The Instigator
ConserativeDemocrat
Con (against)
Losing
0 Points
The Contender
BWhite
Pro (for)
Winning
2 Points

Does the website "PolitiFact" have a liberal bias?

Do you like this debate?NoYes+0
Add this debate to Google Add this debate to Delicious Add this debate to FaceBook Add this debate to Digg  
Post Voting Period
The voting period for this debate has ended.
after 1 vote the winner is...
BWhite
Voting Style: Open Point System: 7 Point
Started: 6/30/2016 Category: News
Updated: 1 year ago Status: Post Voting Period
Viewed: 933 times Debate No: 93257
Debate Rounds (2)
Comments (14)
Votes (1)

 

ConserativeDemocrat

Con

I'm too lazy to write rules.

I have seen this multiple times; the website Politifact haas liberal bias. Evidence? They rate more claims by conservatives false then they do liberals. To me, this doesn't show any bias. It just shows more conservatives say false or ludicrous things. For example, Donald Trump is often fact checked, which will obviously increase the amount of false conservative claims. For example, Donald Trump believes climate change is a Chinese hoax. Liberals just don't say things like that on the same frequency as conservatives. Also, since our political spectrum is so far to the right, we don't see many insane far-left claims, because that position isn't represented in the media or by politicians. And since our spectrum is on the right, more conservatives will make claims then liberals, so more conservatives will get fact checked, as they have more claims. I believe I have madey point. Good luck to whoever accepts this short debate.
BWhite

Pro

I offer thanks to "Conservative Democrat" for extending this challenge. PolitiFact's leftward lean does not yet receive the full attention it deserves, so I appreciate the opportunity to make the case.

The fact checkers at PolitiFact, referring to PolitiFact National, have a liberal bias. My argument will make this clear based on a preponderance of the evidence. While that should prove sufficient for the argument to succeed based on the terms of the debate, this argument should also convince the audience beyond a reasonable doubt that PolitiFact National has a liberal bias.

What's the evidence?

First, American journalists as a group are to the left of the general U.S. population (1). Second, PolitiFact was created by the St. Petersburg Times (now the Tampa Bay Times), a newspaper with a notable history of partiality to the left. For example, the newspaper has never in over 100 years endorsed a Republican in the general presidential election (2). The most important evidence, though, comes from PolitiFact's content, not from the likely predispositions of its writers and editors or the editorial history of its parent newspaper. Three lines of evidence together make a strong case that the liberal lean at the Times manifests itself in PolitiFact's fact checking. The first two lines come from research. The third derives from PolitiFact's poorest attempts at fact-checking, which tend to unfairly harm Republicans while sparing Democrats.

The first research project (3) uses PolitiFact's rating scale to show unjustified harm to Republicans compared to Democrats. The research focuses on the difference between the "False" and "Pants on Fire" Truth-O-Meter ratings.

A continuing research project I designed provides a powerful evidence of the bias at PolitiFact National. PolitiFact's failure to provide any objective means of distinguishing between its "False" and "Pants on Fire" ratings strongly suggests the ratings reflect an editorial opinion about select statements PolitiFact finds false. PolitiFact decides its ratings based on the vote of three (possibly more) editors (4). The research project randomly surveyed about a quarter of PolitiFact's "Pants on Fire" ratings in search of an unstated objective principle justifying the ratings. The findings showed nothing close to a consistent result, supporting the hypothesis that PolitiFact has no objective justification backing its "Pants on Fire" ratings (5).

The 2012 study showed the apparently subjective "Pants on Fire" rating was 74 percent more likely to fall on a Republican false statement than a Democrat false statement, covering the years from 2007 through 2011. At the end of 2014, updated statistics showed that Republicans were still 74 percent more likely to have their false (by PolitiFact's reckoning) statements rated "Pants on Fire" than those of Democrats (6).

The same research approach was applied to other PolitiFact franchises, including its state and "PunditFact" franchises. The franchises that deliver the most robust volume of data show fairly consistent trends which do not necessarily track with the trend at PolitiFact National. PolitiFact Wisconsin, for example, tends to give Democrats' false statements harsher treatment than those from Republicans by more likely giving them the "Pants on Fire" designation (7).

