to get married nowadays. That's the subtext of the piece.
Let me translate this to another domain: Let's say back in the 70's, half of everyone who ate 4000 calorie steak and potato dinners every night died of a heart attack. But now, only 15% do.
Would you say, apparently 4000 calorie steak and potato dinners are safer than they used to be? Or would you ask, did everyone who is not athletic enough to burn off the calories actually stop eating 4000 calorie steak and potato dinners entirely, leaving only the fitness nuts?
More to the point, if you're not athletic, does it make it any safer for you to go out and eat like an athlete, just because those people are all doing fine?
Dropping the metaphor, I actually think the focus on family and marriage is one of the few ways society still manages to try to guide people towards being happier and less lonely. (And, you know, more financially stable, motivated to work, and creating the next generation, but you know. Happy and not lonely too.) I'm just bothered by the "declaration of success" this very skewed reading of the numbers reflects. Society hasn't managed to increase advisable marriages, just to decrease inadvisable ones. Maybe that's something.
posted by effugas at 5:27 PM on December 2, 2014
Let me translate this to another domain: Let's say back in the 70's, half of everyone who ate 4000 calorie steak and potato dinners every night died of a heart attack. But now, only 15% do.
Would you say, apparently 4000 calorie steak and potato dinners are safer than they used to be? Or would you ask, did everyone who is not athletic enough to burn off the calories actually stop eating 4000 calorie steak and potato dinners entirely, leaving only the fitness nuts?
More to the point, if you're not athletic, does it make it any safer for you to go out and eat like an athlete, just because those people are all doing fine?
Dropping the metaphor, I actually think the focus on family and marriage is one of the few ways society still manages to try to guide people towards being happier and less lonely. (And, you know, more financially stable, motivated to work, and creating the next generation, but you know. Happy and not lonely too.) I'm just bothered by the "declaration of success" this very skewed reading of the numbers reflects. Society hasn't managed to increase advisable marriages, just to decrease inadvisable ones. Maybe that's something.
posted by effugas at 5:27 PM on December 2, 2014
The divorce rate looks like a nested conditional, conditioned on the probability of marriage. Changes in both rates are interesting for different reasons. I think it is relevant whether you assume the general population is using the divorce rate to estimate the expected outcome of marriage.
posted by Svejk at 5:31 PM on December 2, 2014
posted by Svejk at 5:31 PM on December 2, 2014
mikeh, pretty much entirely agree with you, down to the 7/50 number being significant. You'd have written a much fairer piece.
posted by effugas at 5:32 PM on December 2, 2014 [1 favorite]
posted by effugas at 5:32 PM on December 2, 2014 [1 favorite]
Amusingly, there's all sorts of sloppiness in this space. Saw a great article about where all the single people were, based on marriage stats from the census. As if not being married meant being single.
posted by effugas at 5:35 PM on December 2, 2014
posted by effugas at 5:35 PM on December 2, 2014
Can you point to a sentence in the article that you think is false or is it just a general feeling you have?
posted by MisantropicPainforest at 6:07 PM on December 2, 2014 [2 favorites]
posted by MisantropicPainforest at 6:07 PM on December 2, 2014 [2 favorites]
The "Married in the 70s" sample probably contains a decent number of people who didn't want to have sex before marriage, and therefore got married.
The "Married in the 90s" sample probably doesn't contain nearly as many people who got married out of a desire to have sex.
In that case, either you're supplying an explanation for the change in the divorce rate, which is fine but doesn't make the statistic at all misleading (and which explanation is actually well-covered in the last part of the article), or you're arguing that you are measuring two different populations and the rates shouldn't be compared at all. Which, I guess maybe, but it seems reasonable to compare people in the 60s and 70s to people today even if our attitudes have changed a lot.
Guys. It's an advocacy piece, responding to people's fears of divorce by saying, heh, it's better now, look how much safer it is to get married nowadays. That's the subtext of the piece.
I've reread it twice now and I still disagree. They're saying, "if you decide to get married, you're more likely to stay together now than you were if you got married in the sixties, and people who cite rising divorce rates are incorrect." They cover all sorts of possible reasons for the change in divorce rates: people marrying for love instead of economics, people marrying later, people not having the time or money to marry, socioeconomic factors, feminism and women working full-time, people living together before they get married, etc. But they aren't saying marriage as an institution is on the rise or anything - in fact they even state that directly in the piece:
Some of the decline in divorce clearly stems from the fact that fewer people are getting married — and some of the biggest declines in marriage have come among groups at risk of divorce.
Effugas: Society hasn't managed to increase advisable marriages, just to decrease inadvisable ones. Maybe that's something.
