Difference between revisions of "Talk:Liberal quotient"

From Conservapedia
Jump to: navigation, search
(Where did "12" come from??)
(Technically…: Sig)
 
(34 intermediate revisions by 15 users not shown)
Line 150: Line 150:
 
: Not that small.  Several hundred at least.  Probably a higher percentage than most polls of the American public.--[[User:Aschlafly|Aschlafly]] 01:22, 28 April 2007 (EDT)
 
: Not that small.  Several hundred at least.  Probably a higher percentage than most polls of the American public.--[[User:Aschlafly|Aschlafly]] 01:22, 28 April 2007 (EDT)
 
:: Sampling error isn't significantly affected by the percentage of the population included in the sample -- a representative sample of 1000 is as reliable in the USA as it is in Monaco. The problem in this case is that it's difficult to justify the assumption that those who self identify as liberal or conservative are representative of the community of Wikipedia editors. It may well be true that there are six times as many liberal Wikipedians than conservative, but it's not statistically reasonable to conclude that from a self-selected sample such as this.--[[User:Jalapeno|Jalapeno]] 04:52, 28 April 2007 (EDT)
 
:: Sampling error isn't significantly affected by the percentage of the population included in the sample -- a representative sample of 1000 is as reliable in the USA as it is in Monaco. The problem in this case is that it's difficult to justify the assumption that those who self identify as liberal or conservative are representative of the community of Wikipedia editors. It may well be true that there are six times as many liberal Wikipedians than conservative, but it's not statistically reasonable to conclude that from a self-selected sample such as this.--[[User:Jalapeno|Jalapeno]] 04:52, 28 April 2007 (EDT)
 +
:: The last time I looked at Wikipedia, the same size was 274 (for conservatives + liberals).  This is not a large sample.  The total percent sample size is about half that for a Harris poll (that it admits that it has errors that gives that it cannot give a good confidence[http://www.harrisinteractive.com/harris_poll/index.asp?PID=548]), however the sample for wikipedia is self selected meaning that you cannot use it to make statistically responsible statements.[http://www.aapor.org/pdfs/2006/samp_err_stmt.pdf] --[[User:Mtur|Mtur]] 22:30, 9 May 2007 (EDT)
  
 
2. In the absence of its inverse, the conservative quotient, defining the Lq is silly knee-jerk reactionism.
 
2. In the absence of its inverse, the conservative quotient, defining the Lq is silly knee-jerk reactionism.
 
: It's baffling, and a bit amusing, how liberals just won't admit they are liberal.--[[User:Aschlafly|Aschlafly]] 01:22, 28 April 2007 (EDT)
 
: It's baffling, and a bit amusing, how liberals just won't admit they are liberal.--[[User:Aschlafly|Aschlafly]] 01:22, 28 April 2007 (EDT)
 
+
:: It is baffling that [[User:Aschlafly|Aschlafly]] won't admit that he did bad math. He prefers to call people liberals instead, proving that he ran out of proper arguments.  --[[User:Order]] 10 May.
 
3. Judging by what Ashfly calls "liberal" on this site, the US is about 60% or so liberal and 30% or so conservative.  Leaving 10% "sheep".
 
3. Judging by what Ashfly calls "liberal" on this site, the US is about 60% or so liberal and 30% or so conservative.  Leaving 10% "sheep".
 
: This is false and doesn't make any sense.--[[User:Aschlafly|Aschlafly]] 01:22, 28 April 2007 (EDT)
 
: This is false and doesn't make any sense.--[[User:Aschlafly|Aschlafly]] 01:22, 28 April 2007 (EDT)
Line 193: Line 194:
 
What occurs to me is all that if Perry is correct, with the obsession with Liberals/Conservative points of view, many contributors on this site do not seem to have developed intellectually much further than their freshman year. Hm. Think I'll write an article about this...--[[User:CatWatcher|CatWatcher]] 03:55, 28 April 2007 (EDT)
 
What occurs to me is all that if Perry is correct, with the obsession with Liberals/Conservative points of view, many contributors on this site do not seem to have developed intellectually much further than their freshman year. Hm. Think I'll write an article about this...--[[User:CatWatcher|CatWatcher]] 03:55, 28 April 2007 (EDT)
  
 +
 +
See [[Perry Theory]] for the article --[[User:CatWatcher|CatWatcher]] 19:55, 29 April 2007 (EDT)
  
 
== Where did "12" come from?? ==
 
== Where did "12" come from?? ==
Line 199: Line 202:
  
 
:It would be informative if somebody with access to the appropriate data could calculate the LQ for Conservapedia, for purposes of comparison? --[[User:Jeremiah4-22|Jeremiah4-22]] 12:16, 28 April 2007 (EDT)
 
