1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

A thought experiment

  1. Dec 21, 2008 #1
    I just thought of a little thought experiment for those considering colleges for undergrad

    Suppose there is a HS senior student A (2100+ SAT ,3.8+GPA,research,science fairs) decides to go to a top 7 school not particularly known for grade inflation ie. Caltech,Cornell,MIT. Choose Caltech as example . All classes regardless of the students composing it have a grade distribution centering around a B especially for schools not known for grade inflation. Assume student ends up on the C side most times but not in all classes composed of other students that are IMO,Intel Sci,Putnam high scorers and such. Senior year comes around and student A has 2.3GPA and some research no publications. This is followed by 820PGRE and assume 1400+ general GRE. Very likely student A could have 3.0+GPA in a different school. I suppose A could work even harder but then this would displace a student so that he gets the C in which case make him student A ie someone has to get a C at least some of the time otherwise it would be perceived as grade inflation which goes against the initial assumption.

    Student B is a slightly above average student (3.4GPA 1870SAT no research science expt).
    Attends school around 100(University of Kentucky, Iowa, Hawaii). Graduates with some research experience and a 3.7 GPA and 800 PGRE.

    Both apply to school X in top 10 like Columbia, UIUC. Who is more likely to get in?
    Interestingly enough some schools have GPA requirements to even apply like UCBerkeley so student A is immediately disqualified.
  2. jcsd
  3. Dec 21, 2008 #2

    I was basically student B and I didn't get accepted to several schools much worse than the top 10.

    Anyway, as you said, A is likely dead in the water with that GPA, but the professors at top-10 places, at least within their subfield, all know each other professionally; they might not have any idea how to interpret the letters of recommendation for B at all though. Both of these students sound like good candidates for a top-30 program, but they shouldn't get their hopes up about doing a Ph.D. at Columbia.
  4. Dec 21, 2008 #3
    UCI and USC which are not top 30 have GPA requirements, even some companies have GPA requirements like defense companies.
  5. Dec 21, 2008 #4


    User Avatar

    2.3 GPA is very unlikely to get accepted to any grad school, let alone somewhere in the top 10.
  6. Dec 21, 2008 #5
    It was an extreme case, I think a more fitting comparison is for a grad school around the 40-60 range or that range where A and B are both likely to be admitted ie almost like the situation where only one can be admitted. Whole point being somebody is going to have to get the short end of the stick in A's case without something like grade inflation ie. you could theoretically have a physics undergrad program that only accepts five people works them like grad students but grades on a curve unless the people on the short end of the stick score extraordinarily well on PGRE they are most likely screwed its as if they studied in Bangalore the PGRE stakes are raised higher for grad schools outside top 30. Just something to consider.

    Kind of my point in conjunction with the fact that student A could of very likely received a 3.0+ GPA at student B's school and therefore been accepted into a grad school instead of none regardless of ranking and quite possibly one in top 30.
    Last edited: Dec 21, 2008
  7. Dec 21, 2008 #6
    I strongly doubt that a C average student at Caltech would be significantly above mediocre (around 3.0) at a more lax state school. This is a myth mostly believed by mediocre students at top schools.

    As it is, I think student A squandered a lot of opportunity.
  8. Dec 21, 2008 #7
    Ive met a transfer student who told me about going from 3.5 Pitzer Math/Physics to 2.4 Cornell EE and a significant drop in social life. The transfer students for any of the schools like in A usually are a good example of the reason why this "myth" has validity because they get to experience both but also because courses are curved to some extent at 90% of colleges. There is also the fact that different books are used at different colleges grades will reflect the difference between being taught from Serway and Kleppner/Kolenkow for Mechanics(the two are undeniably not equal in terms of difficulty) or Purcell for E&M.

    Ironically since grad school is graded in a more favorable way regardless of institution, the undergrad environment is not reproduced in grad school.
    This kind of goes against the point that someone has to be on the lower end regardless. If all the students tried their hardest somebody would still need to take the C otherwise it would be considered grade inflation.
    Last edited: Dec 21, 2008
  9. Dec 21, 2008 #8


    User Avatar

    That's absolute ********.
    This whole thing is a myth, and it isn't that much harder to get good grades at a top 10 university than it is at a state university (or rather, it's equally hard everywhere). And getting a C at MIT doesn't show anything except that you failed to grasp most of the material, and should never be understood to mean that you know more than an A student at a state school.

    Better education doesn't mean harder tests.
  10. Dec 21, 2008 #9
    Its not about education or even the tests, its about who is also taking the test. For example imagine if all PGRE were scaled based only on PGRE taken outside the USA, this would lower every test taker score. Thats a pretty clear example of what im referring to.
    There is also the fact that a decent amount of students at top 10 schools have taken courses at state schools as HS students and done well in them to support their apps. The boards like college confidential have tons of HS seniors who have taken even Linear Algebra and physics classes while in HS and have received good grades in them.
    Last edited: Dec 21, 2008
  11. Dec 21, 2008 #10


    User Avatar

    What do you mean, it's not about education? Why do you think you're graded, if not to get a measure of how much knowledge you have? The whole point is to get educated.

    And it's not about who is also taking the test, since most universities (especially the top ones) don't scale their scores in either direction.

