Atlas IMDB User Ratings Free Association Test

Posted by WesleyMooch 12 years, 2 months ago to Culture
21 comments | Share | Flag

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.
SOURCE URL: http://www.imdb.com/title/tt1985017/ratings


Add Comment

FORMATTING HELP

All Comments Hide marked as read Mark all as read

  • Posted by freedomforall 12 years, 2 months ago
    The graph is not the actual vote, its IMDB's secret "weighted average" of the vote, which converts the median actual rating of 9 (mean of 6.9) to the displayed IMDB rating of 4.8. They refuse to reveal their formula. Since it is done in secret, I conclude it is politically biased against liberty. You cannot judge the voters on this because it is NOT their vote being displayed. Its more like the major US political parties taking 50 public opinion polls with slightly different phrasing of questions on each, and then publishing only the one that they decide the public should see. It is Bullshit Propaganda.
    Reply | Mark as read | Best of... | Permalink  
    • Posted by lostinaforest 12 years, 2 months ago
      As someone who works with statistics on a daily basis, I cannot see how they would have arrived at a weighted average of 4.8, given the data shown here: http://www.imdb.com/title/tt1985017/rati...

      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?
      Reply | Mark as read | Parent | Best of... | Permalink  

FORMATTING HELP

  • Comment hidden. Undo