Monday, February 29, 2016

PolitiFraud


We need a name for the so-called "PolitiFact."  PolitiFraud?  PolitiCrap?

I'm not a fan of the "Fact Check" sites. They have an air of "being the last word" and of being objective, but in my experience they are just as unreliable as any news source and far from objective.
Here is one of many examples about Ted Cruz where it's claimed that he said something false. Give it a quick read.  Now note that most statements by Cruz on PolitiFraud are rendered "false."  Could it be that they only focus on some statements and not others?  Could it also be that they are rigging the game with "degrees of truth", whatever that means?

First, in a story about Reagan and Cruz where there is no mention of Clinton, why feel the need to point out that blacks did better under a Democrat? While true--and fair game for politicians to use--that doesn't have the air of objectivity about the particular facts at hand.

Second, while it's fair game to point how how (e.g.) the economy did under a president, that alone tells us nothing about to what extent the policies of the president played a role. Presidents inherit the regulations of predecessors as well as whomever happens to be in Congress, etc. So Cruz is guilty of this fallacy, but then again, so is pretty much every politician I can think of.

Third, when I first looked at their story, what struck me is why they didn't mention why they first said "true." What was it that made them say true in the first place?  They don't say.  That raises a red flag.

I can only speculate since the story has been redacted but here is what I think was the case.
They decided in the second story to use the adjusted income based on inflation in 2014 dollars. (Click on the word "Black" on the chart they provide and you'll see what I'm talking about.)  That seems to me even more misleading. (Why not use the 1988 adjusted income according to inflation? That would tell you how much real earning power was acquired during the 8 years.)

It looks like Cruz is simply using the real time dollars in 1980 and 1988. In 1980 the median income was 10,764; in 1988 it was 16,407 (so actually over $5000 and closer to $6000). If you took the mean income it's an increase of c.$8500 to $10,000 depending if you go with real time or adjusted to 2014 dollars.

So, unless I'm wrong about where the stats are coming from, a more accurate story would have been to clarify why they said true and then note that it depends on whether you are looking at current dollars or dollars adjusted for inflation from 2014.  Again, why use adjusted figures when that's even more misleading?  This appears to be a very underhanded way to say that he was asserting something false.

Related.

2 comments:

  1. They basically admit on their website that their opinions have no credibility: http://www.politifact.com/truth-o-meter/article/2013/nov/01/principles-politifact-punditfact-and-truth-o-meter/

    "Because we can't possibly check all claims, we select the most newsworthy and significant ones."

    By whose opinion is something determined to be newsworthy? Liberal journalists? Hmmm, I wonder if that results in more controversial claims by conservatives being fact-checked.

    "Is the statement leaving a particular impression that may be misleading?"

    What is an "impression"? What plays into the impression a statement leaves? Could it be one's preconceptions or one's a posteriori political leanings? Could that possibly have an impact on whether a statement is viewed as worthy of being checked?

    "Would a typical person hear or read the statement and wonder: Is that true?"

    Who chooses the definition of "typical person"? The liberal journalists checking the statements? Could they be biased in any way to think typical people think the same way they do, and therefore have a harsher perception of conservative claims?

    Essentially, what prevents the writers and editors of Politifact and other fact-checkers from succumbing to confirmation bias? Nowhere in their "About Us" or descriptions of their process do they give any indication that they are even AWARE of the risk of confirmation bias, much less any measures they put in place to mitigate it.

    ReplyDelete