August 8, 2011

"The best way to eliminate grade inflation is to take professors out of the grading process..."

"... Replace them with professional evaluators who never meet the students, and who don't worry that students will punish harsh grades with poor reviews. That's the argument made by leaders of Western Governors University, which has hired 300 adjunct professors who do nothing but grade student work."

That's awfully nice for the professors who don't have to grade. Should we feel for the adjuncts who have a job consisting of the part of the job that the regular professors are freed from? The photo at the link depicts "Emily Child, an evaluator at Western Governors U., [who] grades 10 to 15 assignments a day, six days a week, working at home before her three children are up or while they nap."

So... that's either awfully nice for Ms. Child, who gets to work at home and arrange her hours around household responsibilities... or a restoration of the subordination of women. (Yes, yes, there are some men who keep house and take care of children... especially they've trained in an academic discipline that affords few career opportunities and their spouses have out-of-the-house jobs.)

I can't picture law schools adopting this method of grading, partly because I'm habituated to the burden, which balances what is otherwise the overwhelmingly pleasure of teaching law. In law schools, we avoid grade inflation by imposing a curve, with a restricted range for the average and a required distribution of grades (forcing you to at least give some Cs, even if Ds and Fs are optional). We also "blind grade," identifying the students with code numbers, and it is, in fact, easier to be critical when you don't have an individual's name on the paper.

41 comments:

Shouting Thomas said...

So... that's either awfully nice for Ms. Child, who gets to work at home and arrange her hours around household responsibilities... or a restoration of the subordination of women.

Or, you're just being a crank.

KCFleming said...

Soon, colleges won't be able to afford superfluous hiring like this.

The way to stop grade inflation is to fire professors who do it, and promote the ruthless.

Shouting Thomas said...

Do you foresee a time in the future, Althouse, when a woman will be considered lucky to get any job she can get, without reference to the great feminist cause?

You know, like the rest of us?

Are we required to provide women with jobs that enhance their status and pave the way to social justice?

Or, should the ladies just thank God they've got a paycheck?

Dennis said...

As much as I would have loved freedom from grading when I taught law school, farming out law school grading isn't feasible. There are other ways to control grade inflation even if individual teachers are too cowardly to demand success before awarding grades.

One is the old British system of using outside reviewers to examine a sample of exams to make sure grading at all schools is relatively consistent. That's not perfect, and it is somewhat clumsy and costly, but it helps. (It also helps that student evaluations don't control a faculty member's career.)

A second and much simpler method is for the law school itself to impose a tough curve or at least a modest average grade for larger classes. All classes would then follow the rule and no individual teacher would be at a disadvantage. The difficulty is that law faculties have collectively raised average grades, usually with some beneficent objective like making graduates look better on the job market or avoiding having to give bad grades to poor-performing students. If law faculties are unable to insist collectively on high standards, there's not much hope that they will do so individually.

bearing said...

I don't think we should feel bad for ANYBODY who has a job, especially not a viable one that allows them to work from home and care for one's own children while keeping at least a few professional skills sharp. That "keeping women in the home" trope is entirely misplaced here.

I keep house and take care of children. I also have a PhD in a technical field (my first two kids were 4 and 6 months old when I was graduated). For a little while, when the kids were smaller and before our choice to homeschool them grew into my full-time job, I enjoyed a brief stint working from home to help edit the theses of foreign students in my field -- get the English usage up to par.

I really liked that work. I got to find out what was going on in my field; I had to understand the technical details in order to make usage recommendations; I believed in the work. I wish I could do more of it. (What stops me is that I don't want to deal right now with the business parts of freelancing or independent consulting.)

But it rather strikes me that I was more connected to the economic world outside my home when I was working from it instead of purely in it. Your "subordination of women" reflex is overactive.

Now, if you want to talk about this job being a tool to keep people down in the long run, talk about the fact that they're adjuncts, not the fact that they're women.

Shouting Thomas said...

I'll turn it on its head.

I'd bet that grade inflation is precisely correlated to the takeover of colleges by women. You know (as your post states), we're supposed to be ever on our toes lest we hurt the poor dears' feelings or set back social progress.