:It would be informative if somebody with access to the appropriate data could calculate the LQ for Conservapedia, for purposes of comparison? --[[User:Jeremiah4-22|Jeremiah4-22]] 12:16, 28 April 2007 (EDT)
 +
 +
== An X for a Y, for an L, for a C ==
 +
 +
This article has a few problems.
 +
 +
#X is defined as number of liberals, and Y is defined as number of conservatives. X/(X+Y) is '''the same''' as L/(L+C). The article argues that they are different, proving that the author doesn't know how to work with variables. And it proves that the author doesn't read his own definitions.
 +
#The criticism that a measure is between 0 and 1, shows that the critic doesn't know that such a measure is equivalent to a measure that ranges between 0 and infinity. Take for example the negative logarithm of the first measure. It just show that the author has not the faintest idea about functions over an infinite domain.
 +
#The remark that an increase from 90% to 99% is ''just'' a 10% increase, shows that the author doesn't know how to interpreter percentages. People who do work with percentages know that the difference between 90% and 99% is an order of magnitude.
 +
#The comparison with other metrics is flawed, because the author didn't compare it with any metric that is actually used in literature. Earlier versions of this article pointed to alternative definitions of liberal bias. None of them is mentioned.
 +
#Finally, this article repeats the old misinformation that the ratio liberal in the wikipedia is 3:1, while it is 1:2 in the American public. It doesn't mention the sources, nor that, if you look at the sources, this comparison is based on nothing but quick sand. It just shows that the author has no idea about statistics.
 +
 +
As a summary, this article proves nicely that in conservapedia conservative bias is more important than facts. Even basic mathematics, which is neither liberal or conservative, is canned to make a poor argument. [[User:Order]] 10 May
 +
 +
I saw that Aschlafly fixed some optinionated part of the article, without actually fixing any of the fundamental flaws. I would suggest to fix the fundamental flaws first. It, for example, still argues that X/(X+Y), where X are Liberals, and Y Conservatives, is different from L/(L+C), where L stands for Liberals, and C for Conservatives. If you want to fix something, fix that first. [[User:Order]] 11 May.
 +
 +
== Update ==
 +
 +
I extended and updated the article on ''liberal quotient''. If you find typos or factual errors, please feel free to fix it. [[User:Order]] May 13.
 +
 +
== How conservative is the US? ==
 +
The article says "According to a Harris Poll from 2005 [1], 40% of all Americans identify as moderates, 35% as conservatives, and 18% as liberals. This gives a liberal quotient of about 1:2. These numbers have been fairly constant during the last decades.", but according to Aschlafly, Jack Abramoff wasn't a conservative<ref>http://www.conservapedia.com/index.php?title=Talk:Deceit&diff=158024&oldid=158009</ref>, despite his ties to Grover Norquist, Ralph Reed, Oliver North; his membership on G. W. Bush's transition team, etc.  Rudy Giuliani made a statement held up as a "liberal falsehood of the day"<ref>http://www.conservapedia.com/index.php?title=Template:Tfa&diff=154374&oldid=153131</ref>, so is presumably not a conservative either.  Given this, it certainly doesn't appear to me that merely self-identifying as such is enough to be considered a 'conservative' for this site, and the Harris Poll shouldn't be used to determine how conservative the country is. --[[User:Jtl|Jtl]] 00:42, 13 May 2007 (EDT)
 +
:You make a good point, that will go ignored here. [[User:Human|Human]] 01:17, 13 May 2007 (EDT)
 +
 +
this may be picky, but wouldnt' this liberal quotient be more accurately called a liberal conservative ratio?  Something about this jangles the statistics part of my brain.[[User:Prof0705|Prof0705]] 13:09, 15 May 2007 (EDT)
 +
 +
: That's redundant.  It would be like requiring everyone to say how hot versus cold it is outside, rather than simply "how hot is it?"--[[User:Aschlafly|Aschlafly]] 13:15, 15 May 2007 (EDT)
 +
 +
Not really: temperature is the correct term.
 +
And while heat is real, cold is just the absent of heat (absence of energy).
 +
 +
While Conservatism and Liberalism are both real political movements: there is no reason why this couldn't be called the conservative quotient, or the liberal/conservative ratio.
 +
 +
Anyone ever considered that wikipedia might be more liberal than the American public, because it has a lot of non-American editors?
 +
And despite the gossip on wikipedia, a large part of its edits are scientific/mathematical in nature, and highly educated people simply tend to be more liberal than average Joe...
 +
[[User:MiddleMan|MiddleMan]]
 +
: This why the liberal quotient is considered so useful. If wikipedia would have for 99.6% articles that are NPOV, for 0.3% with a liberal POV, and 0.1% articles with a conservative POV, its liberal quotient would still be 3:1. It doesn't matter how neutral Wikipedia is, as long as the biased fraction is biased towards liberals. You might argue that if Wikipedia's articles would be for 99.6% NPOV, Conservapedia would have no reason to complain, but Conservapedia thinks differently. And it should indeed be called liberal/conservative quotient, to be precise. [[User:Order]] May 16, 9:50 am (AEST)
 +
:: If that was to be the measure, then conservapedia would have an infinite conservative quotient which means that it is also infinitely more conservative than the American populace because there are no articles that have a liberal bias.  --[[User:Mtur|Mtur]] 20:03, 15 May 2007 (EDT)
 +
 +
==References==
 +
<references/>
 +
 +
== Is the denominator the total number of individuals in a population or the number of non-liberals? ==
 +
 +
I see people don't agree on the definition of this term. When I hear the term for the first time, I imagine the denominator is the total number of individuals, which is the sum of the number of non-liberals and the number of liberals. Otherwise I'd see it as a confusing term. --[[User:Pepsi-Cola|Pepsi-Cola]] 22:44, 26 June 2007 (EDT)
 +
 +
: The moderates add nothing to the mix.  Including the moderates in the denominator results in the loss of information, as 1 liberal and 99 moderates is very different from 1 liberal and 99 conservatives.--[[User:Aschlafly|Aschlafly]] 23:20, 26 June 2007 (EDT)
 +
 +
:: Hence moderates do add to the mix. You say it yourself, it matters if there are 99 conservatives, or 99 moderates. [[User:Order]] July 3
 +
 +
== Are these figures out of date? ==
 +
 +
I followed the links in the article, and it seems there are now 81 Liberal Wikipedians and 71 Conservative Wikipedians, with 26 moderates. This reduces Wikipedia's liberal quotient from 3:1 to a mere 1.15:1. If these statistics are so prone to change, can they really be trusted? --[[User:JonathanDrain|JonathanDrain]] 18:48, 24 July 2007 (EDT)
 +
 +
== Identical criticisms ==
 +
 +
Although most of the stuff under "Criticism" relates to different things, there are two parts whihc appear to be the same:
 +
:Critics of the liberal quotient point out that it discounts groups identifying neither as liberal nor conservative, such as the 40% of moderates in the American public. A group with 97 moderates, 2 liberals, and 1 conservative has the same liberal quotient as a group with 1 moderate, 66 Liberals and 33 conservatives, although some might claim the latter group is more liberal.
 +
And this:
 +
:An encyclopedia with 99.7% unbiased articles, 0.2% liberally biased articles, and 0.1% conservatively biased articles, has the same liberal bias as a product with 66.6% liberally biased articles, 33.3% conservative articles , 0.1% unbiased articles. However, the first would be considered a reliable source of information, while the second would be highly unreliable. And this despite having the same liberal quotient.
 +
Both make the point that the use of ratios can distort the data range. Merge them? [[User:PeterS|PeterS]] 20:33, 30 September 2007 (EDT)
 +
 +
: Sure, you can merge them, but both criticisms are defective.  The first fails to recognize that only "undecideds" are in the true middle, and they should be discounted because they will just follow the stronger side.  The second fails because the [[liberal quotient]] is about people, not articles or things.  But go ahead and merge as you like.  Godspeed.--[[User:Aschlafly|Aschlafly]] 20:49, 30 September 2007 (EDT)
 +
::They are different, because articles that are unbiased do not follow the stronger side. The points it criticizes are (1) that it discounts in groups the influence of the middle group, and (2) that the middle group in an encyclopedia, the unbiased articles doesn't swing. If the latter wasn't made sufficiently clear we can try to make the difference more obvious. [[User:Order|Order]] 06:20, 1 October 2007 (EDT)
 +
 +
== Technically… ==
 +
 +
A quotient ''does'' range from 0-1. Sorry. --[[User:SimonA|SimonA]] 16:48, 18 December 2007 (EST)
 +
 +
:Actually, it can be any value, however, in the context… --[[User:AngryCommunist|<font color="#ff0000">Angry</font><font color="#000000">Communist</font>]] 17:49, 2 January 2008 (EST) (approx.)

