Announcement

**Jonas** · Sep 20, 2012, 12:43 AM

Thanks for sharing. Since I brought O'Rilley up in another thread I really should listen to this!

O'Rilleys political views aside I still find him very annoying - a loud white male in born in the 40's who consider himself to know best. I meet to many of those in my daily work and I've had it up to my ears with that attitude.

He may be invite all kinds of views but he constantly interrupts people which I find highly annoying.

Anyway thanks for sharing I think this could be interesting due to the election coming up, and if not for a political discussion it'll be interesting from a debate-psycological view.

oh, the link is broken - there's a [/url at the end

edit, and this is hilarious - it's actually more likely that you are less informed and have less knowledge watching Fox News then if you don't watch any news at all! ROFL (this is ofc statistics and should be taken with a pinch of salt)

http://www.poynter.org/latest-news/m...orst-informed/

b.r

Jonas

PS; .. and I don't find you closeminded - who could? when you have Kubuntu as Second Language!

EDIT; PS2; come think of it - it would be really interesting to hear Bill O'Rilleys thoughts Linux and Free software lol

**woodsmoke** · Sep 20, 2012, 05:08 PM

Hi jonas.

Thanks for the heads up, I fixed the linky!

I COMPLETELY AGREE that O'Reilly interrupts people, however, it doesn't bother me "most" of the time because,

a) "most of the time", he is interrupting them because they are not answering the question, they are making smoke and acting like they are answering the question.

b) and, he also interrupts people when he has a "hard break" coming up, and he says that is why, and forces another question in the thirty seconds left so he can get in all of his questions.

However, yes, sometimes he does just interrupt people, and yes it is irritating.

As to the survey, there have been other versions of the same kind of survey and they always put Fox at the bottom, however....

What follows is mainly for people who might want to know how the displayed numbers of the survey may not actually reflect the "real numbers" for the survey.

If one does not want to read that one can skop to the last two sentences.

a) it was conducted on landlines and also cell phones, so that is good, and we hope that they made assumptions about the statistical "constants" which will take into account that "nowadays";

i) the cell phone is used heavily by "younger" people
ii) while the people who maintain a landline, and will be home to answer it, are "older".

b) However they “oversampled” Republicans, I will come back to that.

c) If one looks at the survey particulars, I find that there was a VERY pregnant sentence in the writup;

They mentioned BOTH Fox and MSNBC as being "idologically" driven.

So that would seem to me to mean that they are at least admitting that there are news sources "other than Fox" that are "ideologically driven".

They also have Larry Sabato's book about Bush. It really was a relatively unbiased look at Bush's first six years, and Sabato is regularly on Fox News as a pollster and has always been relatively correct in his polling. That would mean, to me that “the University itself” is not “against” Fox News.

After going through an odd link, ( the NAME of the FDU university ) I found the original article, a pdf.

Here is the relevant outtake of what they actually said about the methodology.

The effects were calculated using
multinomial logistic regression, a technique that allows researchers to isolate the
impact of one variable on an outcome. The results described control for the effects
of partisanship, age, education and gender, all factors, which commonly predict vote
choice.
The FDU PublicMind study was based on a poll of 1,185 resident adults nationwide,
including an oversample of Republican voters, and was conducted by landlines and
cell phones from Feb. 6 through 12, 2012, and has a margin of error of +/-3
percentage points.

I've used MLR myself and am familiar with it.

The operative terms from the next sentence are:

a) "control for the effects" and b) "commonly".

"a)" is a mathematical assumption made by the person doing the statistics. The person puts “fudge factors” into the equations to “control” for things.

Notice that the MLR is "logistic", that means that it is used to try to “compromise” (as it were) between the “main set of numbers” and the “outlying numbers”.

One of the caveats of mlr is this:

If the multinomial logit is used to model choices, it may in some situations impose too much constraint on the relative preferences between the different alternatives. This point is especially important to take into account if the analysis aims to predict how choices would change if one alternative was to disappear (for instance if one political candidate withdraws from a three candidate race). Other models like the nested logit or the multinomial probit may be used in such cases as they need not violate the IIA.

http://en.wikipedia.org/wiki/Multino...tic_regression

Essentially, linear regression and logistic try to "make discernable" the various outcomes numbers that are in the "norm" of a group of answers and the "ends of the tail" of the group.

