<p>$10 per hour is the going rate for undergraduates here. I can imagine the local copy shops at exam time (6,500 of them at Harvard), 20,000+ at a state u.</p>
<p>As for scanning, I've been scanning materials for my own use. A very significant expenditure of time is involved. I cannot imagine scanning exams, response papers, papers for 500 students or more.</p>
<p>My S has had a variety of exams, most of whom are not only more thoughtfully designed to encourage actual learning and, incidentally, make cheating less likely (as well as save copying, etc...). These involve take home exams, devised in such a way that students have to synthesize materials instead of merely regurgitating them; open book exams; exams in which a list of questions are distributed in advance and the actual questions are not known until the day of the exam. This forces students to review all the materials. The exams should be graded and commented on in such a way that a) they are helpful to students as they seek to discover what they did right or where they could have improved; b) make post-hoc cheating more difficult.</p>
<p>
[quote]
I can imagine the local copy shops at exam time (6,500 of them at Harvard), 20,000+ at a state u.
[/quote]
</p>
<p>As I said, there are plenty of services that specialize in copying documents. If there are 6500 undergraduate students at Harvard, and they each write 50 bluebook pages at exam time, that's 325,000 pages of documents or about 50 banker's boxes worth. I suspect that there are plenty of services that are capable of handling 100 boxes per day.</p>
<p>
[quote]
As for scanning, I've been scanning materials for my own use. A very significant expenditure of time is involved. I cannot imagine scanning exams, response papers, papers for 500 students or more
[/quote]
</p>
<p>With modern equipment, scanning is just as fast as photocopying. I have a $500 HP all-in-one machine, and I hardly ever photocopy anything anymore. I just scan it and then print it out when I need it.</p>
<p>Of course a document service would have much faster and more reliable equipment.</p>
<p>
[quote]
$10 per hour is the going rate for undergraduates here.
[/quote]
</p>
<p>You don't have to hire undergraduates to staff a duplication service.</p>
<p>Yes, scanning is as fast as photocopying, which is to say, not as fast as one would wish. Try photocopying 800 bluebooks (that's for one course). I have a 3-in-1, and make use of all three functions.</p>
<p>The excerpt I quoted from the Harvard handbook does not seem to suggest that copying exams and papers is a concern there. I wonder why it is such a concern at other institutions. Is cheating that rampant elsewhere? Are profs not bothering to make comments? Are they so reliant on multiple choice exams and other exams where it is easy to cheat?</p>
<p>A dedicated 2-sided scanner is an entirely different level of performance from a 3-in-1. We've gone over to scanning our files in house once a case is over, and the sheets pretty much fly through the scanner. Of course, the bluebooks would have to be disassembled first. I do, however, think that Marite is right - a simple unique mark (initials? stamp?) on each page- which also serves as evidence of the fact that all pages of the exam were actually read by the grader - is a good, and adequate, solution.</p>
<p>Back in the day I had a TA job grading exams (short essay answers) for an undergraduate business law class when I was in law school. It was a surprisingly simple task, which required little expertise to perform with consistency. A less concrete subject might well be more subject to individual grader's variables, however.</p>
<p>
[quote]
Yes, scanning is as fast as photocopying, which is to say, not as fast as one would wish. Try photocopying 800 bluebooks (that's for one course).
[/quote]
</p>
<p>With modern eqipment, 800 bluebooks could be copied (or scanned) in something between 5 and 50 man-hours. The cost would be negligible compared to the tuition for the class. The storage space would be negligible.</p>
<p>I don't know how to make this any clearer.</p>
<p>How much do you think an outsourcing company would charge, per page to scan a large number of bluebooks? Do you agree that 10 cents per page is a reasonable estimate for the most you'd be charged?</p>
<p>How many bluebook pages does each student create during the course of the semester? Do you agree that 100 pages is a reasonable estimate for the maximum? (And note that 2 pages can be easily scanned at a time).</p>
<p>If you agree with my estimates, that's 10 bucks per head, MAX. Chump change compared to $5000 in tuition.</p>
<p>As for storage, do you agree that it costs less than 1 dollar per gigabyte to store data now?</p>
<p>Do you agree that 1 gigabyte of memory can be used to store at least 2000 pages of scanned materials?</p>
<p>If you agree with all of the above, then you have to agree that all of a student's bluebooks could be scanned and stored for an amount of money that's negligible compared to the cost of tuition.</p>
<p>Perhaps you're right about the scanning. I'd have to look into it (I estimate, by the way, that my S produced about 130 pages per course last semester). </p>
<p>But it seems to me that there are easier, less time-consuming, and more pedagogically sound ways to go about ensuring that there is no tampering with exams after they're graded. I've provided some.</p>
<blockquote>
<p>Obviously, your S's college must not have come upon this way of cheating much, if at all. But if a school like Yale has had problems...<<</p>
</blockquote>
<br>
<p>Of course, I know that our kids are at the same school (I think they've had a class together actually--your S was one of the "really smart kids"). I guess I just didn't want to point out that Harvard either doesn't have this problem (or isn't aware of this way of cheating) as Yale seems to.</p>
<p>Until the issue becomes a problem, I doubt many schools will go through the problem of scanning all blue books.</p>
<p>I suppose one could draw lines through the answer with a red pencil which would smear if you tried to erase it...</p>
<p>
[quote]
But it seems to me that there are easier, less time-consuming, and more pedagogically sound ways to go about ensuring that there is no tampering with exams after they're graded. I've provided some.