I suspect that grade inflation is tied to the entitlement mentality of young women on campus. Don't they comprise 60% of students?

I read a weepy piece about young adults who can't find jobs a few days ago. My favorite... Young woman goes $150,000 in debt for a degree in film studies at NYU! And, doesn't find a job.

Who could have predicted that?

Questions: Who in the hell is stupid enough to go $150,000 in debt for a degree in film studies? What kind of society actually lends that kind of money to a silly young woman?

Scott M said...

or a restoration of the subordination of women.

Did you wake up on the ERA side of the bed this morning?

ken in tx said...

The problem with 'forced choice' type curve evaluation systems is that it is unfair to people who self select for elite or advanced classes or groups. This was a problem in the officer corps in the Air Force when I was on active duty. Some members in an elite group had to receive lower evaluations even though they were selected for that group because they were considered outstanding. For an outstanding performer, it was better to be a member of a mediocre unit, as far as getting promoted was concerned. This sometimes led to horse trading, where the guy next up for promotion got the highest rating. Then next time, the next guy up for promotion was picked—regardless of performance.

TMink said...

Actually, you could just promote the accurate! Grading on the curve is really pretty easy and defensible once you know how to do it.

Trey

Bob Ellison said...

Curve-based grading can cut the other way, too. In a perfect world, sometimes the average grade really will be an A (if the students are smart and work really hard), and sometimes it will be an F (if the students are lazy and don't learn the material). The latter case does happen.

Grading theory is fascinating. I'm convinced hard grading is absolutely necessary, but what method is best? Probably just the competition from class to class and school to school helps more than anything else.

Gabriel Hanna said...

Because every physics class builds on the ones before it in a straghtforward way, I grade on an absolute scale. If you do well enough that you are prepared to move up to the next class, you deserve a B.

All my students can get As, but what is more likely is that they all get Cs.

One year I had to justify an exam I had written to the department. Some of the dear snowflakes had thought it was unfairly hard and had complained to their parents, who had complained to the university. I showed them that much of the exam was conceptual questions, or questions for which they had to draw a picture, with no intricate calculations. That ended the matter.

I've always tried to write exams so that a student who understands the material will be able to answer the question in a few minutes without having to do any tedious calculation--but there's still only ever a small percentage who get an A. Most of the students just try to use random formulas that might contain some of the parameters they see in the question, and get Cs.

Gabriel Hanna said...

But the hiring of professional graders is just one more way in which the bulk of the teaching work is increasingly done by adjuncts, who do not get the pay and benefits and job security of tenured faculty.

Teaching doesn't bring in money. Research brings in money. Of every research grant awarded to a professor, the university takes 50% before the professor sees any of it. And expects the professor to pay part of his own salary out of the grant, besides buying equipment and funding the education of his grad students. Why WOULD universities want to pay a lot of money for teaching?

Maybe we as a society need to figure out if we want to separate the research and teaching functions of a university into different institutions.

Firehand said...

Let's see, she finds a job that lets her avoid using daycare and such, that probably pays pretty well, and that is part of a restoration of the subordination of women? In what world?

Scott M said...

Of every research grant awarded to a professor, the university takes 50% before the professor sees any of it. And expects the professor to pay part of his own salary out of the grant, besides buying equipment and funding the education of his grad students.

Doesn't the mob work along those same lines?

Skyler said...

Only "some" C's? C's should be the most common grade.

It's an interesting idea, perhaps it will actually get professors to actually teach material. I don't think it could work, though. If you're going to have centralized grading, then that implies centralized lesson plans and that can't really work in many classes that are not standard. If you standardize classes it makes it too hard to explore new subjects.

roesch-voltaire said...

I think evaluation and feedback is part of teaching and not to be looked down upon as many do. Over the years I have suggested that professors exchange essays written in their classes to get second opinions and establish bench marks for grading-- unfortunately few are willing to do so. I found another approach to grading while evaluating a department in Denmark, where a panel that consisted of the teaching professor, another professor and a student met with each student to review the semester work and determine the grade. A very democratic and time consuming process that seem to work, but I doubt if our universities have the time or money to do this given the pressure to win grants that pay for our salaries and provide monies to the college.

Scott M said...

If you're going to have centralized grading, then that implies centralized lesson plans and that can't really work in many classes that are not standard. If you standardize classes it makes it too hard to explore new subjects.

This. From a layman's point of view, surely, but still, given my experience in college...this.

Ann Althouse said...

I've been grading on a curve -- as required -- for 25 years, so it doesn't even make sense to me to speak of something being "really" an A or "really" a C. I view the exams comparatively, and the students themselves set the standard. If the students were better and harder working, it would simply become a harder class. If the students acted collectively... what could they do? Everybody crank it down a notch? That would create an opportunity to curve bust, and wouldn't you go for it?

JimMuy said...

You know as well as I that the blind code is a sham if it's an essay test. Especially in the 2nd and 3rd year classes (and 1st year with the gunners). They've had enough interaction with the student to know their "voice" when they are writing.

Also, why not outsource the grading?
The bar exam--the most important law school test--is outsourced.

Oh, and the "grade on a curve" thing is another lie. It has little to do with preventing inflation. It is mostly in place to ensure a method of flushing out the lower performers so the law school can, on one hand, keep their admittance and application rates high, while on the other keep their bar passage and employment rates up.

Amy Schley said...

In law schools, we avoid grade inflation by imposing a curve, with a restricted range for the average and a required distribution of grades (forcing you to at least give some Cs, even if Ds and Fs are optional)

Pray tell, what is the mandatory median for a 1L UW class? I have a hunch it is far closer to Columbia's 3.2 than UMKC's 2.2. What is the lowest grade that must be given? At Columbia, it's a 2.8. UMKC requires 10% of the class to have a 1.8 or lower.

The more prestigious the institution, the more inflated the grades, on the apparent theory that just getting in means one deserves a higher grade. (Which is a double whammy for lower tier students.)

Rich Rostrom said...

the overwhelm[ing] pleasure of teaching law...

This is perhaps the second most perverse sentence I have ever read on the Internet.

(#1 was "a place with large helicopters perpetually circling overhead fits my preconception of 'heaven' rather well")

Bob Ellison said...

"curve bust"-- is this a standard term among faculty? It's a good, descriptive one.

A friend of mine recently took a university science course with absolute grades. Her fellow students seemed to average around 65% or so, earning Ds. The professor responded to this looming calamity with a flurry of grade-boosting techniques: extra-credit homework, extra-credit test questions, "throw away the lowest test grade", etc. It worked as sort of a back-door curving system. At least one student ended up over 100% for the final grade.

The interplay among the main actors-- students who have an incentive to bust the curve, students who work extremely hard, professors who want to teach and grade fairly, parents who will just about sue for a better grade, and the school that has to balance its reputation against its desire to see students thrive-- is complex and amazing.

Bob Ellison said...

RE: my friend's science course-- I forgot to mention the most significant grade-boosting technique, which was that about half of the students dropped the class before the final grades became unerasable. That's a good thing. Simply flushing out the inadequate students benefits everyone.

MikeDC said...

Curved grading is fundamentally pretty stupid though, because the comparison between students is arbitrarily defined across a single year.

When people go out and work for a living, they're compared to peers who went through school before and after them.

While the law of averages kicks in at some point, any particular class, like any particular person, is likely to not be precisely average. Thus, a student who might get an A in a class full of meat heads will only get a C in a class full of smart students who know the material.

That sort of outcome doesn't strike me as very useful to anyone except the professor, who's relieved of the requirement to take ownership of their evaluations, or form any conclusions about the quality of his or her teaching over successive classes.

Unknown said...

"So... that's either awfully nice for Ms. Child, who gets to work at home and arrange her hours around household responsibilities... or a restoration of the subordination of women. (Yes, yes, there are some men who keep house and take care of children... especially they've trained in an academic discipline that affords few career opportunities and their spouses have out-of-the-house jobs.)"

If you contradict your final phrase (with a back-hand to men of course) in the next sentence, why write it? Only a feminist could see having a job as a woman being subordinated.

sorepaw said...

I don't think we need to worry that hiring adjuncts to grade papers and tests is going to bring back the subjection of women.

I do think we need to worry that professors who don't do their own grading will be relieved of a significant part of their responsibilities. And will be hastening their own obsolescence.

