The way I present a lot of my data is as a 'surface' of values (stiffness, in my case) with respect to two independent variables (angular displacement). It's really the only way you can show the data to someone and have them understand it quickly (tables of the data would be too large and intractable).
One aspect of my dissertation is that I use collected stiffness data to predict what the stiffness would be at an unmeasured set of angles, so I have to come up with some way to do that. One way would be to interpolate from the measured data, but that would make any software I write pretty ugly, since it would have to include the big, intractable tables in the code. A better way is to use a statistical method called regression to find the best equation that fits my data, and then use that equation to do any prediction. Unfortunately, most regression software packages (and descriptions in textbooks and on the web) only deal with scenarios where you have one independent variable. Since I am using two, it was beginning to look like I would have to go the ugly route.
The update for tonight is that I found an easy way to do it, in Excel nonetheless. I'm very pleased about that.
