Assessment for inclusion: rethinking contemporary strategies in assessment design

ABSTRACT Assessment has multiple purposes, one of which is to judge if students have met outcomes at the requisite level. Underperformance in assessment is frequently positioned as a problem of the student and attributed to student diversity and/or background characteristics. However, the assessment might also be inequitable and therefore exclude students inappropriately. To be inclusive, assessment design needs to be reconsidered, and educators should look beyond simplistic categories of disability or social equity groups, towards considering and accounting for diversity on many spectra. This article introduces the concept of assessment for inclusion, which seeks to ensure diverse students are not disadvantaged through assessment practices. Assumptions in assessment design are problematised from this point of view, and three central concerns relating to assessment traditions, assessment expectations, and academic integrity are interrogated. Contemporary design strategies of authentic assessment, programmatic assessment, and assessment for distinctiveness are then harnessed to illustrate approaches to assessment for inclusion. Assessment for inclusion therefore builds on the synergies between inclusive practice and good assessment design.


Introduction
Assessment in higher education should not disadvantage diverse students because of characteristics or abilities extraneous to the outcomes being judged. All students should be supported to achieve and demonstrate capability in an equitable manner. In the move from elite to mass higher education, student diversity has increased: students from equity groups including disability, and/or non-traditional backgrounds have been encouraged to pursue higher education (Marginson, 2016). Previously, underperformance in assessment attributed to equity group membership has been largely viewed as a problem of the student (O'Shea et al., 2016). This has been addressed with accommodations and/or additional support. The positioning of the student as deficient ignores a problematic possibility: that the assessment is not fit for its multiple purposes. To credential learning (i.e., assessment of learning), assessment needs to discriminate between those who have and who have not met the outcomes at the requisite level. Assessment should also challenge students to move out of their comfort zone to promote their development (i.e., assessment for learning) (Boud, 2007). However, assessment might also discriminate against and exclude students through requirements that are irrelevant to the outcomes being judged . If this is indeed occurring, it calls into question the validity of assessment. Therefore, assessment design must be reconsidered from the perspective of inclusion.
Inclusion has been equally used to apply to disability inclusion, and social inclusion in higher education (Stentiford & Koutsouris, 2021). These terms encompass many equity groups which are usually named in relation to disability access (including physical disabilities, learning disabilities, and mental and physical health conditions) and widening participation initiatives (e.g., students from low socio-economic backgrounds, women in nontraditional areas, First Nations peoples). Despite specific equity labels, the people represented by these types of groupings are not homogenous, and the potential for 'equity sub-groupings' are numerous: one study identified 17 sub-groups from research with 35 participants (Willems, 2010). However, rather than dwelling on students' particular membership of equity groups, action to improve inclusion should focus on commonly experienced underlying problems in conventional assessments. Indeed, it has already been pointed out that in one country, the specific institution had a greater impact on student success than characteristics of students themselves such as age, gender, type of attendance, mode of attendance, or socio-economic status (J. Ryan et al., 2020). This suggests that what happens at an institutionincluding assessmentis more important than who attends.
For inclusion to become a more central notion within assessment design, the nature of assessment tasks must be problematised and examined deeply. Possibilities for inclusive assessment are unlikely to be easily identified by those responsible for assessment without explicit attention, since they are prone to see them from the perspective of people like themselveslikely those who have thrived on dominant assessment approacheswho share similar values and ideas of success (Leathwood, 2005). Furthermore, trying 'harder' with more of the same may not always help inclusion. For example, clarity of assessment instructions alone will not promote inclusion, since it is a myth that if only we specify in enough detail, everyone will know what is being required (Bearman & Ajjawi, 2021).
This article focuses on designing assessment for inclusion: assessment that recognises diversity in student learning, and endeavours to ensure that no student is discriminated against by virtue of features other than their ability to meet appropriate standards. We aim to prompt conversations about how this might be achieved by assessment practitioners and researchers. First, we consider what assessment for inclusion is. We review aspects of assessment that create unnecessary barriers to student achievement. Common assessment design concerns are identified and interrogated. Then, we explore how contemporary assessment approaches could be better harnessed towards assessment for inclusion. Throughout this work, we draw on relevant literature across disability and social inclusion to move discussion beyond accommodating designated equity groups toward identifying overall directions for assessment design.