Exactly! That's exactly the point.
posted by dialetheia at 6:16 PM on December 2, 2014 [5 favorites]
The "Married in the 90s" sample probably doesn't contain nearly as many people who got married out of a desire to have sex.
In that case, either you're supplying an explanation for the change in the divorce rate, which is fine but doesn't make the statistic at all misleading (and which explanation is actually well-covered in the last part of the article), or you're arguing that you are measuring two different populations and the rates shouldn't be compared at all. Which, I guess maybe, but it seems reasonable to compare people in the 60s and 70s to people today even if our attitudes have changed a lot.
Guys. It's an advocacy piece, responding to people's fears of divorce by saying, heh, it's better now, look how much safer it is to get married nowadays. That's the subtext of the piece.
I've reread it twice now and I still disagree. They're saying, "if you decide to get married, you're more likely to stay together now than you were if you got married in the sixties, and people who cite rising divorce rates are incorrect." They cover all sorts of possible reasons for the change in divorce rates: people marrying for love instead of economics, people marrying later, people not having the time or money to marry, socioeconomic factors, feminism and women working full-time, people living together before they get married, etc. But they aren't saying marriage as an institution is on the rise or anything - in fact they even state that directly in the piece:
Some of the decline in divorce clearly stems from the fact that fewer people are getting married — and some of the biggest declines in marriage have come among groups at risk of divorce.
Effugas: Society hasn't managed to increase advisable marriages, just to decrease inadvisable ones. Maybe that's something.
Exactly! That's exactly the point.
posted by dialetheia at 6:16 PM on December 2, 2014 [5 favorites]
Furthermore, a big contributor to the rise of divorce in the 70s (at least in the US) was the adoption of "no-fault divorces."
Oddly just today I was reading about divorce laws in my country and I raised an eyebrow at this:
posted by ricochet biscuit at 6:55 PM on December 2, 2014 [1 favorite]
Oddly just today I was reading about divorce laws in my country and I raised an eyebrow at this:
Prior to 1968 there was no federal divorce law in Canada. In Newfoundland and Quebec, where there was no provincial divorce legislation either, persons had to seek the passage of a private Act of Parliament in order to end their marriages.As a Canadian who has been divorced, I am very glad it did not require an Act of Parliament.
posted by ricochet biscuit at 6:55 PM on December 2, 2014 [1 favorite]
It's a good thing I'm not a statistician, since I look at those curves on the graph and think "huh, they all look pretty much the same to me."
posted by Lexica at 7:08 PM on December 2, 2014
posted by Lexica at 7:08 PM on December 2, 2014
dialetheia,
"if you decide to get married, you're more likely to stay together now than you were if you got married in the sixties, and people who cite rising divorce rates are incorrect."
Exactly. If _you_ decide to get married, you're more likely to stay together. The divorce rate problem is gone, here's why.
Except it's not gone. Divorces just now happen before marriages, i.e. there's pre-marriage breakup, and maybe even suppression of the initial relationship.
Suppose ten people got married, that wouldn't otherwise have. Would they experience the 15% failure rate or the 50% failure rate?
I admit there's a fig leaf with the one paragraph that talks about fewer people getting married, but it's not a _small_ number fewer that only applies to a few groups. The _entire_ population that used to get divorced is covered by the population that doesn't get married in the first place. That's actually amazing.
I think Svejk has it right: " I think it is relevant whether you assume the general population is using the divorce rate to estimate the expected outcome of marriage."
It's a reasonable assumption, and it's what the article is responsive to.
posted by effugas at 7:13 PM on December 2, 2014
"if you decide to get married, you're more likely to stay together now than you were if you got married in the sixties, and people who cite rising divorce rates are incorrect."
Exactly. If _you_ decide to get married, you're more likely to stay together. The divorce rate problem is gone, here's why.
Except it's not gone. Divorces just now happen before marriages, i.e. there's pre-marriage breakup, and maybe even suppression of the initial relationship.
Suppose ten people got married, that wouldn't otherwise have. Would they experience the 15% failure rate or the 50% failure rate?
I admit there's a fig leaf with the one paragraph that talks about fewer people getting married, but it's not a _small_ number fewer that only applies to a few groups. The _entire_ population that used to get divorced is covered by the population that doesn't get married in the first place. That's actually amazing.
I think Svejk has it right: " I think it is relevant whether you assume the general population is using the divorce rate to estimate the expected outcome of marriage."
It's a reasonable assumption, and it's what the article is responsive to.
posted by effugas at 7:13 PM on December 2, 2014
Divorces just now happen before marriages
That's not possible. Unless you redefine marriage to be a relationship, then you just called one thing another.
If _you_ decide to get married, you're more likely to stay together.
This is exactly the point of the article.
B=married
A=divorced
P(A|B')=0
P(A|B)= lower than it used to be.
The probability of B isn't the thing we are concerned with, we are concerned with the probability of A.
posted by MisantropicPainforest at 7:25 PM on December 2, 2014 [4 favorites]
That's not possible. Unless you redefine marriage to be a relationship, then you just called one thing another.
If _you_ decide to get married, you're more likely to stay together.
This is exactly the point of the article.
B=married
A=divorced
P(A|B')=0
P(A|B)= lower than it used to be.
The probability of B isn't the thing we are concerned with, we are concerned with the probability of A.
posted by MisantropicPainforest at 7:25 PM on December 2, 2014 [4 favorites]
Divorces are a special case of breakup, as marriages are a special case of relationship. Breakups can and do happen before marriages; I know more than a few people deeply hurt that their 3 to 5 year relationship did not yield marriage. If you think they are not feeling a subset of the pain of a divorce you don't know people.
Here's the bottom line question: Do you believe the nature of marriages has changed, such that relationships that previously would have led to divorce now will not? Because that's a thing the article is definitely trying to propose, when actually the entire differential is explained by a very specific sort of sampling bias: Survivorship bias.
posted by effugas at 7:46 PM on December 2, 2014
Here's the bottom line question: Do you believe the nature of marriages has changed, such that relationships that previously would have led to divorce now will not? Because that's a thing the article is definitely trying to propose, when actually the entire differential is explained by a very specific sort of sampling bias: Survivorship bias.
posted by effugas at 7:46 PM on December 2, 2014
"if you decide to get married, you're more likely to stay together now than you were if you got married in the sixties"
What you're missing here is that, if this data and your interpretations are correct, it's more likely now than in the sixties that I won't decide to get married unless I'm in a relationship that is likely to stay together.
So even assuming your criticisms are correct, it is true that if you decide to get married, you're more likely to stay together now than you were if you got married in the sixties.
posted by straight at 8:24 PM on December 2, 2014
What you're missing here is that, if this data and your interpretations are correct, it's more likely now than in the sixties that I won't decide to get married unless I'm in a relationship that is likely to stay together.
So even assuming your criticisms are correct, it is true that if you decide to get married, you're more likely to stay together now than you were if you got married in the sixties.
posted by straight at 8:24 PM on December 2, 2014
Wow, I've heard the expression "half of all marriages end in divorce" tons of times and I don't think anyone ever nitpicked it to this extent. Despite the fact that it's never even been true.
posted by leopard at 8:49 PM on December 2, 2014 [1 favorite]
posted by leopard at 8:49 PM on December 2, 2014 [1 favorite]
As a Canadian who has been divorced, I am very glad it did not require an Act of Parliament.
I dunno..It might have been neat to have your divorce decree signed by the governor general. Or, if you timed it right so that she was in town, you might even get it signed by the queen! I'm no monarchist, but that would be pretty cool.
posted by If only I had a penguin... at 8:53 PM on December 2, 2014
I dunno..It might have been neat to have your divorce decree signed by the governor general. Or, if you timed it right so that she was in town, you might even get it signed by the queen! I'm no monarchist, but that would be pretty cool.
posted by If only I had a penguin... at 8:53 PM on December 2, 2014
Honestly, this is not about fucking survivorship bias. This is showing off the NYT's highly skilled data team by putting them to work on the widely discussed topic of divorce rates, which are often falsely believed to be rising. The piece clearly points out that "Some of the decline in divorce clearly stems from the fact that fewer people are getting married — and some of the biggest declines in marriage have come among groups at risk of divorce." I haven't dug into the figures, but presumably this is based on analysis indicating that at least *some* of the decline is based on something other than falling marriage rates.
When people talk about the divorce rate, they are talking about the divorce rate. Declining rates of marriage are a related but separate topic. Yes, there are a lot of complicated factors in play, which is why the article is more than two sentences long. You may as well rail about how every marriage is unique so the very idea of looking at statistics to understand this topic is a Big Lie.
posted by leopard at 8:59 PM on December 2, 2014 [4 favorites]
When people talk about the divorce rate, they are talking about the divorce rate. Declining rates of marriage are a related but separate topic. Yes, there are a lot of complicated factors in play, which is why the article is more than two sentences long. You may as well rail about how every marriage is unique so the very idea of looking at statistics to understand this topic is a Big Lie.
posted by leopard at 8:59 PM on December 2, 2014 [4 favorites]
Do you believe the nature of marriages has changed, such that relationships that previously would have led to divorce now will not?
Yes, obviously something has changed. Even if your theory explains 100% of the change (and I'm pretty sure it doesn't), the fact that couples who are more likely to break up are now less likely to get married is a change in the nature of marriage. Yes, it counts.
