Bryan Caplan  

Grades Are So Money

PRINT
A Kingdom for a Frame... Pleased as Punch...

A Typical Blog rejects my equation of grades and money in my post on pity grading. His main objections, and my replies:

1) 'Merit-based' grades seem to be a lot more merit-based than 'merit-based' wealth. All the students these professors encounter in their classrooms have the basic tools necessary to succeed in the class (textbooks, etc) and must perform the same basic tasks in order to get an A.

Not in my experience as a teacher! Lots of students arrive poorly prepared. And whether or not they have the "basic tools" depends on their past experience, which has a strong element of luck and family resources.

As a rule, people think that success in the fields they understand depends on merit, whereas success in the fields they don't understand depends on luck. People who listen to rap think Eminem is amazingly talented; people who don't think he's lucky. I've been in school my whole life, and still see a strong element of luck - especially in college courses where two exams determine your grade.

2) Income and wealth are a lot more important to people's lives than grades. Getting a C is a lot less stressful than being unable to buy food or basic health care.

A Typical Blog notes my counter-argument that income has a relatively small effect on happiness compared to e.g. job satisfaction. At least in developed countries, redistribution is not saving the poor from starvation.

3) Unlike with income, redistributing grades largely destroys their value! As Caplan himself acknowledges, "If employers, other schools, and parents knew that pity grading went on, it would make all grades less informative." So grades open up the doors you want to walk through now, but in the world of redistributed grades they wouldn't open jack.

Not quite. Complete equalization of grades and income destroys the value of both. But moderate redistribution destroys the value of neither. The U.S. income tax has not turned America into Haiti, and basing 5% of our grades on pity rather than merit would not wreck our educational system.

But like I said, I still refuse to raise students' grades because they need them. And my challenge to pro-redistribution professors stands: Give me a good reason why you grade as meritocratically as I do.



COMMENTS (10 to date)
Ben Crost writes:

Even if you are in favour of redistribution, a good argument against pity grades (from an economic perspective) is that they are a very inefficient way of redistributing income and status.

First of all, we shouldn’t overestimate the causal effect of grades on professional success. People with higher grades are on average more successful but I would think that this is mainly because they also have the abilities they need to succeed in their jobs. The grades are just an indicator of these abilities and don’t confer them.

Like I said in a comment to your previous post, companies are already spending a lot of money on testing applicants and applicants a lot on preparing for these tests. So even if you give bad students good grades, chances are that they will fail an application test. Even if they get a good job they might not be as successful in it, be less likely to get promoted and so on. So in order to get a reasonable amount of redistribution you would have to give a lot of good grades to a lot of bad students. This in turn would force companies to spend even more money on testing applicants and the applicants to work even harder for passing the tests. Basically you end up in a vicious circle, in which the number of students that benefit from pity grades becomes smaller and the costs on the economy higher. These costs are of course additional to the ones that come from possibly having ill-qualified people in important jobs (not that this isn’t happening already).

So why should we try to redistribute income by giving pity grades if there are so many other ways of doing so that are so much cheaper for the whole economy? How about taxing income and spending the money on a good and equitable system of public education? That way instead of simply giving away good grades to disadvantaged students, we can give them the skills they need to actually get these grades themselves. (By the way I’m not saying that taxing income doesn’t have economic costs, I’m just saying that they are way lower than the costs that would come from systematically giving people the wrong grades)

dsquared writes:

Look, this is a category mistake. A grade isn't something you "award", any more than a timber grader decides to reward timber by giving it the top rating. What a timber grader does is to try and give a grade which summarises the state of the timber, and what you do is to give a grade which summarises the difficult to assess, but in principle objective, facts about the state of knowledge of the pupil.

Aaron writes:

I found this article on Finland as compared to the US.

http://www.washingtonpost.com/wp-dyn/content/article/2005/08/05/AR2005080502015_pf.html

It sort of relates to this topic because it discusses the differences between Finnish and American Education.

As a graduate student I can see the importance of grades. But some of my favorite classes I did not do so well in. I still remember the concepts, but the subjectivity of grading is a huge obstacle sometimes.

I agree that not all students are equally equipped. Some do not know how to 'show what they know" in the manner required on an exam. There is a skill to test-taking, just as there is a skill for writing. Economics is full of excellent test-takers who can't write papers. At least where I come from.

Grading purely on exams, I think, Is Bullsh*t. It depends on the course, but assignments result in minor collusion. But the costs of collusion are less than the benefits of seeing errors beforehand and working with others. A good marker will spot assignment collusion. A good exam will weed out colluders as well. The best courses offer a paper on a topic that is either free or assigned as well as a final exam where one has choice of questions.

I find in economics that the percentage of the course grade having to do with the first mid term and even the assignments doesn't matter. Profs look at improvement, so if you got a 40% on the mid term and a 100% on the cumulative final, that seems to impress some professors.

They sometimes seem to look at the differential between the final and the Midterm as a measure of successful learning.

I don't think the grade distribution matters so much as grade inflation.

Grades aren't money, but they are a form of currency (gold, diamonds) so perhaps your disagreement with A Typical blog is in definitions. Grades are traded for admission and scholarships. But not always jobs.

Acad Ronin writes:

Some comments re grading.

In B-school, it is common for "Class participation" to count substantially towards the final grade. This is a highly subjective element and provides room for pity, though many of my colleagues try to impose some structure on their assessment to make their grades more consistent across students.

Some problems with pity grades are assessing veracity, adjusting for character (the squeaky wheel problem; some people who deserve pity don't ask for it), and comparing deservedness across stories.

Finally, many instructors give an easy (high mean; low variance) midterm before student evaluations and then a tougher (lower mean; higher variance) final after evaluations are in. When they don't standardize the resulting scores (i.e., subtract the mean and divide by the standard deviation), the result is that the student's final grade depends more heavily on the final than the nominal weighting (40:60 is common) would suggest.

English Professor writes:

A grade is an assessment of performance. This is what makes pity grades (and grade inflation overall) so pernicious. A pity grade is an official assessment of performance that falsifies the instructor's professional judgment. As such it is morally indefensible.

Grades are only important, though, for those who want to move from up the educational ladder to a specific profession. I doubt that your average employer pays much attention to college grades. But if you want to go to law school or grad school, college performance, of which high grades are one primary indicator, becomes a qualifying factor.

At each stage of education, though, the populations sort themselves out again. Consider students who get into Stanford Law. Those who get the highest grades there (law review, etc.) can hope for jobs with the most prestigious law firms or perhaps Supreme Court clerkships. At this level, achievement recorded through grades will buy a high-paying job or prestige or both. The average Stanford Law graduate will probably still do quite well, but his law school grades aren't enough to gain him the highest prizes. That student's performance as an undergraduate, though, which got him into Stanford Law, made it possible for him to achieve this significant level of professional and economic well-being, though not the highest level. So grades are important IF you want to move up to the next educational level, and they can help you achieve the highest rewards if you earn them at the highest level of professional education. High grades are like an exchangeable asset. At the lower levels of education, they can often be exchanged for entree into the next level. At that level, though, your past grades are typical of everyone in the group, so they lose economic value. If you do well again, you gain another set of exchangeable assets, and so forth.

As for grade inflation--that is the true scourge of modern academic life. Because it makes high grades available to mediocre students, it obscures the process by which students should be chosen to move up to the next educational level. So what has been the result? Greater reliance on standardized tests. Nowadays, a student's test scores are a much better indication of his abilities than his grades. Everyone applying to grad school today has great grades, but many of them are mediocre intellects. With test scores, at least you see one competitive assessment of students, all of whom have been told by their teachers that they are A students.

econ professor writes:

English Professor misses one important point:

Grade inflation is not uniform across schools.

So it is well known that grading at Stanford or Harvard is easier than grading at MIT or Caltech.

At the top this means that good grades at the latter places are given serious consideration.

But there's a problem. Since it is easy to get C's at Caltech, a person at the latter school might get rejected by many a grad school, even if he were quite capable of A's or A- grades at other high-ranking institutions.

This is true even within schools. It is well-known in the signalling literature that a C level engineering student might prefer to get A's from a major known to be worthless that gives out all A's. Better an uninformative A than an informative C.

What we really need are really good, tough, national tests -- sort of the LSATs and GREs on steroids -- and then base the bulk of admission to grad schools on test scores and rec letters, and NOT grades I realize this cuts against the grain of modern PC thinking. But then I'm not an egalitarian, I just want the academically best to succeed.

Timothy writes:

A grade is an assessment of performance.

So is a salary. Good workers get promoted, earn more, etc. Mediocre workers stay in the same position or move sideways, and only ever earn the standard COLI, crappy workers get fired. See also performance-based incentive compensation plans like the one one I administer: have an exceptional year, get a big bonus, have a mediocre year, get a small bonus, have a crap year, get no bonus. We don't give pity pay, and there shouldn't be pity grades.

Glen Smith writes:

School is not a place to learn anything. Well, maybe the discipline of showing up to class on-time with your reading done. School is mostly a place to get your tickets stamped so you can get a job that allows you time to learn. Standardized tests are just another ticket to be punched, may not provide as much experience in disciplined environments and are often easy.

English Professor writes:

econ professor is right that grade inflation is not uniform across schools; in fact, it's not uniform within schools. The Humanities are generally inflated almost everywhere, but in many schools, the science departments give uninflated grades. There's an interesting economics paper on grade inflation that argues that it actually skews student choice of major. Here's how: A student who gets a B in English and a C+ in math probably assumes that he is better in English than in math; yet his B might put him in the lower half of the English class while the C+ put him at the median or higher of the math class. He is now, though, more likely to pursue English where he has no real aptitude.

See Richard Sabot and John Wakeman-Linn, *Grade Inflation and Course Choice,* Journal of Economic Perspectives (Winter 1991), pp. 159-70.

The article has some good info on grade inflation from 9 colleges over a period of 20 years--interesting stuff.

econ professor writes:

English professor elaborates on the exact point I made about the signalling literature and uninformative grades (re: engineering vs. the humanities) within schools.

But I am grateful for the journal citation. I believe related papers on the adverse selection problem of uneven grade inflation have appeared in the Journal of Economic Education and elsewhere.

Thank God I still have the right to give out grades as I see fit in my classes!

Comments for this entry have been closed
Return to top