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!

What exactly do exams test?

  1. Sep 28, 2013 #1
    Over time I have realised that exams do not actually test intelligence (I can't define intelligence here but please bear with it).

    I'll take a specific example to explain this: -
    There are two people are I know, one of them always used to score more in math and the other less somehow.
    But the other person was clearly better at math. I knew it. When I used to talk to him, he used to give me wonderful insights and ideas. His abstraction ability was brilliant and he was a genius.
    The first person however focused more on the details and problem solving.

    So for me exams only test abilities like attention to detail, training your brain to solve specific problems etc. But clearly understanding the subject more deeply, getting insights etc. are more important

    By the way - the question is that if I am right how should we take exams and if I am wrong then why?
  2. jcsd
  3. Sep 28, 2013 #2


    User Avatar
    Science Advisor

    If I had a dollar for every time a friend of mine made this exact complaint after getting a bad grade on an exam, I would be one rich man.
  4. Sep 28, 2013 #3


    User Avatar
    Gold Member

    You are right of course.I don't think there would be any way to test intelligence.For example,even if you don't understand the concept,you could still score full marks just by rote memorizing in subjects like biology. :(
  5. Sep 28, 2013 #4
    I'm not complaining but I'm very curious about this thing. And it's not even about me. Seriously there is a friend of mine who is a math genius (pretty much like sheldon cooper). I'm perplexed how he scores less than some of the guys who I believe are not as good as him.
    This makes me question on what exactly do exams test?
  6. Sep 28, 2013 #5


    User Avatar
    Science Advisor
    Gold Member

    Exams are usually designed to test the material that has been presented. A good exam will also serve as a diagnostic tool, showing specific deficiencies in the preparation of the student, which can then be worked on.

    For example the student may understand the problem, but may use an incorrect method to solve it. Or may be careless and omit some aspect that is crucial to obtaining the correct answer. Or the student may be slow with the techniques, and simply run out of time. Or perhaps the brilliant student is not interested in this material, and so does not bother to study.

    If you want to take an IQ test, see a psychologist. Exams are not IQ tests ... they serve only to tell how well you are doing in the current course, and should be measured against the course syllabus.
  7. Sep 28, 2013 #6
    I've heard this same thing probably literally hundreds of times.
  8. Sep 28, 2013 #7


    User Avatar
    Gold Member

    The world is divided to two parts:
    1. Those that need a frame to learn stuff, like school or a degree.

    2. Those that don't need any frame to learn stuff, and can learn by themselves.

    I reckon your first friend belongs to 1. and your second to 2.

    I myself got good grades along not so great.
    The tests check that you give the correct answer, for example I took in 2012 a course in Analytic number theory, in one of the HW the lecturer gave he expected a specific answer from the students, but I gave a different and correct proof for the claim. The lecturer wrote on my assignment that that's not expected answer he wanted to hear though it's correct. In graduate courses not all students have the same background, so it's a problem.
  9. Sep 28, 2013 #8


    User Avatar
    Homework Helper

    Exams are necessary because there needs to be an end point to study. Any course will have a syllabus and the end point is to know the syllabus. Exams are there to confirm that you know the syllabus. If exams fail to test everything they should, we need better exams, but never no exams. The more comprehensive exams are, the better, is my view.

    An argument against exams is that they are stressful and performance is affect by the stress levels. But I have two points against this. If it is stressful, performing when the situation is stressful is a great examination of one's ability or knowledge. My second point is that there is no reason for exams to be stressful, if one knows the syllabus, one will do no worse than anyone else, so there is nothing to worry about.
  10. Sep 28, 2013 #9


    User Avatar
    Science Advisor

    Sometimes exams just test how well you can take an exam so I'm not necessarily disagreeing with you. In my experience, physics prelims have basically just been harder versions of homeworks which I think is fine because homework is there to test how well you know the material. You can't claim to have any insight or understanding of a certain topic in physics if you can't do the homework right?
  11. Sep 28, 2013 #10
    Exams are supposed to be an indicator of the class's (or an individual's) understanding of the material. However, sometimes an exam fails to fairly represent the material that was taught in the class. In some cases the students simply study the wrong things or have studied the right things incorrectly. In other cases the professor (or test writer) misjudges the difficulty of the material (or maybe it's supposed to be difficult).

    I have had the opportunity to take a large number of exams throughout university. I find that if I'm comfortable with the material I am still nervous about the exam - but I think this nervousness helps me concentrate on not making the mistakes I would normally make (in other words, being slightly nervous lets me focus on not making silly mistakes like addition errors etc.). However, when I'm not so comfortable with the material (which, I'm sad to say, has happened more often than it should) I'm much more nervous - to a level where I begin to doubt myself while solving (basic) problems.

    However, I have had some exams that were a poor representation of the material we were responsible for. If the class average is ridiculously low on a given test, I think it's safe to say that the test was too difficult for that level (even if there were a few individuals who scored high). For example, in my first year of study the third test in calculus (the dreaded integral test) was by far the most brutal test Ive ever taken. I scored 23% on this test while the class average was 17%. Was I as prepared as I could have been? No, definitely not - but the fact that the average was so abysmally low should be an indicator that the test was too difficult for that level.

