How to determine if a linear regression best fit is significantly different from 1:1

[microbegirl77]

Hello,

I am comparing data obtained by two different methods that I've plotted and gotten a linear best fit line for, and I'd like to find out if the slope is significantly different than 1, and if the Y-intercept is significantly different than 0. I'm not super familiar with formal statistical analysis, though, and am unsure how to do this. Can anyone offer pointers? I am proficient in Excel and have the StatPlus package (I have a Mac) and also have R and Matlab, though I'm not very experienced with them.

Thanks!

[Dragan]

Hello,
....the Y-intercept is significantly different than 0. I'm not super familiar with formal statistical analysis, though, and am unsure how to do this.
Thanks!

Why? That is, what is so important to your study that you would want to test a null hypothesis that the Y-intercerpt is zero?

[microbegirl77]

i want to test whether or not there is significant bias high by one method or the other.

[woa]

How about using 'Concordance correlation coefficient'?

http://rgm2.lab.nig.ac.jp/RGM2/func....d=epiR:epi.ccc

http://www.medcalc.org/manual/concordance.php