(no subject)

[infryq]

Via Making Light: Paper Mills, and the inability of college grads to form a coherent sentence.
http://chronicle.com/article/article-content/125329/

Reading articles like these makes me ponder ways to counter. If paper mills are plausible because we grade results and not skills, how can an educator arrange grade what she actually wants to measure? How can we address students whose writing abilities are sorely lacking? How can we give ESL students the practice and exposure they (sometimes desperately) need? Some thoughts.

1. Don't grade the 10-page research paper. Or, rather, mark the essay, but call it "training" and have it count for only 10-40% of the course grade. Instead, grade a "debrief" essay written in class, on the same topic, or on a subtopic covered in the larger paper, or critiquing their original essay.
   
2. Do a 20-minute in-class essay on *something* every week.
   
3. During research papers, write me a 20-minute in-class on the research you did for your paper this week, or the edits you made, or the theses you considered. This accomplishes two things: (1) students without writing experience, get some; (2) even if they're having their research paper ghost-written, they have to learn enough to write competently about their topic anyhow, even if it's only in short bursts. If people memorize 20 minutes' worth of essay to write out verbatim every week, I think I'm okay with that, too.
   
4. Recycle student content as editing training, whether as part of lecture or as a stand-in for in-class original content. Lots easier to grade than an essay. Teaches people how to improve arbitrary bad writing, so they can apply the same skills to their own work. Might be particularly effective, because a lot of ESL students I know can read English well enough for research but are crap at writing it. Maybe they can't distinguish things like word order and tense and plurality and spelling and are just working out the meaning of sentences from context? So it's no surprise the examples in the article read like a bag of words. "Critical" reading, reading with the intention of catching errors, is something I think is passed over a lot when you learn a new language, but is crucial for fluent writing.
   

Comments

[perspicuity.livejournal.com]

perhaps part of 4 and 1-2, maybe 3:

have people trade essays (or random draw), and they read that essay, correct it, and then write the debrief/review?

in college, i was forced to take the English 102 course, because it was required, and there was no opt-out unless you took extensive English in HS/pre-college... the teacher had some golden model for what he wanted, but would not provide us the rules/templates for this model, and it felt like being clicker-trained into what he wanted. of course, this is a singular complaint of mine for leadership/teaching in general - if the student doesn't have a clear idea of the outline and outcome via models/templates, the learning process is more about guessing what the rules are, than learning. which seems a harsh waste of everyone's time.

#

[infryq]

Upside: more variety, less possibility for answer key-style cheating;
Downside: more variety... (harder to grade n corrections of n essays than n corrections of 1 essay; more possibility for social engineering-style cheating [c'mon man, give me a break, I need to pass this class]; harder to describe in class the "gold standard" for n essays than the "gold standard" for 1 essay)

Plus, writing a debrief of someone else's essay tests your reading comprehension and pressure writing skills, but doesn't offer any evidence for whether you wrote the long-form paper you turned in. The problem to be solved here is that you want to give people who can't write the opportunity to learn to do so, and you want to have them show you what they've learned in a way that is both robust (or at least resistant) to paper-mill cheating and isn't a PITA to grade. I'm trying to find a system where instead of putting the burden of proof on the prof to show the student cheated, the burden of proof is on the student to show they have fluent command of the content they put in their paper (which, if they wrote it, is trivial, and if they didn't, makes them learn the material anyway, which is almost as good)... and have the new system be politically acceptable.

The other tough thing about teaching the subject (re:the 102 prof) is that the art in writing is often in breaking the rules. Establishing a One True Answer then for the essays you want to see from your students is arguably difficult, especially given the extreme time restrictions in a college class. Clicker training is a lot easier, and could be more responsive; I can imagine he spent a lot less time on lesson plans, less time going over content everybody already knew, more time going over things people were actually struggling with. That said, rubrics and examples of good and bad executions of a principle are going to be more useful to blank-slate students coming in and to everybody as they move on to other classes (as a reference); you just have to balance that with having enough discussion time of the actual content you're studying so that people have something to say. I think a lot of english classes (particularly in college) focus too much on the latter, since it's the fun part. Despite evidence to the contrary profs still prefer to assume their students come into their classes knowing how to write.

[rmitz.livejournal.com]

I would have hated (2), and I did hate it on exams and such. First of all, it really doesn't have much relation to anything anyone will ever have to do OUTSIDE of an academic testing environment. The flip side is that you could switch the focus more to presenting the material that has been learned; this not only is a valuable skill in its own right, but it does mean it's harder to bluff.

I don't think testing is the answer. I think it's going for *more* project focus, not less. Get people involved in primary research more quickly.

[rmitz.livejournal.com]

I left out a key bit: the presentation should be crafted and focussed based on the research and the (previously written) paper. It should be possible for the instructor to tell if the paper was written by the presenter, perhaps if he questions the presenter on topics brought up in the paper that were not covered in the same detail in the presentation.

[infryq]

Again with the scalability, though. Projects and presentations are a bitch to grade. English profs already spend 1.5x or more times on their courses than they are paid for. Presentations are a great way to catch the bluff on content, but how do you verify that someone can write?