When I have to watch cable news networks present and "discuss" polling results I feel the way doctors must feel when they watch E.R. or lawyers would feel watching a cut rate Law & Order knockoff courtroom procedural show. The urge to yell "THAT'S NOT HOW THAT WORKS" at the TV screen borders on overwhelming at times.

When CNN assaulted the data from their Nevada Caucus Entry Poll to declare "Trump wins Latino vote in Nevada" any freshman student in math, statistics, or survey research could tell you what's missing from this picture:


Gee those are some nice lookin' percentages. By the way, how many Latinos voted in the Nevada Republican caucus? Because I'm guessing it was about 80.

That's an exaggeration, of course, but common sense dictates that it isn't going to be very many, what with Barack Obama winning 76% of the Latino vote in Nevada in 2012. Furthermore the NV Caucuses were a legendary shitshow that descended into chaos in more than a few locations, so the question of the validity of those Entry Polls certainly cannot be assumed. Who showed up, who actually voted, and who was some asshole Trump supporter who thought it would be funny to pretend to be Latino are all open to interpretation. Given the low turnout in the NV Caucus overall and the small number of Latinos identified – about 100 – by the polling agency suggest that statewide no more than 2000 Latinos, an estimate that appears generous, participated.

Nevada has at least 750,000 voting-age Latino residents according to the Census Bureau. It might have been helpful to point this out when declaring that Trump "won" a tiny subset of a small polling sample, or perhaps to have one person working at the network on election day who has some idea of how numbers work. But who cares as long as it sounds alarming, right?

23 thoughts on “LIES, DAMN LIES, ETC.”

  • Political pundits/consultants exist to blow smoke rings up delusional rich people's asses.

    Somehow writing political fanfiction with erotic overtones became more profitable than journalism. Who knew?

    Except with the current GOP civil war, some political fanfiction is coming true. People are predicting long range consequences for the GOP over that, but no matter the result, I predict in 2018 the GOP will be fully united against La Femme Dictator and all the bad feelings of 2016 will have been washed away with soma.

  • Condolences Ed.
    They know HOW to talk about the thing in the flashy, alarming cable news way.
    They just don't know, or say, what the THING is.
    Zero on the meaning and value scale.
    Resembles Trump and that ilk in so many ways.
    Flashy and alarming, not much there.

  • I've noticed that NPR does much the same thing nowadays. Seems to have started a few years ago but many's the time I've had to turn the radio off 'cause I just can't stand the incomplete journalism.

  • I think it's even more misleading than what Ed points out. There will be no way to do this, since it's not a general election, but the other point that makes those numbers look big is the nature of the activity. Separate from the small sample size all that is being discussed and tallied (poorly) are self identified Hispanics that voted for Republican candidates. And so, of course, the numbers that get presented look impressive, it's baked into what is being counted. I think it speaks badly of the public's ability to discern the context in which these things exist (which Ed points out very well) at a level deeper than the total votes cast.

  • Actually, with the benefit of another couple of minutes of reflection, I take it back; Ed did say everything I did, just…better.

  • Hmmm…. [drags statistics class skeleton from closet] Two Latino-sounding names totaled 47% in that poll, which, if I am not mistaken, is a higher number than 44%.

  • Dave I am with you on National Pentagon Radio. I think it is wrong to label it 'incomplete journalism'. They know exactly what they are doing, completely. Bias journalism is what I would call it instead. Perhaps just pedantic?

    Basically all the establishment outlets are trying to scare the hell of everyone about Trump. Whatever they can report to make it 'scary', they will. The latino example is an example.

  • "I feel the way doctors must feel when they watch E.R. or lawyers would feel watching a cut rate Law & Order knockoff courtroom procedural show."

    Imagine me watching anything with airplanes.

  • Bill Gates walks into a bar and the average customer is a millionaire.

    Crunch the numbers in enough ways and any number of meaningless things can be quantified. Its done partly to push a narrative but mostly(I think) to fill up airtime. So much airtime. God forbid they actually go out and cover the news.

    As an aside, the show with the worst scripts and most implausible plots is Scorpion. The other person in the house loves it so I am stuck with it. If it were so easy to violate the laws of physics in real life.

  • Skepticalist says:

    For a goof like Trump the old adage that all publicity is good publicity fits perfectly. Cable news has nowhere to go.

  • Now, that Michigan polling error brings up an interesting point. Back in 2012, some (Republican) analysts were claiming that the polls were "skewed" and would not reflect how people would actually vote. Of course the "unskewing" turned out to be a bunch of hot air, and in the general election, people voted pretty much how the polls said they would.

    Is something different this year? Or is the Michigan polling data an outlier? Of course it's worth noting that it was not just one poll, but pretty much every poll in Michigan, which showed Clinton winning the state, by a greater or larger margin.

    And I'm really skeptical of the theory that a lot of potential Clinton voters cast their ballot for Trump in the primary. Until someone comes up with credible evidence to the contrary, I don't think that most rank-and-file voters would care enough to game the system like that.

  • Ms. Ann Thrope says:

    Speaking of NPR's "journalism", I woke up this morning to NPR going on and on and on, I swear 7-10 minutes yakking about the winners of yesterday's Republican primaries, interviewed some Rubio guy or something (I was barely awake) and then at the end it was, Oh, by the way, Sanders won Michigan. Which woke me up. Yuuuge upset, but NPR didn't seem interested.

  • NPR? The people who do that 'tiny desk' music show do news, too?

    But seriously, if you're not already all over it, everyone should check out some Tiny Desk
    Sessions on YouTube or whatever. Good stuff.

  • On something of a tangent – one of the most ridiculous TV courtroom devices, imo, is when the accused is /tricked/ into confessing on the witness stand, due to his or her own bloated ego. Srsly, has that EVER happened in the entire history of the western judicial system? Or any other system, for that matter. It happens several times a week from Law and Order through Matlock.

  • The newscasts are looking to sell advertising to eyeballs. The meaning of statistics doesn't come close to entering the equation. People who care about what is actually happening won't go to the newscasts, and people who go to the newscasts don't actually care about what 's going on – they either want to see what the latest gossip is on everyone's favorite new reality show (As the Oval Office Turns), or they want validation for their unfounded biased opinions.

    There is no profit in facts. The newscasts will not show facts for that reason.

Comments are closed.