3 ms·

You seem to be falling prey to Simpsons's Paradox: https://en.wikipedia.org/wiki/Simpson's_paradox https://en.wikipedia.org/wiki/Simpson's_paradox The chart i

by adwf 11y ago

You seem to be falling prey to Simpsons's Paradox: https://en.wikipedia.org/wiki/Simpson's_paradox https://en.wikipedia.org/wiki/Simpson's_paradox The chart is very clearly downward, with a couple of upward slopes in the middle. If you cherry pick the evidence like that crimeresearch.org link does, you'll get the upward slope. If you look at the data as a whole, you get the long term downward trend. This is pretty much the definition of Simpson's Paradox. The downward trend is the correct one.

jjoonathan 11y ago
Cherry picking isn't the same thing as Simpson's paradox. As for cherry picking: right back atcha. But even that doesn't work for the Ireland, Jamaica, DC, and Chicago graphs.
- adwf 11y ago
  Simpson's Paradox frequently arises because of cherry picking though. Marrs is specifically looking at an upwards slope over 3-4 data points after 1998. Whereas the overall slope of the chart from 1996-2009 is downwards. Clearly falling foul of the paradox. Not to mention, as has been repeatedly stated in this thread - that peak around 02/03 was solely due to the 172 Harold Shipman serial killings being added to the statistics in that year only. And they weren't even anything to do with gun crime... As for Ireland and Jamaica, the increase in homicides are very explainable given the Troubles and rise of Carribean drug trafficking in the 70s.
marrs 11y ago
If anyone is falling prey to Simpson's Paradox, it's you. You're the one implying a causation based on those statistics. I'm asserting that there's not enough information there to make any such connection, and I'm highlighting on the graph a short-term upward trend that contradicts the assertion that you are trying to make. Ignore that part of the data if it makes you feel better; I still wouldn't accept that there's enough data there to support your theory. Another interpretation of the data, also just a theory but no less valid, is that handguns rose in popularity as the weapon of choice within gangs, and the police responded by targeting gun crime. Again, I wouldn't want to state it as fact, but it would be a useful starting point for additional research. The reason frankly why I would need so much convincing that banning hand guns had any effect gun crime rates in the UK is because I just don't think there were enough gun owners to be statistically significant in the first place. Handgun shooting was a niche sport - a very niche sport.

by adwf 11y ago

jjoonathan 11y ago

Cherry picking isn't the same thing as Simpson's paradox. As for cherry picking: right back atcha. But even that doesn't work for the Ireland, Jamaica, DC, and Chicago graphs.

adwf 11y ago
Simpson's Paradox frequently arises because of cherry picking though. Marrs is specifically looking at an upwards slope over 3-4 data points after 1998. Whereas the overall slope of the chart from 1996-2009 is downwards. Clearly falling foul of the paradox. Not to mention, as has been repeatedly stated in this thread - that peak around 02/03 was solely due to the 172 Harold Shipman serial killings being added to the statistics in that year only. And they weren't even anything to do with gun crime... As for Ireland and Jamaica, the increase in homicides are very explainable given the Troubles and rise of Carribean drug trafficking in the 70s.

marrs 11y ago

If anyone is falling prey to Simpson's Paradox, it's you. You're the one implying a causation based on those statistics. I'm asserting that there's not enough information there to make any such connection, and I'm highlighting on the graph a short-term upward trend that contradicts the assertion that you are trying to make. Ignore that part of the data if it makes you feel better; I still wouldn't accept that there's enough data there to support your theory. Another interpretation of the data, also just a theory but no less valid, is that handguns rose in popularity as the weapon of choice within gangs, and the police responded by targeting gun crime. Again, I wouldn't want to state it as fact, but it would be a useful starting point for additional research. The reason frankly why I would need so much convincing that banning hand guns had any effect gun crime rates in the UK is because I just don't think there were enough gun owners to be statistically significant in the first place. Handgun shooting was a niche sport - a very niche sport.