[/quote]
</p>
<p>I checked back and found this proposal of yours:</p>
<p>
[quote]
A much better scenario is for comments to be made in ink on each exam so that if a student altered the content, the comments would no longer look appropriate.
[/quote]
</p>
<p>The trouble with this approach is that in situations where the grader makes a mistake, the comments won't look appropriate either. So it may be difficult to tell the difference between a mistake by a grader and cheating by a student.</p>
<p>But if the grader makes a mistake then it will be all the more reason for a grade change! And graders, whether TAs or profs, DO make mistakes and these mistakes are the basis for a grade change!</p>
<p>In a problem set, for example, graders often deduct points for incomplete or wrong answers and circle the answers as they write down how many points are deducted. At the end of each set, they also write the total earned. In essays, graders make comments of all kinds--about the writing, about the argument, about the evidence or lack thereof, in margins, and at the bottom of essays. TAs are advised to summarize the argument, point out the strengths and weaknesses and suggest improvements. All of this would make it more difficult to drastically alter an exam. We're not talking about changing a line or two, but making enough alterations to warrant a grade change.</p>
<p>LMNOP:
Thanks for the kind words about S. I did not realize they'd been in a class together!
I was thinking that with scanning, one would also need to identify each item for retrieval. When I scan a document, I have to give it name. That would be another time-consuming factor. You are right that, unless colleges think the problem is rampant, they will not want to go to the trouble of photocopying or scanning. As far as I know, the problem they continue to be concerned about is plagiarism-- and there are published guidelines about intentional or unintentional plagiarism.</p>
<p>
[quote]
But if the grader makes a mistake then it will be all the more reason for a grade change! And graders, whether TAs or profs, DO make mistakes and these mistakes are the basis for a grade change!
[/quote]
</p>
<p>Of course. That's the whole point!!!</p>
<p>How do you tell the difference between a bluebook where the grader made a mistake and a bluebook where the exam was altered?</p>
<p>Here's an example (and this really happened to me): On one exam, Problem #2 contained 4 parts: a, b, c, and d. I answered all 4 parts. However, the grader didn't notice that I answered Part C. So I got zero points for that part. </p>
<p>If I showed the bluebook to the TA and demanded points, how can I prove that Part C was not added later?</p>
<p>Or here's another example: Suppose that while writing down a solution in a bluebook, I take a wrong turn, but then go back, cross out part of my work, and proceed. I don't get the final answer, but my work should entitle me to partial credit.</p>
<p>However, the grader (who is in a hurry), doesn't notice what I did right and gives me zero points. If I show the bluebook later and demand partial credit, how do they know whether the changes I made were done while taking the exam or after the fact?</p>
<p>
[quote]
When I scan a document, I have to give it name. That would be another time-consuming factor.
[/quote]
</p>
<p>A numbering system would be quick and easy to use. I imagine that commercial grade scanning equipment could number files automatically. </p>
<p>Ideally, each bluebook should be given an identifying number that is assigned to the student before the exam starts. This allows for anonymous grading and would also let all the bluebooks be scanned into a single computer file for the class.</p>
<p>I understand your concerns. And things like that DO happen. But in my experience, there is less of an adversarial relation between student and TA or student and prof. I would expect that if a student was consistently doing acceptable work, the prof would be more likely to give that student the benefit of the doubt when the student claimed that the TA had overlooked part of an answer.</p>
<p>Believe me, no system is foolproof. I can share stories of wrongly graded exams, students challenging grades, etc... None I know of involve exams that were altered after they were graded.</p>
<p>My chief concern, however, is that all the protective measures proposed are all CYA measures and have little pedagogical purpose. It is much better to devise exams that are harder to cheat on and advance learning at the same time.</p>
<p>
[quote]
But in my experience, there is less of an adversarial relation between student and TA or student and prof. I would expect that if a student was consistently doing acceptable work, the prof would be more likely to give that student the benefit of the doubt when the student claimed that the TA had overlooked part of an answer.
[/quote]
</p>
<p>And there are also situations where the student is suspected, or even accused, of cheating. And there are also situations where the student does cheat and gets away with it.</p>
<p>Why not spend a negligible amount of money and remove all doubt?</p>
<p>
[quote]
Believe me, no system is foolproof.
[/quote]
</p>
<p>That's true, but it's not a reason to not put controls in place. </p>
<p>
[quote]
My chief concern, however, is that all the protective measures proposed are all CYA measures and have little pedagogical purpose.
[/quote]
</p>
<p>What about requiring students to write exams in ink? Are you opposed to that? It's just a "CYA measure with little pedagogical purpose" no?</p>
<p>What about paying a proctor to sit in the exam room with students? Isn't that just a CYA measure with little pedagogical purpose?</p>
<p>I've never claimed that all practices MUST have pedagocical purpose. I've merely pointed out that those that do exist seem to satisfy Harvard; and there are ways that enhance both integrity and learning. You, on the other hand, seem to be keen to avoid discussing how to devise exams in such a way that cheating would be less easy. I wonder why.</p>
<p>But enough of this discussion. If you want 100% security, I doubt you can achieve it.</p>