Divorce Statistics and Studies Blog

Specious accuracy: We need to be approximately right, not precisely wrong - Edward Tufte

By Edward Tufte

Headlines show how journalism distorts statistics & studies

Headlines from a Mathematically Literate World by Ben Orlin at mathwithbaddrawings.com

Big caveat: Some of the examples make points that go outside the author's actual expertise -- he's solid on mathematics and statistics but those necessarily overlap all kinds of other fields, which in turn depend on statistics but also on other branches of science. His example about the minimum wage is uniquely outside the scope of his expertise and of the whole point of his article, because it's actually about microeconomic analysis and prediction, not an empirical, statistical  analysis of outcomes. 

 

6 Stats Not to Trust: 6 Graphs, 6 Cartoons

Why Not to Trust Statistics

By Ben Orlin at mathwithbaddrawings.com

See also:

• What journalists, weather forecasters, and Han Solo get wrong about probability
• Headlines from a Mathematically Literate World 
• What Does Probability Mean in Your Profession?

 

Divorce/Margarine, Drowning/Marriage ... This Cluster of Correlations Couldn't be Coincidence

Divorce and margarine went in and (deservedly) out of fashion together, as everyone knows who thinks about it for a moment and is old enough to remember. And now, as marriage continues its long decline, Kentucky fishing-boat drownings do too, probably since marriage makes you want to go out alone in a small boat with nothing but a case of beer and your fishing gear. Seriously, how many of those guys out there fishing do you think are single? Single guys are probably somewhere where they're more likely to meet single women.

And yet Tyler VIgen refers to these and many others as "Spurious Correlations." These are among a top-ten list that also includes murders with hot objects, death from getting tangled in the bedsheets, Nicholas Cage movies, cheese-eating, moon-landing, suicide, Miss America,video games, killer spiders, spelling bees, and Ph.D.s.

Spurious Correlations

10 more here:

The 10 Most Bizarre Correlations

And here's ten that riff on another commonly reported form of statistic or study, not a correlation of change over time, but studies where a subset of respondents who have a particular response also show a different mixture of responses to a completely different question than is representative of the entire study pool. Stuff like, 60% of divorced people preferred chocolate ice cream compared to 80% of the general population.

10 Crazy Correlations Between Unrelated Things

By Gus Lubin and Tony Manfred

 

But wait, there's more correlations, marked in some damn good and genuinely important articles:

How a TV Sitcom Triggered the Downfall of Western Civilization

BY DAVID HOPKINS

 

And cutting a wider swath but coming to about the same conclusion, although blaming not sitcoms but the reality shows that mostly replaced them ...

Postapocalyptic hellscape by Jacob T. Levy

 

(h/t Jason Rylander and Todd Seavey) 

The Neuroscience of Marital Success or Failure

The science of a happy marriage

By Christian Jarrett on cnn.com,  August 17, 2015

via The Association of Family and Conciliation Courts

62% of psychology studies' results aren't as strong when experiments repeated

"Psychology Findings Not Strong as Claimed", by Benedict Carey in The New York Times 

 

To be clear, this replication study of 100 psychology studies did not reexamine their data and determine that the results were overstated; rather, it re-ran the experiments and found that 35 studies turned out the same, 62 produced the same outcomes, correlations, tendencies etc. but not as strongly as the original studies, and 3 were not counted because it was unclear how they turned out.

The stronger the results in the original study, the more likely they were to remain as strong when redone. The experience and prestige of the authors did not correlate positively or negatively with consistency of results.

The Times story also quotes some criticism of the replicators' methods and questioning of their rigor. But it does seem significant that 62% of retests showed weaker results, and apparently none showed stronger results,* and none showed completely different results, either.*

This certainly does seem to show some kind of tendency to overstate results, although it is hard to see how, since they are based directly on the data. And it appears that that tendency kicks in when the results are not that overwhelming in the first place. Or perhaps there are other factors that make the results stronger the first time an experiment is run and published? Or maybe results that are not so overwhelmingly one-sided are more likely to vary in different samples?

Perhaps more significant is that the experiments practically always ended up with the same conclusion, just not as strongly.*

"Social science isn't rocket science, it's harder",  Edward Tufte said when sharing the article on Facebook. But it's important to remember that this study was really designed for medical sciences, and was applied here to psychology studies. However, many social sciences, especially demographics, do not involve those kinds of studies, and sometimes reproducing results is not a possibility in many social science fields. While psychology studies are a significant share of the items on this blog, many more are demographic studies, polls, or in other disciplines of the social sciences.

__________

*Items marked with * are what I assume from the Times article, but it does not say so as explicitly as I would hope. I didn't look at the original study it reports on. I hope someone will do so, and post a comment saying whether these interpretations of the article are correct.

Wise questions on divorce stats & Goldilocks Theory: Are 30-something newlyweds really doomed?

Reporting on a correlation that was actually discovered several years ago, Joanna Rothkopf of Jezebel.com says what I've often thought but ne'er so well expressed, and which is rarely expressed at all, but deserves serious thought:

"But for real, how seriously are we supposed to take this? Studies keep claiming bullshit indicators are the keys to lasting marriages, including the price of the engagement ring (don’tspend between $2,000-4,000), amount of individual pre-marriage wealth, the size and price of the wedding and the presence of the honeymoon. So, I don’t know, just live your life."

The eminent marital statistician Nicholas Wolfinger describes what sound like very dramatic correlations between divorce and one's age at marriage:

“My data analysis shows that prior to age 32 or so, each additional year of age at marriage reduces the odds of divorce by 11 percent,” he writes. “However, after that the odds of divorce increase by 5 percent per year.” How come? Wolfinger isn't sure. But controlling the data for demographic and personal characteristics such as race, education, religion, sexual history, family background, or the size of the cities survey takers lived in didn't change the results, suggesting none of those factors could explain it. Ultimately, the professor suspects that there's a lot of self-selection at play: The sorts of people who wait a very long time to say "I do" just might not really be the marrying types, whether they realize it or not. Or, even if they are, their dating pool might have been whittled down to people who aren't.

Study: People Who Get Married After 32 Are More Likely To Get Divorced

By Joanna Rothkopf on Jezebel.com, 7/16/15

Based on:

• " Want to Avoid Divorce? Wait to Get Married, But Not Too Long" by @NickWolfinger on Family-Studies.org July 16, 2015
• " The Goldilocks Theory of Marriage" by Jordan Weissmann on Slate.com
• "Age at marriage and marital instability: revisiting the Becker–Landes–Michael hypothesis" by Evelyn L. Lehrer in the Journal of Population Economics, April 2008, Volume 21, Issue 2, pp 463-484, also dated 01 Sep 2006

 

Social & human sciences worse than useless from ignoring that "statistical significance" isn't PRACTICAL significance!

The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives, by Deirdre McCloskey and Stephen Ziliak, University of Michigan Press, argues that statistical significance in its all too common meaning, philosophical possibilities uncalibrated to the sizes of important effects in the world, is useless for science.

"Yet in medical science, in population biology, in much of sociology, political sciences, psychology, and economics, in parts of literary study, there reigns the spirit of the Mathematics or Philosophy Departments (appropriate in their own fields of absolutes). The result has been a catastrophe for such sciences, or former sciences. The solution is simple: get back to seeking oomph. It would be wrong, of course, to abandon math or statistics. But they need every time to be put into a context of How Much, as they are in chemistry, in most biology, in history, and in engineering science.

"In many of the life and human sciences the existence/whether question of the philosophical disciplines has substituted for the size-matters/how-much question of the scientific disciplines. The substitution is causing a loss of jobs, justice, profits, environment, and even life. The substitution we are worrying about here is called "statistical significance"—a qualitative, philosophical rule which has substituted for a quantitative, scientific magnitude."

I think this results from "scientism" -- from people in the social sciences and humanities, and in politics and the media, wishing to SOUND scientific and as a result failing to BE scientific, putting way too much stock in scientific terms they don't actually understand.

I know and admire Prof. McCloskey from her inspiring, interdisciplinary teaching at an Institute for Humane Studies seminar on "Liberty in Film and Fiction", a life-changing experience I'd recommend to any student.

ACS marriage/divorce data is crucial: leading family demographer @justinwolfers explains why

It's great to see my opinion on the proposed elimination of marital status from the long-form census confirmed, with much more understanding, context and background, by an eminent statistician and demographer, Justin Wolfers, in the New York Times! 

"Census Bureau’s Plan to Cut Marriage and Divorce Questions Has Academics Up in Arms" (NYT 12/31/14)

Wolfers reviews what the American Community Survey and its predecessors do about marriage and divorce, what other sources of data are out there and how they work (and don't work), and why the data is crucial for both progressives and conservatives who are concerned about families.

Beware numbered lists of "myths"

Mythunderstood: Five Myths About “Five Myths About” Stories

By Gene Weingarten in the Washington Post, 11/30/14