Anonymous K joined in and replied with this 15 years ago, 2 hours later, 4 hours after the original post[^][v]#226,227
@OP > no errorbars > extremely small sample size > the top result has a sample size of 500 > no information on how the samples were taken or even what the question was
2/10 bud
Anonymous E replied with this 15 years ago, 15 minutes later, 5 hours after the original post[^][v]#226,235
@previous (K)
If the sampling is done properly, the sample size is fine. Lower sample sizes for places such as Iceland, Luxembourg, Malta, etc. is fine for areas representing smaller populations. Those numbers aren't anything different then you would see in most professional opinion polls.
(Edited 2 minutes later.)
Anonymous K replied with this 15 years ago, 7 minutes later, 5 hours after the original post[^][v]#226,242
@previous (E)
The problem is there's no way to actually pick random samples because this was likely done via some type of internet survey, so they've already excluded large portions of people who don't have internet access. With non-random sampling you need a much larger sample size to compensate for the selection biases.
Simple error bars would help address this, but without them we are left in the dark entirely about the validity.
Anonymous L joined in and replied with this 15 years ago, 5 minutes later, 5 hours after the original post[^][v]#226,254
@previous (K)
True, not to mention an internet survey introduces voluntary bias based upon the fact that only certain groups of people might visit that website (such as Fox News hosting a survey is going to appeal to more conservative people with a more conservative answer), but I'm sure you probably already knew this.
green joined in and replied with this 15 years ago, 3 minutes later, 5 hours after the original post[^][v]#226,258
Turkeys are an evolutionary throwback to more sophisticated birds like the pheasant and peacock.
Anonymous N joined in and replied with this 15 years ago, 1 hour later, 6 hours after the original post[^][v]#226,291
@OP
Is it just me, or is it that all of the countries high on the list are in deep shit with debt, or because they belong to a union that is in deep shit?
Anonymous E replied with this 15 years ago, 5 hours later, 12 hours after the original post[^][v]#226,418
@226,242 (K)
You can roughly guess the standard error at a few percentage points for most of the countries just going by the sample size. If you are assuming the sampling was done reasonably well, error bars wouldn't be all that enlightening.
> The problem is there's no way to actually pick random samples because this was likely done via some type of internet survey, so they've already excluded large portions of people who don't have internet access. With non-random sampling you need a much larger sample size to compensate for the selection biases.
Yeah, that would be more of a worry. Anyone can do a FOX News-style poll where you ask your biased audience biased questions and then report the result as if it meant something. You would probably have to dig more into the sampling methodology and poll questions to get satisfactory answers about possible bias.
hurr joined in and replied with this 15 years ago, 57 minutes later, 13 hours after the original post[^][v]#226,451
> Comparison of public acceptance of evolution theory in 34 countries. > 34 countries > rule 34 > no porn
well this thread fails especially hard
Anonymous H replied with this 15 years ago, 6 minutes later, 13 hours after the original post[^][v]#226,459
@226,227 (K) > >no information on how the samples were taken or even what the question was
This chart depicts the public acceptance of evolution theory in 34 countries in 2005. Adults were asked to respond to the statement: "Human beings, as we know them, developed from earlier species of animals." The percentage of respondents who believed this to be true is marked in blue; those who believed it to be false, in red; and those who were not sure, in yellow. http://news.nationalgeographic.com/news/bigphotos/21329204.html