The "statistical" result will be affected by the "assumptions" that the worker makes in feeding "constants" into the equation and in this comes from the next part of the sentence:

factors, which commonly predict vote choice.

The worker "makes assumptions" about what is "common".

Those assumptions dictate the choice of the "constants" etc. that are in the equation and those assumptions then affects the amount of data which ends up in what "part" of the output.

So the outcome of that is that if the article is produced by someone with a "left bias" then the assumptions of what is "normal" will affect the "constants" etc. that go into the equation and that will "skew" the regression toward the "outlying" numbers or more toward the "norm", Same will happen if a "conservative" takes the numbers and churns them and makes an article.

So, now let us get back to the “main idea” of the article.

What one SEES in the included table is not just a simple division of the number of respondents into the various questions.

What one SEES is a “percentage” that has been put through the “sausage maker” of the MLR.

A GOOD thing is that they provide the actual hard data, which is below the tables, and the questions.

One can only get at this by clicking the “name of the university” at the top of the article which is:

Fairleigh Dickinson University

One gets a .pdf file and that is a good thing, except:......

When one clicks the link for the “methodologies” one gets looped back to the same original page upon which is buried the article itself:

However, again, here is what is provided in the .pdf.

The margin of error for a sample of 1185 randomly selected respondents is +/- 3 percentage points.

The margin of error for subgroups is larger and varies by the size of that subgroup.

Survey results are also
subject to non-sampling error. This kind of error, which cannot be measured, arises from a number of factors
including, but not limited to, non-response (eligible individuals refusing to be interviewed), question wording, the
order in which questions are asked, and variations among interviewers

The meaning of the bolded sentence above, is, quite simply, that the margin is over “3” and they do not say what the number is, it might be 3.3 or it might be 15.

The “non-sampling” error “cannot be measured”.

What? There is a built in error that cannot be measured?

So, what does one do to “compensate” for that?

The worker puts a “fudge factor” into the equation to “take care of that”.

BUT IT IS A GENERAL RULE that the error of a measurement should not be larger than the error of the calculation. (for folks with statistical background one can see the humour in that statement!

)

So.....since we do not know....

a) what the fudge factors are
b) what is considered “normal” responses
c) how the oversampling of Republicans was “adjusted” so that the oversampling did not “affect” the overall numbers....
d) how they dealt with “errors that cannot be measured”......

It is quite possible that the “raw numbers” said that Fox News watchers actually did better on the questions.

However, again......since they used the terms that both "MSNBC" and "Fox" are " ideologically driven, they "may" have actually put fudge factors into MLR that really did compensate for "both extremes".

That would be a good thing, seemingly, but we don't know because they did not actually provide the fudge factors that supposedly compensated for that.

But.... since there were two pregnant things.

a) they provided the sentence that both MSNBC and Fox are “ideologically driven” then possibly they do not have a “bias” against Fox.
b) They don't “mind” having a book by Sabato associated with them,

A look at the Wikipedia site shows what seems to be a rather “middle of the road” college that is more interested in “getting people an education” than being some hotbed of anti-conservatism or anti-liberalism. But then one does not know what is really going on if one is not there.

There is a link to this study at the bottom of the Wikipedia page.

http://en.wikipedia.org/wiki/Fairlei...son_University

This “rate your college” site has about even splits on whether the “college sucks” or the “college is just sooooo great”... but no complaints about it being “left wing” or “right wing”....

student reviews

A last caveat is that the college seems to pride itself about it's "surveys".

There are two possible implications for that:

i) The college really does want to provide a real, valid service which can showcase the quality of the college.
ii) the college is hurting for money and is going to try to leverage some studies into a position whereby they can make money by "providing predictable results" for a group with an agenda.

