Matthijs J. Warrens

Accuracy assessment of classified imagery is an important task in remote sensing. A commonly used approach for assessing classification accuracy is based on the error matrix. The error matrix can be summarized by overall and category accuracy measures, e.g. the overall agreement and the user’s and producer’s accuracies. Two relatively new measures are the so-called quantity disagreement and allocation disagreement. In this article, relative versions of the category-level variants of these measures are defined. It is also shown how the new category measures are related to the overall quantity and allocation disagreement measures.

Cohen’s kappa is a standard tool for the analysis of agreement in a 2 × 2 reliability study. Researchers are frequently only interested in the kappa-value of a sample. Various authors have observed that if two pairs of raters have the same amount of observed agreement, the pair whose marginal distributions are more similar to each other may have a lower kappa-value than the pair with more divergent marginal distributions. Here we present exact formulations of some of these properties. The results provide a better understanding of the 2 × 2 kappa for situations where it is used as a sample statistic.

Kappa coefficients are standard tools for summarizing the information in cross-classifications of two categorical variables with identical categories, here called agreement tables. When two categories are combined the kappa value usually either increases or decreases. There is a class of agreement tables for which the value of Cohen’s kappa remains constant when two categories are combined. It is shown that for this class of tables all special cases of symmetric kappa coincide and that the value of symmetric kappa is not affected by any partitioning of the categories.

Cohen’s kappa is a widely used association coefficient for summarizing interrater agreement on a nominal scale. Kappa reduces the ratings of the two observers to a single number. With three or more categories it is more informative to summarize the ratings by category coefficients that describe the information for each category separately. Examples of category coefficients are the sensitivity or specificity of a category or the Bloch-Kraemer weighted kappa. However, in many research studies one is often only interested in a single overall number that roughly summarizes the agreement. It is shown that both the overall observed agreement and Cohen’s kappa are weighted averages of various category coefficients and thus can be used to summarize these category coefficients.