[Lcal]

I'm running a kinetics lab where I'm graphing Ln(absorbance) on the y-axis and the time in seconds on the x-axis

My question is...I need to find the error in the slope of the linear trend line...in excel i did Tools>Data Analysis>regression i selected the data range 95% confidence level and pushed OK

How do i interpret the information given so i can find the error to report the slope as (slope) +/- error? or is there an alternative way to calculate this?

this error will be used in further calculations to figure out the error in rate constants etc.

[Martingale]

I have excel 2007 and I don't know how to do the regression analysis you did, but if you can post what the excel program is giving you I might be able to help.

[Lcal]

what excel gives me is what they call the coefficients, standard error, t stat, p-value, lower 95% and upper 95%

that is both for the intercept and the x variable

what the lab is asking for specifically is to 'determine 95% confidence levels of the slopes and intercepts'

possibly my approach to find the error is incorrect but the next step is to take this slope (and its error) and divide it by a concentration (and its error) to then use to determine another set of points, the slope and intercept of that relationship (for my rate constants)

[JGK]

This might help with the stats

http://cameron.econ.ucdavis.edu/excel/exregfn.html

[Martingale]

The confidence interval for the slope is given by...

(b-t*s,b+t*s)

where b is the estimate for the slope (the one that excel found)

t is the student-t statistics (also given by excel)

and

s=sqrt([SSE/(n-2)]/[sqrt(n-1)s_x])

S_x is the standard deviation of X

SSE = sum of squared error