    Also, just because a HS student got a good score in a freshman linear algebra course at a state university doesn't mean that this state university's fourth-year QM final exams are any easier than those at a top 10 university.
  12. Dec 21, 2008 #11
    Thats why it was assumed the school did not have grade inflation ie not everyone can have an A therefore someone has to get a C or B which means the people you are taking a test with does matter .
    Fourth year grades dont matter for grad school admission especially since your GPA is primarily first/second year grades so it is relevant an A in a lower division courses is received in HS.
  13. Dec 22, 2008 #12
    Sorry to interrupt, but this line just seems illogical to me.

    So, you are saying that it weights more an A in Calculus 1,2,3, Linear Algebra, Abstract Algebra, as a freshman/sophomore student rather than a C in Real Analysis 1,2,3, Abstract 2,3, Advanced Lin. Algebra as a Junior/senior student to the grad school admission cometee?

    I think it is completely the reverse. THe more advanced courses one takes as a junior or senior, weight much more than the ones u took as a freshman.
  14. Dec 22, 2008 #13
    I didnt say anything close that. I just said that when calculating your GPA (the number) for grad school freshman/sophomore grades contribute a greater amount because the amount of courses you have taken unless you take twice as many courses per term your junior year or your university uses a system where Real Analysis A counts as a 5 or 6 when calculating your GPA like AP classes do in some schools in High School.
  15. Dec 22, 2008 #14
    Actually everyone can have an A without grade inflation - it's not like if everyone completely understands the material, the ones who have slightly less mastery get a C for A-quality work. I've seen tests from MIT; the material in the advanced classes wasn't significantly different from what we did at my unknown state school. If you got a C on it there, you would have gotten a C on it anywhere.

    So, yeah, a student with a C average, even from a top school, is most probably not adequately prepared for graduate work at a top school.
  16. Dec 22, 2008 #15
    You pretty much missed the point of the last few posts ie the actual test administered is not as important as the people your taking the test with. This could cause 85% to be an A at one school and 78% to be an A at another school and on and on. The exams can be different look at Cornell or Caltech. If only the material mattered then one can say that honors classes are not different from regular classes and all classes are created equal ie there is no point in administering the PGRE or GRE or any type of standardized test. Or you could think about the foreign student PGRE example. The PGRE is not the greatest measure for physics but you can surely say that someone with a score in the 400/500 PGRE is not prepared for grad school regardless of GPA. It is not unheard of someone with 3.5GPA+ having a 400-600 PGRE score domestically that has to say more than ones GPA which is not standardized. The fact that someone can get 400-600 PGRE with a 3.5+GPA supports my point because the PGRE is standardized. A person with a score above 650 in PGRE and any GPA in any school is better prepared for grad school than a person with a high GPA and 400-550 PGRE because the difficulty of ending up in the 400-500 range or 900+ range. The fundamentals are important and there may not be a difference for score +/-100 but there is between 500 and 700 or 600 and 800 but probably a bigger difference for the 500 to 700 case because one can do only the mech and e&m questions to get above 500 in PGRE.

    I hope your not taking it personally. I am just constructing what really amounts to a worst case scenario. That's why is was more of a thought to show that people dont have to be so obsessed with prestige or taking honors courses when they dont matter nearly as much as they did in HS where AP classes would give your GPA a 5 or the fact that you really should take AP classes because it is pretty much required if offered whereas in college you dont have to take grad classes to get into top 10. I also have been debating about the bailout.
    Last edited: Dec 22, 2008
  17. Dec 22, 2008 #16

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    Well, spot, I think this is starting to look a lot like "I have this friend, see..."

    While I understand that you probably don't like what will.c is telling you, I don't think that means he missed the point. It means his experience is different than...um...your friend's.

    For whatever it's worth, my experience is similar to will.c's. An A student at a Big Ten school may be an A- student at an Ivy, but C students are C students pretty much wherever they go.
  18. Dec 22, 2008 #17
    I guess this is one of those fence issues you either stand on one side or the other like the SAT or any type of test that standardizes some are going to say its rubbish others are going to defend it. The whole concept of standardization and standardized test lies at the heart of the issue and some believe they are pointless others dont. Its also becoming too anecdotal. I guess i should of figured as much before posting I could imagine if this was posted in college confidential boards you would get totally different responses in the Princeton board than the SUNY board. Its the internet people.

    Thats pretty spurious. I could be anyone from a top 10 student with 3.4 and 850PGRE bitter about being in 15th school to a senior HS student justifying state school choice or just a college student doing the same or just anybody who is bored its the internet and I could assume anyone who disagrees with me is a 3.5+GPA state school grad with a low PGRE.
    Last edited: Dec 22, 2008
  19. Dec 24, 2008 #18
    I'd like to see how your little gedanken turns out. You've now got me worried that doing well in an aussie place that clearly doesn't compare to top 10 US unis will not be enough to get into a good grad school.
  20. Dec 24, 2008 #19
    I would like to point out, having skimmed all responses, that A is a total idiot and clearly had a ton of potential, but squandered a huge amount of it and did nothing with his time. Thus, he isn't likely to make it to the school he wants, and will be sad. Which may cause worse grades.

    Sad. Too may A's out there.
  21. Dec 25, 2008 #20
    I just graduate from Berkeley and I do not remember one instance when the curve was used against any math student. Average math scores were around 65-70%, usually 80-85%+ is an A. So I highly doubt that the curve hurt most students, if everyone did well enough to get 90% and above we would all have A's.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: A thought experiment
  1. Thoughts on Msc (Replies: 1)

  2. Thoughts on C++ (Replies: 15)