What you're really arguing is that marriage and divorce rates are irrelevant because "marriage" is just a piece of paper and a label with no real-world impact, and what really matters is "relationship status" changes, and you're pretty sure that those are stable over time, and that this research into changes in marriage/divorce rates is really just investigating relationship labeling conventions. And this is pretty obviously untrue, because divorce rates rose rapidly in the 1970s and it wasn't because marriage rates were skyrocketing at the time. So yeah, marriage and divorce are actual things.
posted by leopard at 9:11 PM on December 2, 2014 [6 favorites]
Yes, obviously something has changed. Even if your theory explains 100% of the change (and I'm pretty sure it doesn't), the fact that couples who are more likely to break up are now less likely to get married is a change in the nature of marriage. Yes, it counts.
What you're really arguing is that marriage and divorce rates are irrelevant because "marriage" is just a piece of paper and a label with no real-world impact, and what really matters is "relationship status" changes, and you're pretty sure that those are stable over time, and that this research into changes in marriage/divorce rates is really just investigating relationship labeling conventions. And this is pretty obviously untrue, because divorce rates rose rapidly in the 1970s and it wasn't because marriage rates were skyrocketing at the time. So yeah, marriage and divorce are actual things.
posted by leopard at 9:11 PM on December 2, 2014 [6 favorites]
Here is a link to the Wolfers/Stevenson paper on marriage and divorce that is also linked in the NYT article.
Abstract:
Abstract:
We document key facts about marriage and divorce, comparing trends through the past 150 years and outcomes across demographic groups and countries. While divorce rates have risen over the past 150 years, they have been falling for the past quarter century. Marriage rates have also been falling, but more strikingly, the importance of marriage at different points in the life cycle has changed, reflecting rising age at first marriage, rising divorce followed by high remarriage rates, and a combination of increased longevity with a declining age gap between husbands and wives. Cohabitation has also become increasingly important, emerging as a widely used step on the path to marriage. Out-of-wedlock fertility has also risen, consistent with declining "shotgun marriages". Compared with other countries, marriage maintains a central role in American life. We present evidence on some of the driving forces causing these changes in the marriage market: the rise of the birth control pill and women's control over their own fertility; sharp changes in wage structure, including a rise in inequality and partial closing of the gender wage gap; dramatic changes in home production technologies; and the emergence of the internet as a new matching technology. We note that recent changes in family forms demand a reassessment of theories of the family and argue that consumption complementarities may be an increasingly important component of marriage. Finally, we discuss the welfare implications of these changes.posted by leopard at 9:18 PM on December 2, 2014 [3 favorites]
It's actually more interesting than I thought.
The divorce rate is actually relatively constant at the seven year mark. No, really, look at that chart again. 1960's is at 10%, 1970s-1990s is at 15%, 2000s is at 12.5%. No cohort really approaches 50% until 80's couples, 30 years in. Things stabilize around 35% 18 years in, around when kids are on their way out.
So, being aggressively fair:
1) The "half of all marriages end in divorce" line is arguably false; it's more "a third do after the kids are gone"
2) The divorce rate hasn't really changed all that much for the present cohort, they just haven't been married long enough. In fact, there's this amazing paragraph where they compare "about 70%, about 65%, and nearly two thirds" as if these were all wildly different values.
But what has changed, significantly, is the marriage rate. Half of marriages are in fact not occurring in the first place. Way more marriages are not happening than divorces that would end them.
posted by effugas at 9:25 PM on December 2, 2014
The divorce rate is actually relatively constant at the seven year mark. No, really, look at that chart again. 1960's is at 10%, 1970s-1990s is at 15%, 2000s is at 12.5%. No cohort really approaches 50% until 80's couples, 30 years in. Things stabilize around 35% 18 years in, around when kids are on their way out.
So, being aggressively fair:
1) The "half of all marriages end in divorce" line is arguably false; it's more "a third do after the kids are gone"
2) The divorce rate hasn't really changed all that much for the present cohort, they just haven't been married long enough. In fact, there's this amazing paragraph where they compare "about 70%, about 65%, and nearly two thirds" as if these were all wildly different values.
But what has changed, significantly, is the marriage rate. Half of marriages are in fact not occurring in the first place. Way more marriages are not happening than divorces that would end them.
posted by effugas at 9:25 PM on December 2, 2014
Mod note: effugas, I think you've worked this angle plenty in this thread, maybe it's time to step back and let the conversation flow?
posted by mathowie (staff) at 9:39 PM on December 2, 2014 [2 favorites]
posted by mathowie (staff) at 9:39 PM on December 2, 2014 [2 favorites]
In fact, there's this amazing paragraph where they compare "about 70%, about 65%, and nearly two thirds" as if these were all wildly different values.
Paragraph in question:
Half of marriages are in fact not occurring in the first place.
