Originally Posted by **Nrt** Imageshack - adszzx
thanks, but my problem is i dont know how to get my straight line from F=kx. I dont know k so i cant give values and draw.
Is the best fit line that im supposed to draw? |

,

Originally Posted by **r.samanta** you draw the line so that it passes through the maximum number of points. also make sure that the number of unused points above this line equals the no of unused points below this line. then find k by my method. |

I may disagree with r.samanta (depending of the following). One of the most used method for interpolating datas in physics are the least squares method (see

http://en.wikipedia.org/wiki/Least_squares or just type it in google and look how to get the equation y=ax+b, where a is the slope. Basically you've found y's and x's, it remains to find out a and b precisely by the Least squares method). Now let me ask you a question Nrt : Have you been introduced to the Least square method? If no, then maybe the tip of r.samanta is good enough. If yes, then you'd have to apply what you've been introduced!

Sidenote : In fact what r.samanta proposed seems something similar to the least square method, but many times the line won't pass by any points. I've made many lab reports and I don't remember a single one in which the line passed by a point I had measured. I even wrote a program (with my friend) in Fortran 90 in order to calculate "a", "b", their errors (or uncertainties) and many more things related to this.

Edit: Maybe take a look at

http://www.shodor.org/unchem/math/lls/leastsq.html in order to plot your data to compare with what you did. Search for "linear least squares".