3 thanks: Thanks for checking gaussEP out on gnumeric. I'm pleased to
hear that it works. Thanks for your tips. I set this up 15 years ago
in MathCad and never implemented it in the classroom because I set my
sights too high--with all the histogram plots started, I wanted them to
grow before the user's eyes with the values being used for one
calculation of the function, and the resulting value of the function,
"lighting up" while the calculation was taking place. Your web document http://www.av8n.com/physics/uncertainty.htm
motivated me to set it up again, this time on a spreadsheet. Thanks for
making such a well-written, well thought-out document available.
I used gaussEP in lab for the first time yesterday. Beyond error
analysis, it turns out that using it results in an immediately practical
reason for finding an algebraic solution to a problem (rather than
working with numbers from the start). Figuring out what to type into
cell A2 of the first worksheet is a lot easier when one has, on paper, a
correct algebraic expression for the function. One can easily see the
one-to-one correspondence between what one has written and what one
needs to type into A2, which, instead of =sqrt(2g(h-h')) was typically
=sqrt(2*g!a2*(h!a2-hp!a2)).
I read about your "crank three times" method in your uncertainty
document but I appreciate the reiteration. I think I'll probably
implement that in lab later this semester to help convey the idea of
what the spreadsheet is doing.