Technologies that enable individualization for students have significant potential in special education. Computerized Adaptive Testing (CAT) refers to digital assessments that automatically adjust their difficulty level based on students' abilities, allowing for personalized, efficient, and accurate measurement. This article examines whether CAT performs differently for students with and without special educational needs (SEN). Two simulation studies were conducted using a sample of 709 third-grade students from general and special schools in Germany, who took a reading test. The results indicate that students with SEN were assessed with fewer items, reduced bias, and higher accuracy compared to students without SEN. However, measurement accuracy decreased, and test length increased for students whose abilities deviated more than two SD from the norm. We discuss potential adaptations of CAT for students with SEN in the classroom, as well as the integration of CAT with AI-supported feedback and tailored exercises within a digital learning environment.

Context: Students with special educational needs in the area of learning (SEN-L) attend vocational trainings to be provided with qualifications for the labor market. Competences in arithmetic operations and comparing quantities such as weights and lengths are indispensable for obtaining a vocational qualification. Therefore, the study investigates whether students with SEN-L in Germany convert job-relevant quantities and master arithmetic operations accurately and with a certain speed before beginning vocational training. Methods: 152 students with SEN-L in pre-vocational training programs in southern Germany were examined using curriculum-related computer tests for formative assessment. Three skill-based tests were constructed as progress monitoring tests comparing weights and lengths and arithmetic operations including a speed component. Results: All newly developed tests meet the requirements of the Rasch model. Students' performance on the lengths test and the weights test correlates with a significant linear relationship (r = 0.64). Only weak correlations (r > 0.29 < 0.45) are found between the performance in the arithmetic operations and the weights and lengths tests. For items in the weights and lengths tests, participants show on average significantly higher probability of solving than in the arithmetic operations test. Furthermore, students with SEN-L show slow processing speed (median processing speed weights test 6.37 seconds, lengths test 6.26 seconds, arithmetic operations test 33.97 seconds on average per item). The students solve more items of the weights and lengths tests correctly than of the operation test (median for weights 25 items, for lengths 24 items, for arithmetic operations 7 items). Conclusion: The majority of the students with SEN-L were not able to solve the expected number of items correctly and show that the participants' accuracy in dealing with arithmetic operations and quantity comparison is low. It can be assumed that this might lead to difficulties in mastering mathematical items in the context of vocational training. This is the case, in particular, where a certain speed is required (fluency). Math instruction in pre-vocational programs needs a stronger focus on building up and expanding arithmetic operations and quantities and has to be complemented by monitoring the learning process of students with SEN-L.

Although inclusion is the declared goal, the transition from a system based on special schools to an inclusive school system has only been progressing very slowly in individual countries. In an evolving school system, the existing special schools keep struggling to justify their existence. This study investigates the regional distribution effects based on official school data and shows the influence of a pull effect on special schools as a distance effect of special schools affecting the placement of students. For this purpose, official school statistics including all students at special and regular schools in the years 2010, 2015 and 2020 (N = 11 280 040) are evaluated in a spatiotemporal comparison using Educational Data Mining. In a hierarchical regression model on school placement in inclusive schools, the distance between primary and special schools has the highest influence ( β = 0.48) on the inclusion rate (i.e., the proportion of students with special needs who are educated in regular schools in relation to all students with disabilities), along with the size ( β = −0.14) and the density of special schools in a district ( β = −0.12). The effects differ according to the population density of the region and are stronger in large cities. When the proportion of students with and without SEN in regular schools is considered (support rate), the density of special schools has the greatest impact on school placement ( ß = 43.44). Self‐preservation of schools, traditional funding systems and regional differences between urban and rural areas are discussed as possible reasons.

Background: Computerised adaptive tests (CATs) are tests that provide personalised, efficient and accurate measurement while reducing testing time, depending on the desired level of precision. Schools have different types of assessments that can benefit from a significant reduction in testing time to varying degrees, depending on the area of application, but for which the loss of measurement accuracy has a different impact. The implementation of CAT can take several approaches, each of which can potentially affect the resulting test length and accuracy. Objectives: We compare the methods of estimation‐based CAT and binary‐search‐based CAT to determine the extent to which they are suitable for school assessment in terms of their length and accuracy. Methods: This study uses simulations based on empirical data from a cohort of pupils with and without special needs ( n = 400) to examine the effects of probabilistic estimation‐based CAT and deterministic binary‐search‐based CAT on the length and accuracy of an adaptive reading test for pupils with different ability levels. Results and Conclusions: Estimation‐based CAT leads to a 40% test reduction with an average accuracy of r = 0.96, while binary‐search‐based CAT leads to a test reduction of up to 88% with an average accuracy of r = 0.83. Both methods demonstrate the applicability of CAT in educational environments. Practical advantages and disadvantages of both methods for learning environments are discussed, as well as which method is best suited for specific assessment needs.

Inclusive education is a human right that involves restructuring the school system and closing special schools. However, there is uncertainty about whether all or only some students in an inclusive system should attend regular primary schools and whether all or only some special schools should be closed. This question is usually only discussed in terms of the best possible care and education and the challenge of transportation to school and travel distances is neglected. This article attempts to take both into account. The federal state of Bavaria in Germany does not yet have a fully developed inclusive school system. Using school statistics data for the entire Bavarian primary school population in 2020 (N > 445,000) and geographical data, we simulate deterministically what the Bavarian primary school system could look like with various underlying inclusion models. We analyse the resulting student numbers, support rates of primary schools and school travel distances of former special school students in an urban-rural comparison in four different simulated inclusion models. The results show that all models are potentially feasible. The student numbers at each primary school resulting from the closure of special schools would in most cases not change significantly, whereas the support rate (i.e., the percentage of students with SEN per primary school) would. Schooling close to home favours students in rural areas in particular and significantly reduces their journeys to school by up to 30 km, while for students in urban areas, all inclusive models would be possible. The pedagogical and systemic consequences of the simulated inclusion models are discussed.

Despite widespread political and societal consensus on the importance of inclusive education (Amor et al. 2019), its practical implementation remains challenging across diverse school systems (Haug...