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!

Do it yourself!

  1. May 9, 2013 #1
    1. The problem statement, all variables and given/known data

    I was curious, why it is considered so important to solve the problems yourself.
    There are a number of puzzles you will get stuck with and trying to brew over them is in my opinion much less productive, then analyzing the solution and getting to know new concepts.

    What's your opinion?

    2. Relevant equations

    3. The attempt at a solution
  2. jcsd
  3. May 9, 2013 #2


    User Avatar
    2017 Award

    Staff: Mentor

    It can be useful to study other solutions as an introduction, or if they use tools you do not have yet. But you cannot learn how to solve problems based on that - you have to solve problems yourself.
  4. May 9, 2013 #3
    Hi igorronaldo. Welcome to Physics Forums!!!!

    What are you going to do when you get out in the real world, and your boss gives you a problem to solve that's not exactly like the ones in your textbook?
  5. May 10, 2013 #4
    for example in chess you learn by analyzing the games of extraordinary players, in physics I guess it is similar you learn by trying to do it yourself, but your time is invested more wisely, if you analyze the solution and try to do it yourself, rather then brewing over the exercise for hours without making any progress, I guess.
  6. May 10, 2013 #5
    Reading and analyzing solutions will not get you very far. It's possible the analyze 1000's of physics problems, but still be stumped when you actually have to solve one yourself.
    Solving a problem yourself is much more than just finding the solution. It's also visualizing things, problem solving, trying to construct steps that lead you to the solution. When you simply read a solution, you miss most of this.
  7. May 10, 2013 #6
    IMO, this is a huge problem for people like me who are trying to self-study math and physics. Of the thousands of tutorials you can find online, almost the entirety are of the "Witness the presenter solve the problem" variety. This is of course, great, but there needs to be a follow up after this where you are given some problems to solve on your own. At this point, then the presenter should walk you through the solution, fully worked. I'd almost go so far as to say that watching someone solve a problem, thinking you understand it, and then moving on to another topic without doing some of those problems yourself is practically useless. However, that's pretty much all that's available for online tutorials. MIT's recitation videos are a notable exception, and they have really helped me. You get a full lecture on a subject, and then a recitation session where you are given a few problems to solve yourself before the TA comes out and walks you through the solution. Here's an example:


    Edit: Btw, if anyone knows of other videos of the kind I posted above, please let me know!:smile:
    Last edited: May 10, 2013
  8. May 10, 2013 #7


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Are you asking because you want to discuss the forum rule that says that you have to show your work when you ask for homework help, or because you want to discuss what's the better way to learn?
  9. May 10, 2013 #8
    well,it is very important to slove problems by yourself. Because if you are to answer a question that is application level type,the person who studied similar questions without attempting to answer cannot answer the application level type. Only the one who tries to slove problems by himself would accomplish this one..
  10. May 10, 2013 #9
    good question!! I think it is the first one!!
  11. May 10, 2013 #10

    Yep, I think bolded text is a dead giveaway. Lol:tongue:
  12. May 10, 2013 #11
    The most important point for me is the fact that only trying to do a problem yourself you are able to identify where you are stuck, and what you did not understand well actually. If you just analyze a solution, you can say "yes of course this I understood, so it is clear why they do so". It is only trying it by yourself that you understand what you really didn't understand... and then you have much clearer ideas on what to search or study again... or ask (here!) :wink:
    My two cents
  13. May 10, 2013 #12
    It's the same thing with science or technical writing. I've had concepts or models I had worked out in my head, or schematically on paper, and thought I had the model completely worked out. I would then proceed to write up a paper on it thinking I'd have it done lickity-split. Then comes the moment of truth when you hit that block and can't even get through the introduction because you realize you have no idea what you're talking about. The great thing about the process of writing is that is really lays bare and specifically identifies the areas you need to work on. It's the same with understanding maths and physics problems, as Mattia illuminates. You need to do them yourself.
  14. May 10, 2013 #13


    User Avatar
    Science Advisor

    It's not enough to just do problems. You have to do the hard ones. I can guarantee you that you will have an infinitely better understanding of freshman mechanics if you work through all (if not most) of the problems in Kleppner vs say the ones in Halliday and Resnick. Plus, the problems in Kleppner are really fun (I can't begin to tell you how much pacing they got me into, it was quite an amazing book in terms of problems and exposition - some of them really get you thinking).

    You also have to do a bunch of problems filled with physical insight and/or physical subtleties. I think Griffiths' Electrodynamics book is a great example of these kind of problems, as well as Purcell's Electricity and Magnetism (maybe more so the latter than the former).

    As a higher level example, I can tell you that to some extent the problems in Wald's General Relativity (GR) text will most likely not help you actually understand GR. The plethora of tensor calculus problems can be fun (and at times just mind wracking to figure out) but they really don't help you learn GR itself and Wald has these kinds of computational problems in abundance in every chapter (except one of them). There are also times when the problems in the text are just absurdly easy. My point is that if you actually want to learn the physics of GR, especially by doing problems, Wald won't really help you with that. You would need some other book (like MTW or Ohanian for example) in order to accomplish that. However in the case of graduate texts like Wald, it might be understandable that he expects you to have done the "physical" problems beforehand in undergraduate GR classes.

    The overall point I'm trying to make is that it isn't enough to just do any old problem, you gotta do the hard ones that actually get into the nitty gritty of the physics.
    Last edited: May 10, 2013
  15. May 10, 2013 #14


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper
    Education Advisor

    Sure, there are instances where you learn a new concept by reading the solution to a problem. That's why textbooks work out examples for you after all. But in general, it's not a good strategy for learning, and a lot of students make the mistake of thinking because they can understand the solutions, they understand how to solve problems. Many students also delude themselves into thinking they've really tried to learn the material and so now it's okay to look at the solutions without having worked the problems out on their own.

    When it comes to learning, it's what the student does that's the most important. If learning physics simply involved reading a textbook or solutions to problems, then we wouldn't need teachers or courses, would we? You could just sit down with a book for a few hours, and boom! you'd know physics. A colleague likes to use the example of learning how to drive. You can watch others drive, watch videos on how to drive, read manuals, etc., but until you actually get behind the wheel of a car and practice, you don't know how to drive.
  16. May 10, 2013 #15
    If you don't consider a question reasonable, you don't have to comment. I am curious, so I think it makes sense to ask and the metaphor that involved driving a car was very enlivening.

    Comments like yours are very untasteful and poison the atmosphere. It is disappointing.
  17. May 10, 2013 #16


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    All he suggested is that I was right to ask you what you meant to ask in post #1, so I don't understand what you find offensive about the post you quoted.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook