Well, as i said, technically it's illegal to do so at this school.

Not sure why they would bell curve, there are people who deserve 12s and they won't get it because those profs apparently curved the mark.

Originally Posted by Alchemist11

Well, as i said, technically it's illegal to do so at this school.

Not sure why they would bell curve, there are people who deserve 12s and they won't get it because those profs apparently curved the mark.

Oh yeah, I remember math 2Z03 was also curved. Protas gave us the formula he used for marks adjustment.

Also, I think what professors do is a mixture of both, so that people on the higher end also get a fair treatment.

Originally Posted by Alchemist11

Err, a bell curve is never used unless they've been doing it illegally.

Adjustments, however, are not illegal and likely made often to reach the average the professors want for the class.

I think what you mean to say is that we never get "curved down" - profs in the sciences, unlike humanities, can't really 'curve on the fly', so to speak. They have to try and write tests that will get them the distribution they want - sometimes the average is too high, but most of the time, it's too low. If it's too high, there isn't anything they can do but write tougher subsequent exams, while they do 'shift the curve' by reducing denominators, etc. if it's too low.

In other words, we get shifted upwards, but we'll never be 'placed differently' on the curve - so the curve will never be stretched or compressed in any way. Well, most of the time - one of my professors explicitly gave us the function he used to modify our marks, haha

All those people who think bell curve is used are stupid for thinking so. In university they don't bell curve because the method can discriminate people within a certain mark range (i.e. their mark can go down). Therefor what the prof would probably do is either use a formula to bump EVERYONES marks up so that to increase the class average, or just change the marking scheme of the exam.

For example. If people did really bad on one question which was worth 30 marks, he/she may reduce the weighting of the question and increase the weighting of another question where the class did relatively better.

None of the courses at mac get bell curved. People just assume bumping marks up/down (which is what typically done at mac and other canadian schools) is bell curving, but actually its not. Bell curving involves having a predetermined distribution, in which students are assigned grades depending on how well they do compared to each other. AKA, only the top 5% get an A+ even if 10% of the class scored higher than 90%.
http://en.wikipedia.org/wiki/Bell_curve_grading

Originally Posted by Mahratta

I think what you mean to say is that we never get "curved down" - profs in the sciences, unlike humanities, can't really 'curve on the fly', so to speak. They have to try and write tests that will get them the distribution they want - sometimes the average is too high, but most of the time, it's too low. If it's too high, there isn't anything they can do but write tougher subsequent exams, while they do 'shift the curve' by reducing denominators, etc. if it's too low.

In other words, we get shifted upwards, but we'll never be 'placed differently' on the curve - so the curve will never be stretched or compressed in any way. Well, most of the time - one of my professors explicitly gave us the function he used to modify our marks, haha

Although I agree entirely, I think it's a matter of semantics in this case.

When people say "bell curved" I believe they're using it as a verb - that is, the professor is performing an adjustment to people's marks in order to obtain a Gaussian curve that wasn't previously there. In that case, I maintain that doing so is illegal and no one does it because people's marks would likely drop while others would receive an increase, and professors are not allowed to lower someone's mark at this school through adjustment like this.

So they aren't compressing or stretching the curve, they are making adjustments, but professors also cannot generate a "bell curve" by adjusting a test or exam.

Originally Posted by Alchemist11

When people say "bell curved" I believe they're using it as a verb - that is, the professor is performing an adjustment to people's marks in order to obtain a Gaussian curve that wasn't previously there.

Yep - profs aren't supposed to do that. Unfortunately, people tend to use "curve" for "translated the curve" as well. When 'curving' occurs in our school, it's basically a matter of the professor setting a new (higher) mean or median, and then using a linear function to modify the other marks in the distribution accordingly. So, this preserves the shape (without compression/stretching) of the distribution - all that's changed is its 'location'. I think that's where the confusion originates.

In light of this, I think the proportion of tests that people regard as 'too hard' over 'too easy' makes rather more sense - a test that's too hard is easily fixed, while an easy one isn't quite as nice to work with.

Np, my christmas gift to you.