I, personally, don't find anything to indicate that the second implication is present so would opt for the first implication.

So, I will give the study a “neutral” on whether the numbers were “fudged”.

See... I just go where the data takes me!

woodsmoke

**claydoh** · Sep 20, 2012, 06:48 PM

I sometimes have watched O'Reilley, often seeing him bash idiots on the right. I do not consider him a republican, don't know (or care) if he has a party affiliation. Yes, he does yell, but not as often (from my viewings) as he is made out to.

In reality, or rather my opinion, the actual news coverages of the various news outlets are less biased than you may think. It is the Talking Heads analysis stuff that gets all the attention. Which of course is what these commercial entities program.

I stopped watching all the TV talking heads, and most talk radio because almost all of it is pure polarized rubbish. The truth is that most of TV and Old Media print news is left biased. But the vast airwaves in the radio portion of the spectrum are right-leaning. I will venture a guess that radio talk gets many more ears than cable TV news channels get eyes. Which means all of it has a bias.

Then there is the whole blogger-as-news-source crapola and the very scary internet comment streams.

Bill and John debating will be interesting. I will bet that there will be a number of areas where they agree if there is time to discuss a wide range of topics.

If anyone wants an un-partisan (as opposed to non-partisan, if that makes any sense) take on things, check out Dan Carlin's Common Sense

**woodsmoke** · Sep 20, 2012, 07:24 PM

well thank you claydoh, After watching a short while, it looks like a place that really is "un" .....

woodsmoke

**claydoh** · Sep 20, 2012, 08:58 PM

His history podcasts are even better (and longer)!

**Jonas** · Sep 21, 2012, 02:36 AM

@Woodsmoke - I agree with you on statistics (thus the pinch of salt comment)and thank you for your very in depth review on the actual article and numbers.

II think you hit the nail, (possibly without knowing), what Fox News and most media coverage lacks is in depth, for several reasons. For what I've seen of Bill O'Rilley (and this applies to more anchors then him) is that he isn't interested in hearing or having a more in depth discussion on the topics, he want to get to the point because they are there for the viewer rate. And the fact that he is male, white and in the 60's (?) makes him the most heard kind on this planet. Even yet he have to be this loud annoying character. It's possibly hes "style", but I can't stand it.

Every media station reports the "good" or the "bad" cause it sells, the "real" (if one could actually report that) and the in depth don't sell. Fox are very good at selling, and if they didn't give their angle there wouldn't be any point showing it at all.

Interesting that you make such a in depth read of the methodology of the poll, but let media coverage go so lightly ((though I'm not implying you would take any news without a critical stand point).

*A slight sidetrack: I grew up in one of the so called "trouble areas" in Stockholm, and I learned quiet early that the news coverage from national newspapers and TV was covering, was at best angled reports and too many times pure bulls**t, and it always boiled down to selling their report (and angle). *

@Claydoh - thanks for sharing the link, downloaded a couple of shows and will listen to it once I get to my studio.

b.r

Jonas

**woodsmoke** · Sep 21, 2012, 10:47 AM

Well,
there are continually people saying that Fox is not "hard news" and all "right wing opinion".

The White House itself is now saying that Fox News was correct about the killing of the Ambassador, that it was a "terrorist attack" and not a "spontaneous uprising" about a film, that may not even exist.

Fox was showing video of the night of the supposed "uprisings", it is still showing it, that there were some people with rocket propelled grenade launchers and no protestors.

This article reports what the White House spokesman says:

http://www.examiner.com/article/whit...ut-middle-east

woodsmoke

**whatthefunk** · Sep 21, 2012, 11:23 AM

Originally posted by woodsmoke View Post

So.....now the pregnant question is..... why has not a prominent PBS news personality, or a prominent NBC news personality offered to go head to head with a prominent "conservative" personality and to have a "debate" over topics pertinent to the election that is moderated by someone who "just asks a question of interest" and lets the people debate it.