The other research project was based on reasoning that journalists could be expected to make a common math error more often in cases where the error favored their ideology. Using the often-made error of using the wrong denominator when calculating percentage error, the research project used a search routine to collect a set of PolitiFact stories that might contain percentage error calculations. The research approach turned up a small sample of stories with that equation present, yet the results were lopsided. Out of 14 examples reasonably classified as percentage error calculations, PolitiFact used the wrong denominator nine times (8).

Out of those nine errors, six either unfairly harmed a Republican or helped a Democrat. Three either helped a Republican or harmed a Democrat (9).

The third line of evidence comes from an examination of PolitiFact National's poorest work. If PolitiFact National's worst work favors Democrats over Republicans, then it serves as an evidence of PolitiFact National's liberal bias. The strength of that evidence depends on the strength of the anecdotes offered in illustration. I'll provide examples as space allows.

A) PolitiFact Florida ruled that a Republican candidate raised his own pay a number of times while serving in the U.S. House of Representatives (10) A later parallel fact check by PolitiFact Ohio rated a Republican candidate's claim that his Democratic opponent had raised his own pay in the House. The reason? PolitiFact Ohio correctly said the rules prevent representatives from raising their own pay (11). Though it involves PolitiFact Florida, this example contributes to the argument because the writer (Louis Jacobson) principally writes for PolitiFact National (12).

B) A PolitiFact defense of the ACA absurdly counted the ACA's projected deficit reductions as spending cuts to help justify rating the ad of a conservative group "False." PolitiFact claimed the Congressional Budget Office said the ACA would cut spending by $124 billion (13). But the CBO actually said the ACA would cut the budget deficit by $124 billion (14). Spending is not the same as the budget deficit. This PolitiFact Florida item was written by Angie Drobnic Holan, now chief editor at PolitiFact.

C) PolitiFact found it "Mostly True" President Obama's claim that health care was the single biggest factor driving down the budget deficit, despite the fact that health care costs were increasing (15). The only things that can shrink the budget deficit are costs that drop or revenues that go up. Slowing the growth of a category of costs does not decrease the budget deficit (16).

D) During one of his State of the Union addresses, President Obama mentioned the 23-cent gender wage gap, called it "wrong" and said women deserve equal pay for equal work. PolitiFact ruled that Obama was correct that a 23-cent wage gap exists, docking him only for an apples-to-oranges aspect of the statistic. PolitiFact said Obama's call for equal pay for equal work was not part of his gender pay gap claim, excusing the former as "aspirational" (17). The Washington Post Fact Checker disagreed, finding Obama implied the gap was relevant to gender discrimination (18).

In the space remaining, I'll address the argument that PolitiFact gives Republicans worse rating as a group than Democrats.

By itself, more poor ratings for Republicans says essentially nothing about bias at PolitiFact. Bad ratings for Republicans might occur because Republicans make more false statements. However, when we consider Republicans' poor record with PolitiFact in context, it serves as an additional, albeit weak, evidence of PolitiFact's liberal bias.

It is important to note that we have no scientifically supported baseline for which political party makes more false claims. So when PolitiFact rates Republicans more harshly than Democrats, it supports no reasonable prediction about PolitiFact's ratings matching a known tendency. On the other hand, knowing that journalists lean left makes it reasonable to predict that if that bias affects PolitiFact's ratings then it may result in harsher ratings for Republicans. The harsh ratings for Republicans support a prediction associated with the hypothesis that PolitiFact has a liberal bias.

So, it is reasonable to conclude that PolitiFact has a liberal bias because journalists tend to lean left, two research approaches support the idea, and PolitiFact's most glaring mistakes tend to unfairly harm Republicans or aid Democrats.

Conservative Democrat will need to either refute each prong of the argument or instead provide affirmative evidence of either PolitiFact's neutrality or its conservative bias, otherwise it is more reasonable to accept that PolitiFact has a liberal bias.

Thanks again to Conservative Democrat for issuing this challenge.

(1) http://www.journalism.org...
(2) http://web.tampabay.com...
(3) https://drive.google.com...
(4) http://www.niemanlab.org...
(5) https://drive.google.com...
(6) http://www.politifactbias.com...
(7) http://www.politifactbias.com...
(8) http://www.politifactbias.com...
(9) http://www.politifactbias.com...
(10) http://www.politifact.com...
11) http://www.politifact.com...
12) http://www.politifact.com...
13) http://www.politifact.com...
14) http://www.cbo.gov... (Page 3)
15) http://www.politifact.com...
16) http://www.zebrafactcheck.com...
17) http://www.politifact.com...-/
18) http://www.washingtonpost.com...
Debate Round No. 1
ConserativeDemocrat

Con

Thank you for accepting! I hope we can have an interesting discussion.

"First, American journalists as a group are to the left of the general U.S. population (1). Second, PolitiFact was created by the St. Petersburg Times (now the Tampa Bay Times), a newspaper with a notable history of partiality to the left. For example, the newspaper has never in over 100 years endorsed a Republican in the general presidential election."
- True this may be, but this doesn't prove that PoliticFact is biased unless we have other evidence as well.

"The findings showed nothing close to a consistent result, supporting the hypothesis that PolitiFact has no objective justification backing its "Pants on Fire" ratings."
-No, they did state why they give some claims False and some Pants on Fire. They give False to statements that are just wrong, but not ridiculous. They give Pants on Fire to claims that are ridiculous and wrong. For example; False claim: I am leading over my opponent in Poll A - But you were not leading; Pants on Fire: I am leading in every poll - But your opponent actually is beating you in many polls.
http://www.politifact.com...

"The 2012 study showed the apparently subjective "Pants on Fire" rating was 74 percent more likely to fall on a Republican false statement than a Democrat false statement, covering the years from 2007 through 2011. At the end of 2014, updated statistics showed that Republicans were still 74 percent more likely to have their false (by PolitiFact's reckoning) statements rated "Pants on Fire" than those of Democrats (6).
- Regardless of whether the rating is subjective or not, there is still a reason for this. Republicans are generally more religious then liberals, so many don't believe in evolution. Many also don't believe in Global Warming. PoliticFact believes those things are true, so they rate those claims accordingly. Another thing to consider, there are more Republicans then Democrats in congress, so more Republicans will get fact checked, so more of them will make "Pants on Fire" statements. Republicans probably make more bogus claims, so they will get fact checked more.

"The same research approach was applied to other PolitiFact franchises, including its state and "PunditFact" franchises. The franchises that deliver the most robust volume of data show fairly consistent trends which do not necessarily track with the trend at PolitiFact National. PolitiFact Wisconsin, for example, tends to give Democrats' false statements harsher treatment than those from Republicans by more likely giving them the "Pants on Fire" designation (7)."
- Liberals are generally more intelligent then conservatives. So why would the apparently liberal PoliticFact allow PoliticFact Wisconsin to not be liberally biased? What is more likely is that the claims Wisconsin Democrats make are more likely to be false then the Republican statements." http://www.americanscientist.org...

"The other research project was based on reasoning that journalists could be expected to make a common math error more often in cases where the error favored their ideology. Using the often-made error of using the wrong denominator when calculating percentage error, the research project used a search routine to collect a set of PolitiFact stories that might contain percentage error calculations. The research approach turned up a small sample of stories with that equation present, yet the results were lopsided. Out of 14 examples reasonably classified as percentage error calculations, PolitiFact used the wrong denominator nine times.
Out of those nine errors, six either unfairly harmed a Republican or helped a Democrat. Three either helped a Republican or harmed a Democrat."
-First off, PoliticFact didn't use, "The wrong denominator nine times." Your source says they miscalculated 9 times. It also says that of these 14 claims, 2 were ambiguous, and 3 were correct. So we are down to 9 miscalculations. And of those, 3 helped republicans. If this site was liberally biased, why even allow the republicans any bonuses? Plus, we don't get any links to the miscalcuations, probably because PoliticFact fixed their errors, like any site would. Next, PolitiFact has done at least 1000 fact checks. There were 6 calculation errors (apparently) that helped Democrats. This means they have a .9% chance of making an error in favor of a Democrat. That doesn't seem biased at all. Also, your source is "PolitiFactBias." That doesn't seem like an unbiased source.

