« It Should Be Noted... | Main | Home Again... Naturally. »
April 09, 2008
Drives Me Crazy
I have read at least 3 news stories today quoting numbers that are used to support an argument by the article's author. But while people are busy whining about blogs, how is it that they - and more importantly, their editors - getting that all stats are simply not the same.
An AOL poll about television is, a) based on people who are using AOL's site, b) based on a site, AOL, that is owned by Time-Warner, not an unbiased place to start, 3) naturally manipulated by the news, views, and ads on that AOL site, and 4) based on people who opt-in, who are notoriously unreliable.
I didn't run into it today, but my friend, Anne Thompson, runs stats from Fandango each week as though they are news. They are not. Of course there is a natural tendency for more ticket sales to reflect more popular movies. But audiences that buy tickets online, people who hang out on Fandango, the theater chains that Fandango serves, race, age, and sex all are variables that are not in any way counted in this "survey." With all due respect, it is nothing but a slightly interesting artifact.
And in today's NY Times, a lazy reading of the MPAA materials about the last year at the box office leads to a quote of a MPAA "stat," that is actually from the portion of their report that was based on a survey. There is a HUGE difference from a hard fact, based on hard numbers, and a number coming out of a survey... especially when movie surveys come up misleading time after time after time, based primarily on the questions asked, not any ill intent. (As I often comment... could YOU really tell someone on the phone tomorrow how many movies you will go to a theater to see this summer? Without looking at a release chart and estimating? Speaking for myself, I could see 20 mainstream releases... or 30.. or more... in the 16 weeks of the summer season.)
This really isn't about the constant backhanding of the web with cheap disdain... though that raises my fury. It really is about reporters and editors not paying attention to where their information is coming from. Sometimes a banana is just a banana. Sometimes, it is a soda or a pie or a penis or some pudding. That detail matters if you seek to make a legitimate argument and not just to support your position with any "survey" or "stat" that's handy.
Posted by poland at April 9, 2008 11:38 AM
Comments
Come on David - you know there are 3 kinds of lies:
Lies, statistics, and Peter Travers pull quotes.
Posted by: a1amoeba ![[TypeKey Profile Page]](http://www.mcnblogs.com/thehotblog/nav-commenters.gif)
at April 9, 2008 12:04 PM
David acts like he hasn't written a story with a hook in a million years. Actually he hasn't, he writes for a blog.
Posted by: T. Holly at April 9, 2008 12:41 PM
It really is about reporters and editors not paying attention to where their information is coming from.
I have two words for you: Judith Miller.
Posted by: Chucky in Jersey at April 10, 2008 09:05 AM
You know, most motion pictures are fiction. Why should writing about them be any different?
Posted by: doug r at April 10, 2008 01:19 PM
Is a lie a good hook, T Holly?
I know you hate substantive questions. But...
Posted by: David Poland at April 10, 2008 05:13 PM
Was it the AOL poll that said Ellen was the best daytime host? I missed the stories, can you link them?
Posted by: T. Holly at April 11, 2008 07:08 AM
Posted by: David Poland at April 11, 2008 02:56 PM
Seems like Cieply used the
http://www.mpaa.org/MovieAttendanceStudy.pdf
-- the report footnoe is: "All statistics throughout represent data from mid-July to mid-July of the following year - not calendar years."
So it's a survey? Even though,
"Admissions data is calculated from data sourced to Nielsen EDI and NATO."
Posted by: T. Holly at April 11, 2008 04:50 PM
The tip is the footnote in the *by nationality* section, "Survey is conducted only in English."
But it's acceptible to call trend tracking with year-to-year surveys (going back to 2003), statistics.
Let's ask the MPAA to release the questionaire and data about sample size, not pile on the people using them. OK? (sure, sure)
Posted by: T. Holly at April 12, 2008 08:28 AM
Post a comment
Thanks for signing in, . Now you can comment. (sign out)
(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)