Latest revision as of 22:49, January 2, 2008

What a fabulous Encyclopedia entry, with entries like this I don't see why no one is using this site?

Fantastic!! This means we now have a wonderful new method of assessing Conservative Bias, and for the first time it is measurable. To calculate the Conservative Quotient (CQ), all one has to do is to know the Liberal Quotient (LQ), then apply the formula: CQ = 1 - LQ.

A rating above zero presumably means that there is Conservative Bias?

Shurely Shome mishtake?

--CatWatcher 13:39, 5 April 2007 (EDT)


80%?

A source for that might do nicely. DrLib 14:08, 5 April 2007 (EDT)

There is a poll under journalism students suggesting that 80% have liberal opinions. I will look it up when I have time. I saw another research suggesting that journalists are socially liberal, but economically more conservative. Few oppose NAFTA, for example. So its a well-off kind of liberalism. User:Order 7 April 00:45 (AEST)

Liberal quotient of Conservapedia

Not counting the vandals, I'm genuinely curious what our liberal quotient is. MountainDew 14:13, 5 April 2007 (EDT)


Most of the liberals here have been pretty disenfranchised, a few still prawl around the talk pages but they are not really "part of the group" so I don't think its very high. There is what, about 40 active users, maybe a dozen or so are liberals, but I think they have been pretty effectively shut down by bans, warnings, and condescending and insulting sysops. Etaroced 14:23, 5 April 2007 (EDT)

(Inserted comment) I'm opposed to most condescension and nearly all insults. Please help me develop guidelines for a standard of Conservapedia:Civility. --Ed Poor 10:49, 6 April 2007 (EDT)

Sneaky and deceptive as they are by nature, I suspect many liberals are unlikely to freely admit to being liberals, therefore making an accurate measurement of the liberal quotient here difficult to obtain. --Jeremiah4-22 16:22, 5 April 2007 (EDT)

