4 ms·

I think there's a related phenomenon: we have "debates" and "disputes" over the expected effects of economic policies, social policies, political policies, etc.

I think there's a related phenomenon: we have "debates" and "disputes" over the expected effects of economic policies, social policies, political policies, etc. and then we get to see those policies in action. Neither major US political party seems to do better than 50/50 overall, despite having "expert" opinion from professional economists, social scientists, and so on. The media providing coverage of the debates, the arguments, and the policies don't generally do a good job of picking out the best analysis; their track record is far worse than 50/50. But viewers keep watching! Then the same politicians and the same media trot out "scientists" in support of various measures. Sometimes the scientists are legit, and sometimes they're cranks. Sometimes the scientists are legit, but are talking about policy preferences rather than science. And most of the time, the politician or media member giving the "scientist" the podium doesn't have the expertise necessary to understand if the person is legit or totally nuts. And all of this gets ratings, so the media keeps it coming. Throw in a little confirmation bias -- where people remember their side's successes and the other side's failures, and therefore trust of their side's experts and distrust of the other side's experts -- and you have a recipe for divided opinion even where there shouldn't be.

shasta 16y ago
> Neither major US political party seems to do better than 50/50 overall I'm not sure what that even means or how you could measure it. Most political policies don't have tangible win/lose results. There are not often going to be actions that make everything peachy. Usually choices have pros and cons, and it's easier to get the public to notice the cons than it is to get them to evaluate the trade offs involved. Also, if the cons don't materialize within two years, the public will probably not connect the dots. So watch some cable news, form your half-ass opinion and go vote.
- lotharbot 16y ago
  > Most political policies don't have tangible win/lose results. No -- but occasionally politicians will state a particular tangible, directly measureable result they're expecting a policy to produce (either pro or con). They seem to be wrong as often as they're right, across the board. Ideally, we'd have a policy suggestion, true experts would give us a clear understanding of the tradeoffs involved and the actual effects of the policy, and we could evaluate them and make sensible decisions. What we actually get are policy suggestions, statements from both sides' hand-picked "experts" declaring fictional effects of the policy, we vote for whatever sounds the best to us, and half the time the end result is totally different from what we thought we were voting for. And pretty much all the time, both sides cherry-pick statistics that don't really measure what they say they're measuring in order to prove that they were right all along. And then we wonder why people are skeptical...
MichaelSalib 16y ago
Neither major US political party seems to do better than 50/50 overall According to Larry Bartel's research, in some major areas of economic policy, this is not true. See http://rodrik.typepad.com/dani_rodriks_weblog/2008/03/american-politi.html http://rodrik.typepad.com/dani_rodriks_weblog/2008/03/americ... and http://www.washingtonmonthly.com/archives/individual/2005_05/006282.php http://www.washingtonmonthly.com/archives/individual/2005_05... for more details. Of course, insofar as you're talking about public perceptions of news media "debates", that only confirms your point.
- yummyfajitas 16y ago
  While the graph is nice, error bounds would be extremely useful. Comparing two groups based on a sample size of 5 (per group) is an extremely tricky thing to do. It's even trickier when you try to rule out systematic errors by reducing your sample size further. For instance, the blog mentions one systematic error caused by D->R transitions. He then claims this is ruled out by analyzing two R->R and one D->D transition, ignoring the fact that 2 of those three transitions were based on fluke circumstances themselves (specifically Kennedy->Johnson and Nixon->Ford).

shasta 16y ago

> Neither major US political party seems to do better than 50/50 overall I'm not sure what that even means or how you could measure it. Most political policies don't have tangible win/lose results. There are not often going to be actions that make everything peachy. Usually choices have pros and cons, and it's easier to get the public to notice the cons than it is to get them to evaluate the trade offs involved. Also, if the cons don't materialize within two years, the public will probably not connect the dots. So watch some cable news, form your half-ass opinion and go vote.

lotharbot 16y ago
> Most political policies don't have tangible win/lose results. No -- but occasionally politicians will state a particular tangible, directly measureable result they're expecting a policy to produce (either pro or con). They seem to be wrong as often as they're right, across the board. Ideally, we'd have a policy suggestion, true experts would give us a clear understanding of the tradeoffs involved and the actual effects of the policy, and we could evaluate them and make sensible decisions. What we actually get are policy suggestions, statements from both sides' hand-picked "experts" declaring fictional effects of the policy, we vote for whatever sounds the best to us, and half the time the end result is totally different from what we thought we were voting for. And pretty much all the time, both sides cherry-pick statistics that don't really measure what they say they're measuring in order to prove that they were right all along. And then we wonder why people are skeptical...

MichaelSalib 16y ago

Neither major US political party seems to do better than 50/50 overall According to Larry Bartel's research, in some major areas of economic policy, this is not true. See http://rodrik.typepad.com/dani_rodriks_weblog/2008/03/american-politi.html http://rodrik.typepad.com/dani_rodriks_weblog/2008/03/americ... and http://www.washingtonmonthly.com/archives/individual/2005_05/006282.php http://www.washingtonmonthly.com/archives/individual/2005_05... for more details. Of course, insofar as you're talking about public perceptions of news media "debates", that only confirms your point.

yummyfajitas 16y ago
While the graph is nice, error bounds would be extremely useful. Comparing two groups based on a sample size of 5 (per group) is an extremely tricky thing to do. It's even trickier when you try to rule out systematic errors by reducing your sample size further. For instance, the blog mentions one systematic error caused by D->R transitions. He then claims this is ruled out by analyzing two R->R and one D->D transition, ignoring the fact that 2 of those three transitions were based on fluke circumstances themselves (specifically Kennedy->Johnson and Nixon->Ford).