> Has it become acceptable in teaching to just fit a curve to data points
> and not report a correlation coefficient to indicate how well the curve
> fits? American Biology Teacher author Stanley Rice seems to think so and
> justifies his position because "most classes taught by readers of this
> journal would not be prepared for these analyses." ABT 66: 475.
>> A correlation coefficient seems like a simple enough concept even for
> students with little knowledge of statitics. Why bother to fit a curve
> to the data if there is no indication of how well it fits?
I'm confused.
It's not like "they" are fitting a curve to data points by eye. They are
obviously using a program to fit the curve, and therefore some sort of
algorithm is used to make sure the best fit is made.
PPDD
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.760 / Virus Database: 509 - Release Date: 10/09/2004
