How to know which variables are more important in a process? [closed]

I have a process with 15 effective variables. I could record 9 variables to study its effect on process.
I am looking for an appropriate factor to estimate the value of effectiveness of each factor. I believe p-values, standardized coefficients, partial correlation and partial $R^2$ are the potential solutions in this regard.

Which one is more useful in informing the degree of effectiveness of each variable on scale of 0 to 100?

-
Could you say a little more about the details of your context? Is it time series data? are the predictor variables experimentally manipulated or just observed? What is the aim of your analysis: control, prediction, something else? –  Jeromy Anglim Oct 24 '11 at 0:17