Conference Agenda

The increased linguistic diversity in European classrooms has exacerbated equity and assessment validity problems in education. Mathematics assessment tasks often involve a high level of linguistic demand. Although mathematics is often described as a universal language, mathematics assessments typically require extensive use of discipline-specific vocabulary, syntactically complex formulations, dense informational structures, and culturally situated contexts. These linguistic features may act as obstacles for multilingual students who are still acquiring proficiency in the language of instruction. The validity and reliability of digital mathematics assessments in increasingly multilingual European education systems may be seriously undermined when assessment results conflate linguistic proficiency with mathematical ability, systematically underestimating the skills of multilingual learners.

The main research questions guiding this study are as follows: How can linguistic complexity in digital mathematics assessment items be operationalized and quantified systematically? How are various linguistic dimensions related to students’ mathematics performance? What role does linguistic complexity play in the differences between assessment outcomes of multilingual and monolingual learners? These questions are directly connected to European priorities on inclusive education and assessment equity, emphasizing the identification, interpretation, and mitigation of language-related barriers in learning contexts that are increasingly multilingual and digitally mediated.

This study is situated within the Finnish educational system, where DigiArvi, a national digital mathematics assessment platform, has been implemented. DigiArvi provides a rich dataset containing detailed linguistic annotations of assessment items as well as comprehensive background information on students, including language status, prior achievement, socioeconomic indicators, and gender. Finland’s advanced digital assessment infrastructure combined with its increasingly multicultural classrooms offers a timely case study with implications extendable to other European educational settings experiencing similar demographic changes and technological advances.

The theoretical framework draws on sociocultural learning theory and cognitive load theory. Sociocultural learning theory views language as a mediational tool essential for learning and assessing mathematical concepts. Cognitive load theory complements this by suggesting that unnecessary linguistic complexity can limit students’ reasoning abilities. Together, these theories offer a comprehensive framework for investigating how linguistic constraints and supports may coexist in mathematical reasoning throughout the assessment process.

This study contributes to ongoing European and international debates on equitable assessment by expanding the conceptualization of linguistic complexity to a more multidimensional perspective. Unlike traditional readability indices, this approach disaggregates language into distinct components to identify which aspects impede or support equitable assessment outcomes. The findings are expected to influence assessment design by balancing linguistic accessibility with maintaining mathematical rigor, with practical implications for policy and practice. Ultimately, this research aims to support inclusive assessment practices in diverse classrooms throughout Europe.