Justin Grimmer vs. the Hoover Institution commenters

I was curious what was up with the Hoover Institution so I went to their webpage and was pleased to come across a post, No Evidence For Voter Fraud: A Guide To Statistical Claims About The 2020 Election, by political scientist Justin Grimmer, with this clear summary:

We focus on fraud allegations with the appearance of statistical rigor. Trump and allies used statistics to claim some election facts would be unlikely if there had been no fraud. The claims fail either because sometimes the “fact” is inaccurate or it is accurate but not surprising. For example, a viral anonymous report claimed Dominion machines added 5.6% to Biden’s vote share. But, we show that the purported Dominion effect disappears as soon as we control for 2016 results, or make any number of other sensible design choices. Other times this is because accurate claims about the 2020 election simply are not that surprising. Trump and his allies claimed it was suspicious that Biden lost 18 of 19 counties that had correctly picked the winner since 1980. But we show that bellwether counties are bad at predicting future winners. Since these counties went for Trump in 2016, Biden’s low haul of bellwether counties isn’t suspicious at all. Likewise, in a lawsuit filed against PA the Texas Attorney General claimed that Biden had a “one-in-a-quadrillion” chance of winning. The probability comes from a report filed by Charles Cicchetti who examined election-to-election changes and the shift from early-to-late votes. We show Cicchetti’s tests are riddled with errors and vastly understate the probability of change. We apply his test historically and show that vote changes he said had a “one in almost infinite chance” of occurring actually happened in 6% of US elections. Our work is intended to help assess the security of US elections. We think it is important that non-partisan election experts evaluate fraud claims—to either identify suspicious results or reassure the public about the safety of US elections.

Here’s the full article, with Andrew Eggers and Haritz Garro, which has this crisp abstract:

After the 2020 US presidential election Donald Trump refused to concede, alleging widespread and unparalleled voter fraud. Trump’s supporters deployed several statistical claims that supposedly demonstrated that Joe Biden’s electoral victory in some states, or his popular vote in the country, were fraudulently obtained. Reviewing the most prominent of these statistical claims, we conclude that none of them is even remotely convincing. The common logic behind these claims is that, if the election were fairly conducted, some feature of the observed 2020 election result would be unlikely or impossible. In each case, we find that the purportedly anomalous fact is either not a fact or not anomalous.

To put it another way, there’s about as much evidence for massive election fraud in 2020 as there is that beautiful parents are more likely to have girls.

The sad thing is that Eggers, Garro, and Grimmer needed to write this article in the first place. I’m reminded of the scientists in the 1970s who would take time out of their busy lives to patiently explain the flaws in popular arguments for the Bermuda Triangle, Bigfoot, creationism, JFK conspiracy theories, spoon benders, etc., while also trying to understand how these rumors became some widespread and so hard to dislodge.

But, yes, I think this work is necessary, as we can see from the comments to Grimmer’s comments on the Hoover page. The vast majority of the commenters are absolutely sure that something bad went down and that 81 million people didn’t really vote for Biden. Grimmer went in the comments and carefully replied, refuting particular arguments and asking commenters to clarify their more general statements. It doesn’t seem to help, of course, but I applaud him for trying. It’s scary to see all those commenters with their conspiracy theories (either a conspiracy of vote suppression or a conspiracy of political scientists, election officials, and judges in multiple states who somehow think that Biden really did receive all those votes).

P.S. It must feel weird for the regular Hoover writers to know that these are the sort of people who are reading their columns. It’s good to at least try to inform these people and, sure, commenters are not a representative sample of readers, but still. If I had the impression that the vast majority of our readers were so out of it, I’d be pretty demoralized.

119 thoughts on “Justin Grimmer vs. the Hoover Institution commenters

  1. The repulsive creature announced long *before* the election that it *would* be crooked; and that, by implication, he would not concede. In Chinese we say, “The thug accuses first”. No one should have expected anything but. George F. Will was worried enough to write in a column several months before, that unless the defeat (expected) were *massive*, we would be in for trouble.

    • The Trump election fraud claims were too far out there and bizarre even for this simulation experiment.

      Not even the advanced beings from planet KXhghy3n56ZZzzrg could have come up with that one.

      • Joshua:

        Indeed. But what’s scary is that there’s a whole world of Hoover commenters who think it’s obvious, people who are so sure that Biden didn’t get 81 million votes that they’re willing to swallow conspiracy theories in order to maintain this belief.

        • In comparison with a couple of quite respective blogs I read regularly the small sample of comments in the Hoover blog that I read seem rather calm and restrained. Nutty as #$%%^& but pretty restrained overall.

        • Jkrideau:

          Compared to the Marginal Revolution commenters, the Hoover commenters are very polite. Hoover has a big budget so maybe they have someone who filters out the rudeness. In content, though, the Hoover commenters are off the deep end.

        • Andrew –

          >But what’s scary is that there’s a whole world of Hoover commenters who think it’s obvious,

          One thing to keep in mind that that there was deliberate propaganda campaign, enacted by a group of very seasoned hucksters, empowered by vast media entities and the psychological influence of social media, to make this happen.

          This isn’t just some kind of a random or accidental development.

        • Joshua:

          Yes, the starting point is that the Democrats couldn’t possibly have received the clear majority of votes. Everything flows from there. What’s weird about this view is that the Democrats have received more votes than the Republicans in most national elections in the past few decades. It’s the standard pattern.

        • But that’s just proof that Democrats have been rigging elections for a long time.

          After all, in 2016 Donald Trump actually won California’s legitimate vote. He only lost because of millions of votes from illegal immigrants.

          I live in California, and in certain circles that claim by Trump was (and probably is, I find myself avoiding such people nowadays) widely believed.

  2. I wonder if some of this connects to the intellectual dishonesty/politicians stuff on the other post’s comment thread.

    I mean, if you are already predisposed to believe that *essentially everyone in government is fundamentally dishonest* these sort of conspiracy theories are a bit easier to believe.

    I think also some people who buy into this stuff don’t realize how large the conspiracy would actually have to be; it’s much like the ‘overcounting COVID deaths’ thing – it’s one thing to believe that someone in the CDC office in Washington is just writing down whatever number comes to mind; but once you realize that these numbers come from thousands of people in separate hospitals, counties, state health departments etc. it becomes totally implausible.

  3. Anything can be described as a hoax and a conspiracy theory can be concocted about anything. Human information processing capabilities, statistical reasoning capabilities, emotional biases, and political and self-attention-seeking proclivities. allow it, and basically pretty much make it inevitable. A more pressing question is what can be done to minimize the damage (without producing unacceptable negative unintended consequences.) Individually I can think of numerous solutions, but on a collective level, and with highly motivated people incentivized to churn out hoaxes and conspiracy theories, (some of which are just trollery, or at least started out as such) I don’t have high hopes. Who will bell the cat?

  4. Why is it considered crazy to think people work together behind the scenes to achieve a shared agenda? That is normal human behavior. Except when the stakes are high like elections I guess.

    Why does the UN send election observers around the world to ensure fair elections? It just seems that if you think this type of behavior is rare it would be difficult to understand history or how the world works. That would be a serious disadvantage.

    • Anon:

      Nobody here is using the term “crazy” except for you. Eggers, Garrow, and Grimmer are talking about specific claims. They look into them and find no support for these claims. If you want to think that a large number election officials and judges of both parties and in many states colluded in the 2020 election . . . well, you can think that if you want, but you might want to think twice given the low quality of the evidence that’s been offered to support this claim.

      • Democrats and Republicans work together all the time. It is even admired that large numbers of them worked together behind the scenes on the 2020 election against Trump.

        “There was a conspiracy unfolding behind the scenes, one that both curtailed the protests and coordinated the resistance from CEOs. Both surprises were the result of an informal alliance between left-wing activists and business titans.”

        https://time.com/5936036/secret-2020-election-campaign/

        • Anon:

          This offers zero evidence of election fraud. Working together in a public relations effort is different from working together in a criminal conspiracy. Again, all things are possible but nobody has offered any evidence of fraud or irregularity in the election. That’s the point of the article by Eggers, Garrow, and Grimmer: they look at several of the more publicized of such claims and find that there was nothing there. Pointing to evidence of people working together in campaigning or post-election plans does not represent any evidence at all of an election fraud conspiracy.

        • When I use a word,” Humpty Dumpty said, in rather a scornful tone, “it means just what I choose it to mean—neither more nor less.” “The question is,” said Alice, “whether you can make words mean so many different things.” “The question is,” said Humpty Dumpty, “which is to be master—that’s all.”

        • Not election fraud, it shows that large numbers of democrats and republicans work together behind the scenes to achieve a goal. That is not an outlandish idea, in fact it is a common occurance.

          Of course the statistical arguments mean nothing because the premise all elections are samples from the same population is flawed. You can prove anything from flawed assumptions.

        • “Not election fraud, it shows that large numbers of democrats and republicans work together behind the scenes to achieve a goal. That is not an outlandish idea, in fact it is a common occurance.”

          Huh. You started this off by asking:

          “Why does the UN send election observers around the world to ensure fair elections?”

          To try to prevent or, if that is not possible, expose election fraud. Not to expose the well-known fact that sometimes people of differing political beliefs.

          If you’ve not been talking about election fraud, the subject of this entire post, what is your point? Yes, Democrats worked with a small slice of Republicans to defeat Trump, common knowledge as there was no attempt at secrecy. The Lincoln Project had some entertaining ads, after all, and Mitt Romney had voted to convict Trump in the first impeachment trial and very publicly said he would not vote for him in the 2020 election.

          In the context of a discussion about election fraud, so what?

        • The point is that a vast shadowy conspiracy of large numbers of Republicans and Democrats to influence the 2020 US election results is not only plausible, it is admitted to have occurred.

          It is also that such conspiracies to influence election results are not rare, rather they are standard across many countries throughout history. Often using both legal and illegal means.

          Many people consider it the default assumption that an election was not fair unless proven otherwise. Show them that adequete measure were taken to ensure fairness, the burden of proof is on those running the election.

        • The Frosted Flakes advertisements I watched on TV as a child are part of a vast shadowing conspiracy of wealthy Democrats and Republicans to influence what I eat for breakfast. These are superficial parallels in particular keywords, I struggle to see your point about fraud. The fact that large numbers of people can work in groups to accomplish a goal supposed to suggest that large numbers of people worked in groups to do this one thing in particular?

        • There is not much of a “conspiracy unfolding behind the scenes”, and “extraordinary shadow effort” to get you eating frosted flakes. The effort is very in your face, and for good reason. Behind the scenes shadowy conspiracies are generally frowned upon, because why not be open and transparent if you weren’t up to no good?

          However, it has been claimed that the sugar industry is behind some shoddy science that has been used to push sugary cereals over sausage and eggs for breakfast. So even there we find some possible conspiring.

          https://www.nytimes.com/2016/09/13/well/eat/how-the-sugar-industry-shifted-blame-to-fat.html

        • “The point is that a vast shadowy conspiracy of large numbers of Republicans and Democrats to influence the 2020 US election results is not only plausible, it is admitted to have occurred.”

          I actually took the time to read your linked Time article, and it doesn’t say that at all.

          Read it again and quit lying about what it said.

        • > There is not much of a “conspiracy unfolding behind the scenes”, and “extraordinary shadow effort” to get you eating frosted flakes. The effort is very in your face, and for good reason. Behind the scenes shadowy conspiracies are generally frowned upon, because why not be open and transparent if you weren’t up to no good?

          The subject of the article you link is also out there in the open. That’s why it’s in time magazine—as you put it yourself, it’s been admitted to.

        • They call it a conspiracy and extraordinary shadow effort in the article. Are you saying the article does not say that?

        • You can call it whatever you want, words are just symbols, man. But you just said the distinction of concern between Tony the Tiger and your linked piece is that one is secretive and under wraps, while the other is not. The fact is that neither is secretive or under wraps. So-called conspirators gave quotes for the article. Is reporting on yourself in Times magazine secretive?

          > why not be open and transparent if you weren’t up to no good?

          The subjects of the Times article are open and transparent. Many of the actions taken in that article are joint public statements!

        • Apparently that means the conspiracy is over now, not that there was none. It is pretty simple to understand the article.

        • “They call it a conspiracy and extraordinary shadow effort in the article. Are you saying the article does not say that?”

          Nope. I am pointing out that the article doesn’t say the effort was to INFLUENCE the election results.

        • “an extraordinary shadow effort dedicated not to winning the vote but to ensuring it would be free and fair, credible and uncorrupted”

          Rather the opposite of trying to influence the election result, as you claim.

        • It was to counteract trump and his destructive lies. You don’t think that is influencing election results?

          What would be the point if there was no influence?

        • Enough on this particular thread! There were many high-profile accusations of fraud, pushed by Trump, Cruz, etc. As Grimmer and his colleagues demonstrated, these accusations fell apart under scrutiny. Unfortunately the Hoover commentators couldn’t handle this and so kept changing the subject. Which is kinda what is happening here.

        • Andrew,

          I am more worried about the inability to acknowledge a conspiracy even when the conspirators acknowledge they conspired in public.

          That gives secretive methods an advantage over open and transparent sharing of information.

        • One paragraph further:
          “The handshake between business and labor was just one component of a vast, cross-partisan campaign to protect the election–an extraordinary shadow effort dedicated not to winning the vote but to ensuring it would be free and fair, credible and uncorrupted.”
          TIME did not call this a conspiracy ‘against Trump’. The way you implied that it did, in your comment, is arguably intellectual dishonesty.

        • There should not have to be extraordinary measures every election to ‘protect’ the election, because our system has many mechanisms to ensure that elections are free and fair – registration, poll watchers, election judges from opposing parties, provisions for recounts and/or re-canvassing. There should definitely never be a President who repeatedly and unjustifiably undermines public confidence in the fairness of an election with numerous lies. But there was, and people who support continued democracy in the US felt a need for countering those destructive messages.

        • Yes, I agree it was special measures to stop trump and his destructive lies. That is how I accurately summarized it, which you called intellectual dishonesty.

        • Anonymous, you said, “Democrats and Republicans work together all the time. It is even admired that large numbers of them worked together behind the scenes on the 2020 election against Trump”, then cited a TIME magazine article.
          I believe that a large majority of people reading that would take “against Trump” as meaning “against Trump’s re-election”, as I did. That is an empirical claim; you’re welcome to obtain a random sample of American adults and try it out. And the article did not say that. That is why I suggested intellectual dishonesty. If what you meant was “against Trump’s efforts to foment distrust of the conduct of the election”, then I would only say that you were very unclear.

    • Anonymous –

      > Why is it considered crazy to think people work together behind the scenes to achieve a shared agenda?

      That’s a pretty slanted way to think about it. Context matters.

      The question is whether it’s plausible to think that a very large number of people, many of whom take elections very seriously, some % of whom are lifelong Republicans,, would conspire together to violate the law, undermine our election process, to rob Republican voters on a massive scale, and risk serious legal ramifications if just one member of that group stepped forward to offer real proof that the the theft occurred.

      > That is normal human behavior.

      No, actually in context it isn’t “normal human behavior.”

  5. It’s a side issue for sure, but I always find the use of the term “surprising/unsurprising” in this kind of argument strange. By its nature, the term “surprising” is about the relation between facts and someone’s expectations, as such always relative to the expectations. Trump expected to win (at least that’s what he said), so he is surprised that he lost. This doesn’t mean much, but neither does it mean much that Grimmer is not surprised. It’s not an imminent feature of the facts to be surprising or not, surprise is always relative to the surprised person.

    Of course I’m not criticising the analyses, and “not anomalous” is much better wording. Whenever I see statements of the kind that certain facts are in themselves surprising or not, I think that somebody tries to give something essentially subjective an objective status, which looks suspect to me. (Fair enough, I’d probably concede that there is a subjective element in calling something an “anomaly”, too, but still… does that undermine my point?)

    • Christian –

      > It’s a side issue for sure, but I always find the use of the term “surprising/unsurprising” in this kind of argument strange

      As a funny twist, I agree that “surprising” is largely a rhetoric term – as expectations are necessarily subjective – but on the other hand, I don’t think it’s “strange” that people are rhetorical about this. Of course, you did say that *you* find it strange (subjectively) – but do you really? I think it’s entirely predictable that people will frame this rhetorically, for obvious reasons – I think it’s more the norm than an exception or somehow perplexing.

    • Christian—this is a good point. In the paper one of our key points is that to classify something as “anomalous” you need a null distribution or other formal measure of how unlikely the result is. We use “surprising” to be colloquial, but if we’re introducing confusion we can definitely switch to anomalous throughout.

      • On that subject, there’s a recurring family of argument I’m seeing here which is some test statistic differs substantially from one year to the next. Most famously, there’s the economist who tested against the null hypothesis that Joe Biden 2020 is actually Clinton 2016, but there’s also the signature thing here. Rather than explaining why signature verification is easier this year or whatever, I’d venture to say that this entire line of reasoning that expects stationarity in all properties of the election is nonsense to begin with. You want deviations from the null hypothesis that elections are conducted fairly, not the null hypothesis that nothing changes in 4 years.

  6. After a decade of reading your blog and books I’m so far out there that more junior colleagues already shout out “Bayes” before I even start to give methodological advice. Talk about being out of it. Thanks Andrew and colleagues!

  7. I didn’t see where the Hoover article covered the discussions I have seen significant reductions in various places in the % of mail in ballots rejected (compared to prior and sometimes very recent elections) as invalid. From what I have read, most of the rejections have typically been for failure to sign the mail in ballot.
    I personally found it very comforting that our electorate improved its ballot signing performance in such a dramatic way… if indeed this was as widespread as some have suggested. I haven’t seen this addressed elsewhere, and maybe I missed it in the Hoover piece.

    • Great question! The state where this comes up is Georgia. Gabriel Sterling , a republican election official, explains this as a result of a law change in GA that gives citizens a chance to fix their signature if the election workers find a problem. Here is the relevant quote from this NY times article: https://www.nytimes.com/2021/01/04/us/politics/trump-georgia-election-fraud.html?smid=tw-nytpolitics&smtyp=cur

      “The decrease in rejections is attributable to a recently passed law that gives Georgians a chance to correct problems, such as a rejected signature, with their ballots. Both parties had teams roaming the state and contacting voters whose ballots were at risk of rejection, but Mr. Sterling said the Democrats were simply more prepared for the task.”

      I hope that helps!

      • Good points from you and Justin.
        My recollection —perhaps incorrect— was that I had seen the info for Wayne County MI and parts of PA. I did not undertake or check the analyses, but noted them b/c they were provided by a very sharp data analyst I follow for his other quantattive work.
        My question would be, however, if the % of mail in voters increases significantly, wouldn’t one expect an increase in faulty ballots due to lack of familiarity? Put differently, why would we have expected the prior election mail in voters to have been less likely to send in valid ballots? What’s the hypothesis here ??
        And don’t get me wrong here… I am only remarking on the one item I did see in discussions that got me thinking since it struck me as a laughably easy way to steal votes if that was what folks wanted to do in some places. That and a friend had said to me before the election “watch for big changes in % of rejected ballots”. And then a couple of data points popped up on cue.
        All this would —I suppose— be open to empirical testing.

        • Thanks for the follow up. I’d love to see the analysis that made the original claim. One thing that went along with the increase in mail in voting were changes in laws that made it possible for people to fix ballots that would be potentially rejected. This made it possible for the parties to connect voters to potentially rejected ballots and have them corrected. So on the one hand, it does seem intuitive that an increase in mail in ballots might increase signature issues. But on the other hand, the ability to fix signatures suggests that it should decrease substantially.

        • MJM-WA –

          > That and a friend had said to me before the election “watch for big changes in % of rejected ballots”.

          I wonder if your friend had a theory as to why there would be a big change in the % of rejected ballots?

          >… since it struck me as a laughably easy way to steal votes if that was what folks wanted to do in some places.

          My understanding is that there’s (1) no significant evidence of fraud w/r/t changes in the % of rejected ballots, (2) there are procedures to protect against such fraud, and (2) there isn’t evidence those procedures were circumvented on a significant scale.

          So maybe it isn’t so “laughably easy?”

        • “if the % of mail in voters increases significantly, wouldn’t one expect an increase in faulty ballots due to lack of familiarity?”

          Presumably you mean in increase in the percentage of faulty ballots?

          I’m not sure we should have a prior on that issue bcz this election is so anomalous. There are factors that could push the percentage either way. You suggest lack of familiarity of ballots would increase the percentage of faulty ballots, but it’s also reasonable that several other factors about this election could *decrease* the percentage of faulty ballots:

          1) the stakes in the election were very high, so people wanted to make sure their vote counted;
          2) lots of people are leading much less stressful and hurried lives because they’re not commuting as much or not working at all, and many other forms of entertainment are off limits.
          3) Ballots may have come out earlier
          4) there was a substantial effort to educate people on how to use the ballots

        • 5) Household effect. I voted absentee for years, because I worked and remained out of the state during the week. My wife voted in person until 2020. She had no trouble completing the ballot, but if she had, I could have assisted.

        • Rejecting absentee ballots would actually be a very inefficient way to steal votes. From what I have read, most states’ absentee/mail-in ballots are similar to those in my state of Connecticut. Inside the mailing envelope is a second envelope, which must be signed, and perhaps filled in with other validating information. It is the lack of signature, or information, or questions about either, that are the most frequent causes of rejection. And that would occur before any election official viewed the ballot. So excessive rejections would have tended to skew the vote only to the extent that one party’s dominance in the mail-in vote was greater than in the in-person vote in the same jurisdiction. But the argument about 2020 is apparently that too few ballots were rejected, since the expectation was that Democrats were much more likely to vote by mail. That adds in a ceiling effect; you can’t accept more than 100% of the ballots, as against a historical acceptance rate in the high 90%s. And if you accepted all ballots, you would be accepting some votes against your preference, offsetting the gain by the preferred, dominant party.
          If there were large differences in ballot-rejection rates across precincts within a state, that might raise suspicion. I have seen no data on this.

        • I live in PA, am registered dem, and voted by mail for the first time last November.

          There was a concerted effort on the part of local Democratic Party officials to educate voters on how to vote by mail. I was told our process is mildly confusing. I found it straightforward but apparently not everyone did.

          In any event, there was lots of help available and lots of publicity about that help. This hasn’t been true in past elections. So maybe that resulted in a lower percentage of rejected mail-in ballots in PA?

          So I wouldn’t find a lower rejection rate surprising^H^H^H^H^H^H^H^H^H^H anomalous.

        • “So maybe that resulted in a lower percentage of rejected mail-in ballots in PA?”

          And your state also put a lot of effort into educating county-level election officials on how to treat faulty ballots, which by law can be submitted as provisional ballots (not rejected) and “cured” by the voter later.

          Despite what MJM-WA is arguing, the substantive argument by Republican officials (the real kind, not the Rudy Giuliani kind) has not been that the decline in rejected ballots is due to fraud, but rather than the “curing” process favored democrats because Democratic counties were more aggressive in implementation. Actually it wasn’t just Democratic counties that aggressively tried to help voters “cure” their ballots, but even if they did, it was totally legal and if Republican chose not to do so that’s their problem.

          So along with voter education attention was paid to trying to “cure” as high a percentage of mail-in votes as possible, entirely legal.

          The only argument against trying to help as many people as possible have their vote counted is that for some reason vote suppression is a good thing …

        • I checked an article in the Philadelphia Inquirer. In Pennsylvania the voter signs the secrecy envelope; problems with signature matches, missing signatures, etc. are cured before the envelope is opened; so even in Democratic-leaning counties, election officials could not favor a particular party or candidate in deciding whether a ballot should or could be cured. Only in the case of a ‘naked’ ballot (no security envelope) could an election official be able to know the voter’s preference. So even if Democratic-leaning counties were more amenable to allowing ballot curing, the efficiency of that as a vote-skewing mechanism would be attenuated because some number of Republican voters’ ballots would be cured with Democrats’. The differential would depend on party differences in vote-by-mail preference, and in sloppiness.

  8. People who are surprised or scared the Hoover comments need to adjust their priors on human irrational beliefs

    Few if any people act on the basis of reason alone. Every human being holds irrational beliefs, many wildly irrational. Frequently they are obviously wrong, and neither intelligence, education or social status are a defense against believing in nonsense. One person I went to grad school with, specializing in geoarchaeology, believed we should return to a hunter-gatherer culture!! This person is now faculty at a state university, and a fairly typical professor.

    On the one hand it’s good to confront these things; on the other, having been involved in the creation debate on the geology side in the past, I don’t know if it does much good. These things seem to wax and wane on their own and I suspect that as the election fades and emotions fade, the motivation to believe in the election fraud claims will fade too.

    • One famous statistics academic actually believed that the unprecedented number of votes for the dullest candidate ever fielded (far more than the messiah Obama managed to get), in light of the ridiculously thin parking lot rallies, was actually due to silent majority. That people were afraid to show that they were democrats.

      Anything goes for prior, even for Bayesians, as long as it gets to a comfortable result!

      • Chebyshev:

        What you’re doing is similar to what the Hoover commenters were doing, which is to respond rudely without evidence and without acknowledgment that the arguments that were presented all fell apart under examination. This is a form of argumentation in which you start with the conclusion (in this case, that Biden didn’t win the election despite the evidence of the 81 million tabulated votes) and then spew out an endless series of arguments in favor of that conclusion, and when each argument is shot down just come up with another argument. OK, so the dead-voters thing didn’t fly? Try parking-lot rallies. Etc. I do not think this is a useful way to get closer to the truth. To me it seems like a mix of rhetorical strategy and a way of avoiding facing up to unpleasant facts. If you think Trump would be a better president than Biden, fine. 81 million voters can indeed be wrong: they’re just a majority of the people who voted, and majorities can be wrong all the time. But I think we’d all be better off if you’d just make that argument directly rather than grasping at straws and trying to deny what happened during the election campaign.

        • There are “arguments”, there are “rhetorical strategies” … and then there are rhetorical strategies to incite riot and civil-war. The famous prisoners kept cozily in Landsburg in 1923 learned a thing or two from their failure. And they had their arguments published!

        • Andrew:
          What you said holds true for both sides. Both sides have started with a conclusion and proceeded from there, fortifying positions with selective hunting of evidence. Did you really consider all evidence without bias or did you proceed from the assumption that there is no election fraud and start looking for evidence to “support” your truth?

        • Chebyshev:

          What I’d like is for the Hoover commenters to say something like: “Ok, none of the evidence for election fraud holds up. Biden received 81 million votes and nobody’s provided any good evidence to the contrary. But we’re unhappy that Biden won, and we’d like to believe that Trump actually got more votes, even though no serious evidence has been presented to support that claim. For now we’ll have to accept that Biden is the legitimate winner, just as we accepted that Hillary Clinton won more votes than Trump in 2016 (but lost the electoral college) and just as Democrats accept that George W. Bush, who they despised, won more votes than Kerry in 2004. But it’s still hard for us to believe that Bidan was preferred by the majority of voters in 2020, so we remain open to evidence of massive fraud, if such evidence turns up.”

          But they’re not saying this. Instead, they’re just either insisting there’s fraud in spite of the lack of good evidence, or like you they keep talking about how they can’t personally believe that Biden won based on uncalibrated statements about parking-lot rallies or dullness of the candidate or whatever, without accepting that the specific arguments of fraud fell apart under examination.

          In contrast, I think that Eggers, Garro, and Grimmer are open to new evidence if it is presented. They state clearly that they are evaluating the claims that they’ve seen so far. Also I think the saying by Daniel Davies is appropriate, that good ideas do not need lots of lies told about them in order to gain public acceptance.

        • Andrew:

          You expect to much—it’s the comments section of a political site. That’s the most argumentative and extreme sample from the population. To quote Obi Wan, “You will never find a more wretched hive of scum and villany.”

        • “Did you really consider all evidence without bias or did you proceed from the assumption that there is no election fraud and start looking for evidence to “support” your truth?”

          First of all, the assumption isn’t that there is NO election fraud, simply that the rate isn’t high enough to impact elections unless they are very, very close and even in those cases, there’s not evidence in our modern elections that fraud favors one side or another in any consistent manner.

          In fact, there were a few election fraud indictments related to the last election. Of Republicans. Of course, it wasn’t on a scale to tip the election to Trump.

          But even in that weaker form, it is not an ASSUMPTION that there’s not enough fraud to massively swing election results. There is substantial EVIDENCE that in recent decades there has not been substantive election fraud. Remember that Trump claimed that in 2016 he actually won in California, and that he actually won the national public vote as well. He appointed a federal commission, led by a prominent proponent of massive election fraud favoring Democrats, to great fanfare and publicity. A year or so later that publicized effort, despite its partisanship and their stated goal of uncovering fraud they assumed exists, led to a conclusion that election fraud is almost non-existent.

          Just like every other study done on the subject in recent decades.

          You are the one making assumptions with zero credible evidence.

        • > Both sides have started with a conclusion and proceeded from there,

          Did you start with this conclusion and hunt for evidence of it selectively? No, you’ve done the the only truly honest thing, which is abstain from evidence at all and make strong claims anyways.

          More seriously, it’s not stated in this post or in the linked paper that “no election fraud at all took place.” Certainly, election fraud in the sense of >= 1 illegitimate vote being counted did take place, in this election as it did in all the others. This post is about specific pieces of evidence in favor of a specific theory, the theory that election fraud took place on a sufficient scale as to change the result of the presidential election. The linked paper claims that they don’t hold up under investigation and this paper concurs. Do you have a rebuttal to those investigations or a specific piece of counter-evidence?

          On the subject of your turn of phrase “a prior of millions of silent voters,” this is your argument stated correctly:
          “In pre-election surveys, it was projected that x many people would vote for Biden based on voter intentions in pre-election surveys. Y >> x actually did vote, which implies either voter fraud or millions of silent Biden voters. Since you’re still ruling out voter fraud, your prior must have necessarily excluded voter fraud from its domain of positive support to begin with, which is begging the question.” Remember that a prior is what’s unconditional on the observed data—the silent voters is a conclusion necessarily drawn AFTER observing data. If Gelman had a prior on silent Biden voters, then it would have come through in the analysis before the election by forecasting a larger number of Biden voters to begin. If that was the prior, the forecasts wouldn’t have been off. Don’t just throw in words you don’t understand like “prior” or “Bayesian” to buttress your argument with the veneer of mathematical sophistication, you’ll just embarrass yourself.

          The problem with the line of argumentation above is obvious when stated correctly. There are lots of reasons a forecast based on pre-election surveys can fail systematically as it did for turnout in this election, many of which have been blogged about here. You can read them if you want.

          As for my own opinion on the stolen election, the evidence against fraud is that the system itself is constructed for resilience against frauds. The system is decentralized, run independently by the states, many of the states in question are run by Republicans, there are registration, signature, envelope and ballot verifications, election officials have testified, there are major consequences if found guilty of such frauds to deter would-be fraudsters, and so fraud on such a scale is difficult. So yes, a default starting position of believing the election results is perfectly reasonable—to get to that is the point of all the work that’s been done constructing our election systems. There needs to be some evidence of fraud to overcome this initial preponderance of evidence, which has yet to materialize.

          The big steal either happened or didn’t, American citizens have to come to terms with one or the other. Your pet theories about people’s thought processes don’t actually have any bearing one way or the other; do you have something relevant to add onto a topic under discussion, any evidence or defense of a piece of evidence rejected here?

      • Cheby –

        > One famous statistics academic actually believed that the unprecedented number of votes for the dullest candidate ever fielded

        You are assuming, without evidence, that being a “dull” (completely subjective determination, but I’ll take it) candidate is a negative feature for a particular campaign against a particular candidate.

        Biden may have been, in some ways, a candidate uniquely suited to defeat Trump. By being relatively middle-of-the road, he was able to sustain votes from both the progressive and moderate wings of the Democratic Party, as well as independents and lifelong Republicans of the “never Trump” variety. Almost all the other candidates who were running for the Dem nomination would likely have not been able to unify the party and moderates and independents to the extent that he did. Also, it’s entirely possible that many people voted for Biden because after 4 years of non-stop drama, Biden represented a likely return to a less anxiety-provoking norm.

        When making assessments such as these, you should at least attempt to control for your own subjectivity. From all appearances, you have neglected to do that.

        • What’s even more remarkable is that you go from an entirely subjective conclusion (that Biden couldn’t have beaten Trump because he’s “dull”) to making a conclusion that an empirical analysis that contradicts your conclusion must not be correct.

          You don’t present an analytical argument against the empirical analysis, because you deem that your subjectively-derived opinion about the effect of Biden’s “dullness” obviates any need for you to present an analytical argument.

          I have to say, that’s really, really, pathetic. You should take that kind of stuff back to other blogs. Such commentary is really beneath Andrew’s.

        • I am just questioning the prior assumption here that’s all.
          In standard Bayesian texts (e.g. BDA3, the best of them!), there is a healthy discussion on choosing (carefully) the priors.

          Here is an example of a ridiculous prior. Nothing around us justifies a prior that there are millions of silent democrat voters. Nothing we experience justifies. Isn’t orange-man-bad moralism quite cool to wear openly? That is very agreeable and career enhancing.

          Perhaps, the elections were fair. May be there is a tiny probability they were not. I am not too fussed either way. What i am curious about is the way the market place for ideas have been functioning.

          PS
          Is Nate Silver intellectually dishonest? How about Neil Ferguson?

        • Chebyshev:

          We have a lot of prior information based on past elections. This prior information goes into our “fundamentals” model, which predicted that Biden would get approximately 54% of the two-party vote. In the event, he got 52%, so the model wasn’t perfect, but of course we wouldn’t expect it to be perfect either. You refer to “silent” voters. In 2020 as in previous elections, most voters don’t go to rallies etc. That is, most voters are “silent” in that sense. So, yes, lots of experience justifies a prior that there are tens of millions of silent Democratic voters and tens of millions of silent Republican voters, and that the Democratic candidate was favored to win.

          Regarding claims of election fairness: again, there were many specific claims, and they fell apart upon examination. I think the Eggers, Garro, and Grimmer paper is a good example of the marketplace of ideas working well. I think actions such as those of Ted Cruz supporting ridiculous claims are destructive of the marketplace of ideas.

        • plus, using rallies as a proxy for votes in 2020 is particularly absurd…only one of the candidate was crazy enough to encourage mass campaign gatherings during a pandemic…
          you might as well say, “trump had so made more tweets during the campaign, there is no way biden won”

        • > “trump had so made more tweets during the campaign, there is no way biden won”

          Trump lied more. He must have won.

          He also played more golf. Appeared on Hannity more. The list is endless.

        • “Isn’t orange-man-bad moralism quite cool to wear openly?”

          A tragic shame it is, that the Napoleon of the age is not better recognized for his greatness; that he has had to spend all his ample energies in petty squabbles with the vermin that fly about pestering; and so has been diverted, tragically, in spite of his noble efforts to supervene these distractions. That he has not been allowed to carry on according to his destiny, that his noble work has been so shamefully interrupted is an ominous reminder: the work of the archfiend on this earth leads men astray!

        • > Here is an example of a ridiculous prior. Nothing around us justifies a prior that there are millions of silent democrat voters.

          You seemingly don’t know what a prior is

        • And also doesn’t seem to understand that “independent”, in the context of American politics, doesn’t mean “won’t vote for a Democrat”.

  9. I don’t think it’s lack of pushback that has propagated these ridiculous claims. There has been **alot** of pushback against the claims of election fraud, even by Republican officials in the states where these fraudulent activities supposedly occurred.

    IMO most of the fervent believe has been perpetuated by Trump himself. Whatever you like or dislike about Trump, you have to admit he’s an extraordinary individual. Very few people have the cohunes to keep making such blatantly false claims. The fact that he persisted in the face of all the evidence against it probably was a perverse kind of certification of truth for some people.

    • Jim:

      I’ve been thinking about this. I don’t know how many people were convinced by Trump’s statements. But I do think that Trump and other Republican leaders making these claims gave people permission to believe them also.

      • “I do think that Trump and other Republican leaders making these claims gave people permission to believe them also.”

        Yes, when other people like Cruz chimed in I’m sure that seemed also like certification to a lot of people.

      • As Marjorie Taylor Green eloquently put it “[they] were allowed to believe in things that weren’t true.”

        I also agree that Jim’s point is important.

        Scenario A. A major political party engaged in a massive election fraud (with tens of thousands of illegal and/or miscounted votes) that changed who the next president of the United States will be.

        Sure, that seems implausible. But what is the alternative?

        Scenario B. The president of the United States repeatedly spread lies and misinformation about non-existent election fraud every day for months with most of his party backing him up.

        If you posed this question ten years ago, would people necessarily think scenario B is more plausible?
        I don’t think so.

        Of course, some of us “know” that the president happens to have a history of saying easily falsifiable things and have been updating our beliefs accordingly.
        But for some Trump voters, the cognitive dissonance must be unbearable when your choice is between accepting an implausible conspiracy theory or admitting that you supported a politician with minimal interest in truth and morals.
        It might even be “rational” for them to believe the election fraud narrative (not that many of them are making this decision consciously).

        • I think you’re leaving something out. I’ll add in.

          > Scenario B. The president of the United States [who constantly expressed hatred towards people who support another party, people that over a period of many years you increasingly blamed for practically all problems in our country, and who you think are morally inferior and ethically broken] repeatedly spread lies and misinformation about non-existent election fraud every day for months with most of his party backing him up.

          Trump and his PR team aren’t singularly responsible for what has set this up.

          I think that there is probably a non-trivial number of people who don’t necessarily believe (or even particularly evaluate) Trump’s scenario of fraud, but who have been primed for years to believe that Democrats are fully committed to stealing elections. Voter fraud, illegal immigrants voting, illegal immigrants taking away jobs, voter ID, busing in voters, dead people voting, liberal elites, etc, have been in the background and growing for quite a while as a threat to the status and power of Republicans as a minority of the electorate. Trump exploited what was already there in addition to creating his own momentum towards an acceptance of the stolen election conspiracy theory.

        • “But for some Trump voters, the cognitive dissonance must be unbearable when your choice is between accepting an implausible conspiracy theory or admitting that you supported a politician with minimal interest in truth and morals.”

          No. It is hard to face, it offends Americans because they make so much of “fair play” and so on; but the popularity of this creature can be attributed precisely to this “minimal interest in truth and morals”. Precisely to this, which was *never* a secret to anyone. The attraction of the mob to the tens of millions raised on the sopranos and for whom the mild hypocrisies of “fair play” and “the golden rule” no longer govern. The prospect, in 2016, of running with an astonishingly talented demagogue and thug was precisely the attraction. The tends of millions tired of taking guff from their employers, their betters, their wives … just like the little crook-necked book-keeper that used to get kicked around in high-school, but now does the books for the crooked “boss” who holds court behind the grocery on Mayfield and 125th … they didn’t expect the opportunity would ever come round, but it *did* … And. They. Took. It.

    • That’s right. Trump just took the exploitation of the ways ‘real people’ think further than most other politicians in the US in recent times have had the balls or the inclination to.

      In politics, however, these explanations cannot be the whole story. At the heart of the lying-politician paradox is an uncomfortable fact: voters appear to support liars more than they believe them. Mr Trump’s approval rating is 11 points higher than the share of people who trust him to tell the truth. A third of British voters view Mr Johnson favourably but only a fifth think he is honest. Voters believe in their leaders even if they do not believe them. Why?

      The answer starts with the primacy of intuitive decision-making. […]

      • People continue using Twitter even though their political hero was censored and kicked off. They also keep using robinhood even though the service limited them to owning one share total of a stock which prevented them from making money off an opportunity of a lifetime short squeeze.

        People often don’t act in their own best interests because they are using some heuristic to make decisions instead. And there is nothing wrong with that, as long as you can recognize the pitfalls.

      • The real tragedy here is the people who believe that politicians are honest.

        All people are liars – this is a fundamental characteristic of human nature. And politicians have a long record of lying, so to believe they are “honest” is naïve. But there’s another characteristic of human nature that provides a check on lying: most people **hate** being caught lying. Trump is only unusual in that he doesn’t care about being caught, which means his lies can be extreme and outrageous. Most other people try to hedge their lies enough that they have an escape route or some way to “plausify” their lies, and this acts as a brake on the degree of lying.

        I don’t believe anything any politician says. But I have a very good idea of what they’ll do based on the track record and range of views in their party.

        • There are people reared on “the sopranos” and “breaking bad” who *admire* him for his brazenness; for his lies. They make excuses or apologize for their own discomfort with lying; but the admire the ones who haven’t any scruples at all. Gangsters — after all — take no guff. Not even from their wives.

  10. Interesting comments again, and I can check something to see if I can find the info on the rejection rates. But you can also Google the NY primary election —which was held late in the election cycle—and see where I seem to recollect that ~80K of the mail in ballots were disqualified (20% f the vote??). There is a standard of comparison from several months before the general election.

    Comments noted on the changes in rules for curing the ballots, but doesn’t the ability to cure provide additional degrees of freedom to ballot counters intent on stealing votes? While I would suspect the rules vary by state, in many states you not only have to sign (yes, some voters dont) but the signature must match the signature of record which is often the drivers license. So a vote counter can first selectively interpret what matches the signature of record. (Failure to match can easily arise when somebody scrawls their signature as opposed to writing it out carefully). This is why I was thinking one should expect more invalid ballots in general w/ larger #s of mail in votes.
    In some instances, the failure to match is the end of the process and the vote is disqualified. So the counter has a degree of freedom in determining which signatures on ballots he or she considers matching. But if curing is standard or required, then send the ballot out for curing that the vote counter wants to eliminate and hope the voter doesn’t cure it on time.

    • You don’t sign the ballot, you sign the sealed envelope containing the ballot – the US votes by secret ballot. The signature-verifiers don’t see the vote, and the vote counters don’t see the signature. So the total vote will only be skewed to the extent that the parties’ differential in mail-in ballots differs from the differential in in-person ballots, or one party’s voters are sloppier than the others in completing their ballots.

    • “So the counter has a degree of freedom in determining which signatures on ballots he or she considers matching.”

      You seem to be forgetting that both parties participate and that if there was blatant fraudulent activity of this kind, the opposing party representative would catch it if they were doing their job.

      You’re really deep in the weeds … perhaps it would help if you actually spent more time reading about how the election process works rather than reading unsubstantiated claims of fraud based in part on the assumption that the reader won’t know how the election process works?

  11. We know there was no widespread fraud in the 2020 Presidential Election.

    How do we know this to be true?

    We assume that such fraud would be obvious to most serious observers and evidence of such fraud could be readily collectible.
    How valid is that assumption in the highly complex system of differing nationwide balloting, ballot validation, counting, reporting, review, and formal approval processes?

    Do we accept that there was likely some minor fraud in the vast election processes, or would that not occur without some evidence surfacing to our attention?

    • Alcar:

      All things are possible. Eggers, Garro, and Grimmer point to specific claims of fraud and explain why these claims don’t hold up. I don’t see how it makes sense to attempt to nullify an election based on the vague statement that there might have been some fraud about which no evidence has been brought to light. I do think it would be good to have election reforms that make it easier to validate election results, as well as reforms that make it more convenient to vote. Unfortunately, the reaction of Trump, Cruz, etc., after the 2020 elections, along with reactions such as those of the Hoover commenters, suggest to me that whatever reforms are done they will be out there crying foul about any election results that disappoint them.

    • alcar –

      > We assume that such fraud would be obvious to most serious observers and evidence of such fraud could be readily collectible.

      You connect those two claims, but they are separable.

      (1) We assume that such fraud would be obvious to most serious observers

      I don’t know who is making that assumption. Could you elaborate?

      (2) evidence of such fraud could be readily collectible.

      I don’t know about “readily” collectable. But the assumption for many, including myself, is that if there were widespread fraud, given the level of energy and resources devoted to detecting such fraud by the part of those who are heavily invested in doing so, there would have been at least some reasonable level of evidence presented.

      That hasn’t happened. I don’t think that it means that there’s zero possibility that widespread fraud occurred, but it does mean that the claims that widespread fraud occurred are highly implausible.

    • This reminds me of the claim by teacher’s unions in their battles against quantitative assessment: quantitative assessments are unfair because teachers’ secret sauce is so amazing we can’t even see it working!! That’s some kinda sauce!

      Man I’d love to start a fast food restaurant with food that so amazingly delicious it can’t even be detected!! :) Talk about high margins!

      • Without going too far off on a sidebar, I would say the more sophisticated version of the argument you’re mocking is basically true: many important things that good teachers do well are not what tests measure (various social worker-like and therapist-like functions, role model functions, etc.); and most of what we can measure will be completely swamped by factors that teachers can’t control, unless students have sufficiently stable lives outside of school and sufficient preparation from prior grades.

        • Yeah, I would agree with this.

          If the only benefit provided by teachers was to increase students’ test scores/college admission rates/etc, the quantitative assessments would be fine. But that is definitely not the case.

          Even beyond the more social-type functions, which are definitely important, the things I learned from teachers that have given me the most lasting benefits were not things measured on any test, nor are they things that are easily quantified. How do you quantitatively measure ability/willingness to question your starting assumptions? a sense that history is not just dull facts to be memorized and recited but directly relevant to today’s issues? etc.

        • anon –

          > I would say the more sophisticated version of the argument you’re mocking is basically true:

          But where’s the fun in mocking an accurate portrayal of someone’s position?

          When I want to mock an argument, I always distort it beyond recognition. Much more fun that way.

        • “many important things that good teachers do well are not what tests measure (various social worker-like and therapist-like functions, role model functions, etc.)”

          I think it would be pretty easy to put a pin in those claims. I think what many “successful” teachers do is pick students who are bright and help them, which shines a nice light on the teacher. This was patently obvious in my high school, where teachers practically stumbled over themselves to associate with the football players etc.

          The reality is that most of a student’s outcome is controlled in the home, not in the classroom, and has little or nothing to do with the teacher. Nonetheless, teachers claim to be responsible for it, and then have the audacity to claim their excellence is invisible.

        • Jim –

          > The reality is that most of a student’s outcome is controlled in the home, not in the classroom.. .

          Assuming that’s true (I think it’s an oversimplification), the point would still be that you look to see how teachers can influence marginal gains.

        • Joshua:

          Isn’t it true that teachers routinely claim that their excellence can’t be measured? Pretty Trumpian if you ask me.

          If it’s there, Joshua, it can be measured. If it’s not there, it can’t be measured.

        • I do not think measuring these contributions of teachers is per-se impossible: but it would require a *radically* different *form* of measurement/assessment. I do think it is impossible in practice, given the current education system and vast confounders.

          And I am not even sure there would be a net gain over a non-quantitative system.

          (But I don’t think all things that exist can be measured: what about quality of art, for example? Sure, you can poll people, but IMO that’s not really answering the same question.)

        • Jim –

          I dunno. Feels like you have a bone to pick with teachers for some reason

          > Isn’t it true that teachers routinely claim that their excellence can’t be measured

          I don’t know where you’re getting these details about what teachers “claim.”

          Most teachers buy into standardized testing and other typical means to judge student achievement.

          Some minority feel those measures are only a limited window into students full development, and shouldn’t be the sole or even prioritized metric.

          Many teachers think that judging teachers’ skills by measuring students’ growth over time with standardized testing interjects a distorted or counterproductive incentive system.

          Some feel that much of their work with students is not captured by giving students standardized tests.

          As for this:

          > If it’s not there, it can’t be measured.

          There’s a common misperception that just because you put a number on something you’re measuring something useful. In standardized testing there’s the concept of reliability vs. validity. Reliability is like when you get the same score of you give a student the same test mutiple times. But validity means that you’re measuring what you intend to measure. Many times if you give a student a test, what you get is merely a measure of how that student did on that test at that particular time – and not much else of any use.

          How do you think you can measure the value of a teacher?

        • Yeah, the impression I get is that most teachers would merely say standardized tests are limited and overemphasized.

          I think there’s a good possibility that it’s actually much worse than that.

          Especially as now basically all information is available at one’s fingertips – how to think and evaluate information is now much more important than passive knowledge, but I don’t know how you test that in a classic standardized format.

        • (Although those issues only become really critical at a certain point. There’s a certain level of basic knowledge and skills everyone needs, eg basic literacy and math, so perhaps that complaint should be more limited.)

        • ” many important things that good teachers do well are not what tests measure (various social worker-like and therapist-like functions, role model functions, etc.)”

          You had me stumped on this for a bit, but now I see your argument for what it is: a romantic fantasy.

          The obvious reason “therapist/role model functions” don’t show up on quantitative assessments is that they’re rarely significant at the aggregate level. How many students can a teacher play therapist for in a given class year in a way that **distinguishes them above other teachers**? I know several teachers, one of whom actually has distinguished themselves with behavioral management with difficult kids. In that case, attendance and behavioral incidents fell significantly. Those are measurable results.

          Sure some teacher might do something wonderful for one or two kids a year, and that’s a great thing to do. But at the aggregate level it’s irrelevant, and helping one or two kids is great if your kid is one of those, but if not, well, you can see how other parents might not find that all so impressive.

        • This seems like it pretty much boils down to “it’s hard to meaningfully differentiate anyone other than rock stars and utter failures”, which has been true in… every organization I’ve ever worked for, public or private, in several different sectors.

      • You’re disparaging the teachers’ “secret sauce” while buying into the “secret sauce” of the people carrying out “quantitative assessments”. Don’t be so quick to reify the bureaucratic, bean-counting “assessments” used in the real world into something that actually measures improvement in children’s lives or their value to society.

        • ‘You’re disparaging the teachers’ “secret sauce” ‘

          Yes, I am, because secret sauce that you can smell see touch hear or feel or measure in any way doesn’t exist.

          ‘while buying into the “secret sauce” of the people carrying out “quantitative assessments”. ‘

          Nope. Not necessarily. But I am pointing out that the claim that a teacher’s benefits are so immense they cant be measured isn’t a plausible or logically sound counter argument to quantitative assessments.

        • I think the argument is that the wrong thing is being measured: test scores etc are not all that important for overall success in life.

          There are ways to measure *in principle* but probably not *in practice*: there are huge confounders (family situation, etc.), but even beyond that, you can’t really see the results of an education for a couple of decades, so the timescale is all wrong to *use* the measurement.

          And even then, success in life isn’t a simple scale: economic success? happiness? making the world a better place?

        • “test scores etc are not all that important for overall success in life.”

          That may or may not be so. However, I have not referred in any way to “test scores” as a measure of teacher performance, I referred to “quantitative assessment”, which may include test scores but may also include many other measurable features of teacher performance.

          “success in life isn’t a simple scale: economic success? happiness? making the world a better place?”

          If you don’t know what constitutes “success in life” then how can you then say “test scores etc are not all that important for overall success in life.”?

          Furthermore, I don’t agree with the implication of your statement: that the school system is there to provide “success in life”. My understanding is that it’s job is to provide an education because the ability to read, write, do calculations and understand the world we live in are important components of success. And while it’s probably true that social connections are more important than raw test scores (see George Bush), I think it’s safe to say that people who can’t read and write aren’t generally successful in life, and that test scores do generally distinguish these abilities.

        • >>If you don’t know what constitutes “success in life”

          I didn’t say “I don’t know”, I said “it is not a simple scale”.

          >>Furthermore, I don’t agree with the implication of your statement: that the school system is there to provide “success in life”. My understanding is that it’s job is to provide an education

          Of course the school system can’t provide everything needed for success in life.

          Certainly basic literacy and math skills, etc., are necessary and measurable. And if the issue was only measuring these things there would be no problem.

          But the problem is that “an education” includes much more than this – especially in an era where massive amounts of information are quickly available and the skills of how to evaluate information, question presuppositions, etc. are more critical than ever.

          And these skills are much harder to measure – I will not say impossible, but I do not think good measurements could in practice be arrived at within a bureaucratic system.

        • My test scores were so-so but those high-school teachers who used to put up with my preposterous questions gave me the confidence I needed … to continue my studies and get more so-so test scores in college. And as proof of that, I have a stack of material on relativity, which I’ve pulled off the back-shelf yet once again; maybe this time I’ll crack another piece of it? Who knows. Maybe I’m wasting my time (and everyone else’s) but those teachers surely are to get the credit (or the blame) for this persistent rivulet of enthusiasm and of effort.

    • Here’s another annoying rhetorical strategy emerging; make an unrelated, uncontroversially true statement that’s suggestive of one side or another. The discussion is on whether or not Trump actually won according to a legitimate vote, which was covered up by a massive conspiracy of fraud, so chime in by declaring that minor instances of fraud are certainly possible. I suppose the goal is to try and bait someone arguing against the big steal into disagreeing with your obviously true claim, thereby implying that the negation of their claim, that the big steal happened, as true.

      People are impressively clever, even when unaware of what, specifically, they’re doing.

  12. “It must feel weird for the regular Hoover writers to know that these are the sort of people who are reading their columns.”

    You say it as if it’s a bug, rather than a feature.

    There’s the old adage that says that if you encounter assholes everywhere you go, it’s probably because you’re the asshole. The corollary is likely true for Hoover/MarginalRevolution comments. If the comment section on every post you make is full of far-right lunatics, then…

Leave a Reply to Ken Schulz Cancel reply

Your email address will not be published. Required fields are marked *