No One Cared About My Spreadsheets
The most painful part of writing The Case Against Education was calculating the return to education. I spent fifteen months working on the spreadsheets. I came up with the baseline case, did scores of “variations on a theme,” noticed a small mistake or blind alley, then started over. Several programmer friends advised me to learn a new programming language like Python to do everything automatically, but I’m 98% sure that would have taken even longer – and introduced numerous additional errors into the results. I did plenty of programming in my youth, and I know my limitations.
I took quality control very seriously. About half a dozen friends gave up whole days of their lives to sit next to me while I gave them a guided tour of the reasoning behind my number-crunching. Four years before the book’s publication, I publicly released the spreadsheets, and asked the world to “embarrass me now” by finding errors in my work. If memory serves, one EconLog reader did find a minor mistake. When the book finally came out, I published final versions of all the spreadsheets underlying the book’s return to education calculations. A one-to-one correspondence between what’s in the book and what I shared with the world. Full transparency.
Now guess what? Since the 2018 publication of The Case Against Education, precisely zero people have emailed me about those spreadsheets. The book enjoyed massive media attention. My results were ultra-contrarian: my preferred estimate of the Social Return to Education is negative for almost every demographic. I loudly used these results to call for massive cuts in education spending. Yet since the book’s publication, no one has bothered to challenge my math. Not publicly. Not privately. No one cared about my spreadsheets.
The upshot is that I probably could have saved a year of my life. I could have glossed over dozens of thorny issues. Taxes. Transfers. The effect of education on longevity. The effect of education on quality of life. The effect of education on crime. How unpleasant school is compared to work. Instead of reading multiple literatures to extract plausible parameters, I could have just eyeballed and stipulated for every tangential issue. Who would have called me on it?
Don’t get me wrong; The Case Against Education drew plenty of criticism. Almost none of it, however, was quantitative. Some critics appealed to common sense: “Education can’t be anywhere near as wasteful as Caplan claims.” Some critics called me a philistine: “Education isn’t about making money; it’s about becoming a whole person.” Never mind that I wrote a whole chapter against this misinterpretation. A few critics bizarrely claimed that one recent paper had refuted my entire enterprise. But as far as I recall, zero critics ever checked my math.
The most novel feature of my return to education calculations was that I tried to count everything that matters. I took the countless papers that start with the standard return estimates and tweak them with one novel complication. Then I merged all the tweaks that seemed convincing to me to get final policy-relevant numbers. If you wanted to use everything researchers know to craft optimal policy, that is precisely what you would do.
In the end, however, I discovered that the true intellectual problem was not lack of supply, but lack of demand. Education researchers don’t tweak standard return calculations to get the world closer to the truth. They tweak standard return calculations to get another publication – then move on with their lives. If the world handed out attention and tenure for synthesizing everything we know about the return to education, someone else would have done it long ago.
It’s hard to avoid a disheartening conclusion: Quantitative social science is barely relevant in the real world – and almost every social scientist covertly agrees. The complex math that researchers use is disposable. You deploy it to get a publication, then move on with your career. When it comes time to give policy advice, the math is AWOL. If you’re lucky, researchers default to common sense. Otherwise, they go with their ideology and status-quo bias, using the latest prestigious papers as fig leaves. Empirical social science teaches us far more about the world than pure theory. Yet in practice, even empirical researchers barely care what empirical social science really has to teach.
Feb 15 2022 at 9:26am
It does not matter that few people questioned your math——it matters that you seemingly went thru all the proper processes required to get it right.
What matters is that it seems the large majority disagree with your conclusions—-so they might believe the math is correct but somehow misses the point.
Maybe college’s greatest value is keeping 18-22 year olds locked up in a place where they can cause the least damage. Gates did not need it. My electrician did not need it. But they are the kind of people who could add value at 18. Me? I was a moron and filled with insecurity. I needed 4 years to get over it.
Feb 15 2022 at 12:34pm
I needed 20 (at least, still working on it)
Feb 15 2022 at 1:54pm
I don’t know if this is a well-known term but I’ve heard it called ‘warehousing’ teenagers. Or it may refer to a related concept, viz. keeping unruly teenagers adolescents in school doing nothing useful until they’re mature enough to actually learn stuff.
There’s a similar concept I know of from a friend who works in Tanzania – sending teenage girls to school not so much to learn anything, because the schools are too abysmal for that, but just to keep them out of trouble (e.g. sexual abuse by older men).
Feb 15 2022 at 5:40pm
Concentrating teenagers results in more dysfunctional teenage behavior, intended to selfishly impress and harass each other, than in placing them in the workplace. There, the damage they wreck will often come from correctable mistakes in communication and task assignment, and their chief motivations are adjusted away from (though maybe not entirely) showing off for each other to showing off for a paycheck and purposeful work. Even grunt work sweeping the shop floor or shredding old paperwork does more for preparing for the real world than preparing a diorama that will not be saved by your parents next to your baby pictures.
Feb 16 2022 at 10:47am
Indeed. I was able to push the attendant grunt work to my mid-20s with all the psychological humiliation it entailed. It finally forced me to decide what compromises I was going to make in life between passions and finances. I’ve worked with lots of people now who have never done grunt work, and they lack for it. Clarity of vision is refined by compromise. Chock this up to another case of wisdom non-transferrable experience.
Feb 15 2022 at 8:15pm
College causes the most damage. Young adults should work long hours in restaurants and factories and as other kinds of unskilled laborers. When they get home from work, they should be so tired that they immediately go to sleep.
Feb 16 2022 at 10:15am
I did work when I went to college—-during school year and in summers. My most satisfying job was working in a union factory producing packaging for consumer products—-basically giant print machines and giant cutting machines. I was non-Union—-but worked for a 50 something Spanish machine boss. My only desire was to make him think I was useful. And he did think I was useful—-and I felt accomplished. It was hard work—due to speed required—-not backbreaking. I worked 12 hours a day 7 days a week—-and at night went out with my GF—-and then started again the next day.
No doubt I learned about work and responsibility——-but it was the circumstances that made it work. I went another 7-8 years doing other things (grad school!!!!)and not expanding my skills or confidence in any way. I did have a decent 25 year run later——but mostly it felt useless—-even as I was “successful”.
Engage is useful work—-it must be self defined.
Feb 15 2022 at 10:20am
This is true well beyond empirics and into most of the “trappings” of reason. They’re deployed in the service of folks underlying desires. Thus, aiming appeals to reason almost never works. Instead aim appeals to morality.
Feb 15 2022 at 10:21am
Did you consider offering a bounty for finding major mistakes, or something similar?
I don’t know enough about the economics of publishing that sort of book to say where the bounty funds should come from, but I imagine they could come from the publisher, you, or from book sales. Then set a cut-off period after which the bounty money goes to you (or back to the publisher if they paid). Neutral but trusted party could adjudicate the merit of critics’ claims. Maybe it only pays out if the mistake(s) identified change the final estimate of returns by more than some percentage.
It would incentivize people to find mistakes, but also lend you credibility by showing you indisputably have skin in the game (reputational damage is always disputable).
It would be great if you could also find a way to get potential claimants to publicly announce their intention to examine the data as they embark on it. That way, when responding to other critics or doing media appearances you can point to all the people that failed to find fault.
Hell, given the opportunity, I would be willing to make a donation to the bounty fund if getting more money helps to lend credibility to the case. If it means the book actually winds up having any policy impact, I’d (in a perfect world) make the donation back in reduced taxes.
Feb 15 2022 at 10:33am
Maybe a better way to set it up would be to consider the bounty to be just a necessary cost, and to choose a date (~5 years after publication) at which the bounty funds will be given to whoever made the best (most careful/impactful/etc) criticism.
That will initially be a huge incentive for someone to make an initial, weak case against the conclusions. With that established, in order for someone else to snatch the bounty, they only need to make a slightly better case.
With the other bounty-for-first-major-mistake model (as with the no-bounty model), there’s a huge initial barrier to arguing against your case because there’s a ton of work involved. With the bounty-for-biggest-mistake model, the initial barrier is very low, because their criticism only has to be better than no criticism at all. The idea here would be to encourage gradual commitment from critics in an escalating fashion.
Feb 15 2022 at 10:36am
And then in 5 years when the bounty gets paid out, release a second edition correcting whatever mistakes were found! 😀
Feb 15 2022 at 12:03pm
Being thorough and making the spreadsheets public was good nonetheless.
Feb 15 2022 at 12:17pm
Most likely, you were considered to be reasoning from false premises. Once that conclusion is reached, spreadsheets become useless for persuasion.
Feb 15 2022 at 12:33pm
I agree with this. Just because you consider yourself to have counted everything that matters does not mean that your readers are obliged to agree with you. The strongest criticism of your work is probably that you miss non-quantifiable returns to education, which makes your spreadsheet irrelevant.
Feb 15 2022 at 1:59pm
Quantitative analysis is great and has its place. But don’t confuse your map with the territory. Jamming an attribute into a scalar in a model is different than grappling with multi-dimensional reality in its full complexity. I feel like we need to get Russ Roberts over here to talk to Bryan.
Feb 15 2022 at 2:08pm
Bryan has done at least 4 econtalk episodes, Feb 12, 2018 on this book in particular.
Feb 15 2022 at 7:13pm
Not sure who you are responding to. For my part, if one starts with an unpersuasive point, accurate numbers don’t persuade. I happen to believe that institutional education is vastly overrated by most in terms of learning useful things. Unless I am able to persuade in the first place, facts will be assumed as cherry picked in the second place regardless of rigor.
I haven’t read The Case Against Education. I assume the arguments follow the same lines as the extensive posts on open borders which I find singularly unpersuasive. And my wife, employees, and about a third of my suppliers are immigrants. There are strong arguments for immigration reform that are buried under the open borders rhetoric. Others have brought up opposing arguments repeatedly.
Feb 15 2022 at 4:07pm
I would say that’s the weakest criticism of his book. It’s trivially easy to postulate non-quantifiable benefits to something. The argument for spending measurable resources on unmeasurable (and thus unfalsifiable) benefits is almost always weak.
Feb 16 2022 at 10:52am
Qualitative benefits are still quantifiable. Otherwise, art auctions wouldn’t exist.
Feb 15 2022 at 1:38pm
Of course the math was correct as Excel, in my experience, never makes any mistakes. It religiously calculates what you input into it (I’ve done countless numbers of spreadsheets over the years for financial analysis). You are picking the wrong issue to write about. Some of us did read your book with a very critical eye and offered substantial comments on why various segments were either right or wrong (in my case there was much more wrong with the book than right). It’s interesting that I could only find less than a handful of reviews of the book and only one in the mainstream press IIRC (The Washington Post which was pretty scathing “…it offers little more than dangerous, extravagant ideology masking as creative data analysis…”)
Feb 16 2022 at 10:51am
That is an incredibly dumb statement. The question is not about the calculation, but the formulas that were used for the calculation. Did Bryan put in the right formulas? Are his coefficients correct?
Feb 15 2022 at 1:42pm
One reason may be that the results were so, as you say, ultra-contrarian. A quick Google search shows median weekly earnings in Q3 of 2019 at $749 for high school diplomas but $1281 for Bachelor’s degrees. The extra $28,000 per year is impressionable.
Feb 15 2022 at 2:32pm
Bryan doesn’t dispute there are positive selfish returns to education. He found negative *social* returns to education.
Feb 15 2022 at 3:53pm
That’s why I said impressionable instead of impressive. I’m suggesting that many were content to assume the wage difference justified the value of education, and wouldn’t bother looking further into the details.
Feb 15 2022 at 2:22pm
You didn’t waste your time. If you had made your arguments without the spreadsheets-just guesstimating & eyeballing, you would’ve gotten quantitative criticism. A man who successfully deters burglars didn’t waste his money on a security system just because it never got used.
Bruce K. Britton
Feb 15 2022 at 3:33pm
In your introductory slideshow, you don’t say what the numbers are quantities of. Please let me know that.
Feb 15 2022 at 4:13pm
What makes you think that no one checked your math? Presumably, what you know is that no one checked your math and was able to confidently identify an error. I mean, anyone who checked your math and didn’t find any issues isn’t going to make an issue of it.
Absent finding major errors yourself, why would you come to the conclusion that no one checked your math rather than the conclusion that your math was sufficiently compelling as to not make a worthwhile target? Math like that is only going to get challenged *either* if you made mistakes or were sloppy (so someone can simply declare the methodology is sloppy or walk away) or if someone could generate a very different answer using the same general methodology but different plausible estimates. Maybe that’s just not really possible in this case.
(As an aside, I personally tend to agree with your conclusions about the direct benefits of education but are more pessimistic about our ability to avoid those disadvantages (eliminate college and you’d some other for-profit form of credentialism replace it). In particular, I fear that it wouldn’t be possible to capture the social/fun/romantic benefits of sending young people all off to a common location to have fun together without the pretense it was about training even if, in theory, it should be possible to capture those advantages much much more cheaply.)
Feb 15 2022 at 7:09pm
Strikes me as an instance of the Law of Continued Failure, subspecies Law of Continued Disagreement.
When Eric Drexler published _Nanosystems_ after years of painstaking work and calculation at MIT, basically nobody who criticized molecular nanotechnology criticized those calculations and I don’t think it convinced many people who weren’t convinced by _Engines of Creation_.
Why? On my reading, because even though _Engines of Creation_ was vastly less technical, the arguments there were basically correct and obviously so to anyone who could possibly follow along with the technical reasoning in the later book. Anybody who disagreed with _Engines of Creation_ was doing that for reasons that they couldn’t be talked out of by a more technical argument.
Similarly, anybody who can follow along with a complicated spreadsheet about returns to education, who wants to arrive at the true answer and not an answer with some other properties, was already convinced by much simpler arguments that were obviously true. And anyone who wasn’t convinced by those much simpler arguments, is unconvinced for reasons that don’t change in the face of the spreadsheet however true it may be, and doesn’t want to see the spreadsheet, and definitely doesn’t want to fight you on your home ground of the spreadsheet, and will never talk about the spreadsheet.
Feb 15 2022 at 8:07pm
If you don’t know Python, can’t speak Spanish, and don’t play a musical instrument, you’re not an educated person. Get your sons to help you out.
Feb 15 2022 at 8:20pm
You’re so good at explaining your work verbally that I feel less need to crack open the excel sheets.
A little dashboard or data visualization would go a long way for making your work accessible. Maybe I’ll make one for your calculations.
3.Your math persuaded me. I admit, I didn’t dig in to the math (yet). But your thorough process for truth-seeking convince that you are a highly credible person in your areas of expertise. Probably the most credible. Your research, books, and talks have changed my mind dramatically, and entertained me in the process. As one person who is a member of the real world, your brand of quantitative social science
Feb 15 2022 at 11:42pm
In 2019, Tyler Cowen wrote about the State of Alaska cutting state university funding by 41 percent.
I left a comment on Cowen’s 2019 article:
Caplan asks why his skeptics didn’t challenge or even consider his quantitative arguments. I’d turn the question on Caplan: Why haven’t you engaged in any serious advocacy of actual policy changes like you have with immigration. When Alaska was close to a drastic cut in education spending, why were you and your supporters disinterested?
Tyler Cowen is super smart and very interested in the subject and he’s a close friend of yours. Yet, Cowen has the opposite view. Cowen was horrified at the idea of large budget cuts, and actually favors large spending increase. Why can’t you convince him, even just a tiny bit? Or Edward Glaeser, who publicly debated the issue with you: Glaeser debated against cutting higher ed spending. Glaeser is a smart, reasonable, libertarian-sympathetic economist. Why could you not convince him even a little?
Both Cowen and Glaeser are smart, professional economists, whom you respect, and your book and your spreadsheets didn’t convince them in the slightest. Why?
Feb 20 2022 at 10:59pm
Re: “How about free market education?” Government-funded education redistributes wealth from childless taxpayers to those with kids. Most people have kids at some stage – the fertile are in a majority. In a democracy, stealing from the childless and giving to those with kids is a vote-winner – at least in most places. Most politicians realize this on some level – and they uniformly propose it as policy. It is about as simple as that.
Feb 22 2022 at 2:27am
This statement assumes that education is a service provided to parents. If education is supposed to build human capital it would seem more correct to describe it as a service provided to the child. Then it would be redistribution from high tax paying individuals to individuals getting a lot of school – mostly a redistribution across the individuals own lifecycle.
Feb 16 2022 at 2:19am
Why would critics check your math if the whole field already knows you are wrong? If you make an ultra-contrarion statement don’t expect engagement on substantive details.
When you write something, readers only recall a fraction of what you wrote. Anything that doesn’t fit into the reader’s model of the world will be immediately forgotten because the reader is a human.
Written persuasion is hard. A writter cannot use real-time feedback from the reader to tweak the explanation, use techniques to emphasize key points like in speach like repetition or pauses, and the reader will skip sections without knowing what is in them.
I worked in an engineering team spread across the world. I had an idea that would make our product way better. We talked about it for 2 years at the weekly team meeting and I wrote up a detailed 10 page proposal. One day the engineer I needed to implement it was in town. Within 5 minutes of one-on-one face-to-face talking the engineer said, “oh I totally didn’t catch [core, simple concept]. That does solve a huge number of problems. I’ll get it done in a few months.”
This is why we have teachers instead of insisting that the students just read the book. Wierdo autodidacts (maybe anyone that is intensely creative) can read a book and fit most of the ideas into their model of the world. Almost tautalogically, autodidacts already thought they learned most things outside of their education and professors thought that higher education is important. So, the number of people that are critics and can fit most of the ideas into their model is small.
Your critics don’t need to look at your math. They heard you say something like “higher education doesn’t increase earning potentials as much as the cost”. That can fit into their model of education and they know how to counter it.
BTW don’t try to learn Python. If you did any programming when you were younger (I assume Fortran or BASIC ) and now you make complex spreadsheets, you will not have an easy time thinking like a Python programmer. Python is made by programmers for programmers to be able to do anything. Learn MATLAB or maybe R instead because they are built for non-programmers to crunch numbers.
If you want to learn a whole new way of thinking and you don’t care about the results then you should consider Lisp (or another “functional language” that your local Computer Science professors think is “the one true way to program”) to learn other ways to think about computation as well.
Feb 20 2022 at 10:27pm
Re. Python: For what it’s worth, I learned Python on the job after years of using Matlab in undergrad, graduate school, and professionally, and not knowing much if anything of different languages. I found the transition quite smooth, because they syntax is very similar. Python is also free, whereas Matlab is not. … I write patents now, I haven’t programmed for many years.
Feb 16 2022 at 6:25am
My experience is that very few people have the skills needed to do the analysis.
The benefit of your analysis, apart from checking your arguments, is that when somebody disagrees with your analysis you can simply ask them to check your calculations. This will cut off any further argument, which saves you lots of time, but I doubt it will change the mind of the person you are talking to.
Years from now, you will encounter spreadsheets that have obviously been derived from yours, minus any citation.
Feb 16 2022 at 10:03am
I actually did go through your spreadsheets after having read Case Against Education. I really do appreciate you for having posted them.
I remember having found the spreadsheets difficult to understand as stand-alone products. I was especially interested in the Selfish Return by Ability spreadsheet, so I went through that one in some detail. I saw lots of unsourced hard-coded values. Some of the columns were ambiguously labeled (e.g. “Experience Adjustment”). I figured I would need to reread sections of the book to hope to understand them, and I wasn’t ready to put that much work in.
Looking again, I now see that I missed much of the documentation. The documentation for the Selfish Return by Ability was actually in the Background files. Clearly, my mistake, but I wonder if others potentially made that same mistake.
I probably spent ~5 minutes looking through your spreadsheets. But if I had found the documentation more intuitively organized, I probably would have spent much more time going through them in detail. Now that I do see the documentation, I actually might take another look, as I might be interested in playing with your framework to build some of my own models.
Here are some suggestions for the future:
If understanding a spreadsheet requires going through the “Background” files first, clearly indicate so. In each workbook, explicitly recommend that the reader consult the “Background” package and provide a link to where the documentation is located. To be fair, the Selfish Return by Ability sheet does say “All other variables from meta.xls and text,” but I didn’t even know what “meta.xls” was or where it was to be found. I also didn’t know where exactly to look “in the text.”
Expanding on that last idea, in general, it would be helpful to draw a more definitive correspondence between the spreadsheet and the text. For example, make the output tables in the spreadsheet resemble how the output tables appear in the text, and maybe include corresponding page numbers or table numbers in the spreadsheet.
Ideally, include the workbook-specific documentation in the corresponding workbook. For example, looking now, I see the “Background” files includes a doc.xls spreadsheet, which includes a sheet “Doc for Selfish Return.” In my initial review, I was expecting to see something like the “Doc for Selfish Return” tab in the Selfish Return by Ability.
I understand that this is a lot to ask. I would never criticize you or not having done the above things As you know, sharing your spreadsheets at all was an over-the-top display of transparency and research integrity. However, if you were hoping for a higher level of engagement in your spreadsheets, I wonder if some of my suggestions might have helped.
Feb 16 2022 at 10:49am
I agree with this. I looked at the spreadsheets yesterday and it took some searching to find some descriptions. Having the docs as a sheet within the corresponding spreadsheets would be a big improvement.
Feb 16 2022 at 11:33am
This article is not saying what you think it’s saying.
Multiple people (seemingly in positions to understand what you were trying to accomplish quantitatively) told you to use a specific tool, python, in order to make your work easier, more reproducible, transparent and shareable. You choose to use a different tool, Excel. Now you complain that people capable of challenging your quantitative analysis — people probably similarly skilled to those who suggested python initially — didn’t comb through the data, assumptions and functions hidden in the winding path of cells in your Excel workbook. Then you lament the lack of quantitative rigor in the social sciences (amen, unarguable)! It starts by not using Excel! Ain’t nobody got time for that.
Knut P. Heen
Feb 16 2022 at 1:10pm
After reading some of the arguments for the minimum wage during the progressive area (Sidney Webb’s 1912 JPE-article), I have entertained the idea that education is a way to draw people out of the labor market to increase wages. Webb did complain about the presence of children and married women in the labor market reducing the wages of men at the time. Establishing public schools may have been a way to increase wages for men by removing children from the labor market. Public schools must be compulsory for this scheme to work because the temptation to free-ride increases with the increases in wages. Hence, it will look like a bad economic decision to go to school.
The same scheme may work for higher education too if getting a certain degree is compulsory to get a certain job.
Remove higher education altogether and we will see the supply of labor increase by 10 – 15 percent (4-6 years added to a 40-year career). This will reduce the price in the labor market. If the argument is that everyone takes education because everyone else does it, then you have to compare everyone taking education with everyone not taking education. The annual labor income will be different in those two scenarios because of the supply issue.
Feb 17 2022 at 7:08am
Adding more labour supply doesn’t seem to decrease labour incomes.
If anything agglomeration benefits point the other direction. People are richer in big cities than in sparsely populated areas.
Knut P. Heen
Feb 17 2022 at 11:52am
How many drivers do a bus need? More bus drivers without more buses, will drive down the wages.
Feb 16 2022 at 5:09pm
Like a few others here, I glanced through your spreadsheets. I am – with all due modest – very, very good at Excel spreadsheet design. My work often involves reviewing quantitative analysis and modeling that is frequently done in Excel. I’ve been involved in that sort of work for 20 years.
From a professional standpoint, I’m sorry but your spreadsheets stink. If one of my teams sent them to me and asked me to review the analysis, I’d tell them to go talk to some of their peers about how to make the analysis sufficiently clear and transparent before I’d spend my time looking carefully. That is not to say that beautifully-designed spreadsheets would have stood any more chance of being reviewed, but if you actually want people to check your work, there are various ways that you could have designed the spreadsheets to reduce the burden you have put on those folks.
Feb 16 2022 at 7:20pm
Well, maybe the value OF education is to proof read and see that its not the value TO education :-).
See the beginning of your post here for the typo “return to education”
Feb 16 2022 at 11:12pm
I disagree that the conclusion we come to should be decreased education spending. We need to fundamentally rethink what education is for. Postsecondary education shouldn’t be encouraged for all students for it’s perceived benefits (more earning potential, creating more worldly/rounded/knowledgeable people, etc.). It should simply be the next step for those suited to it. Earnings shouldn’t be tied to education level. A 16-18 year-old who wants to go into a skilled trade shouldn’t be discouraged by low wages or burdened by superfluous education. Think electrician, plumber, service industry worker, even many jobs that currently require degrees could probably do without education requirements. It will take a reorganization of the entire labor economy to assign wages to jobs based on demand and value added instead of how expensive the education was to attain. The market could do it on its own, provided it’s nudged in the right direction by education policy. It’s my opinion that the best way to nudge the market would be by removing the cost of education from the equation – by spending more on education to make it tuition-free for everyone and allowing everyone with the aptitude to attain the highest level of education they are able to. We need people to do what they’re good at, not what pays best. That’s efficiency. It may disrupt prices of some foods and services or even entire sectors, but it’s a short-term cost that will have massive benefits in the long term, encouraging of personal/professional fulfillment, social mobility, and even social justice (if you’re in to that sort of thing).
Feb 17 2022 at 7:07am
To be fair, Excel is a reasonable ‘programming language’ to write models in. But it’s almost unreadable, and thus also almost unreviewable.
I used to deal professionally with both spreadsheets and programs written in more conventional languages like Python.
Feb 18 2022 at 1:55pm
As someone currently spending days drilling through 17k order lines for the purpose of generate a single data point for a larger internal narrative, I am deeply sympathetic.
Feb 20 2022 at 12:28pm
For several decades I had a “pursuit of truth” mental model of academia. I assumed that evidence contrary to mainstream academic narratives would eventually have to be addressed – because contradictory evidence was an intellectual embarrassment if not addressed.
But for some time now I’ve been more convinced by the notion that as long as professors can achieve relevant status markers within their domains, there are no pressures to address contradictory evidence outside their domains. Academic disciplines, and subfields within disciplines, drift for long periods of time without addressing contradictory evidence.
The greatest example of this is the fact that Marxist academics can continue to claim that capitalism makes the rich richer and the poor poorer. David Harvey is one of the most highly cited scholars in the humanities,
Yet as of 2014 he was still claiming,
”Everywhere, the rich are getting richer by the minute. . . . By contrast, the well-being of the masses at best stagnates or more likely undergoes an accelerating if not catastrophic (as in Greece and Spain) degradation.”
But rather than focus on the fact that Marxists have mostly ignored empirical economics, I see the academic norm as one in which careers are made by advancing within a particular discourse community while ignoring all contrary research and evidence.
Why would the discipline of education pay attention to evidence that undermines most of their livelihood? They advance within an incentive system that only requires success within their disciplinary discourse community.
I’d like to see a standing list, ideally promoted by elite academics, of unaddressed contradictory evidence in the social sciences. Until there is a price to be paid for disciplinary insularity, most professors will ignore all contradictory evidence unless acknowledging that evidence gives them a reputational boost within their discipline.
Feb 20 2022 at 12:53pm
Looking at the data you do present, it seems as though you are not counting benefits to parents. Government-funded education represents a wealth transfer from taxpayers to people with kids. People with smart kids benefit more. Kind-of like a government-funded breeding program that taxes the sterile and gives to the fertile. So: you didn’t count a bunch of effects of government-funded education that look pretty relevant to me.
Feb 20 2022 at 1:02pm
It seems as though your analysis is very focused on the benefits to the individual students. Government-funded education also offers benefits to the government. They get to influence the curriculum and decide which areas get funding. Without this there would be all kinds of programs designed to make people into pop stars and models. Governments with successful higher education programs can subsequently sell their education programs to foreign students for profit. Traditionally, education has been used to instill values in students such as patriotism and respect for the law. Patriotic students go on to become soldiers and law-abiding students don’t clutter up the prisons. Where are the benefits to the government in the analysis?
Comments are closed.