Using Statistics to Evaluate Process Improvement

The techniques associated with Statistical Process Control (SPC) are very useful, but they are not sufficient alone to
provide inferential comparisons. An additional need is the ability to make valid
comparisons.
This paper suggests two additional techniques to help evaluate differences: inference for
difference between two means (using t tests and confidence intervals for the difference
between the means), and inference for two way tables (using chi-square tests). These basic
statistical techniques should be in our analysis toolbox, along with the traditional SPC tools.