Assessment twins: An approach for strengthening assessment validity in the age of generative AI

Abstract

The rise of generative artificial intelligence (GenAI) is raising pressing concerns about the integrity and validity of higher education assessment. Assessment redesign is increasingly seen as necessary; however, there is a relative lack of literature detailing practical approaches. In this study, we introduce the concept of assessment twins as a practical approach to redesigning assessment tasks. We use Messick's unified validity framework to systematically map the ways in which GenAI threatens content, structural, consequential, generalisability, substantive, and external validity. Following this, we conceptualise assessment twins as two deliberately linked components that address the same learning outcomes through different modes of evidence, scheduled closely together to allow for cross-verification. We explain how the twin approach helps mitigate validity threats by triangulating evidence across pedagogically valuable, yet GenAI-vulnerable, assessment formats. To guide implementation, we propose an assessment twin design process: identifying vulnerabilities, aligning outcomes, selecting complementary tasks, and developing interdependent marking schemes. We also acknowledge the challenges, including resource intensity, equity concerns, and the need for empirical validation. Nonetheless, we contend that assessment twins represent a validity-focused response to GenAI that prioritises pedagogy while supporting meaningful student learning outcomes.