Please show your work here. This article says that 47% of American women over age 15 were married in 2013, compared to a peak of 65% in 1950. I hope you're not getting confused by shifts in age demographics and delays in the age of first marriage.
posted by leopard at 9:41 PM on December 2, 2014 [3 favorites]
Paragraph in question:
About 70 percent of marriages that began in the 1990s reached their 15th anniversary (excluding those in which a spouse died), up from about 65 percent of those that began in the 1970s and 1980s. Those who married in the 2000s are so far divorcing at even lower rates. If current trends continue, nearly two-thirds of marriages will never involve a divorce,Please note that the 65 percent and 70 percent statistics refer to marriage survival rates at 15 years, while the two-thirds figure refers to ultimate survival rates.
Half of marriages are in fact not occurring in the first place.
Please show your work here. This article says that 47% of American women over age 15 were married in 2013, compared to a peak of 65% in 1950. I hope you're not getting confused by shifts in age demographics and delays in the age of first marriage.
posted by leopard at 9:41 PM on December 2, 2014 [3 favorites]
leopard, see National Marriage Project data upthread -- from 76.5 per 1000 women to 37.4 per 1000 women are getting married each year. That would seem the appropriate number to look at, since we're discussing marriages that are occurring now, not that are continuing from another era.
The big surprise to me is that there really wasn't much of a surge in the first place.
posted by effugas at 11:02 PM on December 2, 2014
The big surprise to me is that there really wasn't much of a surge in the first place.
posted by effugas at 11:02 PM on December 2, 2014
No that obviously isn't the right stat to look at if you want to talk about marriages "never happening." A society in which everyone gets married at 30 will have a lower marriage rate than a society in which everyone gets married at 20, even though everyone is still getting married. From what I can tell it remains the case that most people marry by age 40, so I don't see how your "half of all marriages are not occurring in the first place" could possibly be true.
posted by leopard at 4:31 AM on December 3, 2014 [4 favorites]
posted by leopard at 4:31 AM on December 3, 2014 [4 favorites]
The elevated divorce rates of the 70s and 80s was yet another piece of Baby Boomer optimism / self-indulgence / privilege. People now are more cautious of the financial catastrophe that divorce can be.
In terms of gender filing disparity, in many courthouses the petitioning spouse obtains important de facto advantages. Easier to get financial discovery, easier to get ex parte initial orders on child custody, asset control, and temporary support, easier to kick out a spouse who otherwise refuses to leave the family home.
posted by MattD at 5:03 AM on December 3, 2014
In terms of gender filing disparity, in many courthouses the petitioning spouse obtains important de facto advantages. Easier to get financial discovery, easier to get ex parte initial orders on child custody, asset control, and temporary support, easier to kick out a spouse who otherwise refuses to leave the family home.
posted by MattD at 5:03 AM on December 3, 2014
Divorces are a special case of breakup,
But we are measuring the subset of breakups, divorces, not the full set of breakups. We are talking about divorces, not breakups.
Also the data is censored but it is always censored in this case so comparisons are still possible; it would only produce bias in this case if old people get divorced.
posted by MisantropicPainforest at 5:46 AM on December 3, 2014 [5 favorites]
But we are measuring the subset of breakups, divorces, not the full set of breakups. We are talking about divorces, not breakups.
Also the data is censored but it is always censored in this case so comparisons are still possible; it would only produce bias in this case if old people get divorced.
posted by MisantropicPainforest at 5:46 AM on December 3, 2014 [5 favorites]
I think all the decades of handwringing over divorce rates have made a lot of people forget that divorce is actually an important social justice issue.
If anything, it's still too difficult to get divorced. I had friends that recently got divorced, and the mandatory 1-year trial separation and all that it entails just ratcheted up the animosity between the two of them and resulted in dueling restraining orders and a lot of bitterness that's made it difficult for them to share custody of their child.
posted by Ham Snadwich at 8:41 AM on December 3, 2014 [1 favorite]
If anything, it's still too difficult to get divorced. I had friends that recently got divorced, and the mandatory 1-year trial separation and all that it entails just ratcheted up the animosity between the two of them and resulted in dueling restraining orders and a lot of bitterness that's made it difficult for them to share custody of their child.
posted by Ham Snadwich at 8:41 AM on December 3, 2014 [1 favorite]
Maybe the people who would have stayed in unhappy marriages are continuing to stay in unhappy marriages because they love being miserable. Maybe they're all ridiculously religious. Shrug. Who knows?
posted by discopolo at 10:23 AM on December 3, 2014
posted by discopolo at 10:23 AM on December 3, 2014
If anything, it's still too difficult to get divorced. I had friends that recently got divorced, and the mandatory 1-year trial separation and all that it entails just ratcheted up the animosity between the two of them and resulted in dueling restraining orders and a lot of bitterness that's made it difficult for them to share custody of their child.
