Skip to Main Content
Three discrete multivariate analysis techniques were used to assess the accuracy of land use/land cover classifications generated from remotely sensed data. Error matrices or contingency tables were analyzed using these techniques and the results reported. The first technique is a normalization procedure using an "iterative proportional fitting" algorithm that allows for direct comparison of Corresponding cell values in different matrices irregardless of sample size. The second technique provides a method of testing for significant differences between error matrices that vary by only a single variable or factor. The third technique allows for multivariable comparisons to be made between matrices. Each technique is implemented through the use of a computer program.