How would someone study physics effectively?

[Radarithm]

How many problems should a physics student (or self-learner in my case) be able to solve correctly? I assume it's safe to say that all (*) problems should be correct, but what about (**) and (***) problems? How many of them should be done correctly? Is getting the correct answer important (because getting everything wrong is definitely not right)? I'm stumped at the moment.

 

[jbunniii]

Regarding the number of stars an author assigns to a problem, this very much depends on the textbook. In some books, such as Spivak's "Calculus", most problems have no stars. The ones with one star are challenging, and the ones with two or more can be very hard and may require some trick or ingenuity. Failure to get these would not necessarily mean you don't understand the material.

Some authors take this even further, for example Pugh's "Real Mathematical Analysis." He writes: "One star is hard, two stars is very hard, and a three-star exercise is a question to which I do not know the answer." Obviously no one is going to fault you if you can't do one of his three-star problems.

 

[gfd43tg]

From my experience, I think 1 star and 2 star problems are doable if you understand the material presented in its entirety. I think the 3 star problems usually require some outside ingenuity or past principle to be applied to the new situation which understanding the text alone would not suffice to solve. So if you are a self learner and can do the 2 star problems, then I think you are in good shape.

 

[Cosmobrain]

I am a self-learner too. What I can say is that you shouldn't worry much about it. Study physics slowly, you have all the time in the world. Sometimes you should hang around and think of things you don't know about, doubts etc. and then you come back to the computer google your question or start a thread here in PF.
I used to be all worried and stuff because I always thought I didn't know enough until I realized I really don't need all of that knowledge on the internet. I can learn slowly by the basics to the most complex problems and questions.

 

[Radarithm]

What if we're talking about rigorous books (atleast for a guy whose only done physics at the level of halliday) like Purcell or K&K? Should the reader be able to solve in order for him to prove his understanding and advance to more complicated books, eg. 8/10 questions? 6/10?

 

[Cosmobrain]

What if we're talking about rigorous books (atleast for a guy whose only done physics at the level of halliday) like Purcell or K&K? Should the reader be able to solve in order for him to prove his understanding and advance to more complicated books, eg. 8/10 questions? 6/10?

If you want to study methodically, then yes. However, you can pretty much study in any way you want. I don't know neither of those books so I can't give you any advice. I hope someone else does

 

[WannabeNewton]

What if we're talking about rigorous books (atleast for a guy whose only done physics at the level of halliday) like Purcell or K&K? Should the reader be able to solve in order for him to prove his understanding and advance to more complicated books, eg. 8/10 questions? 6/10?

Ideally you should be able to solve all the books in K&K. Purcell is a bit of a different story at least with regards to the 3rd edition because it has a plethora of problems and a relatively large number of them are incredibly difficult. You should attempt all of them, time permitted, but if you can't solve every single one don't worry about it.

No one can quantify this stuff for you. We don't know your schedule and your other obligations so I honestly don't know what you're expecting. There's no magic number written in the sky about how many problems a person should solve from a textbook. Just stick to whatever you find to be the most pragmatic and effective.

 

[micromass]

What if we're talking about rigorous books (atleast for a guy whose only done physics at the level of halliday) like Purcell or K&K? Should the reader be able to solve in order for him to prove his understanding and advance to more complicated books, eg. 8/10 questions? 6/10?

The transition from a book like Halliday to a book like Kleppner can be difficult. First of all, Halliday has an enormous amount of exercises, so you're not expected to do all of them. Moreover, most exercises in Halliday repeat the same thing over and over again. Not that that's bad thing, but books like Kleppner are very different. The exercises there all tend to be quite difficult, but all of them are rewarding. So I would at least attempt each exercise in Kleppner, since they're all nice.
However, you may find that an exercise in Kleppner may take you a lot of time (at least compared to exercises in Halliday). Some people struggle a lot with this and think weeks about one exercise. This is not necessarily bad, but it will make your progress quite slow. I recommend you to think about the exercise for a time, and when you can't find it, ask for help here on PF.

 

[462chevelle]

What books are star coded? I've never seen this in a textbook.

 

[serllus reuel]

From my personal experience, a *** problem in a book like Halliday or Serway is equivalent to a ** problem in KK/Purcell/Morin. I know of at least one problem (finding B due to a rotating charged sphere) to be exactly the same, but *** in Serway and ** in Purcell.

The ** problems in the Halliday-type books are good for practice if the material is new, but they are simple compared to what you see in more advanced books. What I would do is do the ** ones, then move on, let the material sink in, then come back to the *** problems.

 

[Radarithm]

Thanks for the help everyone.
Decided to do every problem (except the redundant ones, and there are hardly any).

 

[guitarphysics]

Hey Radarithm, sorry to chime in late but wanted to give my two cents because I've studied from those exact books, and maybe can help you out. When I was doing Kleppner I struggled a lot but managed to do a good number of the problems (I'd say around 90%). The ones I couldn't do I left for later, and every once in a while I dig back through the book to see if anything's changed (usually I manage to solve at least a few more). As for Purcell, that's a different story. Try to do as many of the problems as you can (the ones with answers in the back, so you can check your work), but don't worry if you can't do some *** problems (I usually can't do them, tbh). Try to be able to do all (or most) of the **, and obviously all the *. Also do some exercises (same as problems but no answers in the back), but if you can do most of the problems you're probably fine (I think there's about 30 problems per chapter, so that should be enough without doing any exercises. It's up to you though). Like Wannabe said, though, there isn't really a magic number or anything for what you should do. Have fun learning from K&K and Purcell, it's really fun.

EDIT: Forgot. As for if it's important to get the correct answer- yes and no. Yes in that obviously, if you get the right answer you most likely understand the concept and can apply it. But also no, because when you get the answer wrong you realize you didn't understand something fully, and can go back and review it. NEVER just look at the correct answer and process straightaway if you get it wrong. Try again, or peak a little at the solution to see the step you may be missing (I think Purcell may have talked about this in the preface, it's really important).