Fairness in assessment is often understood as giving every student the same paper, the same time, the same instructions, and the same conditions. Consistency matters. Without common expectations, it becomes difficult to explain why one response receives a particular judgement and another does not. But identical treatment is not always the same as fair treatment. A task can create barriers that have little to do with the knowledge or skill it is meant to assess.
Consider a task designed to assess a student’s understanding of a science concept. If the instructions contain unnecessarily complex language, the assessment may partly measure reading proficiency rather than the intended science understanding. If the goal is to assess oral communication, a written response cannot provide the same evidence. The first responsibility is to identify the learning being assessed and then examine whether the format allows students to demonstrate it.
Separate the learning goal from the route
Not every assessment should offer unlimited choice. Some skills require a particular form of performance. A handwriting task needs handwriting; an oral presentation needs speaking. The important question is whether a condition is essential to the learning goal or simply a habit of assessment design. When the format is not essential, more than one route may allow students to show the same underlying understanding.
For example, students might explain a historical cause through a structured written response, a labelled sequence of events, or a short oral explanation supported by notes—if the assessment is concerned with causal reasoning rather than writing fluency. The criteria must remain clear. Different routes should not mean different standards; they should mean that an irrelevant barrier does not decide who gets to demonstrate the intended learning.
Designing for fair evidence
Accessibility is not a lowering of expectations
Reasonable adjustments, accessible instructions, assistive tools, additional processing time, or alternative ways of responding are sometimes misunderstood as advantages. A more useful question is whether the adjustment changes the learning goal or makes it possible to assess that goal more accurately. If a student’s disability, language background, or access need creates a barrier unrelated to the skill being assessed, ignoring that barrier does not make the assessment more rigorous.
At the same time, flexibility should be planned rather than improvised as a favour. Teachers need to know which elements of a task are essential, what forms of evidence are acceptable, and how the same criteria will be applied. Students should not have to disclose private information publicly to justify an accessible route. Schools also need to consider time, staffing, assistive resources, and the practical limits of large classes. Fair assessment is a shared design responsibility, not an individual teacher’s burden alone. CAST’s Universal Design for Learning Guidelines 3.0 offer a related design principle: vary and honour methods of response, and plan access to appropriate materials and tools while keeping meaningful learning goals in view.
Audit the task before judging the response
- What exact knowledge or skill is this assessment intended to measure?
- Does the language of the instructions create an unnecessary obstacle?
- Is the response format essential to the learning goal?
- Can students access the materials and tools required?
- Are the criteria consistent across different routes?
- Could a student explain what is expected without guessing the teacher’s preference?
Fairness is not achieved by pretending that all learners arrive with the same resources or experience. Nor is it achieved by removing every challenge. A meaningful assessment should retain intellectual demand while making the route to demonstrating learning accessible. The standard should be clear; the path should not contain obstacles that have nothing to do with the standard.
Illustration: Original AI-generated artwork created for IncludiaTrust for this article. View the image source.
Further reading
- National Academies of Sciences, Engineering, and Medicine. (2018). How People Learn II: Learners, Contexts, and Cultures.
- CAST. (2024). Universal Design for Learning Guidelines, version 3.0.