People are responsible for their own bad divorces and inability to get along, especially if there's a kid involved. I wouldn't blame length of separation of difficulty of obtaining divorce on ppl with emotional immaturity. No one forces them to act out on negative emotions.
posted by discopolo at 10:27 AM on December 3, 2014 [1 favorite]
People are responsible for their own bad divorces and inability to get along, especially if there's a kid involved. I wouldn't blame length of separation of difficulty of obtaining divorce on ppl with emotional immaturity. No one forces them to act out on negative emotions.
posted by discopolo at 10:27 AM on December 3, 2014 [1 favorite]
What purpose does it serve? To keep people from making rash decisions about getting divorced? It would make more sense to do that on the front end, rather than unnecessarily prolonging a process that both parties have already agreed to.
posted by Ham Snadwich at 11:56 AM on December 3, 2014
posted by Ham Snadwich at 11:56 AM on December 3, 2014
But we are measuring the subset of breakups, divorces, not the full set of breakups. We are talking about divorces, not breakups.
okay, but before the 70s the only kind of breakup was divorce. Except maybe among beatniks, hippies or adulterers, a serious relationship meant being married. Now a serious relationship can happen multiple times without marriage. People normally live together, share finances, buy things, even have children without making it official. THis is a bigger change than rates of divorce. It's not that marriage is meaningless, but its meaning has changed. People now have very important, complicated, life-affecting, sexually & financially interactive relationships without the ritual recognition.
Talking about the rates of divorce going down without taking any note of how many people have actually suffered heartbreak and property fights seems disingenuous. These days lots of people get married after they've basically already been married for like five years by old-fashioned standards. It's a different ceremony now. The comparison is apples and oranges.
posted by mdn at 11:58 AM on December 3, 2014 [3 favorites]
okay, but before the 70s the only kind of breakup was divorce. Except maybe among beatniks, hippies or adulterers, a serious relationship meant being married. Now a serious relationship can happen multiple times without marriage. People normally live together, share finances, buy things, even have children without making it official. THis is a bigger change than rates of divorce. It's not that marriage is meaningless, but its meaning has changed. People now have very important, complicated, life-affecting, sexually & financially interactive relationships without the ritual recognition.
Talking about the rates of divorce going down without taking any note of how many people have actually suffered heartbreak and property fights seems disingenuous. These days lots of people get married after they've basically already been married for like five years by old-fashioned standards. It's a different ceremony now. The comparison is apples and oranges.
posted by mdn at 11:58 AM on December 3, 2014 [3 favorites]
Ten years from now, we'll have data including significant numbers of same-sex marriages. And I strongly suspect that those marriages are going to have even fewer divorces, because those couples really had to wait for and fight for marriage. I cannot wait for data showing what a positive impact marriage equality has had on the institution of marriage.
posted by selfmedicating at 12:05 PM on December 3, 2014
posted by selfmedicating at 12:05 PM on December 3, 2014
So marriage has changed. Which I think is precisely the point.
posted by MisantropicPainforest at 12:11 PM on December 3, 2014 [1 favorite]
posted by MisantropicPainforest at 12:11 PM on December 3, 2014 [1 favorite]
leopard, you seem to think there's some sort of mere delay factor happening, i.e. all the weddings that didn't happen in the past must happen in the future, i.e. a compensatory spike is "due" (so that the weddings that didn't happen when people were 20 do happen when people are 30). I don't know why you think this. It certainly doesn't match the long term trend.
posted by effugas at 12:14 PM on December 3, 2014
posted by effugas at 12:14 PM on December 3, 2014
Look at the denominator in your chart. "Per 1,000 unmarried women." A woman who waits a while before getting married spends a lot more time in the denominator than a woman who gets married while very young. This has a huge impact on marriage rates.
Let's say that there are 100 women at each age, and marriage rates are calculated by dividing the number of marriages by the number of unmarried women over the age of 20.
In society A, everyone gets married between the ages of 20 and 21. Then the marriage rate every year is 100/100 = 100%.
In society B, everyone gets married between the ages of 21 and 22. Then the marriage rate every year is 100/200 = 50%.
In society C, everyone gets married between the ages of 20 and 30, and age at marriage is uniformly distributed. Then the marriage rate every year is 100/550 = 18%.
Note that in every one of these societies, 100% of people are married by age 30, and yet marriage rates vary between 18% and 100%.
Now back to the real world. Look at this table. In 2010, 86% of all women aged 40-44 were married or had been married. How on earth is this possible if "half of all marriages aren't happening anymore"?
posted by leopard at 1:37 PM on December 3, 2014 [1 favorite]
Let's say that there are 100 women at each age, and marriage rates are calculated by dividing the number of marriages by the number of unmarried women over the age of 20.
