Slashdot Mirror

← Back to Stories (view on slashdot.org)

Lies, Damned Lies and Cat Statistics

Posted by samzenpus on from the nine-lies dept.

spopepro writes "While un-captioned cats might be of limited interest to the /. community, I found this column on how a fabricated statistic takes on a life of its own interesting. Starting with the Humane Society of the United States' (HSUS) claim that the unsterilized offspring of a cat will '...result in 420,000 cats in 5 years,' the author looks at other erroneous numbers, where they came from and why they won't go away."

11 of 175 comments (clear)

  1. No Don't Ruin This, I Need This! by eldavojohn · · Score: 4, Funny

    When I was a kid, I used to sit there smashing mosquitoes that bit me. Every time I smashed one fat with blood, I relished the idea that I had just killed a female mosquito who was about to lay thousands of eggs. And those mosquitoes would in turn breed and lay thousands of eggs and I had essentially just ended the lives of an infinite number of mosquitoes!

    Please, just let me have this -- your environmental constraints and logical reasoning be damned!

    --
    My work here is dung.
    1. Re:No Don't Ruin This, I Need This! by cappp · · Score: 5, Insightful

      That's pretty much TFA's point, and it's a scary one. Interest groups propogate faulty statistics so as to support questionable claims. Exaggeration, conflation, and the like does nothing more than undermine legitimate concerns. We've seen it with climate change - the desire to effect policy by presenting worst case scenario journalism has just fed the other side.
      I remember when I got the "you're a guy so try not to rape everyone" speech in college. Good underlying point...concent is important, getting concent is complicated, sex under the influence is generally a bad idea. It was totally undermined by the 1-in-4 statistic, and the way in which it was presented, and ultimatly served to offend my friends and I while also instilling the seeds of anti-feminism (ooh those stupid fem-nazis and their crazy ideas....)in a bunch of guys. The stat is wrong, it's been shown to be lacking, and it's still repeated. It has significant utility and so it's not questioned but, ultimatly, it does more to harm a good cause than it does to support it.

    2. Re:No Don't Ruin This, I Need This! by cappp · · Score: 5, Insightful
      Which is somewhat my point. We were specifically told that 1 in 4 women would be raped on campus. Not 1 in 5 in her lifetime. Not with all the significant caveats and modifiers that precede the numbers at the CDC you've referenced (the 1995 numbers). Check out the 2007 numbers - it’s now 20 to 25% either attempted or completed. The Institute of Justice found that

      A survey of college women found that 2.8 percent had experienced either a completed (1.7 percent) or an attempted (1.1 percent) rape within a 9-month timeframe.
      13.7% of undergraduate women had been victims of at least one completed sexual assault since entering college: 4.7% were victims of physically forced sexual assault; 7.8% of women were sexually assaulted when they were incapacitated after voluntarily consuming drugs, alcohol or both; and 0.6% were sexually assaulted when they were incapacitated after having been given a drug without their knowledge
      Finally, a national-level study of college and community based women found that approximately 673,000 of nearly 6 million current college women (11.5 percent) have ever been raped, and approximately twelve percent of these rapes were reported to law enforcement

      I'm not questioning the underlying idea that rape is pervasive and wrong. What I'm getting at is that by dragging out exaggerated, faulty numbers you introduce weakness into an argument. Those men in that room would have been horrified to hear that 13.7% of women had been sexually assaulted on campus - but that numbers not sexy enough for widespread hyperbole. All it took was for one guy to do a little digging into the stats, find the body of literature that criticized the methodology of that one source, and campus rape became a joke to half the community. Instead of disgust we had widespread disdain for the claim itself, and that is extremely damaging.

      There is something extremely patronising, or condescending, which presumes that people cannot be motivated by subtle or nuanced arguments – every problem doesn’t have to directly affect 98.43% of the population to count.

  2. Huh? by Anonymous Coward · · Score: 5, Insightful

    With all the fucking bullshit that ends up on the front page. All the iPhone astroturfing, all the Android FUD, all the iPhone FUD and Android astroturfing, all the dupes and tripes and all the dopes. All the ridiculous "Ask Slashdot"s that could have been solved with 2 minutes Googling...

    This ends up in idle? Mis-reporting of unsubstantiated facts by news-outlets may not be news, but it's stuff that matters, and if ever there was an example of something that shouldn't have been consigned to the idle bin, this is it! Adding some stupid allusion to lolcats does not make it idle-worthy. If someone submitted a story about the pope dying and added "His hat looks a bit like a wang! LOL" does it follow that the story should go on the Funny Pages?

    1. Re:Huh? by CheshireCatCO · · Score: 5, Funny

      ... in 3 million years you get a race of bipeds evolved from cats that are uber cool and fight wars over what colors their hats should be?

  3. More reliable statistics by Bogtha · · Score: 4, Funny

    People who prefer dogs to cats are 4.8 times as likely to prefer Scotch to vodka, but people who prefer cats to dogs are equally likely to prefer Scotch or vodka.

