Using a modification of vector analysis for calculating surgically induced astigmatism, we describe a simple method that divides the induced cylinder into two orthogonal components. This decomposition allows with-the-rule and against-the-rule astigmatic changes to be calculated for individual case analysis, as well as for statistical analysis of aggregate data. Because it is based on the true induced cylinder, as determined by optical principles, this method overcomes some of the problems associated with earlier methods of evaluating surgically induced astigmatism.