Technically, and I don't expect some people on here to understand this, I think you would have to say everyone on here is moderately liberal. We allow for the addition of material that might question our beliefs, could contribute to free and open debate. Those are classicly liberal ideas. Don't like that--go to Saudi Arabia where they aren't so liberal. Flippin 10:26, 6 April 2007 (EDT)
I think that liberals 'decorate' our project rather nicely when they aren't making sly digs. As far as I'm concerned, they are welcome to stay and particapate as equals. I think some of the syosops are liberals, not that I want to out anyone.
But if they are here to disrupt and play tricks, like the Seattle WTO protesters who provoked the police into "excessive force" just so that they could accuse 'the establishment' of 'not listening to them' then they aren't helping anyone. Serious thinkers see right throught that kind of malarkey.
Any liberals who want to preserve their franchise would do well to couch their suggestions in polite terms. Any idea can be brought up, if done in (1) the right way (2) in the right venue (3) at the right time. (My talk page is always open. :-) --Ed Poor 10:30, 6 April 2007 (EDT)

Liberal Quotient will always be a subjective factor, of course. If one assigned an LQ based purely on party association, and started digging into American history a little bit, one would find Democrats who went around lynching blacks, and Republicans who went around trust busting. Teresita 11:15, 6 April 2007 (EDT)

'But if they are here to disrupt and play tricks, like the Seattle WTO protesters who provoked the police into "excessive force" just so that they could accuse 'the establishment' of 'not listening to them' then they aren't helping anyone. Serious thinkers see right throught that kind of malarkey. ' But the police still used excessive force. That is a legal fact. You keep bringing up this accusation, Ed Poor; scared?

Oh, and I've made this. Wikinterpreter

Fox News

Quoting from article: Television news anchors on the major networks had a 100 percent liberal quotient until Fox emerged

Does this mean that Fox News is a Bad Thing as it upset the finely balanced (and therefore unbiased) 50:50 distribution of liberal and conservative anchors?


Formula

I'd rather define it so that 100 percent is all Liberal, 50 percent is half and half, and 0 percent is no Liberals. --Ed Poor 10:21, 6 April 2007 (EDT)

I don't think Order's changes helped the article:
The liberal quotient is defined simply as the ratio of liberals to conservatives in a group. Thus the liberal quotient is zero when there are no liberals in the group. It is infinity if there are no conservatives. A quotient of 100% corresponds a group that has as many liberals as conservatives, while a quotient below 100% means that there are fewer liberals than conservatives. The same number of conservatives and liberals will always result in the same liberal quotient, regardless of the size of other groups, such as moderates or independents. [I marked up the part I didn't like with strikeout formatting. --Ed Poor 10:46, 6 April 2007 (EDT)
You may not like it, but that's how it's defined in the article. If the quotient is ratio of liberals to conservatives in a group, then the liberal quotient is infinity if there are no conservatives. And liberal quotient of 1 amongst news anchors means that there were equal numbers of liberals and conservatives Chrysogonus 13:49, 8 April 2007 (EDT)

Why didn't my changes help. I just tried to clarify the definition. User:Order 7 April, 1:00 (AEST)

You changed the definition. I think 100% Liberal is all liberals, no conservatives; and 0% liberal is no liberals and all conservatives. Using infinity would make the quotient too cumbersome to use. --Ed Poor 11:01, 6 April 2007 (EDT)

I agree w/ Ed. But I also think that we should present 0% as ideal. --BenjaminS 11:04, 6 April 2007 (EDT)

I disagree with the fact that I changed the definition. But for the moment we can keep it like it is. User:Order 7 April, 1:10 (AEST)
I put this in two places, oops. Here is the "correct and useful" formula: Liberal Quotient = L/(L+C), that is, the number of identified Liberals divided by the sum of identified Liberals and Conservatives. Likewise, the CQ = C/(L+C). This results in a number from zero to one in either case, which can be expressed, as in the original definition, as a percentage.
I just edited the article to reflect this. I hope no one minds. Human 17:37, 9 April 2007 (EDT)

Mathematics

There are, obviously, so many flaws in this whole idea that it is hard to know where to begin. First with definitions of Lib and Cons. I suspect than many of us are not simply one or the other. Also, if LQ=L/C then the LQ of, for instance, the Democratic Party is LQ=1000/0=undefinable.

My 2 cents, just dont make me divide them by zero.--PalMDtalk 14:13, 8 April 2007 (EDT)

