Grade inflation, deflation according to the theory of relativity

Colleges and Universities A-Z University of California - Berkeley
1

<p>I've heard these terms used so many times to describe colleges. It seems like the top 3 (HYP) get accused for grade inflation while the remaining top 15 (Ex. Cornell, John Hopkins, etc.) and Berkeley are accused of grade deflation. Why would such policies even exist and if they did exist wouldn't graduate colleges take them into account if the difference was truly that spectacular? What happened to the middle ground, grade neutrality or is there even a term for it? Why won't colleges just grade based on if it's a 50 point test, you lose 2% for each point lost, rather than awkward curves?</p>

<p>PS: I realize none of this has anything to do with Einstein's theory of relativity but I just liked the title so I decided to use it :P.</p>

2

<p>yeah i was wondering that too.</p>

3

<p>If its an easy test, everyone gets 100%. If its hard, everyone does bad. So then grades are based on test design.</p>

<p>The way they measure what you know is by comparing you to other people.</p>

<p>Grade neutrality means average GPA is a C</p>

4

<p>yes grade neutrality would make the average GPA a C, but I would still prefer it since if you study hard and learn all the concepts and applications of those concepts you should do well (at least get an A or B, I mean a professor can’t test you on things you haven’t learned yet right, he may give you a tricky application problem that requires you to think outside the box but it still uses the concepts you have learned, not ones you haven’t right?)</p>

5

<p>Grade neutrality would be by determining a priori what the desired knowledge and capability should be for someone mastering the material, developing tests that have some consistency in rating people against that absolute attainment of the objectives for the course when it was created, and then giving grades based on such a rating. If one section is filled with people who all reach the mastery on an absolute level, they should all get A. If another section is filled with a majority that learn the subject poorly, then most should get C or D grades. </p>

<p>Statistically fitting the two hypothetical sections to the same curve means that section after section, semester after semester, they produce neat distributions of A - F grades, but that the learning done for a given grade varies quite a bit. </p>

<p>Further, the same absolute mastery with the same effort can be rewarded by very different outcomes - due to no fault of the student involved.</p>

<p>Intellectual sloth leads to that choice in pedagogy. Over time, it all smooths out, they might argue, but at the expense of justice or fairness to the individual students, such that the only beneficiary is the college/department/professor who over a large enough sample have consistency in the averages. </p>

<p>As another thought experiment, imagine the course is taught by two professors, one extremely tough and demanding, the other easy and unconcerned. Both produce the same distribution of grades, yet the effort by the students to earn a particular grade band is very different. </p>

<p>Grade deflation usually is asserted for a university that produces very neat bell curve distributions of grades. Grade inflation is usually tagged against a school that has a non-bell distribution with more high grades than the idealized distribution. Neither says anything about the comparability of the same GPA. A school that accepts only the most gifted and hard working students could very well have a huge excess of A grades and be justified in that result, if those students were tested to some absolute scale of required knowledge and skill. Two schools of very different selectivity and average student talent could teach the exact same course in the same way, and if they forced the distribution to a bell curve, would be assigning very different meanings to a C or an A.</p>

<p>It takes considerable effort to produce tests that are consistent and normalized. It takes considerable effort to ensure that every professor and GSI teaches to a consistent standard of difficulty and covers all the material. It takes courage to assign the blame for an underperforming section on the person teaching instead of the students. Bell curves are much easier.</p>

6

<p>wow Rider, that makes perfect sense. Thanks!</p>

7

<p>How do you propose, precisely, to judge whether or not tests are “consistent and normalized”? As I’m sure you’re aware, college tests do not (and should not) have the kind of simple problems where you can judge the difficulty from just looking at it. When you’re writing one essay for your history final, by what possible standard can anyone judge how difficult that one prompt is? </p>

<p>And yes, we all know the problems with curves in theory. In theory, a student who knows the material really well could fail because his class was just incredible. In theory, a student who barely knows a thing could get an A because the rest of his class was even worse. But I have never heard a single credible story about that happening. The people who whine about how the curve screwed them over are usually not nearly as smart as they believe.</p>

8

<p>I agree with amarkov. In classes with over 150 students, the curve is going to work out. You aren’t going to get 150 geniuses or 150 slackers all of a sudden - the class of 2015 isn’t going to be significantly better or worse than 2014’s (2015 might be very different from 1985, but we’re talking about curves relative to students in the same decade or so). </p>

<p>Curves can be hard in classes that are small (approx. 25 students) but professors are usually reasonable graders. You’d have to really do poorly to get a C or lower. Curves are also hard when a class is targeted at a advanced students (there’s always a class or two that’s not mandatory but everyone knows it’s challenging) in which case you do get an influx of smarter kids. Still, there’s consistency from year to year.</p>

9

<p>while amarkov and bsd arguments seem sensible that the curve will generally stay the same, by saying that you assume that the average intelligence and respect and value towards education that a human possesses doesn’t increase, and I must disagree with that logic. I’d like to think that if variable X was time and Y being the sum of human intelligence & potential then the limit of Y as X approaches infinity will be infinity as well, thus in my opinion the idea of a curve staying the same shouldn’t stay the same if we truly are progressing as a species. Although this statement is extremely theoretical and probably doesn’t play out or show in practice in such a small span of time.</p>

10

<p>hmm, what if there was a crack in the bell.</p>

<p>im completely serious here.</p>

11

<p>But why does it matter if your grades match up to those of, say, 20 years ago? You aren’t competing with the people of 20 years ago.</p>

12

<p>That is the point - is a grade an olympics prize based on competition with the other athletes of the moment, or evidence of acquisition of an education? If the engineering students of today are as qualified as the engineers 20 years ago, and master just as difficult a set of subjects, they should have the same GPA. The context of your ‘competition’ is lost years later when seeking jobs, where GPA is compared against other job seekers of different ages, not by comparing you to the field of ‘athletes’ against whom you ran in a particular meet.</p>

13

<p>im so confused.</p>

14

<p>

</p>

<p>Erm… no. I assure you, when you are 40, nobody will ever ask for or care about your college GPA. Who cares how good of a student you were half of your life ago?</p>

15

<p>I’m 26 and nobody asks me about my Berkeley GPA. Whew!</p>