Sunday, January 1, 2012

PolitiFact 2011: A review

Crossposted from Sublime Bloviations.

PolitiFact Bias has now spent approximately a full year highlighting criticisms of the PolitiFact fact checking brand.

Our hopes that PolitiFact would improve its performance in light of outside criticism have gone largely unfulfilled.  Perhaps the biggest improvement was the reconciliation of two differing definitions of the "Half True" rating, but that modest accomplishment occurred without any announcement or acknowledgment at all from PolitiFact.  By contrast, PolitiFact wrote extensively about its momentous change in calling its fourth rating from the top "Mostly False" rather than "Mostly True" even though the definition remained the same.

Here's a rundown of the issues that should keep discerning readers from trusting PolitiFact:

1)  PolitiFact persistently ignores the effects of selection bias.  It simply isn't plausible that editors who are very probably predominantly liberal will choose stories of interest on a neutral basis without some systematic check on ideological bias.  PolitiFact, for example, continues to publish candidate report cards as though selection bias has no effect on the report card data.

2)  PolitiFact continues to publish obviously non-objective stories without employing the journalistic custom of using labels like "commentary," "opinion" or even "news analysis."  Readers are implicitly led to believe that stories like an editorial "Lie of the Year" selection are objective news stories.

3)  PolitiFact continues to routinely apply its principles of analysis unevenly, as with its interpretation of job creation claims (are the claims assumed to refer to gross job creation or net job creation?).

4)  PolitiFact has yet to shake its penchant for snark.  Snark has no place in objective reporting (see #2 above).  Unfortunately, PolitiFact treats it like a selling point instead of a weakness, and PolitiFact's intentional use of it has apparently influenced Annenberg Fact Check to follow suit.

There is a silver lining.  PolitiFact's methods produce perhaps the best opportunity yet to objectively measure mainstream media bias.  Some of those projects will be published at PolitiFact Bias over the coming year, with the study specifics available through Google Docs.

60 comments:

  1. FYI: A rebuttal:

    http://contentinreality.blogspot.com/2012/01/politifact-bias-or-not-defending-truth.html

    ReplyDelete
  2. Thanks for bringing that post to our attention. It is not, however, a rebuttal. It admits that selection bias is obvious. It grants the non-objectivity of "Lie of the Year" stories. It misses the bulk of the supporting information provided with point #3 (and fails to acknowledge the helpful hint offered in the text regarding net vs. gross). It forgives snark as though snark is acceptable in objective reporting. Was the author aware that the "Lie of the Year" story ran on the front page of PolitiFact's home base newspaper?

    Worst of all, the "rebuttal" fails to give the PFB team proper credit for correctly identifying what constitutes evidence of ideological bias. A more thorough read of the PFB materials might have forestalled that false conclusion.

    ReplyDelete
  3. Shorter version of KnocksvilleE's blog:

    "The Blog "Politifact Bias" seems to be a pretty popular blog for conservatives who want to confirm their poorly-justified belief that the neutral fact checking site known as Politifact is actually left-biased...I have yet to look at this blog directly blah blah blah blah blah blah blah blah blah blah blah blah blah blah blah..."

    ReplyDelete
  4. Next week, KnocksvilleE reviews the handling characteristics of the Space Shuttle:

    "While I haven't actually flown the shuttle myself, I'd like to spend a few moments describing to you folks how this puppy handles in orbit when you're sitting in the captains seat, and then expose some errors that I assume it probably has..."

    ReplyDelete
  5. @Bryan
    "Thanks for bringing that post to our attention. It is not, however, a rebuttal. It admits that selection bias is obvious."

    Yes it admits the selection bias is obvious, but if you cared to actually read the article, it also pointed out the bias is in no way liberal bias, as the article suggests. The bias is toward significant statements, as Bill Adair confirmed. Maybe you should read the article more carefully.

    "It grants the non-objectivity of "Lie of the Year" stories."

    Where does Politifact ever call it objective? I'd think the quote "PolitiFact editors and reporters have chosen" shows that it is obviously their opinion.


    "It misses the bulk of the supporting information provided with point #3 (and fails to acknowledge the helpful hint offered in the text regarding net vs. gross)."
    What evidence?! I pointed out the link went to another blog re-posting this same article. Would you please provide a working link?
    again
    I pointed out the link went to another blog re-posting this same article. Would you please provide a working link?
    sorry but i thought if i posted this twice you would notice this time.


    "It forgives snark as though snark is acceptable in objective reporting."
    You have failed to show why it isn't, depending on your definition of snark.

    "Was the author aware that the "Lie of the Year" story ran on the front page of PolitiFact's home base newspaper?"
    Your point being?


    Worst of all, the "rebuttal" fails to give the PFB team proper credit for correctly identifying what constitutes evidence of ideological bias. A more thorough read of the PFB materials might have forestalled that false conclusion.

    @Jeff: Do I need to read all of the crap on this site before I can criticize this shoddy work of a post?

    ReplyDelete
  6. KnoxvilleE wrote:

    Yes it admits the selection bias is obvious, but if you cared to actually read the article, it also pointed out the bias is in no way liberal bias, as the article suggests.

    The review article suggests nothing about whether the selection bias is partisan in nature. We fault PolitiFact for knowing about selection bias but failing to warn its readership about the shortcomings in its report card data. If you jumped to a different conclusion we can only recommend that you rest your legs a bit more.

    Where does Politifact ever call it objective?

    It's implicit in industry standards. Ever wonder why there's an "opinion" page in your typical newspaper? It's an intentional effort to separate opinion from objective news. Other items will carry the label "op-ed" or "news analysis" to distinguish them from objective news because items missing such labels are assumed to fit the objective reporting paradigm.

    I pointed out the link went to another blog re-posting this same article. Would you please provide a working link?

    The link works as I intended it. If you observe closely, the link carries you to a keyword search of the Sublime Bloviations blog, which in turn puts the information I specified on the page if you know how to look for it (starting by not assuming that the top item was what you were intended to see).

    You have failed to show why it isn't.

    Snark isn't tolerated in objective journalism for the same reason that adjectives and adverbs are typically discouraged. This is journalism 101. Is this stuff new to you?
    http://handbook.reuters.com/index.php/A#adverbs

    Does it make sense to you to help ensure journalistic objectivity by avoiding judgmental adverbs while preserving the use of snark?

    Do I need to read all of the crap on this site before I can criticize this shoddy work of a post?

    It might have saved you some embarrassment.

    You mistook the purpose of the post (it was not intended to provide evidence of ideological bias) and your criticisms were uniformly useless in undermining the purpose of the post (to offer examples undermining PolitiFact's implicit status as an fact check outfit following the "objective" news paradigm). I was simply offering brief summaries of the continued problems that should bother news consumers. Over the years you can find journalists at publications like Slate and the Wall Street Journal echoing our concerns.

    You could also have looked up "Ostermeier" on our site in order to investigate the way we treat data that simply show PolitiFact choosing to rate more stories involving one party than the other. Instead it looks like you jumped to another conclusion.

    We're more careful than that here.

    ReplyDelete

  7. "The review article suggests nothing about whether the selection bias is partisan in nature. We fault PolitiFact for knowing about selection bias but failing to warn its readership about the shortcomings in its report card data. If you jumped to a different conclusion we can only recommend that you rest your legs a bit more.


    I'm sorry, but i thought "It simply isn't plausible that editors who are very probably predominantly liberal will choose stories of interest on a neutral basis without some systematic check on ideological bias." Pretty clearly suggested you were accusing them of liberal selection bias. Should I be reading around what you said?

    It's implicit in industry standards. Ever wonder why there's an "opinion" page in your typical newspaper? It's an intentional effort to separate opinion from objective news. Other items will carry the label "op-ed" or "news analysis" to distinguish them from objective news because items missing such labels are assumed to fit the objective reporting paradigm.


    I thought it painfully obvious the "lie of the year" was subjective when those articles said "PolitiFact editors and reporters have chosen..." It is an "award," which by its very nature is subjective, like an "emmy." You don't need an opinion page to tell that. Does Politifact need to write in big bold letters "Opinion" above every such article? Or does the obvious opinion-oriented wording suffice for reasonable people?

    ReplyDelete
  8. I will continue responding to your comments tomorrow. However, since you mentioned that there are other journalists pointing out "problems" with politifact, I should note that I have read a few and found them to be juvenile if not all out dishonest.
    http://contentinreality.blogspot.com/2011/08/weekly-standard-still-thinks-long.html

    http://contentinreality.blogspot.com/2011/06/newsbuster-fail-defending-truth-o-meter.html

    http://contentinreality.blogspot.com/2012/01/weekly-standard-attacks-fact-checkers.html

    http://contentinreality.blogspot.com/2012/01/weekly-standard-attacks-fact-checkers_10.html

    http://contentinreality.blogspot.com/2012/01/weekly-standard-attacks-fact-checkers_9918.html

    ReplyDelete
  9. KnocksvilleE wrote:

    (I) thought "It simply isn't plausible that editors who are very probably predominantly liberal will choose stories of interest on a neutral basis without some systematic check on ideological bias." Pretty clearly suggested you were accusing them of liberal selection bias. Should I be reading around what you said?

    It does help to consider context, and I will assist you.

    I stated that given the journalistic predisposition toward the left, it isn't plausible to take ratings as evenhanded without some check on selection bias. You believe that statement accuses PolitiFact of an actual ideological bias in its selection process. Yet it obviously isn't. PolitiFact may have some unstated mechanism for checking ideological bias in its selection process. We simply don't know whether they have such a check, though we suspect there is none. In this item (nor the others) we do not offer any opinion on whether there is ideological bias in the selection bias. We distinguish between the two types of bias. More importantly, even if PolitiFact is ideologically biased in its selection process it remains perfectly possible for a liberal political figure to receive unfair ratings regardless of ideological bias (simply because of selection bias). The tendency of journalists to lean left (and the public awareness of same) simply provides another reason why PolitiFact needs to mention selection bias any time it publishes a "report card." It is journalistically irresponsible to omit that information. It is telling a half truth, to put things in PolitiFact terms.

    I thought it painfully obvious the "lie of the year" was subjective when those articles said "PolitiFact editors and reporters have chosen..." It is an "award," which by its very nature is subjective, like an "emmy." You don't need an opinion page to tell that. Does Politifact need to write in big bold letters "Opinion" above every such article? Or does the obvious opinion-oriented wording suffice for reasonable people?

    It's obviously opinion and as such an obvious choice to be accompanied by the appropriate journalistic label according to journalistic standards.

    Do you want to play "Let's pretend" and take PolitiFact out of the historical flow of American journalism? Remember when I mentioned that the "Lie of the Year" story ended up on the Times front page? That's the relevance. The reality is that many people don't read the stories. They just take in the headline. And on top of that, people are capable of assuming that "We've chosen" means according to some set of objective criteria. They trust the nonpartisan, Pulitzer-winning PolitiFact that much. PolitiFact editors also choose which "Truth-O-Meter" rating goes with each story, when it comes to that. They vote on it. And then they publish the vote totals. No, just kidding about publishing the vote totals. That's something readers need not know about.

    (S)ince you mentioned that there are other journalists pointing out "problems" with politifact, I should note that I have read a few and found them to be juvenile if not all out dishonest.

    No doubt in your reality-based world it is completely appropriate to respond to mentions of the Wall Street Journal and Slate with your opinion of NewsBusters and the Weekly Standard.

    Bottom line, if you can't answer for falsely accusing us of finding our proof of ideological bias in the breakdown of which party gets the most harsh PolitiFact stories then it's hard to see why we should waste our time dealing with your other opinions.

    ReplyDelete
  10. "They just compiled a list of articles on Politifact over the course of a year and came to the conclusion that, since the worst ratings tend to favor republicans more than democrats, they must be biased in their selection in favor of the left."

    You made it up.

    ReplyDelete
  11. Snark isn't tolerated in objective journalism for the same reason that adjectives and adverbs are typically discouraged. This is journalism 101. Is this stuff new to you?
    http://handbook.reuters.com/index.php/A#adverbs

    Does it make sense to you to help ensure journalistic objectivity by avoiding judgmental adverbs while preserving the use of snark?

    You keep on repeating this claim yet provide zero evidence. Please provide some because I have zero reason to merely take your word for it.

    ReplyDelete
  12. You mistook the purpose of the post (it was not intended to provide evidence of ideological bias)
    Interesting how your first example does actually provide evidence. Your third example does as well (albeit in an unnecessarily distracting and cryptic way). I'm sorry I assume that one should provide evidence to back up their claims.

    ReplyDelete
  13. PolitiFact continues to routinely apply its principles of analysis unevenly, as with its interpretation of job creation claims (are the claims assumed to refer to gross job creation or net job creation?).

    http://subloviate.blogspot.com/2011/07/grading-politifact-virginia-bob.html#more

    is this link a good example to look at?

    ReplyDelete
  14. "I stated that given the journalistic predisposition toward the left, it isn't plausible to take ratings as evenhanded without some check on selection bias. You believe that statement accuses PolitiFact of an actual ideological bias in its selection process. Yet it obviously isn't. PolitiFact may have some unstated mechanism for checking ideological bias in its selection process. We simply don't know whether they have such a check, though we suspect there is none. In this item (nor the others) we do not offer any opinion on whether there is ideological bias in the selection bias. We distinguish between the two types of bias."
    Still sounds like you are accusing them of selection bias to me. Honestly, stop trying to dance around the fact. It is implicit, but when you say "we suspect there is [no check on bias]," you are saying you suspect there is liberal selection bias.

    ReplyDelete
  15. "It's obviously opinion and as such an obvious choice to be accompanied by the appropriate journalistic label according to journalistic standards."
    Does Politifact need to make an Opinion page JUST for the lie of the year?

    In fact your entire argument rests on the idea that, in order to show the "lie of the year" is merely opinion, they have to follow traditional journalistic practices. However, Politifact is not a traditional media source. If Politifact can make it apparent their "lie of the year" is based on their own opinion by simply saying "we have chosen"

    "And on top of that, people are capable of assuming that "We've chosen" means according to some set of objective criteria"
    They are capable, but they would they would be making an assumption not warranted by Politifact anything Politifact said. In addition, Politifact has pointed out that the Lie of the year is chosen from both reader polls and editor decisions over which statement has had the most impact on political discourse that year:
    http://www.politifact.com/truth-o-meter/article/2011/dec/20/how-we-chose-lie-year/

    Politifact cannot help it if readers make unwarranted assumptions about their selection process when they have the criteria for these processes readily available to them.

    ReplyDelete
  16. "No doubt in your reality-based world it is completely appropriate to respond to mentions of the Wall Street Journal and Slate with your opinion of NewsBusters and the Weekly Standard.
    I'm sorry, given that you have posted articles from the weekly standard as recently as a Christmas. I thought you were merely giving a few examples of publications that echo your concerns. If you can point me to specific articles, I may check on them. However, given the lack of quality in the conservative attacks on politifact, I find it hard to believe these would have any better quality themselves.

    ReplyDelete
  17. KnocksvilleE's going serial ...

    You keep on repeating this claim yet provide zero evidence. Please provide some because I have zero reason to merely take your word for it.

    You're hilarious. You could just admit that you know virtually nothing about journalism. :-)

    http://journalism.about.com/od/ethicsprofessionalism/a/objectivity.htm

    I'm sorry I assume that one should provide evidence to back up their claims.

    Selectively, I note (given that you still haven't backed up the statement I quoted from you above). It is poor writing to back up every claim every time. At some point we have to have a common knowledge base from which to work. Your knowledge base apparently includes little information about the methods of American journalism, or else you'd probably have an easier time digesting the point.

    (I)s this link a good example to look at?

    It is.

    Still sounds like you are accusing them of selection bias to me.

    We all remember how important it is to provide evidence in support of one's claims.

    You're correct that when we say that we suspect that there is no check on the selection bias it indicates that we believe that there is no check on the selection bias. But a smart person like you would have realized that the statement you're quoting does not come from the post in which you say that we're claiming an ideologically based selection bias.

    Or am I giving you too much credit?

    ReplyDelete
  18. You made it up.

    "By levying 23 Pants on Fire ratings to Republicans over the past year compared to just 4 to Democrats, it appears the sport of choice is game hunting - and the game is elephants."

    Seems I didn't.

    ReplyDelete
  19. You're hilarious. You could just admit that you know virtually nothing about journalism. :-)

    http://journalism.about.com/od/ethicsprofessionalism/a/objectivity.htm


    It seems I misunderstood your use of the slang word, "snark." Based on this article, it seems you are actually talking about biased language:

    "Objectivity means that when covering hard news, reporters don’t convey their own feelings, biases or prejudices in their stories."
    Please provide an example where Politifact has violated this principle.

    ReplyDelete
  20. "given that you still haven't backed up the statement I quoted from you above"
    I may be a bit slower at posting, but its up there

    ReplyDelete
  21. "It is poor writing to back up every claim every time. At some point we have to have a common knowledge base from which to work.
    Yes there are items which need no backing up. Items that have virtually no effect on the outcome of the thesis for one's post, as well as incredibly obvious items need no backing up. However, the assumption that, in order for me to critique this summary post, I need to laboriously comb through a large set of previous posts just to find evidence, is a pretty underhanded way of shielding this post from criticism.

    ReplyDelete
  22. "You're correct that when we say that we suspect that there is no check on the selection bias it indicates that we believe that there is no check on the selection bias. But a smart person like you would have realized that the statement you're quoting does not come from the post in which you say that we're claiming an ideologically based selection bias."

    Why the tautology? Also, yes that quote did not come from this article, but you did provide a link that said practically the same thing.

    ReplyDelete
  23. "By levying 23 Pants on Fire ratings to Republicans over the past year compared to just 4 to Democrats, it appears the sport of choice is game hunting - and the game is elephants."

    Seems I didn't.


    It seems you did, and you make yourself look even worse by ignoring the context. Ostermeier uses a key piece of reasoning that you seem to have completely ignored. He notes that the number of ratings for Democratic Party and GOP political figures is roughly equal. If the same editorial sense were in play in selecting statements, then we should expect the ratings to break down along similar lines. Otherwise, we should expect to see considerably more ratings for the members of the party that supposedly tells more falsehoods. And, of course, you're taking the kicker line out of context to try to make it seem like it means more than it does. Ostermeier does say that PolitiFact produces the appearance of going after Republicans. But it isn't the conclusion of his study. Ostermeier allows, as do we, that PolitiFact might have some type of check in place. They're just not telling.

    "But this potential selection bias - if there is one at PolitiFact - seems to be aimed more at Republican officeholders than conservatives per se."

    "The question is not whether PolitiFact will ultimately convert skeptics on the right that they do not have ulterior motives in the selection of what statements are rated, but whether the organization can give a convincing argument that either a) Republicans in fact do lie much more than Democrats, or b) if they do not, that it is immaterial that PolitiFact covers political discourse with a frame that suggests this is the case."
    http://blog.lib.umn.edu/cspg/smartpolitics/2011/02/selection_bias_politifact_rate.php

    ReplyDelete
  24. "3) PolitiFact continues to routinely apply its principles of analysis unevenly, as with its interpretation of job creation claims (are the claims assumed to refer to gross job creation or net job creation?)."
    I finally get a chance to take a look at what you are talking about. The problem with your post is you merely list each case and state whether it was helped or harmed by the inclusion/exclusion of "gross jobs." The problem is you fail to investigate context. There are some statements where it is appropriate to look at net jobs, such as a statement that looks at "jobs added" as a whole. Conversely, there are statements where it is appropriate to ignore lost jobs. When a politician states the stimulus did not create one job, it doesn't matter how many jobs were already lost as a result of the recession, just so long as at least one job was directly created by the stimulus. Since the mechanism for lost jobs and the mechanisms for gained jobs are two different things, it is more than appropriate to ignore what has nothing to do with the mechanism under question. For example, If I went to a village under attack by militants and pulled a single child to safety, thereby saving it's life, should I not be credited with saving that kid's life just because more than one life was lost in that village from the attack? Any reasonable person would find this absurd.
    Please investigate context first.

    ReplyDelete
  25. Please provide an example where Politifact has violated this principle.

    "With the 2012 election in their sights, Republican candidates spend most of their time trying to prove that President Barack Obama and the Democrats will make the economy worse."
    http://www.politifact.com/truth-o-meter/statements/2011/jul/21/newt-gingrich/newt-gingrich-says-dodd-frank-law-requires-20-perc/

    ReplyDelete
  26. KnoxvilleE wrote:

    However, the assumption that, in order for me to critique this summary post, I need to laboriously comb through a large set of previous posts just to find evidence, is a pretty underhanded way of shielding this post from criticism.

    You can critique all you like. But a good critique doesn't substitute assumptions for things that are known.

    ReplyDelete
  27. It seems you did, and you make yourself look even worse by ignoring the context. Ostermeier uses a key piece of reasoning that you seem to have completely ignored. He notes that the number of ratings for Democratic Party and GOP political figures is roughly equal. If the same editorial sense were in play in selecting statements, then we should expect the ratings to break down along similar lines. Otherwise, we should expect to see considerably more ratings for the members of the party that supposedly tells more falsehoods. And, of course, you're taking the kicker line out of context to try to make it seem like it means more than it does. Ostermeier does say that PolitiFact produces the appearance of going after Republicans. But it isn't the conclusion of his study. Ostermeier allows, as do we, that PolitiFact might have some type of check in place. They're just not telling

    Yes he does not, in most of his article, explicitly say there is selection bias. However, after presenting the data, he treats it as the new null hypothesis so long as Politifact does not provide evidence otherwise. His final statement makes this position painfully obvious. He has shifted the burden of proof on Politifact to provide evidence it is not liberally biased. He does use the term "may," yet he makes it abundantly clear, given the context of the article, as well as the final statement, that this "may" is the most likely explanation, without any evidence that remotely controls for other factors.

    ReplyDelete
  28. "With the 2012 election in their sights, Republican candidates spend most of their time trying to prove that President Barack Obama and the Democrats will make the economy worse."
    Is this not a universally recognized fact? Do you deny the fact that Republicans candidates spent most of their time (before their debates) attacking Obama? Where are the reporter's feelings? Where are his/her prejudices? Where are the unnecessary adjectives? What about this is not objective?

    ReplyDelete
  29. "You can critique all you like. But a good critique doesn't substitute assumptions for things that are known."
    Where did I do this?

    ReplyDelete
  30. The problem is you fail to investigate context. There are some statements where it is appropriate to look at net jobs, such as a statement that looks at "jobs added" as a whole. Conversely, there are statements where it is appropriate to ignore lost jobs.

    There you go making assumptions again. Weren't you telling me earlier how important it is to back up assertions with evidence?

    When a politician states the stimulus did not create one job, it doesn't matter how many jobs were already lost as a result of the recession, just so long as at least one job was directly created by the stimulus.

    Isn't the speaker's intent important at all? Or have you ruled from On High exactly how job claims regarding the PPACA ought to be handled?

    How about some examples of PolitiFact explaining its decision as to whether to consider the question one of net or gross jobs?

    Since the mechanism for lost jobs and the mechanisms for gained jobs are two different things, it is more than appropriate to ignore what has nothing to do with the mechanism under question. For example, If I went to a village under attack by militants and pulled a single child to safety, thereby saving it's life, should I not be credited with saving that kid's life just because more than one life was lost in that village from the attack?

    Sure, but if somebody's criticizing you for the way your action killed 10 bystanders for a net loss of 9 lives it's fair to say that, also. It comes down to speaker's intent to a large degree. And PolitiFact often seems uninterested in such things. Perhaps they are led by their ideological bias on occasion in sorting through such questions.

    ReplyDelete
  31. "With the 2012 election in their sights, Republican candidates spend most of their time trying to prove that President Barack Obama and the Democrats will make the economy worse."
    Is this not a universally recognized fact?


    LMAO

    Let's just suppose that they don't spend more time sleeping than they do talking politics.

    We're in the season of the GOP presidential primary. Say we've got eight candidates and four of them make the case that they'd handle the economy better than President Obama.

    For whom do you vote?

    lol

    ReplyDelete
  32. PolitiFact needs to mention selection bias any time it publishes a "report card."

    I would agree that a disclaimer would not be a bad idea when publishing these report cards. However, it isn't necessary. Practically every article written by Politifact makes it clear they are not randomly choosing statements by using statements like "our readers asked us to look into this" or "this statement caught our attention." In fact, it is unclear why a reasonable person would think the statements are actually selected randomly. What sense would it make for Politifact to select a bunch of random, mostly inconsequential statements. Given the fact that statements can't exactly be doled out into discrete sets and thrown into a hat for a reporter to randomly pull, what sense does it make for any reasonable to think this way?

    ReplyDelete
  33. "There you go making assumptions again. Weren't you telling me earlier how important it is to back up assertions with evidence?"

    I thought it would be painfully obvious to anyone who read the article.
    "What if we were to look at cases where PolitiFact has ruled on jobs claims and where considering the jobs claims in terms of gross jobs rather than net jobs might have affected the ruling?
    The following list is not scientific. It just represents a survey using two different sets of search terms. I culled from the results cases where the fact check dealt with claims overtly mentioning net jobs.

    Helped by omission
    Rick Perry (True)
    Terry McAuliffe (True)
    Carolyn Maloney (True)
    Brian Moran (True)
    Mitt Romney (True)
    Austan Goolsbee (Mostly True)
    Ed Gillespie (Mostly True)
    Joe Biden (Mostly True)
    Nancy Pelosi (Half True)
    Debbie Wasserman Schultz (Half True)
    Rick Scott (Half True)
    Tim Pawlenty (Pants on Fire; harmed by omission?)
    John Boehner (False; harmed by omission)

    Harmed by inclusion:
    Texas Public Policy Foundation (Half True)
    Rod Smith (False)
    Rick Scott (Pants on Fire)
    Rick Perry (False)

    In cases where neither the speaker nor PolitiFact gave attention to gross job creation and the mention of same might have pulled the rating down, we have seven Democrats and four Republicans. Two additional cases where adding gross jobs to the mix may have helped the rating both involved Republicans."

    where did you investigate context again? You just listed these articles. Granted you attempted an investigation into the context of choosing based on the word "created." I will give you credit for that, but you did not investigate whether or not Politifact was consistent when the mechanism that "created" jobs was the same as the mechanism that caused jobs to be lost. I will give you some credit for the attempt, but your critical thinking cap may have only been half-on. As the stimulus example shows, your list definitely ignores important context.

    "Isn't the speaker's intent important at all? Or have you ruled from On High exactly how job claims regarding the PPACA ought to be handled?
    Politifact has to rate these statements based on how a reasonable person should interpret them (not all readers are journalists, btw). If the speaker's intent in the case of the stimulus was to talk about net jobs, he chose an incredibly poor phrase to convey this. I am not sure even the most charitable interpretation would even give him this. If he was talking about net jobs, he should have said "the stimulus has failed to bring job levels back to pre-recession levels" or something to that effect.

    ReplyDelete
  34. "How about some examples of PolitiFact explaining its decision as to whether to consider the question one of net or gross jobs?"

    "So, because more jobs were lost overall than gained, that's the same as saying not one job has been created?

    Not according to Rebecca Rust, senior economist for the Agency for Workforce Innovation. In fact, it's standard to talk about jobs gained, even when jobs lost may be greater.

    "There's churning in the job market," she said. "Employers are adding and cutting at the same time. … Just because we have a job loss, you can't say we haven't gained any stimulus jobs. What you could say is that the job losses could have been greater."

    Sean Snaith, a University Of Central Florida economist who has been critical of the effect of the stimulus, agrees that saying "the stimulus has not created one private sector job" is inaccurate."

    http://www.politifact.com/florida/statements/2010/sep/17/rick-scott/rick-scott-says-stimulus-create-single-private-job/

    ReplyDelete
  35. "Sure, but if somebody's criticizing you for the way your action killed 10 bystanders for a net loss of 9 lives it's fair to say that, also"

    But then you would have to make the case that the mechanism that saved jobs actually cost a significant number of jobs as well? If that were the case, you would have to try to quantify these losses and separate them from losses that occurred due to other factors, such as the recession. In the case of Rick Scott, citing the unemployment rate did not control for jobs lost by the recession.

    "It comes down to speaker's intent to a large degree. And PolitiFact often seems uninterested in such things."
    see my last post

    "Perhaps they are led by their ideological bias on occasion in sorting through such questions."
    Or perhaps they occasionally make mistakes. Or maybe they didn't do a good enough job explaining their reasoning. Such things are completely expected given the fact that they are human. Maybe, out of hundreds of posts, one or two instances of bias slipped in by mistake. Just because Politifact isn't perfect, does that mean one is justified in being just as skeptical about them as they are about the Huntington post, or media matters, or newsbusters, or random bloggers?

    ReplyDelete
  36. "Let's just suppose that they don't spend more time sleeping than they do talking politics."
    Great waste of text.

    We're in the season of the GOP presidential primary. Say we've got eight candidates and four of them make the case that they'd handle the economy better than President Obama.
    And your point being? So you are admitting that Republican Presidential candidates spend most of their time making the case that they'd handle the economy better than President Obama?

    ReplyDelete
  37. More serial Knocksiousness ...

    Where did I do this?

    You do it often, with your summary of Ostermeier's process perhaps the most outstanding example. But here's another good one from your next post:

    (Y)ou did not investigate whether or not Politifact was consistent when the mechanism that "created" jobs was the same as the mechanism that caused jobs to be lost.

    Your evidence, please?

