Kelley’s Discrimination Index (DI) is a simple and robust, classical non-parametric short-cut to estimate the item discrimination power (IDP) in the practical educational settings. Unlike item–total correlation, DI can reach the ultimate values of +1 and ‒1, and it is stable against the outliers. Because of the computational easiness, DI is specifically suitable for the rough estimation where the sophisticated tools for item analysis such as IRT modelling are not available as is usual, for example, in the classroom testing. Unlike most of the other traditional indices for IDP, DI uses only the extreme cases of the ordered dataset in the estimation. One deficiency of DI is that it suits only for dichotomous datasets. This article generalizes DI to allow polytomous dataset and flexible cut-offs for selecting the extreme cases. A new algorithm based on the concept of the characteristic vector of the item is introduced to compute the generalized DI (GDI). A new visual method for item analysis, the cut-off curve, is introduced based on the procedure called exhaustive splitting.

Progress monitoring of academic achievement is an essential element to prevent learning disorders. A prominent approach is curriculum-based measurement (CBM). Various studies have documented positive effects of CBM on students’ achievement. Nevertheless, the use of CBM is associated with additional work for teachers. The use of tablets may be of help here. Yet, although many advantages of computer- or tablet-based assessments are being discussed in the literature (e. g. innovative item formats, adaptive testing, automated scoring and feedback), there are still concerns regarding the comparability of different assessment modes (paper-pencil vs. tablet). In the study presented, we analyze the CBM data of 98 fourth graders. They processed the exact same computation items once with paper and pen and once in a tablet application. The analyses point to comparable results in the test modes, although some significant deviations can be found at item level. In addition, the children report perceived benefits when working with the tablet.

Although Goodman–Kruskal gamma (G) is used relatively rarely it has promising potential as a coefficient of association in educational settings.  Characteristics of G are studied in three sub-studies related to educational measurement settings. G appears to be unexpectedly appealing as an estimator of association between an item and a score because it strictly indicates the probability to get a correct answer in the test item given the score, and it accurately produces perfect latent association irrespective of distributions, degrees of freedom, number of tied pairs and tied values in the variables, or the difficulty levels in the items. However, it underestimates the association in an obvious manner when the number of categories in the item is more than four. Towards this, a dimension-corrected G (G2) is proposed and its characteristics are studied. Both G and G2 appear to be promising alternatives in measurement modelling settings, G with binary items and G2 with binary, polytomous and mixed datasets.