3 ms·

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

gpanders 6y ago
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?
- rodiger 6y ago
  More representative, but difficult to draw any "real" predictive power
Someone1234 6y ago
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, not the actual subject of the statistical question (which is a country, not an individual cabinet minister). It also isn't in a peer reviewed journal at all.
- teej 6y ago
  N = 15 is a lot larger than you think when the highest possible N is ~200.
  - Someone1234 6y ago
    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.
  - marcosdumay 6y ago
    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.
  - Arnt 6y ago
    I think the limit is closer to 400 since some federal states have states as big as many countries, but very good point.
rsynnott 6y ago
I think it may not be entirely serious (or at least the news article's coverage of it isn't).

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?

rodiger 6y ago
More representative, but difficult to draw any "real" predictive power

Someone1234 6y ago

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, not the actual subject of the statistical question (which is a country, not an individual cabinet minister). It also isn't in a peer reviewed journal at all.

teej 6y ago
N = 15 is a lot larger than you think when the highest possible N is ~200.
- Someone1234 6y ago
  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.
- marcosdumay 6y ago
  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.
- Arnt 6y ago
  I think the limit is closer to 400 since some federal states have states as big as many countries, but very good point.

rsynnott 6y ago

I think it may not be entirely serious (or at least the news article's coverage of it isn't).