One thing to begin with would be to use the definition in the article itself, rather than the alternative that it tries to replace. User:Order 9 April, 21:25 (AEST)
What do you mean? The definition in the article is liberals/conservatives, which gives an undefinable answer when there are no conservatives. Chrysogonus 08:15, 9 April 2007 (EDT)
Oh, dividing by zero is a minor problem. It has a few other problems, such as that 997 moderates, 2 liberals, and 1 conservative have the same liberal quotient, as 1 moderate, 666 liberals and 333 conservatives. To avoid dividing by zero, we can agree that the value is then infinity.
What bothers me most is it that of all the Liberal quotients mentioned in the article none of them was computed according to the definition. And this is the article defining the liberal quotient. And if you look an this talk page, you see that few people do actually understand the definition. Most think its the fraction of liberals or something similar. User:Order 9 April, 22:45 (AEST)
I have to agree with User:Order's assessment. I mean, are people just making these up? That doesn't sound very useful, and it certainly undermines Conservapedia's usefulness as a source! Sterile 13:44, 9 April 2007 (EDT)
Why can't we have people self-identify? I know Richard tried this at one point and got into some hot water (which I won't expand on)but the underlying principle was valid--if this wiki is supposed to contain only conservative content how can we not ask people to tell us the truth about where their allegiance lies? Flippin 13:55, 9 April 2007 (EDT)
Self-identifying won't work, especially not on this site. Andy and his crew are so conservative that almost everyone is a liberal to them (if you don't believe me, look at Andy's posts and how everyone who disagrees with him is labeled a liberal). In most circles I'd be considered a conservative, but here I'm nothing but a dirty liberal hippie (and that's a good thing as I'd rather not be grouped with some of the wingnuts here). Jrssr5 14:03, 9 April 2007 (EDT)
Here is the "correct and useful" formula: Liberal Quotient = L/(L+C), that is, the number of identified Liberals divided by the sum of identified Liberals and Conservatives. Likewise, the CQ = C/(L+C). This results in a number from zero to one in either case, which can be expressed, as in the original definition, as a percentage. Human 17:05, 9 April 2007 (EDT)
You changed the definition. It is now no longer consistent with the long discussion on bias in Wikipadia [1][2][3][4][5]User:Order 10 April, 11:52 (AEST)
Why do you ask people to self-identify in this context? The problem is that Aschalfy first defines the liberal quotient as Liberals/Conservatives and then doesn't use it when he computes e.g. the LQ of news anchors. This is not a matter of liberal or conservative, it is a matter of math. User:Order 10 April, 8:25 (AEST)

A new term coined by Conservapedia?

A new term coined by Conservapedia? I find that hard to believe and suspect it was infact Aschlafly who coined the term. WhatIsG0ing0n 13:59, 9 April 2007 (EDT)

The term definitely existed before. Simple google search proves that. However, I assume that the term has its own, special meaning on this site. Not sure, but it's possible. --Sid 3050 14:04, 9 April 2007 (EDT)

Andy got a lot of comments on the way he arrived at his measure of Liberal bias of Wikipedia; one main point was that using liberal/conservatives to measure bias is confusing. Research uses typically other measures. So, Andy decided to declare his alternative way to compute bias an 'innovation'. User:Order 10 April, 8:35 (AEST)

Is it actually possible to classify individuals as being absolutely Conservative or Liberal? With the exception of a few especially pure characters (and I salute your dedication, guys) most people are hopelessly flawed and (at best) somewhat intermediate. So to calculate the LQ of a group accurately, a method is first needed of assessing the Liberality of all individuals included in the party. Otherwise a further level of inaccuracy is necessarily introduced into the calculation.
(As an example, although George W Bush is considered Conservative by most people, any good Conservative can see right through him to his pernicious Liberal tendencies - to call him a true Conservative is clearly erroneous, when in fact he is probably only 70-75% Conservative (Just a wild guess here on my part...).)
If the LQ is to be a useful measure, it needs to be calculated with rigor; a roomful of right-leaning people clearly shouldn't have the same LQ as a roomful of true Conservatives, which seems to be the situation at present. --Jeremiah4-22 21:06, 9 April 2007 (EDT)

Why does formula not matter?

Ed Poor just deleted the section on calculating the LQ and said "(formula is not the point)" in his comment. I don't see why not, especially when numbers are quoted in the last paragraph. Please justify this removal. To call something a "quotient" by definition means that some numbers are being divided. If there is a good reason to leave it out, fine. If not, I will replace it, and, if I can find good examples, add a few references. User:human 21:15, 9 April 2007 (sig added aroound 02:15, 10 Apr.)

If you can come up with a formula as precise as IQ, fine. Otherwise, please let's all stop teasing Andy about this. --Ed Poor 22:17, 9 April 2007 (EDT)
Andy came up with a definition. And Human too. They were both very precise. Or do you claim otherwise? You cannot hide the defintion, just because few understand it, and few use it correctly. User:Order 10 April, 12:22 (AEST)

This article gets better and better. Now it states that the definition is not clear (which is fair enough) yet still has figures for the liberal quotient of journalists and news anchors. I guess that at least makes it clear that these numbers were more or less plucked out of thin air. Chrysogonus 16:24, 10 April 2007 (EDT)

