Liberal Media Bias Proven--UCLA Political Scientist

fishin' wrote:

btw, here is a detailed critique of the study:
http://itre.cis.upenn.edu/%7Emyl/languagelog/archives/001169.html

and the study author's response to it:
http://itre.cis.upenn.edu/~myl/languagelog/archives/001301.html

Just in case anyone is really interested.

fishin

I think you've got an older study there.

A couple of relevant points. First, these two guys are Olin Fellows. That is, they have been funded by the Olin Foundation. The Olin Foundation funds NOTHING BUT conservative movement members/supporters. That is the foundation's stated mission.

An earlier study, the one noted above, used a methodology which placed Fox as centrist. A rather odd outcome. The present study, using another methodology, has as a part of its outcome as described by Milyo on Monday night on Tucker Carlson's show on MSNBC...here's a portion of that transcript...

We devote an entire table, Table 2, toward comparing the median and means of the entire Congress to the means of each party. As we note, the Republican mean is 11.2. Meanwhile the Democratic mean is 74.1. By no stretch of the imagination is 39.0 in the middle of the Republican party. In contrast, it is almost exactly equal to the midpoint of the middles (means) of the two parties.[/QUOTE} Yet when you go to look at table 2 it cites the avg democrat as
[quote]average Democrat 84.3

Interesting, I was reading the response by the study authors here
http://itre.cis.upenn.edu/~myl/languagelog/archives/001301.html
but the information they talk about there doesn't really relate to their paper that they link to.

For instance, they state...

Quote:

We devote an entire table, Table 2, toward comparing the median and means of the entire Congress to the means of each party. As we note, the Republican mean is 11.2. Meanwhile the Democratic mean is 74.1. By no stretch of the imagination is 39.0 in the middle of the Republican party. In contrast, it is almost exactly equal to the midpoint of the middles (means) of the two parties.[/QUOTE} Yet when you go to look at table 2 it cites the [color=red]avg[/color] democrat as

Quote:

average Democrat 84.3

Somthing is fishy when they can't even cite their own study correctly. Did they change the numbers after the answered the critique? Or are they out there in their critique?

Even more interesting... All of these "Press Releases" are nothing more than blogs.

Media Matters has a link to the full report but for some reason they stripped off the 1st page so their copy is undated.

The actual report (with the title page) can be found here: http://www.missouri.edu/~milyoj/ with the annotation that it is "forthcoming in the Quarterly Journal of Economics". (Yippie!)

It's dated December 2004.

So basiclly this entire thing is just fodder for the blog-set. The report is a year old and there is only one. Why the hell are people getting their panties in a twist over this? It's a year old and someone just finally noticed it or what???

parados wrote:

Interesting, I was reading the response by the study authors here
http://itre.cis.upenn.edu/~myl/languagelog/archives/001301.html
but the information they talk about there doesn't really relate to their paper that they link to.

For instance, they state...

Quote:

We devote an entire table, Table 2, toward comparing the median and means of the entire Congress to the means of each party. As we note, the Republican mean is 11.2. Meanwhile the Democratic mean is 74.1. By no stretch of the imagination is 39.0 in the middle of the Republican party. In contrast, it is almost exactly equal to the midpoint of the middles (means) of the two parties.[/QUOTE} Yet when you go to look at table 2 it cites the [color=red]avg[/color] democrat as

Quote:

average Democrat 84.3

Somthing is fishy when they can't even cite their own study correctly. Did they change the numbers after the answered the critique? Or are they out there in their critique?

Psst! A "mean" and an "average" are two different beasts.

In Stats when you refer to a means as a midpoint you take the highest and lowest values of a set and divide by 2. The average can vary from that quite a bit depending on where the majority of values fall within that set and how many individual values are in the set.

If you have a set of 1,000 people and the low = "1" and the high is "100" the mean is 50. If the other 998 people = "2" your average is 2.097.

Fun with numbers.