Friday Fun Links 5-31-13

Well, we hit 94 degrees in Boston today, and I’m parked on the couch with some banana peanut butter frozen yogurt and a deep thankfulness that the weekend’s upon us.  Here’s some fun links to keep you amused:

First, Jezebel covers the best/most ridiculous acapella group names.  Aural Fixation from my alma mater made it, but I was a little surprised to see that the MIT Logarhythms  didn’t get a mention.

This link actually wasn’t that fun…it was annoying.  It’s allegedly a list of “disappointing facts for geeks“…but almost the entire lists consists of “movie x made more than movie y” with none of the numbers adjusted for inflation.  My favorite from the comments section “good news for geeks…we know bad math when we see it!”

This is very cool:  Pangea with modern political borders.  Apparently I would have been living right next to Morocco.

This is my new favorite website, btw…the Dictionary of Obscure Sorrows.  An excerpt:
sonder

n. the realization that each random passerby is living a life as vivid and complex as your own—populated with their own ambitions, friends, routines, worries and inherited craziness—an epic story that continues invisibly around you like an anthill sprawling deep underground, with elaborate passageways to thousands of other lives that you’ll never know existed, in which you might appear only once, as an extra sipping coffee in the background, as a blur of traffic passing on the highway, as a lighted window at dusk.

Alright, now some analytics you can really use…do you live in IHOP America or Waffle House America?

If that’s too lowbrow for you, how about this:  could you pass Eton’s admissions exam?

Thursday Quickies: Moms as breadwinners

I’ve seen a few headlines in the past few days about the study that showed that moms are breadwinners in 4 out of 10 households.  It’s based on this Pew Research study, and I feel like there’s a few nuances not made clear in the headline:

  • The denominator was not women or couples, the denominator was “households with children under 18”.  Thus any women without children or whose children are over 18 were not counted.
  • 63% of the female breadwinner households were single mothers
  • Of the 37% who were not single mothers, the only requirement was that they out-earn their husband.  There is no mention of a minimum gap…so a wife who earns $1000/year more than her husband is counted the same way a wife earning $50,000/year more is counted.  
Also interesting:  married households with female breadwinners have an income of four times more than households with a single female at the head ($80,000/year vs $23,000/year).  Households with a male breadwinner have a median income of about $78,000/year.  This shows some interesting selection bias…my guess is that women who earn high salaries and out earn their husbands are less likely to quit/drop to part time when kids come on the scene.  Since part of the normal debate around working/not working post-baby is “does my salary cover daycare costs”, it would make sense that women who could answer a resounding “yes” would be more likely to stay on and keep the family income higher.

Thursday Quickies: Stats, law, college and sexual assault

Eugene Volokh had up an interesting article that touch on the intersection of stats and law.  It was on the topic of campus tribunals that hear sexual assault cases, and I thought it showed a fundamental principle of stats fairly nicely: when in doubt, put it in words.   He does this with 3 legal standards for evidence: beyond a reasonable doubt (95% confidence), clear and convincing evidence (75% to 80% confidence) and a preponderance of evidence (51% or more confidence).  He then says to determine the standard we should convert this in to words:

  • Better that 19  students  guilty of sexual assault remain at the university, with no discipline imposed, than one innocent student be expelled or otherwise disciplined
  • Better that 4 students guilty of sexual assault remain at the university, with no discipline imposed, than one innocent student be expelled
  • These outcomes are about equally bad for both students and the university
There’s some other interesting legal discussion in his post, but I thought the conversion of legal standards and probabilities in to clear sentences was a particularly helpful way to frame the discussion.

Thursday Quickies: Patient "engagement" plus cost

First, I got in to a Twitter discussion today with patient engagement advocate Dave deBronkart about his article that criticized this study.  The study was being advertised under the headline “When doctors and patients share in decision making, hospital costs go up”.  The issue is the headline implies that the study looked at what happened when patients and doctors made decisions together.  In reality, the study asked admitted patients if they felt they should make decisions with the doctor, not whether or not they actually did.  It turns out that the costs for the individual hospital stay was about $860 higher for those who wanted to collaborate (median cost was $14,000 to begin with).  None of this was tied to outcome or future costs, so we have no idea if this was $800 that saved money in the long run, or $800 wasted.  It’s pretty insidious, because it’s being used to justify a rather paternalistic model of medicine that many people (including Dave) have been working hard to get away from.

High School Rankings

John Tierney has an interesting piece up at the Atlantic about how national high school rankings are not only meaningless, but actually harmful.

He doesn’t quibble much with local rankings, and agrees that if done correctly they can provide good information for residents.  As for national level rankings though, he says this:

let’s call national rankings of high schools what they are: nonsense. There is no way to say, with any degree of accuracy at all, where any given high school ranks in relation to others in terms of how good it is or how challenging it is. 

