Wednesday assorted links


5......"retired all of their elephants."
Hope that's followed by the retirement of the Republican circus.

#1. Given the affirmative action of the past decade or more in this field, should the lesser ratings be a surprise?

If you want to make a sexist implication it helps if you actually read the article so as not to look foolish.

"On the other hand, SET are significantly correlated with instructor gender (male students gave higher SET to male instructors, Figure 2) and with students’ expected grades. This adds evidence to the hypothesis that instead of promoting better teaching, SET contribute to grade inflation. We find no evidence that male teachers are more effective than female teachers. If anything, students of male instructors perform worse on the final exam"

marginal revolution troll bot says:
Something something affirmative action, something something liberals. Make America Great Again!

Nothing grates like mr troll bot.

make america grate again

This is like the Nth study that tells us we shouldn't used subjective student evaluations. We should use value-added. Now set to work finding a good measure for that.

I'm increasingly of the opinion that unless we can directly measure something, there will always be factors that confound. An example would be fruit: we buy fruit for a particular type of pleasure but we select it based on proxies like color or firmness, and the store charges is based on weight. Pleasure itself can only be determined after the transaction has been committed. But, by then, it would be too late.

I've yet to see anything that adequately tracks teacher efficacy. It may turn out to be unmeasurable.

Standardized tests make it easier to grasp at value-add in k-12. Who knows how much value was added in a fairly unique college class? The prof can give A's and done.

Of course, that *might* be the way it works now.

I believe MOOCs instrument themselves mainly for student completion, and they try to tweak little things to get better effect in that regard.

You can believe anything you want. MOOCs lead to exceptionally few students completing them.

OK, I admit to stirring the pot a bit, but this is a pretty crappy study. 'Perceived' teachers? Mixed results - sometimes the females rated males higher, sometimes not, is clear evidence of bias?

Also, "This might suggest that gender balance in a field affects gender stereotypes and might reduce bias against female instructors." supports my original trolling comment that artificially controlling the gender balance influences results.

The premise of the argument may be correct, but the study does little to prove it.

This research looks like a good start, if you read it these were online classes where the instructors sometimes listed correct sex and sometimes the other sex. The questionable part was that in the French case the men rated the male instructors higher and in the American case the women rated the men higher. To me this says that there might be issues with the samples of students and this research should be done over. In particular this might be a result that pertains to students of some backgrounds but not others and so we may not be able to generalize.

#4 Steve Sailer has a theory that it's better to have a GOP President because the press will do their job in an attempt to take down the GOP President.

Like the job being done now in taking down the GOP congress?
May be when we move away from " taking down " , there will be scope for a better functioning democracy.

And by better functioning democracy you mean what?

"And by better functioning democracy you mean what?"

I'm pretty sure he means a "democracy" that delivers the results he desires, regardless of any actual democratic process. The GOP congress is just as democratic as any before it, it's just not very left leaning.

I thought it was Ryan Avent.

Tomorrow it will be Thomas Sowell.

Not just the press, but congress and also a large portion of the american people will be invested in reducing executive power. On the other hand a Clinton presidency will be much more willing than Trump to take and use additional powers, and will also probably be more effective at it than Trump, with significantly less resistance from the press, congress and the people.

Forgot to add whichever one wins, Trump and Clinton will both likely be 1 term presidents and you have no idea who will come after. My bet is that we'll get a "good" democrat before we get a "good" republican which is another reason to vote for Trump.

I don't think your recommendation of a Trump vote fits your framework. If we are to have a 1-term president, it is much more likely that the successor president will be of the opposing party than deposing the incumbent via primary. If you want Paul Ryan or his ilk or whatever you consider a good Republican in 2020, you would be better off voting Hillary in 2016.

Actually is there even the slightest chance Trump wants to be a 2 term president? No doubt he wants to win this time, and get the accolades, etc. But I can't see him wanting to go through all of the effort and expense of running again when he's already made his point, and will be 74 years old, etc.

He won't like the job itself much either, too many constraints and too public, and he won't like that he can't just wave a wand and make things happen. I wouldn't be surprised if he tells his vice presidential nominee that he intends to step down in 2 years and let the VP have it. Would attract a better class of VP that way.

Regardless of what Trump wants today, it's likely in 4 years he would be quite likely to agree to another 4 years. How many Presidents have ever decided that they were done?

Trump is pretty different. And LBJ didn't run again when he could have, neither did Polk. Not to mention George Washington.

#7...Good stuff from.Beckworth, as expected.

3. I suppose the faculty debate over who gets the best parking spaces is worse than the debate over the meaning of neoliberalism. Why can't everybody be friends. Isn't that what Walter Lippman wanted? Of course, ecumenism in economics has been about as successful as ecumenism in Christianity, and for the same reason: my truth is truer than your truth. From what I read in today's papers about Trump University, neoliberalism doesn't come to mind; fraud does, but not neoliberalism. Phishing for Phools with a large net. With Republicans quickly coming home to support Trump, terms like "conservative" and "neoliberalism" have no meaning: they mean just what Trump's supporters choose them to mean. If Pinochet can be a model for freedom, why can't Trump.

Hellary will make America grate again!

3. I was aghast, aghast I tell 'ee, to see an Oxford chap reduced to using "critique" as a verb. Low stuff indeed.

As to the pendants who gleefully run around correcting spelling and grammar: let's do them before the lawyers.

Before the end of your days, dearieme, try to spend several not being this guy.

Dear dearieme,

Kindly ignore the reply from the other heathen...your commentary is consistently entertaining.


Won't it be ironic when one day elephants and Orcas are extinct. If only Ring Ling Bros, and Seaworld had prevailed the future could be one of opportunity for them, instead we relegate them to the commons to die out. Cows number at 1.4 billion, but has anyone ever seen an auroch from which they were bred? They died out in the 17th century.

“This was about appropriations law,” he said. “For those of us who work in financial management, when it comes to the Antideficiency Act, which has criminal penalties associated with it, we take it very seriously.”

if the NYT had opened comments, the most liked comment would have said something like: "The Republicans have shown that they are willing to abuse the powers given to Congress under the constitution in order to work to the detriment of the people - solely to disrupt the plans of the president and line the pockets of doctors and insurance companies. Given the circumstances, the actions of the White House lawyers are entirely justified."

3. I'm still not quite sure what "Neoliberalism" is or why it takes up so much mindspace in anti-economists. As far as I can tell, neoliberalism just means "the set of policies that the IMF and World Bank pushed in the late 90's". I don't see any other common thread through the separate policies of embracing free markets, austerity, etc.

#1. Wonder why TC thinks that such a weak piece of drivel merits note? US study was of 4 student sections, so I presume there were 4 instructors. This kind of blogging borders on the irresponsible. Was the French study published in a peer reviewed journal? I'm finally beginning to realize that TC is more of a gadfly than a serious student of the world.

I think you should put a much lower bar for "assorted links' than a blog post by TC. The "assorted links" is often the best part of the blog , covering a wide range of matters. How much effort does it take a reader to ignore something that doesn't meet her or his standards?

Number of data points?:

"The database includes a total of 22,665 observations (12,847 evaluations by female stu-
dents and 9,818 evaluations by male students), including 4,423 different students (57% female
students and 43% male students), and 372 different teachers (33% female teachers and 67%
male teachers)."

6. The China-Pakistan nexus and its strength.

Better China than the US trying to make peace.

It does not surprise me that student evaluations of teachers that rate female instructors lower than males would be labeled biased/sexist. Of course they would, duh!. On the other hand if it had been the other way around the exact same people who now label the results biased would praise those results, again, duh!

I suspect that some female teachers received very high ratings from their students. I also suspect that some male teachers received very low ratings. This seems logical and virtually inevitable as the distribution of sucky and superb teachers would exist for both male and female teachers. I suspect that difference that we are talking about here is small, that is a slightly greater percentage of female teachers received sub-par ratings than did male teachers. I very much doubt that the difference is enormous or universal. We are probably talking about a small statistical difference. Why, then, would anyone jump to the conclusion that it is gender bias? The odds of it being perfectly equally distributed are enormous to off the chart so there had to be a "winner" and a "loser". Probably if you looked for results by age or weight or accent or attractiveness... that you would still find disparities. That is better looking teachers both male and female would get higher ratings or younger (perhaps old) teachers would get higher ratings. Again it is virtually impossible for the rating to be perfectly equally distributed regardless of which criteria you compared. So there must be differences; someone must rate higher and someone lower. Why all the angst?

The researchers start from a position that rankings should be based on the teacher's ability to get their students to do well on the final exam. However there may be any number of other factors that make a student like or dislike a given teacher. So it's not necessarily bias that would cause a male teacher to be rated over an "equally effective" female one- perhaps the male achieves the same results with a higher entertainment value for example. Or perhaps students simply prefer male teachers. If students prefer it for whatever reason, should their preferences be catered to? What if they prefer trans bathrooms (haha!)


A) No one claims SET are related to student performance, or quality of instruction in terms of outcome. SET only serve as a means of feedback for instructors, and as a weak signal that you aren't a completely incompetent instructor. They also serve as signals for other students on whose course to take.

SETs don't make a lick of difference to the instructor, other than at very low-low-super-low ranked institutions where the goal is retention of students (since they have extreme drop-out rates), rather than any sort of teaching quality.

B) The results show that there is...EQUALLY...a bias against male teachers by female students. Why should we only be concerned about female teacher's getting their feelings hurt?

C) The results show that in the US female students also give HIGHER ratings to male teachers. So it's not bias, at least in the US setting, that comes from only male students. Female students also seem to prefer male teachers.

D) Figure 2 shows that for both male and female students, there is a bias (overall) in favor of male teachers.

E) The results for female students seem to be conditional on...DISCIPLINE...of the course. Which would be expected due to self-selection of female students into more BS disciplines, where gender bias is the expected norm.

PS: Not sure how they are making the claim that this leads to "grade inflation", if they are talking about...expected grade...rather than actual grade. Since they show no difference in actual grade, how is this grade inflation?

3. I wonder how SWL is still taken seriously. Just take thhis example
"Another example concerns the minimum wage. The neoliberal view would be that this a bad policy which will only reduce employment. It is economists who have both gathered the evidence and developed the theory to show otherwise."
Now people may legitimately decide (politically) that they want min. wage and are willing tompay the price, but it remains bad economic policy. The few studies 'showing' otherwise don't actually, and believing them mostly shows confirmation bias. He couldn't have chosen a worse example

Comments for this post are closed