Grading on a curve presumes that some distribution of grades to be given is known in advance. The distribution of levels of aptitude or degrees of motivation among the students taking a course is not known in advance.

It makes much more sense to define in advance what constitutes A-quality, B-quality, C-quality, and D-quality work, then give out grades according to the quality of the work that the students actually produce.

Unknown said...

There's nothing more obvious as a tool to up the college's stats than the sliding curve.

It was always my goal to fuck those suckers up with as close to 100% scores as I could get. Which meant of course, that those who got F's didn't get drug up into a C or better.

As for the trick of enhancing the grade with extra credits, I once took a required enviro-commercial which had as extra credit turning in an insect collection. I simply went home and scoured my collection for deteriorating specimens and got enough extra credit for an A in the course sans tests. I took them anyway, but I didn't have to.

sorepaw said...

Gabriel Hanna is taking either the viewpoint of a professional administrator, or of a faculty member in a discipline where most research is expensive to conduct.

Teaching doesn't bring in money. Research brings in money. Of every research grant awarded to a professor, the university takes 50% before the professor sees any of it. And expects the professor to pay part of his own salary out of the grant, besides buying equipment and funding the education of his grad students. Why WOULD universities want to pay a lot of money for teaching?

Like hell teaching doesn't bring in money.

The more an institution charges for tuition, the more revenue is being brought in by undergraduate teaching.

What's more, departments that teach a lot of undergraduates, other things being equal, tend to cross-subsidize departments that teach few undergraduates, and of course they help to support the central administration.

The reason administrators talk as though only extramurally grant-funded research brings in money is because grants with "indirect cost" components increase the funds under the discretionary control of administrators.

What comes in to pay for instruction in, say, English, either gets spent on the salaries and benefits of the English instructors, or gets transferred behind the scenes to pay someone else's salaries and benefits, on terms that no university administration wants anybody to spend much time examining.

So if you're a dean or an assistant associate academic vice-president or whatever, you want more grant funding on account of what you can rake off from it. Secondarily, you may want bragging rights for your unit, but the rake-offs come first.

Gabriel Hanna said...

@sorepaw:Gabriel Hanna is taking either the viewpoint of a professional administrator,

Like hell I am. Simply describing the facts. Nothing I said could be construed as an endorsement.

At the land grant university I attended, grants brought in more than tuition:

http://hrs.wsu.edu/utils/File.aspx?fileid=1206

One professor can do little to bring students and their tution money. One professor can bring in a lot of money from research.

That's simply how it is. Whether I personally disapprove has nothing to do with it.

Geoff Matthews said...

Full disclosure, I used to work for WGU in the Institutional Research department. Great institution if you are a self-directed learner.

Having said that, I think that separating the grader from the instructor is probably the best way to maintain integrity for a university. Apart from sympathy for students, instructors have altered the material that needs to be covered, dumbed-down tests, and curved grades upwards. All of these effect the integrity of the course.

WRT law school, it sounds like the grades don't tell students how well they are learning, but rather how well they are learning in relation to their classmates. Is the bar exam the only assessment they get that gives them a true score of their ability?

Geoff Matthews said...

This is probably an application of the industrialization of education. Separating the various functions of instructors into the separate parts, which allows them to do them faster and quicker (just like machine-graded tests (ie, multiple choice) is a time saver).

I used to work at WGU. The people who worked with the students (much of the learning, ideally, is self-directed, but there is faculty interactions with the students on at least a weekly basis (which is one-on-one) were not the ones who set the curriculum, the syllabus, the rubrics, etc.

The advantage to this is that it allows for good evaluation of the faculty for that class. It reduces the amount of training that faculty need, which reduces the amount of pay they get, but more importantly, the amount that it costs to attend. To hell with whether it provides good paying jobs, the important thing is whether it provides affordable education.

Geoff Matthews said...

I did find it interesting that Ann focused on Emily Child, but not her boss, Diane Johnson (who does go into the office to work), or her boss, Janet Schnitz (as well).

Kirk Parker said...

Bob Ellison,

I'm pretty sure the term "curve bust" has meaning outside the academic arena, too.

ken in tx said...

"Curve busting' is what the unions call 'rate busting'. It means to work harder than you have to and make everyone else look bad.

wv-ductied=tied with duct tape.

Derve Swanson said...

WRT law school, it sounds like the grades don't tell students how well they are learning, but rather how well they are learning in relation to their classmates.

She also forgets to mention the existence of "exam banks", in which members of a group -- often a racial or ethnic group -- have access through their clubs to old tests and class notes.

Since some profs change their materials little through the years, and we're all graded "compared to each other" not on a strict merit scale, then this provides uncalculated advantages/disadvantages to those with years worth of sneak-peeks into the professors testing methods and preferred answers.

Plus, JimMuy at 9:27 is spot on regarding "blind" grading curves. Sorry, but you do get to recognize a voice in writing styles, if you're awake during your classes and when you're busy reading responses.

Synova said...

All I can think of is all of that *work* figuring out a particular professor and what they're after tossed down the toilet.

That's why you go to class, right? You go so that you get a feel for your professor's approach to the material as well as their personality.

Certainly, being good at figuring that out is a valuable skill and ought to be graded too.

Going "blind" by numbers or something on the part of the professor seems like a good idea, because at least it's still the same professor.

If you want a stranger to grade your stuff from home, there are no end to on-line correspondence colleges.

Geoff Matthews said...

Synova,

Ideally, the professor should be separate from the class content. Yes, it is not likely with the upper-division classes, but with intro classes, the content should be set.
And if students are writing to please the professor more than demonstrating knowledge, than all the more reason to separate the grade from the instructor.

sorepaw said...

At the land grant university I attended, grants brought in more than tuition:

http://hrs.wsu.edu/utils/File.aspx?fileid=1206


Yes, according to the charts in this presentation, Washington State expects to pull in 18% of its revenue from tuition and 22% from Federal, state, and local grants during its budgetary period of 2009-2011.

While direct state appropriations as a percentage of operating revenue have declined at WSU, as they have at state universities in general, WSU is pulling in a higher percentage of its revenue from direct appropriations (29%) than is now the norm.

So how of that 29% does the Washington state legislature think it is kicking in to support undergraduate instruction?

I can't answer that question from where I'm sitting, but some other charts in the presentation suggest that the WSU administration has at least a rough idea.

And 29% + 18% in direct state appropriations plus tuition substantially exceeds 22% in grant funding.

sorepaw said...

Gabriel Hanna,

One professor can do little to bring students and their tution [sic] money.

One professor who teaches a large number of student credit hours every semester, and does so at a relatively low cost per student credit hour?

One professor can bring in a lot of money from research.

One professor, in a discipline where research costs a lot to carry out, can bring in large amounts of grant money to fund research that he or she couldn't otherwise carry out.

Which isn't quite the same thing, is it?

Could you be subscribing to the career administrator's de facto definition of research as "whatever is funded by grants"?

sorepaw said...

So how *much* of that 29%...

Gabriel Hanna said...

@sorepaw:One professor who teaches a large number of student credit hours every semester, and does so at a relatively low cost per student credit hour?

One professor I worked routinely brought in 7 figures worth of grants every year.

Another professor I know well brings in about the same.

If either of those men taught 5 classes of 300 students apiece for an entire year--which is about 5 times the work load ever expected of any professor--that might be worth the same amount of money they are bringing in in grants.

But a grant is tied to a particular professor, and tution from students is not. One professor who excels at research can personally bring in millions every year, which would only have been given for his work. One professor who excels at teaching does not bring extra students every year. Undergraduates don't have any idea, before they get to college, which particular professors are good. They select colleges based on general reputation and what they can afford.

Could you be subscribing to the career administrator's de facto definition of research as "whatever is funded by grants"?

Um, no. I'm a physicist, I have a great deal of difficulty in understanding how career administrators think, and my ideas of what constitutes 'research' are based on what kind of research I am familiar with.

One of the professors I mentioned oversees about 20 physicists in a applied field and their experiments are very expensive, and largely funded by DoD and DoE. The other I mentioned does biological systems and agricultural research. This isn't Queer Semiotics of White Privilege or whatever. I have no idea what kind of point you are trying to make with your snark about my supposed shared views with "career administrators".