I consider the sig fig rules just fine for introductory classes--especially
my liberal arts class. If nothing else, it breaks them from the 'read off
the number from the calculator' habit. We spend time with simple
calculations (such as finding the area or volume) to show WHY we have such
rules, then the first lab of the year deals with finding volumes and
densities and using these to try and identify the materials from which the
objects are made. In this exercise we can clearly show that the calculator
answers imply way too much precision while the sig-fig rules keep the
answers 'reasonable.' We also deal with expressing differences (say
between the measured density and the suspected material density) as
percentages--despite students wanting to always talk in absolute
differences. IMO, sig fig rules and dealing with percent differences
(never error) are useful tools for these students.
In the science major's class, we start out the same but eventually get more
sophisticated about reporting uncertainties.
I'm sure there is data out there that doesn't 'fit' the sig fig model, but
for what we do in intro labs, sig figs seem to work OK.
Rick
***************************************************
R.W.Tarara
Professor of Physics
Saint Mary's College
Notre Dame, Indiana
Free Physics Instructional Software
Windows & Mac
www.saintmarys.edu/~rtarara/software.html
***********************************************