    Politifact has to rate these statements based on how a reasonable person should interpret them (not all readers are journalists, btw). If the speaker's intent in the case of the stimulus was to talk about net jobs, he chose an incredibly poor phrase to convey this. I am not sure even the most charitable interpretation would even give him this.

    1) Do we get evidence supporting your assertion that the choice of words was "incredibly poor"?

    2) If the issue is the way the reasonable person interprets things then why did you choose in your next reply to show us PolitiFact using an "expert" source informing us that it is "standard" to talk of creating jobs while considering gross job creation? Did PolitiFact apply that standard consistently?

    But then you would have to make the case that the mechanism that saved jobs actually cost a significant number of jobs as well?

    I thought according to you I could just go with how a "reasonable person" takes the claim. Reasonable people realize that it isn't practical to make such claims about economic mechanisms. The complicated relationships defy easy analysis (one eventually has to choose between Keynesianism and a rival system). Who decides on the views of the "reasonable person," btw?

    Great waste of text.

    Great defense of the writer's factual inaccuracy.

    And your point being?

    Answer the question you skipped over and I'll explain the point if it isn't abundantly clear to you by then.

    As a matter of enforced courtesy, you will be permitted single posts from this point forward. You will wait for a reply before posting again (no more serial posts, in other words). Posts subsequent to the one will be deleted at our discretion. Plan accordingly.

    ReplyDelete
  38. "Your evidence, please?"
    This may take a few posts, but I will go ahead and do the job you should have done:
    “PolitiFact, you see, pays special attention to the words politicians use. If they say "we created x jobs" it does not necessarily mean net jobs. And that's misleading.”
    You failed to take a look at what mechanism was creating the jobs, instead using the vague title “we”. So you should not have included Rick Scott in the list of statements “Harmed by inclusion” without first investigating whether job losses were relevant. I have already dealt with this previously.
    Under the “Harmed by Inclusion” list, you also gave the example of a Rod Smith statement, the mechanism chosen by democrats was Rick Scott. Using net jobs would be inappropriate because net jobs account for more than just jobs directly created or lost by Rick Scott. Politifact justified the false rating:
    “ "Has destroyed" also makes it sound like Scott's policies have eliminated 100,000 jobs that already existed. But that's only the case for some of the jobs included in Jotkoff's tally, such as cuts leading to layoffs of state and local employees. Others, from high-speed rail to highway construction to biotech jobs, didn't yet exist — and may never have materialized.”
    In addition, the Perry article listed under “Harmed by Inclusion” does not point to the right article. The listed article was given a “True” and based its claim on “net jobs.” However, there was a linked “false” article, which you actually mentioned earlier:
    http://www.politifact.com/texas/statements/2010/jan/12/rick-perry/perry-claims-texas-accounted-70-percent-new-jobs-2/
    To this, you said: “So when you said in that article “Right. Perry did not mention the term "net" while talking about job gains. He just talked about job gains, which people tend to understand as net job gains. That interpretation is so ubiquitous that PolitiFact typically favors it without discussion, as shown by the preceding examples. So Perry gets "debunked" by making an exception to the normal conventions of English interpretation.”
    Now usually Rick Perry’s statement could be checked using net jobs. However, Perry made it clear he was talking about “New Jobs,” (gross) as opposed to net jobs when he backed his claim, citing a report from the Texas Workforce Commission that singled out new/gross jobs. If anything, this is Perry’s fault. I should also note that the statement was ambiguous about “who” created the jobs, merely saying the “were created.”
    So when you said in that article “Right. Perry did not mention the term "net" while talking about job gains. He just talked about job gains, which people tend to understand as net job gains. That interpretation is so ubiquitous that PolitiFact typically favors it without discussion, as shown by the preceding examples. So Perry gets "debunked" by making an exception to the normal conventions of English interpretation.”

    ReplyDelete
  39. As for the statement from the Texas Policy Foundation, also listed under “Harmed by includion.” Gross jobs had no effect in the ruling. In fact, nearly half the article was dedicated to justifying why they used net jobs in this analysis. The reason for the half-true was partly because, as a percentage of population, the jobs gained in Texas were not as good as others: “However, comparing the states individually, two had larger percentage increases than Texas: North Dakota (10.6 percent) and Alaska (5.7 percent).”
    The confusing statement may be “But the foundation’s analysis disregards the 40 states where millions of jobs were created but were outnumbered by losses.” On the surface it may appear they are talking about losses within those same states. However, noting that the Texas Policy Foundation statement added together all other states, states with a net loss would cancel out gains from states where jobs were added. One state could actually have more net gains than Texas. However, the losses from other states would cancel it out, meaning it is possible that the Texas Policy Foundation could say Texas had greater gains than all other states combined, even when it didn’t have the highest number! Now I’m not saying this is the case here, but it is a huge flaw with the statement which can be incredibly misleading.
    So, as we can see, every article listed under “Harmed by exclusion” either was justified in not using net jobs (for reasons you did not explore, such as different mechanisms), or they did actually rely solely on net jobs. Now if you did explore any of these justifications, please point them out to me. Or do I need to go down this article line by line and say “don’t see it here”?

    Now onto the Virginia article. Notice that Bob Mcdonnel used a similar claim to the Texas Policy foundation. Instead of just comparing state to state figures, they looked at Texas’ gains as a percentage of the population as a whole. If they were to use net numbers, they would fall into the same fallacy as the Texas Policy Foundation, wiping out the gains in some states with the losses in others. Based on that alone, Bob Mcdonnel would lose points for being misleading. Politifact Virginia decided to look at all the work Politifact Texas did on this issue:
    “Our research took us down a road traveled by PolitiFact Texas in examining various claims from Perry that most of the jobs created in the U.S. are in the Lone Star State.”
    It seems that Politifact Virginia may have misunderstood what Politifact Texas was saying. As a result, they justified their ruling incorrectly. You should be able to indetify with this because you did the same thing. As I have mentioned before, occasional mistakes do happen. So out of your 5 examples of Politifact equating jobs created with something other than net jobs, this may be the only one with any validity as a problem. However, I should note that, due to the fact that Bob Mcdonnel made the same Mistake the Texas Policy Foundation did, consistency shows the statement should have been given a “Half True.” The statement could also be docked for crediting Rick Perry with the job increase, which a fallacy. So it should not be completely unexpected that this article would get a “Mostly False” if Politifact had been more careful in their analysis.

    ReplyDelete
  40. "Do we get evidence supporting your assertion that the choice of words was "incredibly poor"?"

    Rick Scott said. "the stimulus has not created one private sector job." HE SAID THE STIMULUS! THE STIMULUS! THE STIMULUS! Once again, net job creation takes into account MORE than just the stimulus. So the fact that he SOLELY mentioned the THE STIMULUS, and NOT THE RECESSION should make any reasonable person think he IS TALKING ABOUT THE STIMULUS AND NOT JOBS LOST FROM THE RECESSION.

    I'm not sure if I can break it down barney-style any more than that...

    ReplyDelete
  41. Please note that you have been asked not to post more than once at a time before receiving a reply.

    ReplyDelete
  42. "If the issue is the way the reasonable person interprets things then why did you choose in your next reply to show us PolitiFact using an "expert" source informing us that it is "standard" to talk of creating jobs while considering gross job creation? Did PolitiFact apply that standard consistently?"

    This is absurd. Have you ever heard of sociologists? political scientists? Psychologists? Experts are often used to gauge how the public will react to or interpret a given saying. Is this a serious question or are you sending me through the gish gallop?

