Ask Skewz.com Founder About Detecting Media Bias

Skewz.com is not the Microsoft-funded Blews experiment that is supposed to help detect rightness and leftness in stories based on blogs that link to them. Instead of detecting blog links, Skewz relies on readers to submit and rate stories, and even tries to pair stories that have "liberal" and "conservative" biases so that you can get multiple takes on the same event or pronouncement. The Skewz About page explains how it works. The site has drawn a fair amount of "media insider" attention, including a writeup on the Poynter Institute website. But what does all this mean? Where is it going? Can Skewz.com help us sort our news better and make more informed decisions? We don't know. But if you post a question here for founder Vipul Vyas, maybe he'll have an answer for you. (Please try to follow the usual Slashdot interview rules.)

Re:So what is liberal or conservative?

[morgan_greywolf]

And how much difference does it make?

For example, [Ll]ibertarians don't tend to see politics in this light. They see things as "statist" or "non-statist". Any viewpoint favoring the rights of the individual over the power of State intervention is non-statist. To a [Ll]ibertarian left and right can both be wrong, as they may, and oftentimes DO, both represent a statist viewpoint.

Re:Self Selection = Inaccurate Data

[faloi]

This will merely attract the obsessive ultra-right crusaders to dump "left wing bias" en masse on everything.

Actually, a quick look at the site makes it look like the "far left kool-aid drinkers" (I think that's the right way to put it) are dumping "right wing bias" en masse on everything.

You are right, though. It's still not an accurate measure of bias. Some of the new stories appear to be filtered primarily by source rather than any particular bias. And some of the stories exist in the gray area, and don't have a really discernible bias.

Skewz me?

[Jeffrey Baker]

Skews makes no sense. Take this article as an example:

http://www.breitbart.com/article.php?id=080401184532.kxjxy7xo&show_article=1 [breitbart.com]

It's an AFP wire story with completely straight, factual reporting about high school graduation rates in the USA. There is no commentary from the author whatsoever. However Skewz users rate the story as "Liberal", giving it 2.5 out of 5 points on the Liberal scale. I'm having a hard time seeing the logic there. How can a purely factual report on this topic possibly be considered leftist?

Re:Why is everything about "bias"?

[Lendrick]

Case in point:

skews.com has this article [breitbart.com] rated as "liberal" -- it looks to me like it's just the result of a (somewhat alarming) study on education. This article here [foxnews.com] appears to have been labeled "conservative" just because it came from Fox News.

Re:Accounting for Regional Disparity

[Lemmy Caution]

NPR only has a left slant to it if you have an American notion of the left/right spectrum, and if you ignore the show "Marketplace", which often displays a market-fundamentalism that would make Ludwing von Mises blush.

Also, most public radio stations buy shows from a variety of sources, not all of which are NPR. American Public Media is another producer of public radio content, and is often chosen by public stations with more conservative demographics.

Re:Are all americans one dimensional

[Anonymous Coward]

And they will. They will stabilize you enough to ship you out. If you happen to survive once shipped out, then the lawyers take over and attach your ass from here to eternity.

Re:Are all americans one dimensional

[Mr. Slippery]

The "Left" is for strong central government--as you say, "top down"--(i.e. Federal government). The "Right" is for strong local control--as you say, bottom-up--i.e. States' rights.

So neocons who have striven to extend the power of the federal government are leftists? And Greens who work for more local control are right-wingers?

No. Federalist versus anti-federalist is a different dimension from left versus right.

The political terms left and right date to the French revolution, when nobility sat on the right and commoners on the left of the legislature. In modern terms, they refer to Labor and Capital. To be in favor of the interests of investors and owners is to be on the right; to be in favor of the interests of workers and ordinary citizens is to be on the left.

It doesn't matter whether you're an Maoist who believes in dictatorship of the peasants, or a anarchist who believes in no government and thus no private capital, you're a leftist; and it doesn't matter if you're a plutocrat who believes that the rich should control the government, or a libertarian capitalist in the minimal government that can enforce strong property rights, you're a right-winger.

Various alliances made over the years have obscured this, to the point where people think of gun control, censorship, abortion, foreign policy, and many other issues in left/right terms, but that's fuzzy thinking. Politics is multi-dimensional, and left-right is just one axis.

Re:Are all americans one dimensional

[chunk08]

So the government exists "to look after society?"

Yes.

No. The government exists to defend the life and liberty of its citizens. Looking after society is best left to religious groups (I subscribe to one in particular, but many seem to do a good job) and individuals.

Not to protect us and our liberty?

That you think the two are separate objectives speaks volumes.