    --
    Bogtha Bogtha Bogtha
  4. Actually... by Moraelin · · Score: 4, Insightful

    Actually, if you actually read the article to the end, they do say it applies to just about any kind of statistics. E.g., an example they use is a statistics which supposedly said that men prefer dating secretaries than female managers -- and you can see how that helped fuel that prejudice that women who pursue a that kind of career won't get laid, and probably are cold hearted bitches who don't have time for love anyway -- but then when someone actually got to the bottom of it, the poll didn't actually ask that.

    Or you can take the myth that a woman who's not married by 35 is even less likely to marry than to be killed by a terrorist. Not only it turns out it was BS unsubstantiated hyperbole, but the perpetrators actually eventually apologized for it. Hey, better a few decades later than never, right? To get an idea how bogus that was, not only didn't the calculated numbers add up to "less likely than being killed by a terrorist" (they even admitted they made that up for sensationalism sake), but it was based on the critically flawed assumption that a woman would _only_ marry older men. But it's been echoed all over the place and taken for a fact.

    And what they say is that basically not only some numbers pulled out of some PR bullshitter's ass get taken for gospel, but basically they become nearly impossible to debunk. You'd have to spend the equivalent of several episodes to debunk one sound bite that takes just 5 seconds to mindlessly repeat all around. And even then, you won't get as much exposure as the mass of idiots repeating the falsehood because they heard it somewhere, and even to a lot of those who hear you debunking it, you'll just sound like some conspiracy-theorist for attacking what they know for a fact.

    And I think that shouldn't be dismissed as just some idle lolcat joke. Especially in IT and CS, we see the same phenomenon every day. There are a ton of "X is better than Y" or "A is 10% more scalable than B" pseudo-facts thrown around, that everyone just repeats and nobody questions them.

    Especially almost nobody in management who heard it in some IT-for-managers ragazine _and_ from the nice salesman using it to sell his snake oil. So it must be true, right?

    --
    A polar bear is a cartesian bear after a coordinate transform.
    1. Re:Actually... by MightyMartian · · Score: 4, Insightful

      The fundamental problem is that most people are credulous morons.

      --
      The world's burning. Moped Jesus spotted on I50. Details at 11.
    2. Re:Actually... by Moraelin · · Score: 4, Insightful

      In a nutshell, yes.

      Though IMHO it is a bit more than that. It's not just that people will believe some BS or another. It's that one piece of BS can get so circulated around, that it becomes basically common knowledge. It becomes something that "everyone knows". Even people who wouldn't just believe it the first time, start falling for it when they hear it from 10 different sources as common knowledge.

      Plus, as they say in TFA, eventually it even gets picked by some newspaper, or repeated by some politician trying to support some bill, and it kinda becomes official.

      Even basically "[citation needed]" doesn't help there, because some piece of BS (with statistics or not) that's been bouncing around for 30 years, can be a bitch to track to the actual source. Publication A cites official report B (see the politicians using them above,) which in turn has a footnote pointing at newspaper article C, which points to out-of-print book D, which even if you find a copy and read it, in turn points out to some study that's behind a paywall, and if you got even there, you find out it's really a meta-study quoting the numbers published in yet another article E.

      Most people will give up somewhere along that chain, and assume it's actually a valid and proven claim. Some right at the first step, because, hey, it does point to a source.

      And sometimes even if you make it all the way to the root source, you'll have trouble convincing anyone that that common knowledge is false. I mean, hey, what are you, some conspiracy theorist? Everyone knows X is true. Plus, supposedly some scientist said that (though usually he actually didn't, and some PR department or journalist mis-represented him), and who are you to question scientists??? You can even see that kind of idiot on Slashdot. There are several people around who seem to thrive on posting basically "who are you to question TEH SCIENTISTS???"

      And even if you got past that, you often find that

      A) they have the same gross misunderstanding of statistics as the journalists who mis-represented it in the first place, so good luck getting them to see why it doesn't actually say that, or

      B) you need to first teach them what an equivocation or amphibology fallacy is, before they're even equipped to understand why the study doesn't actually say what they think it says

      C) you'd need to first teach them a lot about the psychology and pitfalls of polling, i.e., that basically you can produce vastly different results from the same people and to essentially the same question, by just exploiting the tendency of people to say "yes" more than "no", or pick the answer which sounds more agreeable, or just pick the first one more in multi-choice polls. Serious polling companies know and compensate for that, but a PR agency can deliberately exploit that to skew the results.

      Etc.

      And again, try to do that without sounding like a CT-er inventing reasons not to trust those guys, and without falling into "tl;dr" range either. Good luck with that.

      Basically at some point some falsehoods have taken off so well, that you don't even have to be a gullible moron to just take them for granted.

      --
      A polar bear is a cartesian bear after a coordinate transform.
  5. Huh? by martin-boundary · · Score: 5, Insightful
    I don't know what kind of maths they do in the Wall St Journal (yeah, I actually RTFA), but according to the maths that I do from the unfashionable side of my mother's basement, the actual number of offspring after 5 years starting with a single unspayed female cat is zero.

    Now, if one starts with two cats however, the number could well be higher...

  6. Comment removed by account_deleted · · Score: 4, Informative

    Comment removed based on user account deletion