Atlas IMDB User Ratings Free Association Test
Posted by WesleyMooch 11 years, 7 months ago to Culture
I gave myself a completely unscientific but perhaps instructive test, using the results found on IMDB's Atlas Shrugged: Part 2 user ratings page.
Using free association, I attached a single word to the bars on the graphic which caught my eye (category, number of voters, rating, free association word):
Males under 18, 2, 10, typical
Males Aged 30-44, 54, 2.9, castrated
Females Aged 30-44 , 8, 8.4, reality-based
Lastly, without belaboring the obvious too much, a couple of observations:
There's a distinct gender gap both in number and perception of quality (or resonating worldview).
There's also a prominent age gap which would suggest seniors on the fence at best, but probably on the side of the State, when it comes right down to it.
Using free association, I attached a single word to the bars on the graphic which caught my eye (category, number of voters, rating, free association word):
Males under 18, 2, 10, typical
Males Aged 30-44, 54, 2.9, castrated
Females Aged 30-44 , 8, 8.4, reality-based
Lastly, without belaboring the obvious too much, a couple of observations:
There's a distinct gender gap both in number and perception of quality (or resonating worldview).
There's also a prominent age gap which would suggest seniors on the fence at best, but probably on the side of the State, when it comes right down to it.
I can understand their motivation with regard to avoiding vote stuffing, but that still does not explain how their weighted average is well below both the mean and median of the actual data. Normally, a weighted average would be somewhere between the mean and median; not vastly different from both.
If they are not counting multiple votes from the same IP adress, then why not just say that? And even still, would that actually be fair anyway, given that two people might potentially share the same computer and both have similar opinions of a film?