Yes. It says that I am discerning and able to spot the difference. I don't want government forcing me to pay for someone else's healthcare, choosing my healthcare for me, forcing me to comply with insane "environmental" regulations which have no basis in fact, and telling me that I'm not smart enough to figure out how fast I can drive. I do want it defending our soil against invaders of any kind (especially "religious" extremists) and protecting my life, liberty, and property. That's what it's there for.

Re:Accounting for Regional Disparity

[Gat0r30y]

I think you missed what is most certainly the most influential - Parents. Democrat parents raise democrat kids, republican parents raise republican kids. Just like with religious beliefs , you gotta get em early. Liberals generally place a higher value on going to a liberal arts college and the whole higher-ed process, so their kids will want to please their parents by going to said liberal arts college.
On another note completely - about professors, there are generally two refuges for conservatives in higher education, Engineering and Business. As an engineering student I only experienced 1 prof with a liberal bias (this was primarily because he got his paycheck from NREL the national renewable energy lab) which gets cut by every republican president, and hasn't hardly gotten a budget increase since its inception under Carter. As for my business school classes. they would have made Ayn Rand blush for the most part.

Re:Skewz me?

[uncadonna]

Reality has a liberal bias, that's all.

Not this again...

[statemachine]

That study's conclusions are flawed and it even says so.

1) Right-wing bias of the study's authors including or excluding data:
Wanting to make sure the ACLU appears left-leaning by excluding data:

While most of these averages closely agree with the conventional wisdom, two cases seem somewhat anomalous. The first is the ACLU. The average score of legislators citing it was 49.8. Later, we shall provide reasons why it makes sense to define the political center at 50.1. This suggests that the ACLU, if anything is a right-leaning organization. The reason the ACLU has such a low score is that it opposed the McCain-Feingold Campaign Finance bill, and conservatives in Congress cited this often. In fact, slightly more than one-eight of all ACLU citations in Congress were due to one person alone, Mitch McConnell (R.-Kt.), perhaps the chief critic of McCain-Feingold. If we omit McConnell's citations, the ACLU's average score increases to 55.9. Because of this anomaly, in the Appendix we report the results when we repeat all of our analyses but omit the ACLU data.

Wanting to make sure that RAND appears left-leaning by including data:

The second apparent anomaly is the RAND Corporation, which has a fairly liberal average score, 60.4. We mentioned this finding to some employees of RAND, who told us they were not surprised. While RAND strives to be middle-of-the-road ideologically, the more conservative scholars at RAND tend to work on military studies, while the more liberal scholars tend to work on domestic studies. Because the military studies are sometimes classified and often more technocratic than the domestic studies, the media and members of Congress tend to cite the domestic studies disproportionately. As a consequence, RAND appears liberal when judged by these citations. It is important to note that this fact--that the research at RAND is more conservative than the numbers in Table 1 suggest--will not bias our results. To see this, think of RAND as two think tanks: RAND I, the left-leaning think tank which produces the research that the media and members of Congress tend to cite, and RAND II, the conservative think tank which produces the research that they tend not to cite. Our results exclude RAND II from the analysis. This causes no more bias than excluding any other think tank that is rarely cited in Congress or the media.

You can't pick and choose. Either include all of this type of data or exclude it -- don't just pick what supports your beliefs.

2) Right-wing bias in algorithm selection
Study admits that Fox News is way off in right-field if the actual average of Congress is taken:

Table 3:
Fox News' Special Report with Brit Hume 6/1/98 - 6/26/03 39.7 1.9

Figure 2 shows Fox and Washington Times far right of every other news outlet.

3) Study authors omit outright lies.

Citation 21:
Like us, Mullainathan and Shleifer (2003) define bias as an instance where a journalist fails to report a relevant fact, rather than chooses to report a false fact.

4) Different measures of center would seem to nullify any bias other than Fox and Wash. Times due to wide variances
Citation 34 ... "Since this number is 1.7 points les than the mean-based measure of the centrist voter (50.1), if one believes that it is the more appropriate measure, then our main conclusions (based on the mean-based measure) are biased rightward--that is, the more appropriate conclusion would assert that the media are an additional 1.7 points to the left of the centrist voter."
"Yet another measure ..." " The midpoint is 49.4, which is 0.7 points more conservative than our mean-based measure."

Citation 35 "If instead we use medians, the figure is 54.9"

The results are muddled at best. The authors clearly massage data to their liking (at least they admitted it), but this only serves to shoot down the whole paper. The study is fun to look at for entertainment, but its conclusions can hardly be taken seriously due to all the cherry picking, massaging, questionable data gathering, and just plain inconclusive data.