My apologies, I see that I typed the wrong word earlier. I should have said “And, at high school, the exams were written by an external authority, so if the exam questions weren’t released then how would the teachers themselves know what was being *tested?*” My mistake.

If teachers don’t know what questions are being asked, it adds a random factor into the testing results - the teachers who happen to hit on the precise skills that are being tested will presumably have students who perform better than equally-skilled teachers who didn’t manage to guess right.

One important factor in the testing was the relative weights given to different topics. One purpose of the high school exams was to produce a range of results so that the restrictive university courses had a basis for picking students, so the curriculum would cover more material than an average student could learn to mastery in one year. Therefore where teachers put the bulk of their effort mattered for the class’s average exam results.

If the exams created by that external authority are actually testing a defined subject area, then why would the instructors need the exact questions in order to teach that subject?

On the other hand, I have no problem with instructors creating new tests every year and having all the prior years’ tests available. If the students were able to remember how to solve a particular type of problem, that means they understand that type of problem.

Of course, that’s for real subjects that have real solutions. If you get into softer fields, History, basketweaving and such, then I can see you’d have to design the test to whatever you planned to instruct.

As far as attending class and passing exams, the smarter the student, and the more objective the material being tested (say, Algebra), then the less the person actually has to attend lectures to be able to learn the material. Attendance becomes necessary to the degree that the Instructor’s personal opinions and choices will blur or alter the material. There were lots of college courses that I got A’s in when my attendance was… spotty, shall we say.

Yes, in NZ we did assignments. But the assignments were mostly on different things to written exams (eg labwork, project management, research papers). Quizzes weren’t used much. And, at high school, the exams were written by an external authority, so if the exam questions weren’t released then how would the teachers themselves know what was being taught?

Also, how does the USA have quality control on the exam questions if they are not released publicly? In NZ, if there was a badly written exam question it was front page news for the national newspapers. That’s one hell of an incentive for the examiners to write good questions.

Typically, the past exams would contain enough variety in questions to cover the entire course content. Eg for the science exam I was talking about, air circulation was one of the minor topics and wasn’t asked in every exam, *but* we didn’t know ahead of time if a question would be asked on this topic, so the student going after an “A” had to study air circulation and all the other minor topics anyway. (Major topics were things like velocity and acceleration, there would always be questions on those, but what the questions were varied).

[We keep exams for a calendar year, then shred them.]
So there’s a whole extra year of security risk, relative to releasing them openly after the exam.

As for lazy professors, I don’t believe that NZ professors are any better people than Americans (in fact, a lot of the lecturers at my engineering school were American or Canadian). Releasing the exams publicly meant a big incentive for them to write new exams each year, and they did so. Though a number of lecturers did have one of the questions being the same every year (generally worth something like 5% or 10%), as a graduate student I learnt that at least one benefit of this system was that the lecturers could then at the staff club boast about what percentage of their students couldn’t even get that question right. Perhaps the ability to boast was the main benefit. :)

I am really surprised that US academics are giving exams which your students can pass without actually needing to know the material behind it. What do American exam questions look like? And what’s the point of writing exams that don’t test what’s actually covered in the course? (Allowing of course for some variety in the topics covered by the questions each year, and excluding of course those skills, like labwork, that can’t be tested in a paper exam).

In my view, it’s necessary but insufficient to call students to high values: one should also reduce the driving forces that facilitate poor choices like plagiarism and exam cheating.

I just found this blog … it’s a hoot. I love it!

[Spot studying exams to see what questions will be asked isn’t learning the material. It’s learning what will be on the exam.]

A more important consideration is that inspecting past exams tells the students that certain topics, or certain types of questions, aren’t going to appear on the exams, so they don’t have to bother to study those. But it’s pretty difficult to prevent them from knowing that, whether they can see past exams or not.

[In other words, you’re saying that it’s fine if students do not learn the material — just what will be tested.]

I get the impression that your objective isn’t to maximize learning, but to maximize the fail rate. Surely catching cheaters isn’t the ONLY motivation for teachers’ choices.

[You’re an asshole.]

[No, it isn’t. It’s to give students a way to pass the class without learning the material.]

In the courses I did we were expected to not merely write down what was covered in class, but to be able to apply that information to new problems. For example, for my first science exam at high school, the curriculum said we had to cover air circulation, inter alia, and one of the questions in the exam the year I took it required us to apply our knowledge about air circulation to an underground mine. This required being able to apply the principles of air circulation to a new situation (different situations were used in the past exams). I don’t know about you, but I often find I need to practice a new skill several times before I can answer slightly different questions. When you say “People failed exams because they didn’t study the previous ones” that sounds to me like it should just be truncated to “People failed exams because they didn’t study.”

[Studying the exams is not studying the material, nor is learning only the material that will appear on the exams learning the material.]

Serious question - how do students in the USA study for exams?

And how does providing past exams reduce security? Leaving aside unusual situations like a lock-picking course where the professor used his office safe’s lock as an example in a past exam.

[If you write a wholly different exam every semester, it doesn’t. Few do.]

If you keep past exams locked up there is always a risk that a student will get hold of an old copy, so you risk cheaters having an advantage over honourable students. If all past exams are publicly available, the cheaters don’t get any fair advantage.

[We keep exams for a calendar year, then shred them.]

Actually I passed my environmental economics exam despite it being offered for the first time ever that year. I also passed the macro-economics exam despite it being completely different from the preivous years’ ones. So it wasn’t strictly necessary to study previous exams to pass the course.

More generally, if people don’t study for exams, what’s wrong with them consequently failing the exam?

As for security, providing prior exams is sheer idiocy. What’s the point of it, except to give students a way to pass a class without having attended?

Providing practice questions for people studying for the exams. How do students study for exams in the USA? Serious question here.

[We don’t just give (or take) exams. By the time students take the exam, they have taken a whole battery of quizzes, and done an even larger number of assignments. Where are you from?]

I don’t know anyone who passed the exams at university without attending classes. But if someone could do that, either the exam is way too simple or we are talking about a mega-genius.

[Not if the professor gives more or less the same exam every semester — which many do — and makes those exams available for students to study. Some do that, too, but a PhD has never been an indicator of high intelligence or principles.]

I’m really confused here. How is it non-secure to provide copies of old exams, if you are presenting new ones every year? About the only hypothetical scenario I can come up with is a course on lock-picking where the professor used his safe’s lock as an example in an old exam. I suppose there could be similar examples from computer security courses.

[See the last answer. It’s extremely common here for faculty to give more or less the same exam every semester, mostly out of sheer laziness.]

Apart from these scenarios, doesn’t publishing old exams make life fairer for students who don’t cheat? There’s always a chance that a student will get hold of an old exam accidentally. If the old exams are publicly available the would-be cheater can get no advantage over the honourable student by that method at least.

This one puzzles me. In NZ at my university all the exams from past years for all the subjects that had exams were kept at the university library for study reference. Our lecturers told us to do past exam questions as part of studying. Similarly, past questions from high school exams were available publicly, including in our textbooks. And of course the exams were generally similar from year to year. Yet people failed exams all the time, and our marks in the final exams generally varied as we expected from our knowledge of our fellow students.

With this background, I can’t figure out why you think that passing out old exams might not be the best idea.

here’s a clip:
“Many drug and alcohol prevention programs make an effort to build self-esteem on the theory that youth who feel good about themselves will have less desire to use alcohol and other drugs,” says Stephen Small, human development and family relations specialist with the University of Wisconsin-Extension and director of the UW-Madison Center for Excellence in Family Studies. “However, research over the past ten years has found that having high self-esteem is not necessarily the panacea that it has often been made out to be.”