Responding to User:Ed Poor above, I have a couple of questions.
First, and simplest, is what does any of this have to do with "teasing Andy"? I'm really lost on that one.
Second, do you know how IQ numbers are determined? You give a large group of developmentally "equal" people (like a given grade in school, or age) a standardized test, that is set up so all of them will get some of it right, but none of them can get it all correct. Then you take the raw scores and create a transfer function to map them to a Gaussian distribution curve (bell curve). The peak of the curve is marked as "100", and the score scale is created by picking a numerical increment to be equal to each standard deviation (sigma). This is typically 15-16 points = one sigma, making it so five sigmas each way yields IQs of 20-25 and 175-180 - very rare figures indeed, since five sigmas corresponds to less than 0.0001% of the population (I forget the exact number, it's actually even smaller).
I do think that the formula presented (which "explains" the numbers in the last paragraph exactly) is simpler. Using it, for example, on the number of self identified liberals and conservatives on wikipedia (212 vs 75 today), the LQ = 212/(212+75) = 212/287 = 73.9% or 0.739. The CQ is 75/287 = 26.1% or 0.261. The network news example starts off pre-Fox at 3 liberal and no conservatives, LQ = 3/(3+0) = 1 or 100%. You can apply this formula to any group where you have a number for each preference (i.e., registered Republicans and Democrats out of all registered voters).
I really don't care whether or not this article makes "sense", but since it is so easy to have it make sense, why not do it? The original text just wasn't thought all the way through, as numerous mathematically astute editors have mentioned above, in an attempt to make this a better article. Human 13:06, 11 April 2007 (EDT)

Black protestants

I'm glad to see that sterotyping is allowed here (see Federal Reserve) Czolgolz 10:24, 10 April 2007 (EDT)

That black protestants are overwhelimingly "liberal" is actually true as this PEW report shows.[6] Order 11 April, 9:50 (AEST)

Dumbest file imaginable.

Until there is a metric for determining the "liberal quotient" (which is not hard to come up with), it is really embarrassing (to conscientious contributors to this site, and its "owner") and stupid to offer numbers that supposedly represent the LQ of certain groups. Check my previous comments, and deleted edit of 4/9/07 AD ([7]) to see what I mean. I mean, come on. I have no problem with the concept, as a bit of a geek I kind of think it's cool. But as that same geek, I think the formula should be presented. Along with more examples. Andrew, I ask you, you made an accident of language in your first formulation, feathers to chickens*, we all make mistakes, it was obvious what you meant. I (and other contributors) just wanted to get it right. Human 02:04, 19 April 2007 (EDT)

*Every disarming epithet I thought of to use here (heck, darn, drat, doggone) I realized were "nice" ways of using profanity, so I made up my own. Human 02:04, 19 April 2007 (EDT)

LQ = zero

I know some people's palms sweat when they imagine the concept. Human 23:05, 27 April 2007 (EDT)

Innumeracy

As I have said, it doesn't matter to me one way or another what this article (or anything on this blog) says, but, I would I would expect those publicly associated with it to strive for relatively sensible versions of their, um, version of reality. (There are precisely 57 six times as many liberals on Wp as in America!)

No one said "precisely."--Aschlafly 01:22, 28 April 2007 (EDT)

1. The metric of "six times" depends on an incredibly small number of self-identified Wpdians.

Not that small. Several hundred at least. Probably a higher percentage than most polls of the American public.--Aschlafly 01:22, 28 April 2007 (EDT)
Sampling error isn't significantly affected by the percentage of the population included in the sample -- a representative sample of 1000 is as reliable in the USA as it is in Monaco. The problem in this case is that it's difficult to justify the assumption that those who self identify as liberal or conservative are representative of the community of Wikipedia editors. It may well be true that there are six times as many liberal Wikipedians than conservative, but it's not statistically reasonable to conclude that from a self-selected sample such as this.--Jalapeno 04:52, 28 April 2007 (EDT)
The last time I looked at Wikipedia, the same size was 274 (for conservatives + liberals). This is not a large sample. The total percent sample size is about half that for a Harris poll (that it admits that it has errors that gives that it cannot give a good confidence[8]), however the sample for wikipedia is self selected meaning that you cannot use it to make statistically responsible statements.[9] --Mtur 22:30, 9 May 2007 (EDT)

2. In the absence of its inverse, the conservative quotient, defining the Lq is silly knee-jerk reactionism.

It's baffling, and a bit amusing, how liberals just won't admit they are liberal.--Aschlafly 01:22, 28 April 2007 (EDT)
It is baffling that Aschlafly won't admit that he did bad math. He prefers to call people liberals instead, proving that he ran out of proper arguments. --User:Order 10 May.

3. Judging by what Ashfly calls "liberal" on this site, the US is about 60% or so liberal and 30% or so conservative. Leaving 10% "sheep".

This is false and doesn't make any sense.--Aschlafly 01:22, 28 April 2007 (EDT)

4. An appropriate acale for a measure like this would create a middle point of "1" or "0", with a clear reflection about that point in each direction. So it should run form infinity to minus infinity, or 10 to -10, or 1 to -1. or, gee, 0 to 1.

No, the sensible "middle point" is unity, or 1. Do you have your same objection to the scientific (Kelvin) scale for temperature?--Aschlafly 01:22, 28 April 2007 (EDT)

5. This is a really dumb article

You didn't help your cause with that conclusion.--Aschlafly 01:22, 28 April 2007 (EDT)

Discuss amongst yourselves. Human 23:43, 27 April 2007 (EDT)

Ah, the Parthian Shot. Fine, I hope you honor it.--Aschlafly 01:22, 28 April 2007 (EDT)
I presume that means I relinquished the right to quibble with your responses. I'll honor that. if there is anything, however, that you want further comment on, feel free to ask me here or on my user talk page. Human 02:35, 28 April 2007 (EDT)

X/X+Y

Andy, I appreciate that you are discussing the formulation in the article. I think it adds a real measure of accountability and honesty. I have one quick comment though - X+Y is not the whole population, it is only the "self identified" portion, so it does not dilute the results with moderates. Multiplied by 100, it makes a percentage. Remembering older edits, the major networks would be at 1 (100%), (3/3+0), then came Fox - making it 75% (3/3+1).