Because PBS doesnt care about ratings. Why do you think these idiotic windbags are on TV to begin with? For their intelligence? Their extensive political knowledge? No. They are on TV because they get people fired up by presenting emotionally loaded half-truths which draw viewers. They give people someone to blame. Thats why we have all this crap about Liberals doing this and Elites doing that....for ratings. Its not unlike what Hitler did really. And what really gets me is that the people who are telling the masses this garbage are the ones who have the most to benefit from it. People seriously need to wake up.

**Qqmike** · Sep 21, 2012, 12:04 PM

whatthefunk is absolutely correct (Post #9). The political commentary programs on cable are there only because they make money for the owners. I posted elsewhere the tip to look up NBC's Rock Center 9-20-12, the first segment with Ted Koppel. In 15 minutes they did an excellent job covering an overview of that segment of the cable business, the heavy hitters on right and left, and drawing some conclusions about it all.

And, whatthefunk's point: "Why do you think these idiotic windbags are on TV to begin with? For their intelligence? Their extensive political knowledge? No. They are on TV because they get people fired up by presenting emotionally loaded half-truths which draw viewers. They give people someone to blame. Thats why we have all this crap about Liberals doing this and Elites doing that....for ratings."

Absolutely 100% correct. What kinds of real, regular jobs could some of those idiots hold down? Where else would they fit into normal society if not for their loud-mouthing and raging?

It is so stinking obvious now what is going on in this country (USA), the public had better wake up.

**oshunluvr** · Sep 21, 2012, 12:40 PM

Both of you are spot-on. Problem is it seems the only people who vote are the lunatics on the outer edges of society - at least that's what seems to sell. We need a serious centerist moderate third party governed by openness, fairness, truth, and common sense. Good luck finding a way to make money off of that, though. Turns out people who take the time to think through the issues don't have any piles of hate-laden, myopic, one-sided rhetoric to spew on the TV.

**woodsmoke** · Sep 21, 2012, 02:15 PM

qqmike, oshunluver, whatthefunk

Correctamundo as far as I am concerned.

I have argued AGAINST a third party for decades for one simple reason...look at all the European countries with multiple parties.

They can get NOTHING done.

And, that once a third party appeared we would then have a hundred parties and get nothing done.

Well, nothing getting done now!

So, I now think that we should "throw (all) the bums out!"

woodalwayswaspragmatist/kindofindependentsmoke

**Qqmike** · Sep 21, 2012, 02:37 PM

I found it, the link to the video of last night's Ted Koppel report, War of Words, on the warring cable "news" channels. Not long, just about 6 minutes, I think, and very interesting. A nice, brief, balanced conclusion. (Hope you have time to play this, Woody.)

Ted Koppel takes on the truthsayers
http://rockcenter.nbcnews.com/_news/...he-truthsayers

Of course, from that same segment came stuff like this:
Rock Center with Brian Williams: Coulter Claims O’Reilly “Loves” Obama (9/20/12 Recap)
http://www.nerdles.com/2012/09/20/ro...a-92012-recap/

She ranted wildly against the bias of the news media, saying they are NOT news. When asked who she thought the real news media is, in the most straight, serious manner, she replied, Fox News. At which point, the reporter (Koppel, I believe), was respectfully silent, letting the interview with her stand on its own.

**woodsmoke** · Sep 21, 2012, 03:33 PM

Thanks for the linkys Qqmike!

woodsmoke

**SteveRiley** · Sep 21, 2012, 04:58 PM

Originally posted by whatthefunk View Post

these idiotic windbags on TV

I call the Sunday morning guys "sabbath gasbags."

Originally posted by woodsmoke View Post

look at all the European countries with multiple parties. They can get NOTHING done.

Oh, America is doing a pretty good job of getting nothing done even though we have only two parties. (Or two shades of the same party pretty much nowadays, but that's another topic.)

Idea: perchance, might our vaunted two-party political system contribute to our two-ideology political landscape? Might it be that, if we had multiple parties, forced on occasion to actually friggin cooperate with each other, lead to less divisiveness and less rigidity of worldviews?

Announcement

Rumble in the air conditioned auditorium

Rumble in the air conditioned auditorium

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment