Thursday, April 08, 2010

Outsourcing Grading

I have to admit getting a good laugh from this.

Apparently, there's a company that employs people in India with graduate degrees to grade papers for American professors. For twelve bucks a paper, they'll give not just a letter grade, but comments. The idea is to free up faculty to focus on instruction (or, more accurately, research), rather than grading. It also saves the university money, since outsourcing the grading allows you to run classes at much larger sizes.

From the comments to the article, you'd think that this had never been done before. You'd think that professors have always done their own grading, and that the grading was a form of deep examination of each student's soul, resulting in unparalleled insight and bonding.

Um, no. And I have the scantron invoice to prove it.

Those of us who went to grad school in the paper-writing disciplines, such as my own, probably did some time as a T.A. or grader for a large lecture class. (I did both.) Students signed up for a class taught by The Sage On The Stage, only to have their grades determined by twenty-three-year-old grad students who had never taught classes of their own. And this was in an auditorium setting without even an internet hookup; the most advanced technology in the room was a microphone. The university in question was a flagship campus of a state university you've heard of, and that has a respected name in American higher ed. The practice was of long standing, and was generally accepted. Students didn't even expect the Sage to recognize them individually, and he didn't.

As near as I can tell, the breakthrough this company offers isn't the separation of grading from teaching, since that has been standard practice at the university level for decades. Nor is the breakthrough the new involvement on non-Americans; anyone who has spent any time on a major public university campus lately can attest that foreign students are nothing new. No, the only real change here is location. Instead of paying local grad students to do the scut work, now we can use email attachments to pay grad students in India to do the scut work.

Of course, one could always object that grading shouldn't be considered scut work, that it should be located at the core of the teaching enterprise. But at the university level, that ship sailed decades ago. Bringing it up now seems a day late and a dollar short.

By contrast, at my cc, we cap English classes in the low twenties, and social science classes in the low thirties. Professors do their own grading. And we charge less. Not coincidentally, the success rates of our grads who transfer to the state university is actually higher than that university's 'native' freshmen. As a result, our prestige is...wait for it...lower.

(Bang head against table here.)

If you're willing to grant the existence of certain basic irrationalities, though, the whole email-your-paper-to-India scheme could actually offer some benefits.

For one, it eliminates the potential conflict of interest when professors do their own grading, and are judged -- one way or another -- by pass rates. Teaching to an external standard can change the role of the teacher from 'judge' to 'coach,' with beneficial results. External exams have long existed in some fields -- Nursing and law come to mind -- and they can serve as reality checks on grade inflation. 'Blind' external graders can also put to rest the usual charges of race/gender/personality bias in grading, since someone in Bangalore reading essay 54789 has no flippin' idea who you are.

For another, though -- and I actually like this one a lot -- it offers the very real possibility of finally starting to shrink graduate admissions in badly-flooded fields. I've heard of graduate programs trying to do the right thing by restricting admissions for a year or two, only to discover abruptly that they need the cheap labor. So they open the floodgates again, and the reserve army of the overcredentialed just keeps getting bigger. But if they don't actually need as many T.A.'s or graders, they can actually reduce admissions and keep them down. If we're ever going to get a handle on the exploitation of adjuncts, we simply have to start reducing the supply. Until now, there often didn't seem a realistic way to do that; now there is.

(Of course, one could object that this is simply displacing our problem onto India. But I'm willing to let India worry about that.)

I don't buy the 'quality' argument against it, either. If radiologists in India can read images, and programmers in India can work on developing and fixing incredibly sophisticated software, then surely some smart folks in India can handle some freshman comp papers. Seriously. Other information-based industries have endured outsourcing without the quality of the work suffering. Given the inarguable indifference with which our large universities have handled undergraduate teaching for so long, to suddenly get huffy and puffy about standards is disingenuous at best.

The simple truth of the matter is that universities have engaged in a bait-and-switch with intro undergraduate classes for a long time, and have built an entire economic model on it. This may be a case in which following the model to its logical conclusion actually prompts looking more closely at the entire enterprise, which strikes me as a good idea. In the meantime, we'll keep running small classes with real faculty who actually do bond with their students, and doing it at a fraction of the cost. If folks would sneer at us a bit less, I'd be much obliged.