Flat laxity, optimization, and rounding in the selection of class intervals.

Optimization methods for the selection of class intervals for choropleth maps commonly yield class breaks with more significant digits than the map user would prefer or the accuracy of the data warrants. Round-number class breaks, which are more easily remembered, are readily introduced as constraints on an optimal solution. Specific inherently meaningful breaks may be mandated to provide an even more effective classification. As predicted by the principle of flat laxity, the resulting suboptimal, constrained solutions generally increase classification error only slightly beyond the minimum error of a so-called optimal set of class intervals. -Author