Re: The Beer's Law

tressure@xxxxxxxxx wrote: 
Then measure "A" and calculate "c" (either graphically or
from the best-fit equation, with data from your standards).



Best to do it graphically- I've seen Beer's Law plots that were
quite curved (most notably in the last lab of last semester- only
one student got decent results because everyone else drew a
straight line for the absorbance of nickel ions instead of a smooth,
gentle curve.  An imperfectly-designed experiment which I will
have to fix).

A nonlinear plot tells you that Beer's law isn't working. It doesn't say why.
I always like to see a plot, even if I've fit an equation to the data by a computer program. You can apply statistics to decide if the best fit is linear, but there are few data points in calibration set, and the statistics in that case get pretty tricky. It is the kind of thing statisticians love to argue about.
.

Relevant Pages

  • Re: Deseasonalization and detrending of Keeling curve
    ... fit with a linear gain factor. ... The "fit" is based on a stiff spline ... MODEL STATISTICS AND EQUATION FOR THE CURRENT EQUATION (DETAILS ... Standard Error of the Mean =Standard Dev/ .148530E-01 ...
    (sci.stat.math)
  • Re: Regression significance conundrum
    ... > is a good linear correlation between the two variables. ... > that a quadratic fit does not confer significant additional benefit? ... If you do the statistics properly, then you probably won't have any ... referees) you thought there was some slight curvature. ...
    (sci.stat.math)
  • Re: Sorry! The Phaistos disc.
    ... consisting of elements of two signs. ... But is its meaning the one you think ?.. ... The statistics can be ... >>figure doesn't fit with the conclusion I want to reach ?.. ...
    (sci.archaeology)
  • Re: Goodness of fitting of a distribution
    ... linear combination of a weibull and a normal distribution. ... Kolmogorov-Smirnov statistics is NOT a "goodness of fit" statistic. ...
    (sci.stat.math)
  • Re: OLS fails due to outliers, but relationship is very clear
    ... Pawel, you'll probably get more traction in something like sci.stat.math, but here's my two cents: ... Hard to tell what you mean by a "highly biased" fit, and I'm not sure what your criterion for calling something an outlier is. ... If you bin and take means, you'll be estimating a slightly different model, but an equivalent one, and your ultimate predictions probably won't change much. ... what should I search for to learn about statistics in this sort of "bin analysis"? ...
    (comp.soft-sys.matlab)