Color measurement spreadsheet--addendum

I would like to thank Gawain Weaver, whose sharp eyes and
inquisitiveness led to his discovering a glitch in the Excel
spreadsheets I posted online some time ago (Conservation DistList
Instance: 13:44 Wednesday, February 16, 2000). These spreadsheets,
one for Macs  and one for PCs, are designed to help assess whether
differences in L*a*b* data for color measurements constitute a
statistically significant change, or are due merely to chance.
Gawain noticed an error in the formula for calculating average
values of each of the three variables, L*, a*, and b*. The effect of
this error was that the spreadsheet calculations were somewhat more
"lenient" in assessing differences than they should have been. In
other words, suppose you made one set of color measurements at some
point in the past, then made a subsequent set of measurements to
determine whether a particular color had changed. The error in the
spreadsheet might have caused the calculations to indicate that
there was no significant difference in the color, whereas it is
possible that there might have been. Corrections have now been made
in the color assessment spreadsheet, and as before are available for
download from the following URL. If you previously downloaded this
file, please download the new one and discard the old one.

Sorry for the mixup. Even our wonderful Excel guru who helped design
this spreadsheet hadn't noticed this error. And thanks, Gawain, for
finding this!

For PC users:
<URL:http://www.mfa.org/conservation/downloads/pccolor.xls>

For Mac users:
<URL:http://www.mfa.org/conservation/downloads/maccolor.xls>

Roy Perkinson


                                  ***
                  Conservation DistList Instance 17:1
                   Distributed: Monday, June 16, 2003
                        Message Id: cdl-17-1-007
                                  ***