    Exams are not meant to test an individual's intelligence, but rather to test their understanding of the material being covered on a test.
  12. Sep 28, 2013 #11


    User Avatar

    Staff: Mentor

    In the US, it's also a problem in undergraduate courses, because (unlike some countries) we don't have a national secondary-school system with uniform standards. Even within individual states, which can set standards, quality of education varies widely in different local school districts.
  13. Sep 29, 2013 #12


    User Avatar
    Homework Helper

    It depends on how the exam is written. Take a look at the second question discussed in this post. Does that test only attention to detail and solving specific problems, or does it test deep understanding and insight?
  14. Sep 30, 2013 #13
    Exams are always biased to certain kinds of people though they may be brain wracking, conceptual, testing the proof skills and so on.....No exam is a best exam. Every exam has both pros and cons. Take the example of a written exam(whatever may be the standards). Here the handwriting and presentation of the student plays a major role. A student may know the subject properly but he may not have good handwriting. It may result in low grades.
    In my region in India, high school physics means memorizing some stupid definitions, writing the sentence as in textbook and so on.Thus it tests only rote memory. India has tough entrance exams which are conceptual like JEE ADVANCED. Now it is speculated that the students who crack the exam just mark the objective question by hook and crook...Whatever the exams are, they may always be biased.
    Some of the factors which play a major role in exams are;
    1.Availability of material. A rich student may get materials and resources easily.
    2.The ability to manipulate the system. A person who works only for grades will get the grade by hook or crook. These may include memorising definitions, solutions(as is done in my region), memorizing the techniques(if same problems are asked) and so on.
    3.Handwriting(incase of written exams).
    4.Evaluation(Human beings are varied in behaviour and same applies to the evaluation bu teachers).
    5.Method of testing. Let us assume that the exam is conceptual with lot of problems. Both well prepared student, who knows the concepts and another student who only knows the techniques are present. The exam turns out to be easy with plug and chug problems. Thus both of them may get same marks.

    Exams come and exams go but learning should remain the same. I have made up my mind on ancient Indian wisdom which states "Do your work(learning the concepts and attempting the problems) sincerely without expectation of fruits(Grades, marks,jobs and so on)"...
  15. Oct 4, 2013 #14
    You should not take exams and other junk (such as IQ) as an 'absolute measurements'. It is not even an approximation measurement. In other words, exams are for machines, but you are not a machine.
  16. Oct 4, 2013 #15


    User Avatar
    Science Advisor
    Homework Helper


    Or, more accurately, the course has several objectives that it sets out to teach and the tests are a way to measure whether the student learned those particular objectives.

    I would be surprised to see this occur in a math course, simply because it seems like it would be fairly easy to define the objectives and understanding and performing should be pretty closely correlated.

    It can be fairly typical in "fuzzier" courses, especially when the test is written last. Too often, the test winds up being an assortment of trivia questions pulled from the text instead of focusing on whether the student met the objectives of the course.

    I never had that sort of problem with math tests, but there were lots of high school courses where I'd do a lot better on the semester final exam than I did on the midterms. Just due to the amount of material a final exam covered, it was much more likely to test only the objectives vs the "stuffing" that would occur on midterms just to make them a decent length.
  17. Oct 4, 2013 #16


    User Avatar
    Science Advisor

    I would say both are important. If I'm hiring someone to build a bridge or do my open heart surgery, I'll take the guy with problem solving ability and attention to detail every time. For doing fundamental research, the second guy with the deep understanding is probably the better choice. The reality is that our society needs a whole lot more of the first type than the second.
  18. Oct 8, 2013 #17


    User Avatar
    Science Advisor
    Homework Helper

    there is no answer to such a vague and general question, since different exams are written and graded by many different people with different goals. In general however, in college exams measure whether you do what they tell you to do. In fact often the syllabus handed out on day one tells you exactly what topics will be tested at the end, and even what pages they are found on in the book. Still most students refuse to read those pages, and some of those who try refuse to ask questions about difficulties.

    I flunked out of college with low grades on exams and then got back in and achieved much higher grades ever afterwards, the difference? - I started going to class, doing the reading and the homework.

    when i later taught college i tested exactly what i had assigned. most people did poorly largely because they ignored my advice, or skipped class and did not even hear it, failed to read and respond to emails repeating it, and never came to office hours to ask about problems, until the morning of the test that is. If I ever gave a harder than average problem to test real interest it was as an extra credit problem with hints, and no one ever tried it in general.

    mostly it was a complete denial of reality - a persistent vain hope by most students that maybe they could skate by without ever working hard. even the most brilliant but perpetually lazy student who understood everything easily, eventually began to fail from never doing the required work. Some students would prefer to say they never study, than to pass by working hard.

    in some 40 years of teaching hundreds of students of all abilities, i saw virtually none who failed after actually trying. Essentially all who worked hard got honor grades, even the ungifted, and the truly gifted scored above the grade scale. Still the failure/withdrawal rate was huge. a few people showed up for the final unaware that after not attending either class or tests for so many months they had been dropped from the roll.

    to repeat: in most classes you do not get good grades by being clever or bright, you get them by doing what you are told to do, by learning the material they say will be tested. If it were otherwise it would not be a good test.
    Last edited: Oct 8, 2013
  19. Oct 11, 2013 #18
    There is also an element of pressure involved with a timed test. Some people who would otherwise do well can do poorly due to test anxiety etc.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook