Hey, @St_Rev. Since I'm in a mood, I have a math(ish) question for you. It's silly. I'll make it a thread.
-
Show this thread
-
1) OK. Suppose there is a convergence over time with Siskel and Ebert's movie ratings. Compared with a control group of randomly selected critics, S&E come to increasing agreement in their assessment of movies (I think this might be true, but ... no matter).
1 reply 0 retweets 0 likesShow this thread -
2) Once this is established, it seems that S or E is exacting a dominant influence (over S or E), yes. My math-ish question, then: Is it possible, with only this information, to determine who (S or E) is driving the convergence (or exacting influence) in movie ratings? No, right?
2 replies 0 retweets 1 likeShow this thread -
3) If the answer is obviously "no" (as I suspect it is), please understand that I'm wondering about sociometrics (or econometrics?), applied to a simple trend.
1 reply 0 retweets 0 likesShow this thread -
Replying to @NineBandedBooks
The analogous sociometric/econometric tool (which is better founded) is to see if X and Y correlate better after you time-shift one of the trends.
1 reply 0 retweets 1 like -
Replying to @St_Rev @NineBandedBooks
My associate
@metapotat might have a better/deeper answer1 reply 0 retweets 0 likes -
Replying to @St_Rev @NineBandedBooks
If you have information about the movies themselves, then it should be possible in principle. See what movie characteristics get high ratings from S & from E in time t and t + 1, and compare
1 reply 0 retweets 1 like -
If all you have is ratings, then you could do it by seeing which other critics both S and E correlate with in time t and t + 1. Maybe even cluster the critics and see if S or E (or both) switch clusters over tjme
1 reply 0 retweets 1 like
But if all you have are S and E's ratings, I don't think there's anything you can do. You need to be able to compare both to _something_
-
-
Replying to @metapotat @St_Rev
Thanks. I went to bed thinking it could probably be done by plotting against control critics' thumbs, but not in isolation. My guess, should anyone ever be bored enough to test this, is that S & E did converge, and that E exacted the dominant influence.
0 replies 0 retweets 1 likeThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.