"It is important to note that we have no scientifically supported baseline for which political party makes more false claims. So when PolitiFact rates Republicans more harshly than Democrats, it supports no reasonable prediction about PolitiFact's ratings matching a known tendency. On the other hand, knowing that journalists lean left makes it reasonable to predict that if that bias affects PolitiFact's ratings then it may result in harsher ratings for Republicans. The harsh ratings for Republicans support a prediction associated with the hypothesis that PolitiFact has a liberal bias."
-Look at the democratic chart on this website. That hardly looks like they are helping democrats much. Just under 1/2 of the democratic claims are 1/2 false, mostly false, false, and pants on fire. The republican circle is much more to the negative, but as the source points out, there are many Republicans that make ridiculous claims, and those people are quite popular, so they will get fact checked.
http://www.thewire.com...

A) You didn't show how this is a liberal bias.
B) You are focusing on an unrelated part of this fact check. Anyway, their point is that this legislation won't cost as much as the Republicans were saying.
C) This is out of context. Read the fact check. They are saying that healthcare now will reduce the deficit in the long run. Plus, you just used this example in B, but now you are calling it false? Government spending stimulates the economy.
D) Again, what does this mean? You didn't explain how this proves a liberal bias.

So of these 4, you failed to back up 2 of them, one of these "biases" is out of context, and the other one is nitpicking.

So I believe I have rebutted your arguments. I see in your argument that you want me to provide evidence for my side. I don't need to do this. You have the burden of proof. It is not my job to prove PoliticFact is not biased, when no evidence to the contrary exists. If I worded the resolution as, "PoliticFact is not liberally biased", I would have the BoP. But since you have the positive position, you have to prove your side.

Thank you for accepting! Good luck in the final round!
BWhite

Pro

Addressing the first prong of my argument concerning the leftward lean of journalists generally and the St. Petersburg/Tampa Bay Times specifically, Conservative Democrat wrote:

"(T)his doesn't prove that PoliticFact [sic] is biased unless we have other evidence as well."

The debate challenge does not ask for proof. The debate challenge asks whether PolitiFact has a liberal bias. The bias among journalists and at PolitiFact's parent newspaper offer reasonable grounds for thinking PolitiFact leans left. The first prong is not meant as proof, for the debate challenge does not require proof.

It's a bit late to amend the rules of the debate, such as they are.

Addressing my second prong, regarding PolitiFact's failure to provide a principled distinction between its "False" and "Pants on Fire" ratings, Conservative Democrat wrote:

"No, they did state why they give some claims False and some Pants on Fire. (...) For example; False claim: I am leading over my opponent in Poll A - But you were not leading; Pants on Fire: I am leading in every poll - But your opponent actually is beating you in many polls."

Where I inserted the ellipsis, Conservative Democrat simply repeated the description of the ratings like one I had already supplied. The remainder of Conservative Democrat's answer offers example as a substitute for principle. But why is a string of wrong statements not merely wrong but in addition "ridiculous"? PolitiFact never states any such difference in terms of principle. When pressed on the matter during a 2014 podcast interview, PolitiFact editor Angie Drobnic Holan said "(T)he line between "False" and "Pants on Fire" is just, you know, sometimes we decide one way and sometimes decide the other" (1). Her statement does not provide any principled distinction between one rating and the other.

Providing an example in rebuttal without identifying the principle in the example counts as a failed rebuttal.

Again addressing the second prong, this time in terms of the PolitiFact's tendency to more likely grade Republican false statements "Pants on Fire" than those of Democrats, Conservative Democrat wrote (accounting for the next few sections of quoted text):

"Regardless of whether the rating is subjective or not, there is still a reason for this."

If the reasons follow no objective principle then the ratings are effectively subjective and the disparity in the ratings shows PolitiFact's liberal bias.

"Republicans are generally more religious then liberals, so many don't believe in evolution."

PolitiFact rarely rates claims touching the issue of evolution. If Conservative Democrat can offer as much as one example of such rated "Pants on Fire" then the point would be at least slightly interesting. But even then it does not turn my third prong.

"Many also don't believe in Global Warming."

