Introduction
Mathematical reasoning is a fundamental competence in mathematics education (Khardita et al., 2023). Although there is a lack of consensus on a unique definition of mathematical reasoning (Jeannotte & Kieran, 2017), several measurement tools have nonetheless been proposed, from traditional standardized tests to adaptive digital platforms and AI-enhanced systems (Mustafa, 2024). However, no comprehensive review has systematically examined how mathematical reasoning is defined and assessed, especially regarding validity and fairness (Zumbo, 2009).
Purpose
This systematic review aims to identify, categorize, and critically evaluate tools used to assess mathematical reasoning in educational settings.
Method
Following PRISMA guidelines, a systematic search will be conducted in Scopus, PsycINFO and ERIC to find peer-reviewed studies. Included studies will focus on tools explicitly designed to assess mathematical reasoning at all educational levels. Tools will be categorized by features such as instrument type (e.g., paper-based, computer-based, AI-enhanced), development method (manual, AI-assisted) and task characteristics (e.g. multi-step problems, visual-spatial tasks). Evaluation will follow Zumbo's framework of validity (2009), which treats validity as an integrated concept including reliability and fairness.
Results
The review is expected to reveal a wide range of tools, reflecting differing definitions and characteristics. Variation is also expected in the quality of validity evidence. Findings may highlight gaps in equitable and meaningful score interpretation across diverse populations.
Conclusions
This review will provide a timely synthesis of current definitions and assessments of mathematical reasoning, emphasizing validity and fairness. The findings will guide development of psychometrically sound and equitable tools, supporting ethical educational practices.