Xaranid
Smash Apprentice
- Joined
- Sep 18, 2009
- Messages
- 151
To the people arguing about statistics, I agree that most of you don't seem to have ever taken a statistics course. I have taken several masters level statistics courses, so I have an idea of what I'm talking about.
1) I don't think anyone is trying to immediately equate the scores on this poll to the population distribution of "votes" out of everyone who downloaded Project M. If anything it's a good indicator of this board where the observed sample is coming from. Hell it doesn't even include everyone on this board but it doesn't have to, because:
2) All of statistics is estimation. You never have a sample size that is as large as the population, because if you did there wouldn't be a point in running more complex data - just poll everyone and you'd get the true value. What this poll can do is show trends in a sample of the total population (of 40,000+ or whatever it is), and from that data try to extrapolate trends that you can apply to the population values.
Because the sample size is small you couldn't make a case for "hey 80% of EVERYONE wants Isaac in", but you CAN say that "Out of our sample of 150 user responses, 80% of people voted favorably to Isaac's inclusion", and from there get some idea about the consensus. I'm not going to go further into statistics theory since this obviously doesn't need that much analysis, but you get the idea.
So uh, Masked Man for Project M! (But seriously if you haven't played Mother 3, it has the most tragic and heart wrenching final battle that I've ever experienced.)
1) I don't think anyone is trying to immediately equate the scores on this poll to the population distribution of "votes" out of everyone who downloaded Project M. If anything it's a good indicator of this board where the observed sample is coming from. Hell it doesn't even include everyone on this board but it doesn't have to, because:
2) All of statistics is estimation. You never have a sample size that is as large as the population, because if you did there wouldn't be a point in running more complex data - just poll everyone and you'd get the true value. What this poll can do is show trends in a sample of the total population (of 40,000+ or whatever it is), and from that data try to extrapolate trends that you can apply to the population values.
Because the sample size is small you couldn't make a case for "hey 80% of EVERYONE wants Isaac in", but you CAN say that "Out of our sample of 150 user responses, 80% of people voted favorably to Isaac's inclusion", and from there get some idea about the consensus. I'm not going to go further into statistics theory since this obviously doesn't need that much analysis, but you get the idea.
So uh, Masked Man for Project M! (But seriously if you haven't played Mother 3, it has the most tragic and heart wrenching final battle that I've ever experienced.)