Conservative Democrat might have cited that factoid to PolitiFact. Unfortunately, the example plays to my advantage in this debate. PolitiFact does not bring scientific rigor to its ratings on global warming, and in fact sometimes reverses its rating principles on the topic. PolitiFact uses a "burden of proof" standard in its statement of principles. If politician does not support a claim and PolitiFact cannot support it with evidence, PolitiFact calls the claim false (2). Yet a Democrat can say "It is true that there's virtually no Republican who accepts the science that virtually is unanimous" and PolitiFact gives it a "Mostly True" rating despite the fact that most Republicans have offered no statement of their beliefs on climate change (3). PolitiFact will sometimes simply trash its own principles in favor of a popular public narrative. When a Republican does not offer a position on climate change, does it follow that the Republican does not accept the science? Is that logical? When a fact-checker abandons its principles to favor one side of a debate, it's a solid sign of bias.

"(M)ore Republicans will get fact checked, so more of them will make "Pants on Fire" statements."

The research takes that into account. It does not look merely at which party receives more "Pants on Fire" ratings but instead at which party's false statements are more likely to receive "Pants on Fire" statements. If the proportions were equal, Republicans could have 20 times more "Pants on Fire" statements but the research would show no evidence of bias by that measure. Conservative Democrat misstates the nature of the evidence.

"Liberals are generally more intelligent then conservatives"

What's the relevance, other than affording the opportunity to use a citation? If it's not relevant, it's not an intelligent move to include that factoid.

"So why would the apparently liberal PoliticFact [sic] allow PoliticFact [sic] Wisconsin to not be liberally biased? What is more likely is that the claims Wisconsin Democrats make are more likely to be false then the Republican statements."

How would PolitiFact National stop PolitiFact Wisconsin from favoring Republicans in its selections of "Pants on Fire" statements? There's no use asking why if there's no how. The notion that Wisconsin Democrats are more likely to make false statements lacks any evidence in support (using PolitiFact statistics would beg the question), and more importantly Conservative Democrat should note that both "False" and "Pants on Fire" ratings supposedly indicate false statements. It doesn't solve the mystery to say that Wisconsin Democrats more likely make more false claims. That's irrelevant to a disparity in the chances of the application of the "Pants on Fire" rating.

Turning attention to the third prong of my argument, Conservative Democrat wrote:

"First off, PoliticFact didn't use, "The wrong denominator nine times." Your source says they miscalculated 9 times."

There's no discrepancy. Using the wrong denominator results in a miscalculation of percentage error (4). My source says as much, but in different words ("It is incorrect to divide by the incorrect estimate, 175 lbs" (5)). The thing you divide by is the denominator.

"If this site was liberally biased, why even allow the republicans any bonuses?"

That was explained. Wrong procedures by poor fact checkers are more likely to favor the fact-checker's ideology. "More likely" does not mean "exclusively." Lack of bias would mean no pattern of one party receiving more unfair harm than the other. Any other understanding represents a straw man.

"we don't get any links to the miscalcuations [sic].

The source links to a spreadsheet that identifies every case in question.

"There were 6 calculation errors (apparently) that helped Democrats. This means they have a .9% chance of making an error in favor of a Democrat."

That statement shows a lack of understanding of the research. The search strand provided a suitably random sample of PolitiFact's work looking for one type of calculation. A sample looking for special cases. Your conclusion doesn't follow at all.

"Look at the democratic chart on this website. That hardly looks like they are helping democrats much."

That's unresponsive to the point. We have no baseline from which to judge the fairness of the charts. Your opinion is only evidence of your opinion. Maybe the Democrats made more false claims.

Conservative Democrat dealt briefly with my fourth prong:

"A) You didn't show how this is a liberal bias."

Did. "If PolitiFact National's worst work favors Democrats over Republicans, then it serves as an evidence of PolitiFact National's liberal bias."

"B) You are focusing on an unrelated part of this fact check."

Biased fact-checkers are more likely to commit and sustain a blunder that agrees with their bias. How is that not related?

"C) This is out of context."

Based on what evidence? How can rising costs ever reduce a deficit in any context?

"Government spending stimulates the economy."

The only way that could lead to deficit reduction is when taxes are collected on the increased economic activity (revenue, just like I said). But you don't get back more than you spend, do you?

