Live data from Hacker News

Obese politicians signal corruption, study finds

eurasianet.org

11–20 of 28 posts

Re: Obese politicians signal corruption, study finds

#11
post #4

In other news: > "'I was too fat' Boris Johnson says we must lose weight to fight Coronavirus" Perhaps the new svelter Boris will have the side benefit of reducing Tory expenses scandals.

Expenses scandals are a pointless distraction. The lowest hanging fruit of corruption in the UK is the stuffing of the House of Lords.

Re: Obese politicians signal corruption, study finds

#13
post #6

N = 299. Not sure how meaningful those results are.

I'm no statistician, but since the underlying population in this case (the set of all politicians) is much smaller than the general population, shouldn't that mean that a smaller sample size is more representative?

More representative, but difficult to draw any "real" predictive power

Re: Obese politicians signal corruption, study finds

#15
post #11
post #4

In other news: > "'I was too fat' Boris Johnson says we must lose weight to fight Coronavirus" Perhaps the new svelter Boris will have the side benefit of reducing Tory expenses scandals.

Expenses scandals are a pointless distraction. The lowest hanging fruit of corruption in the UK is the stuffing of the House of Lords.

But expenses scandals result in public perception of corruption (which is what the study's about), so logically, based on this definitely real and very sensible study, thinner PMs should lead to fewer of them!

Re: Obese politicians signal corruption, study finds

#16
In Lithuania we often call them 'beavers'- a standard picture of a man in his mid 50s, fat as hell and equally corrupt. At least we are having fewer and fewer of these. However,these occasionally get replaced by the other type- the 'slick ones'. Men in their mid 30s early 40s, outspoken, well looking and equally full of shit.

Re: Obese politicians signal corruption, study finds

#17

N = 299. Not sure how meaningful those results are.

N is 15, not 299. This is comparing one measure for each of the 15 countries against three other measures (cabinet minister's average obesity Vs. Transparency International Corruption Perceptions Index, World Bank worldwide governance indicator Control of Corruption, and Index of Public Integrity). The 299 is the combined number of cabinet ministers whose weight were averaged, which is a single measure per location,…

N = 15 is a lot larger than you think when the highest possible N is ~200.

Re: Obese politicians signal corruption, study finds

#18
post #17

Earlier quoted context omitted.

N is 15, not 299. This is comparing one measure for each of the 15 countries against three other measures (cabinet minister's average obesity Vs. Transparency International Corruption Perceptions Index, World Bank worldwide governance indicator Control of Corruption, and Index of Public Integrity). The 299 is the combined number of cabinet ministers whose weight were averaged, which is a single measure per location,…

N = 15 is a lot larger than you think when the highest possible N is ~200.

If the available sample is too small, some questions may not be answerable via statistical means, at least with any high degree of confidence.

This one may not be answerable via this methodology. Even if you did all 200 countries, there just may not be enough high quality data to reach a satisfactory conclusion.

I'm not even saying the conclusion is wrong, I am saying the data doesn't exist to determine if it is right OR wrong.

Re: Obese politicians signal corruption, study finds

#19
post #17

Earlier quoted context omitted.

N is 15, not 299. This is comparing one measure for each of the 15 countries against three other measures (cabinet minister's average obesity Vs. Transparency International Corruption Perceptions Index, World Bank worldwide governance indicator Control of Corruption, and Index of Public Integrity). The 299 is the combined number of cabinet ministers whose weight were averaged, which is a single measure per location,…

N = 15 is a lot larger than you think when the highest possible N is ~200.

The question is how many variables have they tested before settling on obesity.

Every single signal points for non-significance of this article. What, of course, is just reason to test it in another population. But instead people will just keep repeating it as if the findings were true.

Post reply on HN