Now this seems pretty sensible to me.  Ranking all the schools in a given state against each other can be meaningful, though more so within ranges than with strict numbers (is there really a meaningful difference between #34 in the state and #35?).  But to pluck a few from around the country?  That’s not even useful.  When my husband and I went to buy a house, we knew the general area we were looking at, and school rankings were one of many factors we looked at when picking a town to buy in.  I’m pretty sure most people  do something similar.  This works of course because we already had the region picked out and knew the trade offs that came with the individual regions.  You’re not going to use national rankings like this.

Additionally, he notes that at least one of the national lists (the “Challenge Index” from the Washington Post) literally uses only one metric to determine a challenging high school: the number of AP (or similar) tests taken by the seniors at the high school, divided by the number of seniors:

Note that the numerator is not even the number of such exams passed, but merely the number taken. So, a given school can rise on the list by increasing the number of its students who take “advanced” classes. Conversely, schools that are more discerning and thoughtful about which students ought to be taking AP classes end up suffering in the rankings. So, the list produces nonsensical anomalies such as high schools with very low graduation rates ranking much higher on the “Challenge Index” than excellent schools that don’t game the ranking system…

This idea of ranking interested me.  Ultimately, we actually picked the school district we did in part because of the options it holds for the not-so-academically inclined.  Don’t get me wrong, it’s in the top 20% of high schools in the state, but not by much.  More importantly, the regional technical high school is here, and there’s opportunities to learn how to make a good living even if college isn’t your thing.  I live in a state with a great educational system, and my town is no exception.  I’m less worried about AP tests, and more worried about school districts that might push kids in to inappropriate classes to keep their numbers up, to the detriment of the child.  While a certain baseline level of knowledge should be mandatory, I want my son to be challenged, but not tortured.  I’m suspicious of schools who try to hard on these lists, because school ranking and the best interest of the child don’t always collide.

Looking further down the line, it’s interesting to note that even more advanced methodologies almost always use the percent of kids headed to college as a judge of the high school’s rigor.  As college costs continue to spiral and become and worse and worse investment, I’m curious if we’re going to see a bigger and bigger divide between rich neighborhoods and poor neighborhoods in terms of rankings.  This could drive people out of the poorer neighborhoods, not because the schools were actually worse, but because the metric used to assess them is so contingent on parents have the cash to send their kids to college.  Things to ponder.

Social media growth

I saw a weird headline last week about how teens interest in Facebook has been waning.

I was curious what constituted “waning”, and it appears that most of the proof is that other social media platforms are growing quickly.  What was interesting is that buried in the report is the Facebook ubiquity level among teams: 94%.

I’m sort of curious how any software that has that level of ubiquity could be doing anything other than “waning”.  It certainly can’t be going up.

Also entertaining: Facebook execs called Facebooks loss of young people an “urban legend”.  I’m pretty sure a company that’s less than 10 years old shouldn’t describe anything as an urban legend.

Just for giggles, here’s the growth chart:

Bah.

Friday Fun Links 5-24-13

Interesting stuff from Juice Analytics…30 days to better data storytelling.

Also, just in time for summer, a whole bunch of science reading.  I just finished Space Chronicles and am starting My Beloved Brontosaurus.

Speaking of dinosaurs, here’s the list of best to worst.

On a fun note, here’s 20 biopic actors and the famous people they played.  I always thought Charlize Theron deserved every bit of that Oscar she got for Monster.

In other news, here’s the top 10 species discovered in 2012.  The Semachrysa jadeis my favorite story.  It was found on Flickr.

It’s hard out there boys

Ann Althouse linked to an article about the struggle of working class male undergrads vs middle-class undergrads:

Combine the “chiselled out of rock” body of actor Ryan Reynolds, the intellectual prowess of writer Christopher Hitchens and the “funny, quirky” demeanour of film star Joseph Gordon-Levitt and you have the perfect role model for male middle-class undergraduates. 

But while bourgeois students can “seamlessly integrate” many types of masculinity, a study at two universities concludes that their working-class peers find squaring the many demands placed on the modern man more challenging.

This looked like an interesting study, and I was all ready to read up on it…but it hasn’t been published yet.  It’s a conference paper.  That’s fine, but I was pretty interested that this article gave pretty much zero proof of the assertion that middle class males were seamlessly integrating different types of masculinity, or that working class ones were struggling.  The only piece of data reported suggested that middle class men weree integrating anything was that they included “well groomed” and “metrosexual” as priorities in being good looking, whereas working class men did not.

Other than that, the article was mostly researcher’s continued assertion that this phenomena occurred…though I question her bias a bit as she stated that working class men’s way of thinking about intelligence “belies an assumption of entitlement to dominance….arguably a refashioning of traditional male hegemony”.

So how much of this is data and how much was spin?  Who knows.  Despite what the journalist is reporting, we might all just have to wait for the paper.