"D) Again, what does this mean? You didn't explain how this proves a liberal bias."

Do I need to repeat myself? PolitiFact's apparent incompetence favored the liberal position. I don't need proof. I just need reasonable evidence, as I've already explained.

As this stage of the debate shows, Conservative Democrat has failed to refute any prong of my argument. PolitiFact's institutional bias probabilistically shows a liberal bias, two research projects show strong evidence of PolitiFact's liberal bias, and Conservative Democrat has no counter to my anecdotes.

Conservative Democrat summarized:
"So of these 4, you failed to back up 2 of them, one of these "biases" is out of context, and the other one is nitpicking."

I've explained why each of your criticisms ring hollow.

"It is not my job to prove PoliticFact [sic] is not biased, when no evidence to the contrary exists."

I said that if you do not refute each prong of my argument then you need your own affirmative evidence. That's perfectly true, despite your (straw man) hint that I improperly shift the burden of proof. You haven't refuted any part of my argument (evidence in my reply above), nor have you provided your own affirmative case. Based on our arguments, it is reasonable to conclude that PolitiFact has a liberal bias. All of the evidence is on my side.

(1) http://www.politifactbias.com...
(2) http://www.politifact.com...
(3) http://www.politifact.com...-/
(4) https://www.mathsisfun.com...
(5) http://www.politifactbias.com...
Debate Round No. 2
14 comments have been posted on this debate. Showing 1 through 10 records.
Posted by BWhite 1 year ago
BWhite
What's the record proportion for views-to-votes for a debate? :-)
Posted by tejretics 1 year ago
tejretics
You misinterpret me.

Read the list. Many of those aren't even "lies."
Posted by ConserativeDemocrat 1 year ago
ConserativeDemocrat
Tejretics, Donald Trump is always in the news and can make his statements in front of many people. That is why he gets fact checked more.
Posted by tejretics 1 year ago
tejretics
I'm tempted to accept this, but I'll pass.

But to realize Politifact's bias, one merely needs to go through this list: http://www.politifact.com...
Posted by ConserativeDemocrat 1 year ago
ConserativeDemocrat
I was reading some politifact factchecks and had the idea to look if was a biased site. I found a lot of conservative accusations that the site was biased, and I disagreed with their arguments, so I made this debate.
Posted by Rightreform 1 year ago
Rightreform
Kilk1 why won't you accept the challenge? It sounds like you know enough about the subject.
ConservativeDemocrat where did you get the idea for this debate? I did make comments about politifact being biased in a debate, but that was so I could discrediting my competitor.
Yet I do believe we need to fact check the fact checkers.
Posted by Kilk1 1 year ago
Kilk1
Yes, one or two examples (like I presented) isn't sufficient to make a general claim. However, the article lists more examples of bias. But it is possible that similar things happen to conservatives generally, and since I'm not the one formally accepting the debate challenge, I guess I'll stay out of it. May the right man win!
Posted by ConserativeDemocrat 1 year ago
ConserativeDemocrat
I guarantee they also didn't fact check some wrong conservative claims. That doesn't prove a liberal bias.
Posted by Kilk1 1 year ago
Kilk1
In addition, why would the other things in the article I just gave be the case if PolitiFact doesn't have liberal bias. Whoever takes the "Pro" position in this debate, good luck. If you'd like, consider the article below.
Posted by Kilk1 1 year ago
Kilk1
Why would Elizabeth Warren's claim to be part Cherokee Indian not be fact-checked by them?
1 votes has been placed for this debate.
Vote Placed by RonaldTrumpkin 1 year ago
RonaldTrumpkin
ConserativeDemocratBWhiteTied
Agreed with before the debate:-Vote Checkmark-0 points
Agreed with after the debate:-Vote Checkmark-0 points
Who had better conduct:--Vote Checkmark1 point
Had better spelling and grammar:--Vote Checkmark1 point
Made more convincing arguments:--Vote Checkmark3 points
Used the most reliable sources:-Vote Checkmark-2 points
Total points awarded:02 
Reasons for voting decision: I believe in terms of arguments both sides were equal- they brought up valid points and were able to provide substantial evidence to back it up. However, pro user more sources and his sources were greater in variety- also more neatly organized at the end of each round.