1
TalkBack / RE:Nintendo Fifth Largest Company in Japan
« on: July 28, 2007, 06:47:19 AM »
Excellent news
This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
Quote
Originally posted by: Gamer DonkeyQuote
Its called a sample. Thats how you take a poll when it is impossible with your resources to question the entire populus. Take the news for example, do you think they ask everyone in America their opinion for their polls? Heck, I didn't get a call. And people do lie, but thats one risk of taking polls. This was not meant in a offensive manner, only in friendly-informative fashion, but a little humor thrown in for spice.
The people sampled in this survey are not equal across three variables, hence you can not compare them to one another. The survey goes:
x number from group k said product a was this reliable.
y number from group l said product b was this reliable.
z number from group m said product c was this reliable.
x,y,z being your results
k,l,m being three independent samples
a,b,c being three independent variables
You can not say, "Fact, all these products are the same across the board". Now if the products were different but the group sampled were the same, well then you could say, "My results show that among this sample this product was this reliable". In aveng's survey, he went to three different forums and asked the same question about three different products. Then he said, there you have it. Fact, all of these products are the same. I have no problem with him doing the survey. Nice effort in fact. But to call this survey the unfalliable proof that these products are the same is inaccurate.
What aveng did is three independent surveys. Fine. But what you can not do is say that the three surveys are identical and here is my conclusion. They are in fact three independent surveys with no correlation. The sample used is independent across the three forums and the product asked about is independent across the three forums.