In society A, everyone gets married between the ages of 20 and 21. Then the marriage rate every year is 100/100 = 100%.
In society B, everyone gets married between the ages of 21 and 22. Then the marriage rate every year is 100/200 = 50%.
In society C, everyone gets married between the ages of 20 and 30, and age at marriage is uniformly distributed. Then the marriage rate every year is 100/550 = 18%.
Note that in every one of these societies, 100% of people are married by age 30, and yet marriage rates vary between 18% and 100%.
Now back to the real world. Look at this table. In 2010, 86% of all women aged 40-44 were married or had been married. How on earth is this possible if "half of all marriages aren't happening anymore"?
posted by leopard at 1:37 PM on December 3, 2014 [1 favorite]
I'd be surprised if it would still work at the literal and exact level of half the marriages, but a more generic "lots of marriages don't happen anymore" and "86% are or have been married by 40" are easily reconcilable to the extent that people have switched from "marry, divorce, remarry" to "don't marry, break up, marry."
I understand effugas's point to be on the order of "If we measured breakups-that-realistically-need-lawyers-to-resolve or breakups-with-kids-involved, which are divorces and things we should treat as divorces, the rate of those things probably hasn't gone down."
posted by ROU_Xenophobe at 1:52 PM on December 3, 2014
I understand effugas's point to be on the order of "If we measured breakups-that-realistically-need-lawyers-to-resolve or breakups-with-kids-involved, which are divorces and things we should treat as divorces, the rate of those things probably hasn't gone down."
posted by ROU_Xenophobe at 1:52 PM on December 3, 2014
"I'm not seeing what sampling bias has to do with this, either. The parameter we're trying to estimate about this population (divorce rate) pertains to the population of married people. It doesn't affect our estimate of the parameter just that this population has decreased, although it probably affects our level of confidence in that estimate by increasing the standard error. To argue for sampling bias, you'd have to talk about a specific problem in the study design that led to bias in the selection criteria - just changing the size of the population doesn't do it unless you're doing something weird like pooling results or trying to do a meta-analysis. "
You're right; I was wrong. I can think of some ways where the rate of divorce may be affected by the rate of marriage (similar to how the rate of drivers' licensing in a state correlates loosely with the number of accidents per mile traveled, because they're likely to interact with each other, i.e. fewer marriages total would mean fewer divorces total, and divorces have an effect on other marriages by normalizing divorce), but after reviewing what Penguin said and the study (since I was also concerned that the large variance coming from young and low-income people could be related to data collection methods, e.g. landline collection, but it's the longitudinal SIPP), I was wrong to snark like that.
posted by klangklangston at 2:04 PM on December 3, 2014 [3 favorites]
You're right; I was wrong. I can think of some ways where the rate of divorce may be affected by the rate of marriage (similar to how the rate of drivers' licensing in a state correlates loosely with the number of accidents per mile traveled, because they're likely to interact with each other, i.e. fewer marriages total would mean fewer divorces total, and divorces have an effect on other marriages by normalizing divorce), but after reviewing what Penguin said and the study (since I was also concerned that the large variance coming from young and low-income people could be related to data collection methods, e.g. landline collection, but it's the longitudinal SIPP), I was wrong to snark like that.
posted by klangklangston at 2:04 PM on December 3, 2014 [3 favorites]
I understand effugas's point to be on the order of "If we measured breakups-that-realistically-need-lawyers-to-resolve or breakups-with-kids-involved, which are divorces and things we should treat as divorces, the rate of those things probably hasn't gone down."
That might be true -- I doubt it -- but it would require some work, none of which has been provided in this thread. I'll put a little more trust in people like Justin Wolfers and Betsey Stevenson, who are highly-regarded economists who have published papers on this topic, than in someone on Metafilter who simply doesn't understand how the marriage rate dropping in half doesn't imply that people are twice as likely to never marry. The Wolfers-Stevenson paper I linked above covers a lot of ground -- cohabitation, birth control, women's employment opportunities -- they don't at all conclude that the change in the divorce rate is a pure "labeling" issue, even though they do say that the decline in marriage rates explains *some* of the drop in divorce.
posted by leopard at 2:19 PM on December 3, 2014 [2 favorites]
That might be true -- I doubt it -- but it would require some work, none of which has been provided in this thread. I'll put a little more trust in people like Justin Wolfers and Betsey Stevenson, who are highly-regarded economists who have published papers on this topic, than in someone on Metafilter who simply doesn't understand how the marriage rate dropping in half doesn't imply that people are twice as likely to never marry. The Wolfers-Stevenson paper I linked above covers a lot of ground -- cohabitation, birth control, women's employment opportunities -- they don't at all conclude that the change in the divorce rate is a pure "labeling" issue, even though they do say that the decline in marriage rates explains *some* of the drop in divorce.
posted by leopard at 2:19 PM on December 3, 2014 [2 favorites]
mdn: okay, but before the 70s the only kind of breakup was divorce.That is completely false. Men and women have been abandoning bad marriages since "marriage" meant selling your daughter for livestock.
