It's definitely a part of the conversation, but let's keep in mind that the NYT, and frankly even (to a lesser magnitude) 538, have a lot of agenda-setting power for what's part of the conversation. So Twitter isn't an exogenous measure of what normal people care about. https://t.co/A8oHUPYJPo— Nate Silver (@NateSilver538) January 16, 2019
You can peruse the current tweets about Gillibrand and Franken here. Two examples:
"Holding Gillibrand accountable for her statements and actions IS NOT the same as 'blaming her for Franken’s problems.' Using the #metoo movement to grab some spotlight, her instinct to govern through political theater are legitimate points of debate and criticism" (link).
"Lukewarm on Gillibrand atm, but credit must be given for alienating Clintonworld by saying Bill should've resigned, and standing her ground on Al Franken" (link).
Anyway, I'm very interested in Silver's out-and-proud encouragement of mainstream-media "agenda-setting" and his resistance to the notion that Twitter is "an exogenous measure of what normal people care about." I love that phrase! It's so weird and revelatory of anxiety.
Exogenous — it makes you think! The oldest usage of "exogenous" is in botany. It means "Growing by additions on the outside" (OED). Can that be the metaphor Silver wants? In pathology and psychiatry, it means "having a cause outside the body." What is the relevant body that Twitter could be outside of and measuring? In geology, it means "Formed or occurring outside some structure or mass of rock."
So... Twitter is on the outside... of what?... doing what? I think "exogenous" should at least have to do with something growing or forming on the outside of something, but Silver is talking about Twitter measuring, so it can't be the right word, and I'm still questioning what Twitter would be exogenous to.
The best I can do to save Silver from the conclusion that he's just tossing out a fancy word without thinking it through is that some people imagine that the normal public mind expresses itself through Twitter, and since "Twitter isn't an exogenous measure of what normal people care about," those people are wrong.
IN THE COMMENTS: rehajm said:
Nate is quantitative and is using exogenous in the quantitative, statistical sense...He didn't say "exogenous variable" or "exogenous factor." He said "exogenous measure." A factor is causal. A measure isn't a cause. I get that it's a jargon word for a statistician, but I still don't understand how it works in his statement.
Exogenous Variable
A factor in a causal model or causal system whose value is independent from the states of other variables in the system; a factor whose value is determined by factors or variables outside the causal system under study.
They are discussing the meaning of why Franken tweets are trending in NY, and his point is twitter trending is not an independent variable from which we can draw conclusions since much of what trends is a result of what NYT and 538 and others choose to promote.
Posts
46 comments:
"Lukewarm on Gillibrand atm, but credit must be given for alienating Clintonworld by saying Bill should've resigned,
She defied omerta. She broke with the thirty plus year conspiracy of silence on BIll. Naturally she is going to have enemies at the NYT who decide which “facts” belong in the conversation. If you limit the number of facts one is allowed to consider, Occam’s Razor falls completely apart, as it is too easily manipulated. That’s what the New York Times does. "Flood the zone” with limited facts that point in the direction we decided a story should go before any facts were known.
While I am happy that Gillibrand called out the Clintons, the way she flipped her positions on a dime when she was elevated to the Senate from her sensible positions as a Congressman representing a rural NY disctrict was enough for me to see her as a crass opportunist. Of course Democrat insiders think that when she turned on Franken and Clinton, that was the opportunism.
I'm just hoping for a big fight that alienates nearly everyone and cripples the Democrat Party for 20-25 years.
One can dream.
Exogenous of media. That's my take. There is Nate and NYT and CNN. You know...real news.
Twitter is not news at all. According to the self proclaimed journalists, like Acosta. Twitter is a growth out side what Nate considers 'news', or I guess more accurately, not news, but agenda setting
From Nate's view, more accurately twitter is a parasite, or in reality, twitter and NYT have a symbiotic relationship. Each providing something of value to the other. Sort of.
True symbiotic relationships exist for each other. One cant survive without the other.
Perhaps he's saying the conversation on Twitter isn't outside of the control of the agenda-setters like himself and the Times.
Now that everyone knows to vote for the woman, we have to make sure they don’t pick the wrong one...
It means penis.
Silver is a statistician, and the word seems to have a precise meaning in that field (described here: http://www-personal.umd.umich.edu/~delittle/Encyclopedia%20entries/exogenous%20variable.pdf ) but having read the explanation, I'm still not sure he's using the word properly.
Silver is saying without the NYT, and to a lesser extent 538, people would wake up in the morning with nothing to Tweet.
I remember when people respected Silver because he was a quantitative genius. Last election night he had to sneak his foot around the cord and pull the plug on his predict-o-meter on live TV because it was so out of line with reality. So now what is Nate good for? Well, he's demonstrated he's susceptible to the pressures of Think Progress, which wanted him to fire Roger Pielke and he dutifully complied. Now he's a good mouthpiece for whatever agenda lefties needs pushed at the moment. He can still sell stinky propaganda by wrapping it in logic fallacies with a cherry picked statistic on top. Fuck him.
Twitter isn't an exogenous measure of what normal people care about.
Sometimes we fall in love with a nifty construction and keep it even though it does not quite capture what we want to say. I think he's got it turned around--he means to say "what is important to people on Twitter is not necessarily an indication of what is important to the world outside of Twitter."
"Exogenous" is the opposite of "endogenous." For someone using statistical evidence to evaluate a hypothesis, an "exogenous" factor would be a factor whose occurrence is not correlated with the occurrence of other factors under consideration. I take Silver to be saying that it might be a mistake for media people to use the occurrence of mentions of Franken on Twitter as evidence of how the general population thinks because the media interest in the Franken/Gillebrand issue contributes to what the Twitter conversation looks like.
The Gillbrand atm? The idea of viewing that porno video makes me want to vomit just a little bit.
THEOLDMAN
I am having a hard time viewing Nate Silver as other than some jumped up statistician who yearns for status as an agent of the Progressive Deep State.
@ rehajm - You said it much better than me! Yes, fuck him!
Normal people only care about the conversation chosen by the agenda-setters, says Nate. That is the body. Twitter is outside, exogenous to the body. Nate and NYT decide what people care about, not Twitter users like Trump, who are no-bodies.
Nate is quantitative and is using exogenous in the quantitative, statistical sense...
Exogenous Variable
A factor in a causal model or causal system whose value is independent from the states of other variables in the system; a factor whose value is determined by factors or variables outside the causal system under study.
They are discussing the meaning of why Franken tweets are trending in NY, and his point is twitter trending is not an independent variable from which we can draw conclusions since much of what trends is a result of what NYT and 538 and others choose to promote.
"Exogenous of media. That's my take. There is Nate and NYT and CNN. You know...real news. "
I considered that possibility. A lot. But in the end I rejected it.
He's talking about how MSM imposes the agenda on the public mind, so it can't be that Twitter is a measurement of MSM. The thing to be measured is what people care about, and he's not talking about MSM as the thing getting measured. MSM is the thing that affects the shape of the public mind.
There's a big problem with using the word "measure" with "exogenous" that really can't work. "-gen-" has to do with growth and causation, not measurement.
I wonder what percentage of the voting public could identify who Nate Silver is.
What Nate is saying is that this is the epistemological heuristic with which power is expressed by hegemonic historiography complicit in the intersubjective modernist project of foreclosed dialectical images. These durational discursive practices ground temporal frames, indexing our thinking to the double exigencies of subjectivity and the backgrounding of the relevant post-neoliberal social imaginaries.
Simply put.
Google says about 19% of Americans have twitter accounts and only about 9% are on twitter every day. So, 91% of Americans do not participate in the twitter conversations of the day, which means the vast majority of people can't be directly influenced by twitter but may be indirectly influenced through the 9% who do participate in the conversations. I would suspect that the 9% is mostly composed of journalists, politicians, pundits, lobbyists, policy wonks and other "opinion makers." That 9% is who Nate Silver believes to be the normal people. This type of thinking no doubt contributed to the polling misjudgements and shock of the '16 presidential election and probably will again. Elites gotta be elite.
For pete's sake. Silver is a statistician:
An exogenous variable is a variable that is not affected by other variables in the system... Exogenous variables are taken as a given in a model.
In this case, the model that Silver is talking about is "what normal people care about." What makes Twitter NOT exogenous is that it is not a given. You can't treat Twitter as an accurate external measure independent of all the other things that influence people.
That raises an interesting question -- what is an accurate exogenous measure of what people care about? There may be no such thing.
I see I'm late to the statistics party.
I like Kevin's formulation:
Silver is saying without the NYT, and to a lesser extent 538, people would wake up in the morning with nothing to Tweet.
Well put.
Exogenous is a big word in economics, and I assume political science also. It means "from outside the system". I think Silver's idea is that twitter isn't purely "grass roots". It isn't what people come up with spontaneously. It's also (maybe mostly?) people reacting to what the "agenda setting" media are talking about.
I think we've all got it, even Ann, who went a little deeper down the rabbit hole than needed.
Trump is Twitter. We await his nickname for the senator. Perhaps Sillybrand
I think you're right on "exogenous." I've heard economists use it. Does some set of data actually provide evidence as to whether a policy is working or not? It might represent an "exogenous" factor, that shows up and down trends that are unrelated to a particular policy. We need to look at another kind of factor--endogenous?--that actually shows the results we are looking for in a meaningful way.
I'm still amazed that Gillibrand and Klobuchar put big knives into Franken's back. No doubt he is even more amazed than I am. Will this fly in a presidential election? Women suffer so many daily injustices, someone like Franken has to be destroyed based on such small evidence? I notice R. Kelly is taking on a lot of water, or with a bullfighting analogy, there are a lot of swords making him bleed. There seems to be far more evidence against him.
This thought triggered by comments on another thread last night
https://ricochet.com/588691/president-trump-and-the-kobayashi-maru/
- Captain Kirk was exogenous to the Kobyashi-Maru scenario.
Ordinary Twitter users are not Captain Kirks.
Only Trump - God-Emperor-of-the-Universe.
What a great post. Forced to learn new words and see of our beliefs survive their definitions.
Twitter isn't real. Tweet storms and Twitter attention isn't what real people care about. You see it over and over. It's why companies should just ignore Twitter outrages.
So, maybe people on Twitter who are still mad about Franklin don't represent the majority opinion out there. But I'd think that's true of most of the things KEG has made her name on. You think real people care about the whole Mattress Girl saga? You think real people care about declarations that "the future is female"?
Now, I'd like KEG to have to answer for some of these. I'd like to see her have to answer for pushing out a fellow senator - perhaps to clear her own path to the nomination. I'd like a journalist to ask her if she's sorry she brought Mattress Girl to a SOTU, or what it means to say "The Future is Female", especially when she has sons. But they won't, because journalists are Twitter. They are all in a little Twitter circle, and they don't want to be exogenous-ised.
So the next time you see polling data from 538, remember it comes from a guy who thinks he has "a lot of agenda-setting power for what's part of the conversation."
Who made this guy such a god?
i think that Nate Silver used the wrong "big word" than what he intended. Surrogate is better, or extraneous or simulacrum would work if he was trying to just say that Twitter is not a true representative of the political thought (or rational thought for that matter). But he said "So Twitter isn't an exogenous measure of what normal people care about." So he was right on the prefix "exo" which in toxic-/pathological terms refers to being sourced from outside normal people. But he used "isn't" so with the negative, he should use "endo" since i think he means twitter doesn't reflect what normal people care about. Or as we toxicologist say, "Twitter megaphone in faecibus exturbandis opitulatur capitibus or Twitter is a megaphone for poopy-heads"
@Darrell. Thanks for clearing this matter up.......All my life, I've been trying to figure out what hermeneutics means. It's a word that pops up quite often, and I've never quite got a handle on it. This is the first time I've seen exogenous. I hope it doesnt catch on. On the plus side, I think I can use solipsistic correctly, not that I ever would.
Gillibrand- Franken in 2020!
Nate Silver: "We make the rules, pal. The news, war, peace, famine, upheaval, price of a paper clip, whether or not Gillibrand's ass should be considered Presidential. We pick that rabbit out of the hat when everybody sits out there wondering how the hell we did it."
As others have said, Silver used "exogenous" correctly. Franken discussion is endogenous to the system. In other words, Silver committed a gaffe--accidentally told the truth.
And with Twitter's SJW's deplatforming voices it doesn't like, the conversations that survive will steer even more leftward, making it even less representative of what the country thinks.
Wasn't there a study years ago that opined that without the left-leaning media the country would be even more conservative? It's the butcher's thumb on the meat scale.
You're making this far more complex than it needs to be. Silver is saying ordinary Democratic voters don't care about the things elite Democrats fret about on twitter and in the New York Times. In other news, water is wet.
Saying "exogenous measure of what normal people care about." definitely shows that Silver is just folks like us.
Ann, Yeah, Silver phrased it poorly, but I think it's fairly clear that when he said, ""So Twitter isn't an exogenous measure of what normal people care about', he means "seeing what people talk about on twitter isn't an accurate measure of what people care about, because what people talk about on twitter is largely exogenous to what they really care about." He wrote too quickly (hey, talking about twitter) and collapsed what should have been two sentences into one.
Or to keep it in one sentence, "twitter isn't an accurate measure of what normal people exogenously care about. Which sounds pretty awkward.
Since you've front paged me...
Can we not infer:
1) Nate believes his explanandum what normal people care about is at least partially explained by Twitter and by NYT and by 538
and...
2) His warning: let's keep in mind that the NYT, and frankly even (to a lesser magnitude) 538, have a lot of agenda-setting power for what's part of the conversation implies he believes Twitter is influenced by NYT and 538, in part.
Yes? If so, Twitter is endogenous within this system since it is influenced by NYT and 538. Twitter isn't exogenous.
QED.
(Have at it...)
I suppose one could argue 'endogenous' and 'isn't exogenous' aren't mutually exclusive since we can have factors that are partially both, but that's neither here nor there if we're quibbling about Nate's use of exogenous.
He didn't say "exogenous variable" or "exogenous factor." He said "exogenous measure." A factor is causal. A measure isn't a cause. I get that it's a jargon word for a statistician, but I still don't understand how it works in his statement.
Then shouldn't your quibble be with his use of measure? I see him using 'measure' as a synonym of 'factor' or 'variable' and as more descriptive synonym to boot, as he has already attempted a (faux) quantification before he offers his warning. I see how one could certainly nit pick this. Do you have a problem with it?
The Tweets:
Goldmacher: In New York, tweets about Al Franken are trending tonight
Silver: A lot of the Franken-related tweets (maybe ~60% of the well-retweeted ones?) are supportive of Gillibrand, though.
Goldmacher: Not sure how to measure how many are positive vs. negative (both are certainly in my feed) but a sign Franken remains a part of the (online) conversation about Gillibrand
Silver: It's definitely a part of the conversation, but let's keep in mind that the NYT, and frankly even (to a lesser magnitude) 538, have a lot of agenda-setting power for what's part of the conversation. So Twitter isn't an exogenous measure of what normal people care about.
I once dated a girl with a great pair of "exogenous variables."
If Trump wasn't using Twitter, it would be almost irrelevant to conservatives -- since so many thoughtful Rep-supporters have been banned for trying to speak the truth, like good Reps do.
Rehajm nails it, "measure" is usually a verb, but can be a noun, like metric, and thus sort of like variable.
For measuring what normal folk say, Twitter trends aren't a good metric. Neither is Facebook, any more; and Google searches are also de-emphasizing what Reps search for.
The elite Dem media & tech doesn't want a true measure, and doesn't allow one.
They fear that if more conservatives knew how much they really are not alone, thee would be even more of them, and fewer Dem folk would "walk away" from the careful Rep demonization being done on so many college campuses.
Truth would damage the chances of Dems winning, thus truth needs to be suppressed, a bit. How much? The maximum that allows plausible deniability, same as with voter fraud -- like the voter fraud that allowed Franken to win his election.
there would be even more of them, and more Dem folk would "walk away" from the careful Rep demonization being done on so many college campuses.
(no edit here?? oops, too fast; need to correct the comment correcting a correcting tweet... isn't life ironic?)
This post was better than any Rule 5 post or otherwise titillating article. I was a Political Science B>S> major at Southern Methodist University and this brought back memories of PLSC 3375 Quantitative Methods of Political Research, one of the most feared yet thrilling (from a statistics manipulation standpoint) courses available in the Poli Sci school. Word like Exogenous were not bandied about lightly! Now excuse me while I escape to mah bunk and reread this article.
Post a Comment