50% Divorce Rate is a Myth, It’s More Like 33% or One Third

divorce drawingThere’s this statistic I often hear about how essentially 1 in every 2 marriages ends in divorce, or the 50% divorce rate. In researching for a future post, I discovered this was a myth in how “divorce rate” was calculated. Or at least I’m going to be kind and call it a myth. Someone either maliciously propagated it for their agenda, or were too stupid to know what they were talking about!

Divorce rate, according to that 50% statistic, is measured as number of divorces filed each year, against number of marriages made each year. The serious problem with this equation is that what’s called the number of divorces is from all the marriages which had taken place in previous years, which only happened to have ended in divorce in a given year. If people stopped getting married for a year, which is only an exaggeration of the increasing trend of common-law living, you could easily end up with more than 100% of marriages ending up in divorce by this calculation. Try this numerical example if you’re not getting what I’m saying.

According to the Centre for Disease Control, which I guess considers marriage and divorce a disease, there were 3.6 divorces for every 7.5 marriages in 2005. Both were measured out of 1,000 people so that base can be eliminated to avoid confusion in presenting the statistic. If you take 3.6 and divide by 7.5, you get 48% and that “50% divorce rate”. Now, if common-law living were to reduce marriages to 3.6 marriages per thousand people next year, you’re looking at a 100% divorce rate… and you can imagine the press would go nuts with the stories!

The correct way to interpret that so-called “divorce rate” statistic would be to say there was one divorce for every two marriages in a year, not one in every two marriages end in divorce. There’s a whole slew of marriages out there they can’t say yet whether or not it would end in divorce so they can’t claim 1 in 2 end in divorce!

So what is the real divorce rate, then?

According to 2004 US Census Data, Table 3 for All Races (click to download Excel file):

75.56 million men ever marry (i.e. married at least once)
22.70 million men ever divorce
30.1% = men who ever marry end up getting divorced

87.32 million women ever marry
26.95 million women ever divorce
30.9% = women who ever marry end up getting divorced

162.88 million men and women ever marry
49.68 million men and women ever divorce
30.5% = men and women, combined, who ever married end up getting divorced

I know 30.5% isn’t a third as that is 33 and 1/3 percent, but for rough value’s sake, a third will do just fine as what the divorce rate should be quoted as. Subsequent marriages to the first fail at a rate higher than first marriages (Rutgers University web page), bringing up the overall rate from 30.5%. But because there aren’t nearly as many of them as first marriages, they wouldn’t skew the rate up by that much. So for rounding purposes, I’d say one third is a good estimate.

Now, if you don’t think that’s a huge difference to go from one half to one third, try this somber experiment. Next time you’re at a house party with many of your friends, especially if you’re old enough many of them are married, look around and predict which one of every two couples will end up in divorce. This is going to be true lest you believe you and your group of friends are somehow so special you would defy this average… and don’t kid yourself on how special you all are! Yes, these happy times with all these happy couples at this party won’t be like this some years down the road given all the divorces you just predicted. Go get another drink or two!

It’s a bit of a sad and shocking experiment, really. But then try it again with which one of every three couples will end up in divorce. The fact you get to “save” some marriages is probably a relief. That’s the difference between a half and a third, my friend! Every single marriage that you can “save” from ending up in divorce should be a relief, cause even if you think the divorce should happen, divorces just aren’t pretty things in life.

Flesch-Kincaid Grade Reading Level: 7.3


4 thoughts on “50% Divorce Rate is a Myth, It’s More Like 33% or One Third

  1. You appear to be underestimating the divorce rate by including people who have only been married for a short period of time. The divorce rate during the first 5 years of marriage is obviously much lower than the rate of divorce during the first 40 years, yet your figure effectively averages those rates. In other words, the census figure you cite includes all married individuals, even ones who haven’t been married very long, so of course a disproportionally small percentage admit to having been divorced yet, but that says little about their chances of eventually getting a divorce in the long term.

  2. Thank you for the comment. Getting a true divorce rate is never possible because you’d have to stop counting new marriages at some point in time, like today, and wait until the last of the marriages which took place up to today, to die out (and some old couple will likely never divorce hitting some huge anniversary that “die out” would be the right term. The best you can do is get a “snapshot in time”, if you will. That’s the way the Census Bureau does it, number of divorces per year vs number of marriages. It comes down to finding the most correct indicator.

    While your argument is correct, it doesn’t make that much of a dent in the stats which would make the 50% more correct than the 1/3 rate I gave. Note I gave a buffer to the 30.5% calculation to compensate for things I saw as the point you mentioned, which I left out not to complicate things more. That was close to a 10% margin of error on the 30.5%, btw. Generous for “population” data.

    Sure, there are a lot of people early in their first marriage reporting they have never been divorced, who will ultimately change. Those people will report they have never divorced. Yes. They will skew the percentage I pointed out down, as you said. But how much of the whole “ever married” population are they is the big question?

    For the sake of simplifying the example, let’s say there were more or less the same number of marriages each year. That would be an over estimate in the long run with common-law living being an increasing present trend. And let’s say people live about 50 years after their first marriage (since you were talking about new marriages skewing things). That’s a fair number, I think, because people first marry at about 25 for about a decade now and the average lifespan is about 75. Those further back marrying earlier, their lifespans were shorter by roughly the same amount. 50 is a fair and round number to illustrate the point I’m about to make.

    Over a 50 year period, the number of marriages each year would only be 2% of the base. To make up the difference between the 30.5% calculation and the 50% rate quoted, you would need every single new marriage for the next 10 years to end up in divorce!

    Or try it another way. New marriages each year have a 2% influence on my calculation. You can go 3% if you want to change the 50 years value to 33 years, but that’s people dying awfully early, on average (55-60 year average life span which is WAY low). If the divorce rate were really 50%, then those people who would “lie” because they had not yet been divorced but will, only has a 1% influence each year. It would take 17 years of them to bring the divorce rate, the way I have calculated it, up to 50% (19 if you don’t include my margin of error given). Theoretically, they’d need divorce over 50% to bring the calculation to 50% but that’s being mathematically anal.

    Either way, the point is this. Neither my calculation nor the Census Bureau calculation is correct. We each provide “snapshots in time”. The way the 50% is calculated by the Census Bureau is far less stable a figure than mine. Common-law living on the rise will only bring that rate up, which, I don’t doubt causes doubt in others to marry… not to mention the bigger costs of divorces as the years go by. It’s a bit of a self-fulling prophecy. Misleading stats that will feed itself via fear it causes. The larger the increase in common-law living, the larger the increase in divorce rates as the number of marriages fall off, making the base smaller. You won’t see the absolute numbers in divorces go down for a while yet (10 years maybe?) cause the increasing common-law living trend hasn’t been around that long that the marriage numbers causing the divorce numbers will have fallen off much.

    Between the two snapshots in time, you won’t be able to rock my stats anywhere near as easily as the Census stats. I never said it was correct, but that it was “more like”, implying truer value, and I completely stand by that given the arguments just made.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.