    ReplyDelete
  43. Bryan, I break my posts up because it makes the posts shorter and decreases the chance that a browser crash will cause me to lose a lot of work. What is wrong with this? Or are you just looking for an excuse to kick me off so you can have the last word?
    Also, if you want my posts to be shorter, stop asking me absurd questions like the one i answered in the last post.

    "Great defense of the writer's factual inaccuracy
    So because the candidate spends more time sleeping than saying they could better handle the economy than Obmama, Politifact can't say they spend most of their time doing that? That has the be the most absurd thing you have said on this post so far. Wow!

    As to your question. Okay I will bite. I would vote for the candidate that said they could do a better job for the economy than Obama.

    ReplyDelete
  44. The supposed evidence that I "did not investigate whether or not Politifact was consistent when the mechanism that "created" jobs was the same as the mechanism that caused jobs to be lost":

    “PolitiFact, you see, pays special attention to the words politicians use. If they say "we created x jobs" it does not necessarily mean net jobs. And that's misleading.”
    You failed to take a look at what mechanism was creating the jobs, instead using the vague title “we”.


    As I was speaking in generalities, it would not be appropriate to consider any specific economic mechanism. There's no evidence for you there.

    Under the “Harmed by Inclusion” list, you also gave the example of a Rod Smith statement, the mechanism chosen by democrats was Rick Scott. Using net jobs would be inappropriate because net jobs account for more than just jobs directly created or lost by Rick Scott.

    Gross jobs added also accounts for more than just jobs created or lost by Rick Scott, doesn't it? Rick Scott a mechanism? Seriously? Your statement here is incoherent.

    Now usually Rick Perry’s statement could be checked using net jobs. However, Perry made it clear he was talking about “New Jobs,” (gross) as opposed to net jobs when he backed his claim, citing a report from the Texas Workforce Commission that singled out new/gross jobs.

    Perry's statement did not use the term "new." That was PolitiFact's paraphrase. And it was PolitiFact that chose the method of interpreting the Texas Workforce Commission data.

    Do you seriously think that any of this serves as evidence in support of the statement at the top? Or do you just enjoy using your keyboard?

    ReplyDelete
  45. Bryan, I break my posts up because it makes the posts shorter and decreases the chance that a browser crash will cause me to lose a lot of work. What is wrong with this?

    It looks spammy and makes the flow of conversation more difficult to follow. I can cut you off anytime I like, so no it isn't an excuse for me to have the last word. I need no excuse to do that. Stay short and to the point and we'll have a back-and-forth that's easy to follow. Thanks for your cooperation.

    ReplyDelete
  46. Oh, and if you're concerned about browser crashes then consider composing on Word and then pasting it into the reply window. Works great.

    ReplyDelete
  47. So because the candidate spends more time sleeping than saying they could better handle the economy than Obmama, Politifact can't say they spend most of their time doing that? That has the be the most absurd thing you have said on this post so far. Wow!

    Sure they can say it. It just isn't accurate. It wasn't even accurate just during the debate.

    As to your question. Okay I will bite. I would vote for the candidate that said they could do a better job for the economy than Obama.

    You only get one vote. Four met that threshold in the scenario I gave you.

    ReplyDelete
  48. "As I was speaking in generalities, it would not be appropriate to consider any specific economic mechanism. There's no evidence for you there."
    As I demonstrated with the rest of the post, those generalities were completely wrong when considering any post other than the ONE post from Politifact virginia.

    "Gross jobs added also accounts for more than just jobs created or lost by Rick Scott, doesn't it? Rick Scott a mechanism? Seriously? Your statement here is incoherent.
    If you cared to think just a little bit about what I said, a the mechanism is the actions of Rick Scott. You are correct that a general count of "Gross Jobs" would in fact account for more than Scott. But if you notice, nowhere did they mention "total gross jobs."

    "Perry's statement did not use the term "new." That was PolitiFact's paraphrase. And it was PolitiFact that chose the method of interpreting the Texas Workforce Commission data."
    It is also the interpretation of all the experts who looked at the data. If Perry had been concentrating on net jobs, the statement would have been incoherent at worst, misleading at best. To show why this is, think of a situation in which every other state but Texas had a net loss of 1,000 jobs over a given time period. Lets say though that Texas added 50,000 jobs. The total net job gain would be 1000 jobs total, meaning Texas would have created 5000% of all jobs in the US. This is pretty incoherent. Once again, it is helpful to think critically about this stuff.

    "Do you seriously think that any of this serves as evidence in support of the statement at the top?

    Sounds as if it does. Before you respond this time, please spend a moment and give a bit of critical thought into your responses.

    ReplyDelete
  49. "Oh, and if you're concerned about browser crashes then consider composing on Word and then pasting it into the reply window. Works great."
    I actually started doing this.

    "Sure they can say it. It just isn't accurate. It wasn't even accurate just during the debate
    If it's inaccurate, it's only in a trivial way. Its inaccurate to the same degree that saying "Most of my post is comprised of words" is inaccurate since most of my post is actually white space. Also, notice the article where this quote came from actually took place long before the debates, btw. Is it true Republicans are spending nearly as much time attacking Obama as they are each other nowadays? Probably not, but the post is not from now :P

    "You only get one vote. Four met that threshold in the scenario I gave you.
    Is your point that, if these four candidates spent all their time attacking Obama, I couldn't pick one over another? As long as they spend some time dealing with each other, that is no problem.
    If that's not your point, Ill pick one of the candidates, candidate A ;)

    ReplyDelete
  50. KnocksvilleE wrote:
    (T)hose generalities were completely wrong when considering any post other than the ONE post from Politifact virginia.

    Then you should have skipped ahead instead of pretending that the statement provided the evidence you were supposed to provide.

    the mechanism is the actions of Rick Scott.

    Inaction can help provide a job-creating climate also. You count that as a mechanism? How do we quantify the results?

    You are correct that a general count of "Gross Jobs" would in fact account for more than Scott. But if you notice, nowhere did they mention "total gross jobs."

    That's kind of the point. The statements contain ambiguity and PolitiFact interprets arbitrarily. But where's the evidence supporting your original statement? Is this the argument from silence or what?

    It is also the interpretation of all the experts who looked at the data.

    All two of them? What could possibly go wrong?

    If Perry had been concentrating on net jobs, the statement would have been incoherent at worst, misleading at best. To show why this is, think of a situation in which every other state but Texas had a net loss of 1,000 jobs over a given time period. Lets say though that Texas added 50,000 jobs. The total net job gain would be 1000 jobs total, meaning Texas would have created 5000% of all jobs in the US.

    It's only incoherent if we accept your equivocal language in the final sentence. Texas would have created 5000% of all the net jobs created in the U.S. There's nothing incoherent about it. Contrived argument, much?

    If there's evidence I ignored the mechanisms in Perry's case it has nothing to do with the coherence of Perry's claim.

    Before you respond this time, please spend a moment and give a bit of critical thought into your responses.

    That's so cute! Apparently I need the type of critical thinking that magically turns a coherent statement into an incoherent one.

    ReplyDelete
  51. Is your point that, if these four candidates spent all their time attacking Obama, I couldn't pick one over another?

    Not quite. You could always draw lots. But you're surprisingly close.

    Apart from the fact that it simply isn't true, it doesn't make any sense for the GOP field to behave as PolitiFact claims. Each needs to make the case that he has a better economic plan than his competitors, not just Obama. The claim from PolitiFact is not factual. But it does send a type of message. What type of message do you get from it? What does it say to you about the GOP candidates that they supposedly spend most of their time talking about how their economic plans are better than Obama's?