Divorce just empowers people to do it more safely, with the chance of economic assurances.
posted by IAmBroom at 2:39 PM on December 3, 2014
I understand effugas's point to be on the order of "If we measured breakups-that-realistically-need-lawyers-to-resolve or breakups-with-kids-involved, which are divorces and things we should treat as divorces, the rate of those things probably hasn't gone down."
The thing that's driving me up the wall about this thread is that one can make (and indeed effugas has made!) a perfectly cogent argument to that effect without having to question or even consider the divorce statistics presented, since those statistics are explicitly concerned with the legal procedure of divorce. If someone wants to argue that we need to change the language we use around long-term relationships to remove the heavy emphasis on marriage, I'd agree. But that doesn't make the divorce rate "misleading," it just means that the divorce statistic can't be used to answer the question you want to ask. Since the authors of this piece make no claims about long-term relationships, only marriages, and since they are careful to constrain their conclusions to the realm of legal marriage (at least by my reading), I think it's unfair to say they're misleading their audience.
However, you could definitely argue that as a society we need to change the way we talk about long-term relationships to include non-marriage options like living together for some threshold number of years (what used to be common-law marriage, and still is in a few states*), sharing finances, having kids together, etc. But since these categories are muddier and don't necessarily overlap (you can have kids together but not share finances, etc) gathering data about those partnerships is much harder and there's a lot more argument about what should really count.
Anyway, to move on with the thread, if people have better ideas about how to identify a 'marriage-equivalent' long-term relationship given all of the socioeconomic changes listed in the article and upthread, I'd love to hear about other potential metrics. If legal marriage or divorce statistics don't capture what you want it to capture, what metric would capture it, and what criteria should be considered?
* I am in an unmarried long-term relationship and we recently moved to a common-law marriage state, and apparently all we need to do to be legally married here is to start calling each other husband and wife. It's kind of weird. Sure got the mother-in-law off my back about a wedding though - "Sorry Ma, we moved to Montana and suddenly we're already married! Guess you missed the ceremony when we registered to vote and changed our residency!" Shh, nobody tell her the part about presenting yourself as a married couple though.
posted by dialetheia at 3:00 PM on December 3, 2014 [3 favorites]
The thing that's driving me up the wall about this thread is that one can make (and indeed effugas has made!) a perfectly cogent argument to that effect without having to question or even consider the divorce statistics presented, since those statistics are explicitly concerned with the legal procedure of divorce. If someone wants to argue that we need to change the language we use around long-term relationships to remove the heavy emphasis on marriage, I'd agree. But that doesn't make the divorce rate "misleading," it just means that the divorce statistic can't be used to answer the question you want to ask. Since the authors of this piece make no claims about long-term relationships, only marriages, and since they are careful to constrain their conclusions to the realm of legal marriage (at least by my reading), I think it's unfair to say they're misleading their audience.
However, you could definitely argue that as a society we need to change the way we talk about long-term relationships to include non-marriage options like living together for some threshold number of years (what used to be common-law marriage, and still is in a few states*), sharing finances, having kids together, etc. But since these categories are muddier and don't necessarily overlap (you can have kids together but not share finances, etc) gathering data about those partnerships is much harder and there's a lot more argument about what should really count.
Anyway, to move on with the thread, if people have better ideas about how to identify a 'marriage-equivalent' long-term relationship given all of the socioeconomic changes listed in the article and upthread, I'd love to hear about other potential metrics. If legal marriage or divorce statistics don't capture what you want it to capture, what metric would capture it, and what criteria should be considered?
* I am in an unmarried long-term relationship and we recently moved to a common-law marriage state, and apparently all we need to do to be legally married here is to start calling each other husband and wife. It's kind of weird. Sure got the mother-in-law off my back about a wedding though - "Sorry Ma, we moved to Montana and suddenly we're already married! Guess you missed the ceremony when we registered to vote and changed our residency!" Shh, nobody tell her the part about presenting yourself as a married couple though.
posted by dialetheia at 3:00 PM on December 3, 2014 [3 favorites]
« Older Unicorn Blood | Meet incredible engineers and learn about their... Newer »
This thread has been archived and is closed to new comments
Gosh, I knew there was something special about us.
posted by Naberius at 1:30 PM on December 2, 2014 [8 favorites]