Wednesday, July 9, 2014

The Art of Dumbfuckery

Commenter TS in comments:
 If Kellermann's data pool were a 50/50 split of urban and rural households, he would never get the results he intended because so many of those households would own guns without anyone being murder.


This is why epidemiologists, pollsters, quality controllers and statisticians exist.  Nothing is fifty-fifty.  And you can't poll everyone and you can't test each and every part and you can't ensure every side is evenly distributed.

The fact is Kellermann controlled for owning v renting, rural v urban, poor v affluent and a ton of other variables.  TS still refuses to read the study which indicates he either cannot read or doesn't understand what he does read.


  1. Jade, the counties he used are over 98% urban combined. And given that he only ended up with 388 matched pairs, there is a decent chance that zero rural households were part of his study.

    When you're Jadegold, 98/2 is close enough to 50/50, so long as it suits your agenda.

    1. TS, you're good at pointing out what you think is wrong with studies that produce results you don't like. But what's your point? Do you deny the findings that having a gun in the home, on average, is a bad idea?

    2. TS simply hasn't read the study. More appalling is the fact he doesn't understand basic probability and statistics.

      Let's suppose we we polled everyone in the US. Everyone in every state. Guess what? Per the 2010 US Census, only 19% of the population lived in rural areas. Today, that number (according to is 17%.

      Now that is population. But what of households? Via the population numbers, this means somewhere in the neighborhood of 4-5% of US households are rural.

      Even if we make the wild assumption that every US rural home owns a gun, it doesn't help TS's flawed pseudo-analysis.

      Just for the record, some farmers and ranchers are renters.

      Yet, TS demands that any study that doesn't include a 50/50 split is biased and worthless.

  2. Why do you keep saying the renting risk is because of other socioeconomic factors and at the same time claim he did a good job of separating those factors? You've said it like four or five times. Even after I point out your obvious contradiction, you keep repeating it.

    1. You haven't pointed out anything. You don't understand risk, logistic regression, or control. So your comments are of no value whatsoever.

    2. Because you keep making an non-issue of it. Look, there are some places where you have to be quite wealthy to rent but generally that's not the case.