Physics Help Forum Photoelectric effect error value?

 Nuclear and Particle Physics Nuclear and Particle Physics Help Forum

Mar 14th 2011, 03:09 PM   #11
Junior Member

Join Date: Mar 2011
Posts: 7
 Originally Posted by topsquark That was the part you didn't tell me. I asked about this in post #4 in this thread. So. You got the slope using linear regression then. Here is a rundown on how the different variables are computed. The error can be tabulated in two different ways: using the sum of the square of the vertical offsets or by using the correlation coefficient. I am assuming that the cause of the larger percent error has to do directly with the vertical offsets. These "move the fit line" around to find the slope and intercept with the smallest sum of squares error. That's why your standard errors (the +- in the voltage measurements) are lower than the % error in h...The sum of squares is not always the best of the fits to use. (The article I referenced mentions the perpendicular offsets, which I think is a slightly better fit. The downside is that the error is harder to calculate.) -Dan
Ok thanks,but I find that vertical offset method very hard to understand, is there easy explanation on how to use the standard errors to get the answer?

Mar 15th 2011, 10:52 AM   #12

Join Date: Apr 2008
Location: On the dance floor, baby!
Posts: 2,811
 Originally Posted by wolfhound Ok thanks,but I find that vertical offset method very hard to understand, is there easy explanation on how to use the standard errors to get the answer?
There's one of my other favorite methods called Propagation of Errors. However as you don't have any error measurements for f I'm pretty certain you can't use that method here. If you have a data set that, say, has errors in Vs as well as well as error measurements for f, then you can use a "chi-squared" fit. (I couldn't find a good link for this because there is something called a "chi squared test" in statistics. I couldn't get the search to work right.) The problem I see with chi-squared fit is that the theory is even harder than the least-squares fit.

Sorry that I can't be of any more help.

-Dan
__________________
Do not meddle in the affairs of dragons for you are crunchy and taste good with ketchup.

See the forum rules here.

Mar 16th 2011, 03:36 AM   #13
Junior Member

Join Date: Mar 2011
Posts: 7
 Originally Posted by topsquark There's one of my other favorite methods called Propagation of Errors. However as you don't have any error measurements for f I'm pretty certain you can't use that method here. If you have a data set that, say, has errors in Vs as well as well as error measurements for f, then you can use a "chi-squared" fit. (I couldn't find a good link for this because there is something called a "chi squared test" in statistics. I couldn't get the search to work right.) The problem I see with chi-squared fit is that the theory is even harder than the least-squares fit. Sorry that I can't be of any more help. -Dan
Thats fine ,thanks for your help anyway

 Tags effect, error, photoelectric

,

,

,

,

,

,

,

# how to find error in plank constant ecperiment by using graph

Click on a term to search for related topics.
 Thread Tools Display Modes Linear Mode

 Similar Physics Forum Discussions Thread Thread Starter Forum Replies Last Post spixmacaw Quantum Physics 1 Nov 3rd 2012 10:38 AM Steve George Quantum Physics 3 May 16th 2012 04:27 PM Akshay Quantum Physics 3 Sep 11th 2009 10:43 AM mola Nuclear and Particle Physics 1 Jun 13th 2009 12:07 PM werehk Light and Optics 1 Aug 17th 2008 05:21 PM