It is a perfectly useful metric, and does not bias anything one way or another. Anything over 50% is "suspect" of liberal bias, and anything under is imbued with conservative trustworthiness (see, I didn't even get mean and say "trusworthiness" - remember who was first to point out the typo to get the logo fixed!).

Please, I ask you, run some numbers from various groups to see how it works (the formula). I think you will see that it works very nicely, and is also very clear and understandable.

Oh, and one last thing, us math geeks would tend to use L and C as the variables, for simple clarity (LQ=(L/L+C)). Using lambda and gamma would hurt our keyboards, but believe me, if it was easy, we'd use the Greek! Human 02:35, 28 April 2007 (EDT)

Who's "Liberal", anyway?

Are Libertarians the equivalent of the square-root of negative one in this formula? How about LaRouchites? Log Cabin Republicans? Natural Law Party activists? Tories? Bull Moosers? Anti-Masonic Revivalists? Pat Buchanan?

In a world of rainbow colors, it can be hard to decide with are "black" and which are "white". <SARCASM>Strangely, it's been my experience that no more than about 5% of Democrats actually have Mark-of-the-Beast tattoos and smell like brimstone, and only about 2% of Republicans actually have wings and a halo.</SARCASM> --Gulik2 03:36, 28 April 2007 (EDT)

Gulik2 is absolutely right. See Perry (1970) 'Forms of Intellectual Development in the College Year: A scheme'. Perry says that intellectual development occurs as follows:

  1. 'Dualism' - a simple interpretation of the world as black/white, us/them , good/bad etc.
  2. 'Multiplicity'; -= an acceptance that there are different points of view and they cannot be simply labelled good/bad etc.
  3. 'Relativism' - requires that the different points of view are contextualiised, and that knowledge/behaviour in one context may be perfectly acceptable and correct, may not be appropriate in another.

for a source see: http://www.csub.edu/TLC/options/resources/handouts/scholarship_teaching/perrystheory.html

What occurs to me is all that if Perry is correct, with the obsession with Liberals/Conservative points of view, many contributors on this site do not seem to have developed intellectually much further than their freshman year. Hm. Think I'll write an article about this...--CatWatcher 03:55, 28 April 2007 (EDT)


See Perry Theory for the article --CatWatcher 19:55, 29 April 2007 (EDT)

Where did "12" come from??

Ed, where did your "12" come from?--Aschlafly 09:22, 28 April 2007 (EDT)

It would be informative if somebody with access to the appropriate data could calculate the LQ for Conservapedia, for purposes of comparison? --Jeremiah4-22 12:16, 28 April 2007 (EDT)

An X for a Y, for an L, for a C

This article has a few problems.

  1. X is defined as number of liberals, and Y is defined as number of conservatives. X/(X+Y) is the same as L/(L+C). The article argues that they are different, proving that the author doesn't know how to work with variables. And it proves that the author doesn't read his own definitions.
  2. The criticism that a measure is between 0 and 1, shows that the critic doesn't know that such a measure is equivalent to a measure that ranges between 0 and infinity. Take for example the negative logarithm of the first measure. It just show that the author has not the faintest idea about functions over an infinite domain.
  3. The remark that an increase from 90% to 99% is just a 10% increase, shows that the author doesn't know how to interpreter percentages. People who do work with percentages know that the difference between 90% and 99% is an order of magnitude.
  4. The comparison with other metrics is flawed, because the author didn't compare it with any metric that is actually used in literature. Earlier versions of this article pointed to alternative definitions of liberal bias. None of them is mentioned.
  5. Finally, this article repeats the old misinformation that the ratio liberal in the wikipedia is 3:1, while it is 1:2 in the American public. It doesn't mention the sources, nor that, if you look at the sources, this comparison is based on nothing but quick sand. It just shows that the author has no idea about statistics.

As a summary, this article proves nicely that in conservapedia conservative bias is more important than facts. Even basic mathematics, which is neither liberal or conservative, is canned to make a poor argument. User:Order 10 May

I saw that Aschlafly fixed some optinionated part of the article, without actually fixing any of the fundamental flaws. I would suggest to fix the fundamental flaws first. It, for example, still argues that X/(X+Y), where X are Liberals, and Y Conservatives, is different from L/(L+C), where L stands for Liberals, and C for Conservatives. If you want to fix something, fix that first. User:Order 11 May.

Update

I extended and updated the article on liberal quotient. If you find typos or factual errors, please feel free to fix it. User:Order May 13.

How conservative is the US?

The article says "According to a Harris Poll from 2005 [1], 40% of all Americans identify as moderates, 35% as conservatives, and 18% as liberals. This gives a liberal quotient of about 1:2. These numbers have been fairly constant during the last decades.", but according to Aschlafly, Jack Abramoff wasn't a conservative[1], despite his ties to Grover Norquist, Ralph Reed, Oliver North; his membership on G. W. Bush's transition team, etc. Rudy Giuliani made a statement held up as a "liberal falsehood of the day"[2], so is presumably not a conservative either. Given this, it certainly doesn't appear to me that merely self-identifying as such is enough to be considered a 'conservative' for this site, and the Harris Poll shouldn't be used to determine how conservative the country is. --Jtl 00:42, 13 May 2007 (EDT)

You make a good point, that will go ignored here. Human 01:17, 13 May 2007 (EDT)

this may be picky, but wouldnt' this liberal quotient be more accurately called a liberal conservative ratio? Something about this jangles the statistics part of my brain.Prof0705 13:09, 15 May 2007 (EDT)

That's redundant. It would be like requiring everyone to say how hot versus cold it is outside, rather than simply "how hot is it?"--Aschlafly 13:15, 15 May 2007 (EDT)

Not really: temperature is the correct term. And while heat is real, cold is just the absent of heat (absence of energy).

While Conservatism and Liberalism are both real political movements: there is no reason why this couldn't be called the conservative quotient, or the liberal/conservative ratio.

Anyone ever considered that wikipedia might be more liberal than the American public, because it has a lot of non-American editors? And despite the gossip on wikipedia, a large part of its edits are scientific/mathematical in nature, and highly educated people simply tend to be more liberal than average Joe... MiddleMan

This why the liberal quotient is considered so useful. If wikipedia would have for 99.6% articles that are NPOV, for 0.3% with a liberal POV, and 0.1% articles with a conservative POV, its liberal quotient would still be 3:1. It doesn't matter how neutral Wikipedia is, as long as the biased fraction is biased towards liberals. You might argue that if Wikipedia's articles would be for 99.6% NPOV, Conservapedia would have no reason to complain, but Conservapedia thinks differently. And it should indeed be called liberal/conservative quotient, to be precise. User:Order May 16, 9:50 am (AEST)
If that was to be the measure, then conservapedia would have an infinite conservative quotient which means that it is also infinitely more conservative than the American populace because there are no articles that have a liberal bias. --Mtur 20:03, 15 May 2007 (EDT)

References

  1. http://www.conservapedia.com/index.php?title=Talk:Deceit&diff=158024&oldid=158009
  2. http://www.conservapedia.com/index.php?title=Template:Tfa&diff=154374&oldid=153131

Is the denominator the total number of individuals in a population or the number of non-liberals?

I see people don't agree on the definition of this term. When I hear the term for the first time, I imagine the denominator is the total number of individuals, which is the sum of the number of non-liberals and the number of liberals. Otherwise I'd see it as a confusing term. --Pepsi-Cola 22:44, 26 June 2007 (EDT)

The moderates add nothing to the mix. Including the moderates in the denominator results in the loss of information, as 1 liberal and 99 moderates is very different from 1 liberal and 99 conservatives.--Aschlafly 23:20, 26 June 2007 (EDT)
Hence moderates do add to the mix. You say it yourself, it matters if there are 99 conservatives, or 99 moderates. User:Order July 3

Are these figures out of date?

I followed the links in the article, and it seems there are now 81 Liberal Wikipedians and 71 Conservative Wikipedians, with 26 moderates. This reduces Wikipedia's liberal quotient from 3:1 to a mere 1.15:1. If these statistics are so prone to change, can they really be trusted? --JonathanDrain 18:48, 24 July 2007 (EDT)

Identical criticisms

Although most of the stuff under "Criticism" relates to different things, there are two parts whihc appear to be the same:

Critics of the liberal quotient point out that it discounts groups identifying neither as liberal nor conservative, such as the 40% of moderates in the American public. A group with 97 moderates, 2 liberals, and 1 conservative has the same liberal quotient as a group with 1 moderate, 66 Liberals and 33 conservatives, although some might claim the latter group is more liberal.

And this:

An encyclopedia with 99.7% unbiased articles, 0.2% liberally biased articles, and 0.1% conservatively biased articles, has the same liberal bias as a product with 66.6% liberally biased articles, 33.3% conservative articles , 0.1% unbiased articles. However, the first would be considered a reliable source of information, while the second would be highly unreliable. And this despite having the same liberal quotient.

Both make the point that the use of ratios can distort the data range. Merge them? PeterS 20:33, 30 September 2007 (EDT)

Sure, you can merge them, but both criticisms are defective. The first fails to recognize that only "undecideds" are in the true middle, and they should be discounted because they will just follow the stronger side. The second fails because the liberal quotient is about people, not articles or things. But go ahead and merge as you like. Godspeed.--Aschlafly 20:49, 30 September 2007 (EDT)
They are different, because articles that are unbiased do not follow the stronger side. The points it criticizes are (1) that it discounts in groups the influence of the middle group, and (2) that the middle group in an encyclopedia, the unbiased articles doesn't swing. If the latter wasn't made sufficiently clear we can try to make the difference more obvious. Order 06:20, 1 October 2007 (EDT)

Technically…

A quotient does range from 0-1. Sorry. --SimonA 16:48, 18 December 2007 (EST)

Actually, it can be any value, however, in the context… --AngryCommunist 17:49, 2 January 2008 (EST) (approx.)