Hacker News Reader

Comment by matte_black | Hacker News Reader

Have school grades ever deflated? Or undergone some sort of correction? No citation needed, anecdotes are fine.

One of the Cal Polys has an engineering program which curves the average grade to a C+ which automatically disqualifies the average from many engineering jobs.

That's similar to Georgia Tech. Undergrad Eng courses are curved to a C+. With the average GPA being a bit below a 3.0.

We used to joke at all the pre med students as to why they chose Georgia Tech as just about every other school had gpa inflation. And grad schools don't take Tech's lower grades into account.

Which was totally fair given how the HOPE scholarship works. Not that I am bitter.

Except now STEM courses are weighted as of 2017 in all GA university system schools:

Effective fall 2017, when calculating the HOPE and Zell Miller Scholarship postsecondary Grade Point Average (GPA), an additional weight of 0.5 will be added to grades of B, C, and D for approved degree level science, technology, engineering, and math (STEM) courses

I think it's good since it doesn't affect their Institute GPA but can help them keep the scholarship. I knew a lot of people who lost HOPE/Zell and the increased financial burden made keeping their grades up that much harder.

That's still stupid. The freshman writing courses at GT are harder than graduate CS at UGA or Valdosta.

Edit: Valdosta doesn't offer graduate CS. But imagine that it did.

A curve distorts grades: they tell you how the person performed relative to their peers not relative to their mastery of the material.

This really depends on how difficult the tests in the course are.

I don't think it is uncommon for tests to be written such that with a student a strong command of the material will not score 100%.

Well it tells you how well they did to the rest of the cohort grade inflation means you cant easily compare candidates across time.

At the high school I worked at some years ago we pushed the margins for letter grades up a little. For example, where a C used to be 70-80% it became 72-82%, up to A which is 93-100%. Whether or not that affected the overall GPA distribution I could not say.

> For example, where a C used to be 70-80% it became 72-82%, up to A which is 93-100%

That sounds something like an older system. I don't remember it entirely, but IIRC, very early in my school career, an A was 93-100% and an F was anything less than 70%.

My grades underwent a correction or two back in the day :)

Yes. At my school we had a real curve. If we all got perfect scores, that's a C for everyone. (More often we all got 15s and they became C's). Teachers who gave too many A's would be benched. We laughed at Ivy League 4.0s which just require a phone call from a parent.

MIT?

My wife went there, and while most STEM grad schools understood that grades at MIT were not inflated, medical schools did not seem to care. One of her friends had to first get inflated grades doing a STEM masters degree before she could get into med school due to her uninflated MIT undergrad grades.

GT. I heard you all got to play with kittens freshman year. In response, GT professors replaced 20% more lectures with exams.

Almost no one at GT does true curves anymore. Just like stack ranking, trying to implement true curves with a non random distribution is a fool's errand and poor knowledge of statistics.

To hell with Georgia!

This is standard practice at a lot of universities. At my university results were scaled based on the class performance in that exam and the performance of everyone in other exams. So if you're all 4.0 students, but the exam is horrendous and everyone only gets 20%, you'll probably end up with a good grade still.

I studied physics. It was possible to game the system to a certain extent. If you took the (subjectively) hardest modules like gauge theory, you could get a high grade even if you only answered half the paper. So if you were smart, but not top of your class, you'd be better off taking the hard subjects because every mark you got was effectively worth more. If you took the easy and popular modules, every mark would count and it might even be normalised down.

Of course this only works until everyone becomes wise to it. Then you're worse off where you started.

That's true, but unlikely. These are not particularly large or popular courses (15/70 people was common). It's not a tactic that would work for everyone, but it would potentially useful if you're a good student, but at a borderline grade.

The reason I liked it was because I hated having to cram for easy subjects because the class average would be so high. I would rather put the same amount of effort in to learn concepts (and have to remember less in the exam) than rote facts.

Reply on news.ycombinator.com