Diagnosing Specific Learning Disorder in Mathematics: Development of a Test Battery Tailored to Luxembourg

To diagnose specific learning disorder in mathematics, multidimensional assessment is required, which includes psychometric testing to identify children’s individual weaknesses and strengths, particularly in mathematics. Most of these tools use language to give instructions and present tasks but test language proficiency is prone to impact student’s test performance (Ugen et al., 2021). This is especially important for Luxembourg and its linguistically diverse school population, as results from the national school monitoring program ÉpStan showed school performance differences based on children’s home language (Greisen et al., 2021; Martini et al., 2021). Even though currently used diagnostic tests generally have a good psychometric quality, they rarely consider linguistic heterogeneity, questioning their reliable and valid use in a multilingual setting. In fact, to reduce the risk of over- and under-identifying learning disorders, difficulties arising from lower language proficiency need to be disentangled from problems resulting from an underlying learning disorder, leading to situation-dependent adaptations in the diagnostic process (e.g., translation of test instructions or approximate interpretation of test results).
To optimize this, we have developed a test battery in mathematics for children in grade 3, specifically tailored to the multilingual education context in Luxembourg. Following diagnostic guidelines and neurocognitive models of number processing, we devised a total of 17 paper-pencil subtests assessing children’s performance in basic numerical and arithmetic skills. The language in the instructions was kept simple and straightforward, and only eight subtests contained language directly within the items. The content was adapted to the national mathematics curriculum. To norm the test battery, we have collected data in 49 classes all over Luxembourg (N ≃ 650 children). Preliminary findings of this data collection will be presented to show the tool’s suitability for Luxembourg’s multilingual education setting.