    ReplyDelete
  52. "Then you should have skipped ahead instead of pretending that the statement provided the evidence you were supposed to provide

    Maybe you should have looked at that whole 2-part post I gave you instead of acting like I provided no evidence.

    "Inaction can help provide a job-creating climate also. You count that as a mechanism? How do we quantify the results?"
    For the purposes of this argument, it doesn't matter if it can be quantified. The problem is that when you chose as evidence of your claim a statistic that includes too many other significant variables, then you have failed to justify said claim.

    "That's kind of the point. The statements contain ambiguity and PolitiFact interprets arbitrarily. But where's the evidence supporting your original statement? Is this the argument from silence or what?"
    Actually I just spent two comments explaining the rationale behind the arguments. Can you respond to them before acting like they don't exist. That is an incredibly dishonest tactic. Is this the kind of dishonesty that you present in this site? STOP SKIPPING MY ARGUMENTS AND PRETENDING LIKE I DIDN'T MAKE THEM. Or are you just a dishonest person? I'm wondering if you are trying to drive me away with absurdity.

    "All two of them? What could possibly go wrong?"
    Do you have any evidence they are wrong? Or are you just speculating and attempting to shift the burden of proof? Because you have failed to provide a half-way decent reason to think net jobs was a factor. I have provided a few.

    "
    It's only incoherent if we accept your equivocal language in the final sentence. Texas would have created 5000% of all the net jobs created in the U.S. There's nothing incoherent about it. Contrived argument, much?"

    I feel like an elementary school teacher. Let me break this down to you Barney style since you seemed so committed to absurdities.
    Lets say someone saw the claim "Approximately 5000% percent of the jobs created in the U.S. from November 2007-2008 were in Texas." Am I off base when I say a reasonable person would assume all jobs created in the US state of Texas are created in the US. Now here's the logic:
    a = All the jobs created in the US
    b = All the jobs created in Texas
    |b| = 50*|a|
    so |b| > |a|.
    Since b is greater than a, a set of all a cannot possibly contain all b. Therefore all Jobs created in Texas cannot also have been created in the us. But that contradicts the first assumption.
    As you can say, the contradiction makes statement "Approximately 5000% percent of the jobs created in the U.S. from November 2007-2008 were in Texas" absurd.

    Just to further show the absurdity that could occur if Perry was talking about net jobs, picture a situation where, instead of gaining 50,000 jobs, Texas instead only gains 48000 jobs. This means the net jobs created in the US would be -1000 jobs. This would mean the statement would say "Approximately -4800% percent of the jobs created in the U.S. from November 2007-2008 were in Texas." If this wouldn't confuse the ever living crap out of a reasonable person, if not make Texas' job gains actually look pretty bad, I'm not sure what would...

    "If there's evidence I ignored the mechanisms in Perry's case it has nothing to do with the coherence of Perry's claim."
    Perry's claim has nothing to do with any mechanism that would exclude using net jobs. The kind of claim makes the use of net jobs absurd at best, misleading at worst, as a measuring stick.

    "That's so cute! Apparently I need the type of critical thinking that magically turns a coherent statement into an incoherent one."
    I feel like you should be smart enough that I shouldn't have to explain elementary logic to you. It seems like the problem is that you are not taking any time to think critically about what i am saying.

    ReplyDelete
  53. Almost missed this, sorry:

    (N)otice the article where this quote came from actually took place long before the debates, btw.

    No, there'd been a debate in S.C. before that and the candidates were in N.H. for a debate on the same day Gingrich made the statement PF was fact checking.

    http://en.wikipedia.org/wiki/United_States_presidential_election,_2012_timeline#2011

    ReplyDelete
  54. "Apart from the fact that it simply isn't true, it doesn't make any sense for the GOP field to behave as PolitiFact claims. Each needs to make the case that he has a better economic plan than his competitors, not just Obama. The claim from PolitiFact is not factual.
    This was a long time before the primaries, half a year in fact, so this excuse carries little weight. And your other reasons for thinking this not factual are, as I've pointed out before, absurd.

    But it does send a type of message. What type of message do you get from it? What does it say to you about the GOP candidates that they supposedly spend most of their time talking about how their economic plans are better than Obama's? "
    I'm sorry but I just don't see a negative message here. All it says is that the GOP candidates are spending most of their campaign time campaigning against Obama instead of each other. Or at least they were back then. Not sure what kind of negative impression that reasonably gives.

    If you want to keep pretending this weak example is a good example of "snark" I obviously cannot change your mind. You claim it is false for either absurd or far fetched reasons.

    In general, I am done dealing with this article. I give you well-thought out critiques and you respond with absurdities. I have broken down too many things barney style for you. And I have no doubt you can go on forever responding with these absurdities. I seriously feel sorry for anyone who wastes their time on this site. You may have a few good points in here but they are buried in a pile of logical fallacies and poorly researched articles. If you must still hold to the delusion that Politifact is liberally biased, it is no surprise. Delusions can be a hard thing to escape. Have fun single sourcing conservative sites with the cognitive dissonance needed to pretend they are no less trustworthy than serious neutral fact checking sites. This poisoning the well fallacy is the worst habit of republicans, but there are some so committed to the delusion, you cannot shake them. I hope whoever reads your article goes over our conversations and sees just how poor some of your work is, as well as the absurdities you hold onto just to continue to live in your delusions.

    ReplyDelete
  55. Maybe you should have looked at that whole 2-part post I gave you instead of acting like I provided no evidence.

    I don't see why. If you provided any evidence it didn't come from the part I suggested you skip. So why waste the space?

    (I)t doesn't matter if it can be quantified. The problem is that when you chose as evidence of your claim a statistic that includes too many other significant variables, then you have failed to justify said claim.

    That problem applies across-the-board with the mechanisms you say I err by ignoring. We end up trusting economists and their models. The Keynesians (and partisans?) outvote the minority. Poof! Truth!

    Actually I just spent two comments explaining the rationale behind the arguments. Can you respond to them before acting like they don't exist.

    We're working through those posts, from what I can tell, and we're through the first one without the type of evidence you need. The Texas Policy Foundation one is next.

    Handily, you summed things up for me:

    do I need to go down this article line by line and say “don’t see it here”?

    Argument from silence? Other than your proposed argument from silence, you're trying to argue that my argument should take consistent treatment of "mechanism" into account. That's your only avenue for keeping your approach from being a fallacy. Unfortunately, you were just wrong (you wrote): Gross jobs had no effect in the ruling.

    Yet here's part of the concluding paragraph of that PF story:
    "But the foundation’s analysis disregards the 40 states where millions of jobs were created but were outnumbered by losses."

    That's docking the statement for not considering gross job creation.

    ReplyDelete
  56. If you want to keep pretending this weak example is a good example of "snark" I obviously cannot change your mind. You claim it is false for either absurd or far fetched reasons.

    Reporters don't write things like that. Columnists do (opinion). Reporters, generally speaking, do not use themselves as the source of information unless they personally observed it. I think you know little of American journalism.

    In general, I am done dealing with this article. I give you well-thought out critiques and you respond with absurdities.

    Hogwash. You botched your interpretation of Ostermeier's post about selection bias and you went downhill from there. Your crowning achievement was finding a spot where I used the wrong link.

    ReplyDelete
  57. I feel like an elementary school teacher.

    Well, you do have an elementary school teacher's grasp of logic, based on the following.

    Let me break this down to you Barney style since you seemed so committed to absurdities.
    Lets say someone saw the claim "Approximately 5000% percent of the jobs created in the U.S. from November 2007-2008 were in Texas." Am I off base when I say a reasonable person would assume all jobs created in the US state of Texas are created in the US. Now here's the logic:
    a = All the jobs created in the US
    b = All the jobs created in Texas
    |b| = 50*|a|
    so |b| > |a|.
    Since b is greater than a, a set of all a cannot possibly contain all b. Therefore all Jobs created in Texas cannot also have been created in the us. But that contradicts the first assumption.


    What principle of logic compels you to ignore the fact that net job creation numbers will vary between Texas and the USA as a whole? Obviously there's no contradiction. A net loss in the USA overall does not mean a net loss in Texas by itself. This argument suggests that teaching elementary school would be too advanced for you.

    As you can say, the contradiction makes statement "Approximately 5000% percent of the jobs created in the U.S. from November 2007-2008 were in Texas" absurd.

    lol
    No, it doesn't. As I pointed out, we'd have to treat the numbers as gross job creation numbers for that absurdity to obtain. Your equivocal language suggested just that. But I called you on it. Unfortunately the correction didn't sink in for you. And you've left us with this stinker of an argument for posterity. Nice.

    ReplyDelete
  58. There are some flaws with PolitiFact's methodology, but at their core, they seem to be serious about their mission, despite some oddball rulings.

    Their 2011 Lie of the Year that essentially defends Paul Ryan's proposal to replace Medicare with a voucher program from liberal attacks that it will "end Medicare". With that editorial choice, they lost a lot of respect from liberals who see that as more opinion than fact -- and especially considering that most of the instances of the attack were phrased carefully to be essentially true or fundamentally opinion, and not subject to "fact checking" per se.

    I personally find the PolitiFact's apparent LACK of a detectable numerical bias more suspicious. Selection bias is sure to crop up at times, and one some issues, one side or the other relies more on false statements than the others [versus statements of opinion or cherry picked facts or principles that argue for their position without the need for any misleading or factually false statements].

    And yet, when you look for a systematic bias by the numbers you find it very difficult to argue for one. I find this fishy, but then I heard it was intentional -- how else to get people from both sides to respect you? How do you achieve such a result intentionally? By carefully introducing bias! First, selecting from a larger pool of facts to check, good and bad ratings are generated and released in such a way as to maintain the appearance of neutrality! As an example, consider Palin's recent "True" claim that Obama "said (the individual mandate) wasn't a tax." Score one 100% true statement for Palin even though the implication that Obama lied is a little misleading. Did someone call this claim into question? The Democratics were pretty consistent about calling it a "requirement" and "penalty" rather than a tax -- I doubt any reasonable person tried to call her out on saying the sky is blue, but that claim shouldn't score points either. That's an issue, but it's not a fatal flaw. I would be okay with that, if that were the primary flaw.

    ReplyDelete
  59. The fatal flaw I see is that when one side goes way too far, a second, more dangerous approach is used to maintain the numerical appearance of balance -- giving or withholding the "benefit of the doubt". Off the top of my head, I can think of two common ways this happens.

    One is when dealing with numbers that can have meaning when normalized against GDP, corrected for inflation, scaled in some other way, or perhaps left raw. PolitiFact doesn't begin such a judgements by considering which scale is the most meaningful in the policy context. PolitiFact does give the benefit of the doubt by selecting whichever approach makes a claim True or they withhold the benefit of the doubt by using absolute numbers to make a statement less true when the more meaningful numbers would otherwise make the statement true [I have never seen them use more meaningfully scaled numbers to call someone's true statement about absolute numbers into question -- which they would if they wanted to be taken seriously in an academic policy discussion rather than just in politics].

    The other way they manipulate "benefit of the doubt" is when a political figure asserts something that is technically true, but implies a falsehood. I just heard a GOP congressional leader note that Obama promised health reform was not a tax, and now we learn that it is -- the way he and others have worded it, it was clear to myself and the other person watching that he MEANT to imply it was a broad-based tax to pay for a program. In fact it is a tax penalty that will only touch about 1% of the population and not generate significant revenue. Lucky for liberals, Rush Limbaugh has already gone overboard with an outright Pants on Fire lie on exactly this question. In this case, though, a pure fact check could rule in his favor. But this is the very definition of Half True or perhaps Mostly True. Based on my experience, if they were to rate it, the odds are that they will rate it True. After all, what he said was true. And yet, if it were a Democrat, there is no chance in hell it would get a pure True rating! In Obama's SOTU address in January, he made a series of factual claims about jobs and growth, two of which were taken on by PolitiFact. After reviewing the facts, they ruled that while the statement was 100% factually accurate, because of an implication about the impacts of policy, it earned a HALF True. Conservatives would likely agree with this rating, after all, they would give his policies no credit at all, but as the article itself noted, they made no judgement of the implication. This is a problem. If the implied point is an opinion that PolitiFact is not judging, why would it detract from the truth of the fact they were judging? Isn't politics mostly about opinion that people sometimes try to support with facts?

    In short, they do good on a lot of individual points, but they and others have created a danger. We now see an environment where it is in the best interests of highly visible ideologues who are willing to take some hits to outright lie in the most extreme ways possible to force PolitiFact and any other media that needs to appear unbiased to tip their scales in FAVOR of the liar's position going forward in order to continue appearing unbiased.

    ReplyDelete
  60. Thanks for sharing your views, Oran.

    A few comments ...

    And yet, when you look for a systematic bias by the numbers you find it very difficult to argue for one.

    I disagree. The thing that makes arguing from the numbers tough is simply the difficulty of separating out different methods of introducing bias. PolitiFact's system powerfully reinforces bias. We find it measurably in the selection process, the writing process and in the rating process. Yet through all of that PolitiFact publishes "report cards" that ignore the potential contamination in all those factors. They can't be that stupid.

    As for the Palin example, that piece of selection bias is easy to explain. The tax/penalty angle is in the news and people love reading about Palin.

    I have never seen them use more meaningfully scaled numbers to call someone's true statement about absolute numbers into question

    They do that commonly, in truth. One example had Palin mentioning the worldwide rank of U.S. defense spending as a percentage of GDP. PolitiFact rated her claim "Mostly False" because supposedly the comparison is most meaningful by only including industrialized nations. Yet looking at the rankings *proportionally* there wasn't much change. The number of countries above as compared to below worked out similarly in each case by proportion.

    This type of thing happens often, and Republicans suffer the harm from it more often than Democrats.

    In fact it is a tax penalty that will only touch about 1% of the population and not generate significant revenue

    It generates significant revenue. Just not near enough to pay for the high cost of the bill.

    "Revenue, it turns out, is the key idea. To fall within the tax power, a law must raise revenue. And the mandate certainly does. The Congressional Budget Office estimated that the mandate will have raised $17 billion by 2019, and that starting in 2017 it will raise approximately $4 billion a year.
    http://www.theatlantic.com/politics/archive/2012/05/the-health-care-mandate-is-clearly-a-tax-0151-and-therefore-constitutional/256706/

    if it were a Democrat, there is no chance in hell it would get a pure True rating!

    There is a chance, actually. Bill Clinton claimed full credit for reducing the budget deficit. PF gave him a "Mostly True" rating for his claim, mentioning absolutely nothing about him exaggerating his role. He was docked for somewhat inaccurate numbers, not for omitting credit to the Republican-controlled Congress.
    http://www.politifact.com/new-jersey/statements/2012/jun/08/bill-clinton/bill-clinton-touts-fiscal-record-president-during-/


    In short, they do good on a lot of individual points

    So do bloggers and high school newspapers. Calling it fact checking calls for a higher standard. PF doesn't cut it.

    ReplyDelete