What is meant by assessment for inclusion?
Discussions on how to improve student equity within higher education assessment have occurred for some time (Kurth & Mellard, 2006;Riddell & Weedon, 2006). Waterfield and West (2006) introduced the idea of 'inclusive assessment', arguing that all student capabilities should be accounted for proactively. This superseded previous approaches of 'contingent assessment' (adapting existing assessment with specific accommodations), and 'alternative assessment' (offering a different assessment). In her report on inclusive learning and teaching, Hockings (2010, p. 34) defined inclusive assessment as 'the design and use of fair and effective assessment methods and practices that enable all students to demonstrate to their full potential what they know, understand and can do.' She identified substantial evidence that assessment did advantage and disadvantage particular groups of students and suggested that providing a choice of alternative assessment methods could mitigate this. However, it has been noted that 'current assessment methods introduce barriers to a wide range of students' (Lawrie et al., 2017, p. 7), and so it is unlikely that these recommendations have translated into widespread improved practice in higher education .
Taking a theoretically informed position, the broader concept of assessment for social justice was introduced by McArthur (2016, p. 968) as speaking to 'justice of assessment within higher education, and to the role of assessment in nurturing the forms of learning that will promote greater social justice within society as a whole'. McArthur (2018) critiqued the assumption that 'fair' treatment of students in assessment processes results in just outcomes. Fairness in assessment is an expectation, but it can be thought of both in terms of procedural fairness (equal treatment) and fairness of outcomes (justice) (O'Neill, 2017). A focus on procedure alone can impair inclusion through continuing to promote narrow conceptions of success. Instead, a multifaceted conception of fairness in assessment, taking complexity into account, should support equitable and just student outcomes (McArthur, 2018).
Drawing on the general philosophy of assessment for social justice, and noting the importance of outcomes, we suggest 'assessment for inclusion' captures the spirit and intention that a diverse range of students and their strengths and capabilities should be accounted for, when designing assessment of and for learning, towards the aim of accounting for and promoting diversity in society. In introducing this new term, we hope to better negotiate praxis: that is, joining together theory and practice, to act as a lever to achieve change in assessment but also through assessment. The overall aim is still to achieve assessment for social justice through focusing on assessment design, since design needs to consider not only the immediate task but the wider course and institutional context (Dawson et al., 2013).
Assessment for inclusion recognises that diversity can exist on many spectra. How these spectra overlapwhat can be termed intersectionality (Crenshaw, 1991)needs to be considered in what and how we assess. Any learner, no matter what their characteristics or background, must only be judged in terms of the necessities of what a task seeks to assess, rather than any feature or ability that is not relevant to these outcomes. There is also a need to continually question if particular characteristics have any fixed association with task capability, and how the task might be discriminatory. Simultaneously, it must also be acknowledged that contextual factors might impact on the demonstration of capability, since assessment exists within a given contextthat is, there are limits to all assessments' generalisability. Additionally, as new social insights emerge, we will become aware of new ways in which we are not yet inclusive. Thus, designing assessment for inclusion should be considered an ongoing, future-facing process, rather than a fixed view on what has been excluded in the past.

Current problems with assessment design
A challenge in assessment for inclusion, common to many fields of practice, is the difference between 'work as imagined' and 'work as done' (Hollnagel, 2015). While it is easy to imagine optimal assessment practices, it is much more difficult to implement them in messy, real-world environments, where budgets, time and departmental politics come into play (Bearman et al., 2017). Any form of assessment is constructed within a particular time and place, so inclusive practices need to take account of this. In this section we explore how assessment operates in practice, focusing on the assumptions that often accompany academic traditions.
The specific requirements for how assessments are designed and conducted in any given discipline or institution are often considered sacrosanct. Contemporary assessment tends towards solo, unaided performance at the expense of working with others (Lipnevich et al., 2021). Further, it removes students from the normal resources (e.g., Internet access, the advice of colleagues) that graduates would typically access in everyday practice. These unchallenged limitations are likely to have more of an impact on the success of those who might gain the most from an inclusive approach to assessment. However, as illustrated in Table 1, on closer inspection, many limitations are not an inherent requirement of a particular task, are potentially excluding students, and thus should be questioned. Examining these assumptions provides useful insights into how educators may need to think differently within their own contexts. Furthermore, promoting more inclusive assessment practices can be a matter of incrementally improving assessment or course design. This interrelationship between good assessment design and inclusivity is exemplified in Table 1, where nine common assessment assumptions (drawn from the authors' experiences in designing and researching assessment), critiques of this assumption both from assessment design and inclusion perspectives, and the potential disadvantages that are a consequence of these assumptions are detailed.
When exclusionary practices in assessment are recognised, they are generally addressed by providing adjustments for those students who need them. This is not inherently problematic, since students are diverse with individual circumstances, so any instance of assessment is unlikely to account for everyone's needs and capabilities. However, while adjustments for some students will always be required, adjustments are not a panacea and often place an additional burden on those who need them. Commonly, teachers only tinker at the peripheries of these assessment assumptions (e.g., providing a break or extra 10 min for students) (Lewandowski et al., 2013), and do not deal Common assumption about assessment Why question from an assessment design perspective?
Why question from an inclusion perspective?
1. Exams should be closed book. Implies that memorisation of content is the valued outcome, which may not be conducive to learning.
Advantages those who can recall information quickly under pressure. These may not be characteristics necessary to demonstrate the specific outcomes being judged. 2. Students should only take a specific set of notes/pages/texts into exams.
Privileges student capability to construct notes in usable form, which may be far from the intended capability under examination.
Advantages those who have greater access to resources, more experience, support or time to construct resources. For example, those who cannot afford the textbook or have little experience of exam techniques may be disadvantaged although these are situational factors rather than performance-related. 3. A task must be responded to in a particular mode, most often written, such as an essay.
Common formats may shift student attention to capability relating to that mode rather than the outcomes being assessed. Further, these formats might be rolled over for re-use since they are familiar or already exist, rather than being optimal for the task.
Often the mode itself is heavily emphasised rather than the task itself. Some modes exclude some students without being relevant to the task at hand. For example, handwriting in exams can be cognitively or physically exhausting; essays privilege a form of writing that may not be a skill required within a particular unit; oral presentation can cause severe anxiety but may not be needed. 4. There must be strict time parameters for assessments such as exams.
Very constricted timed tasks, like exams, generally do not reflect the real world. In a standards-based approach to grading, timing should not matter, unless it is clearly specified as part of the assessed learning outcomes.
Timed tasks advantage students who can concentrate immediately, maintain focus for the duration of assessment, perform the task quickly and/or perform well under stress. Students who have physical or cognitive conditions that prevent them from doing so are disadvantaged. 5. Tasks cannot be substituted, even if alternatives equally assure learning outcomes from those designed Only offering one task reduces opportunities for distinctiveness and creativity, where portrayal of achievement is representative of the experience or aspirations of individuals.
Alternative tasks allow for broader ways of knowing and communicating, for example Indigenous ways of knowing. Moreover, representation of achievement can be more broadly considered in ways that can never be possible by a single individual. 6. The linguistic standards required are those of a sophisticated first language speaker In many cases, language is not the underlying capability the assessment seeks to assure.
Emphasis on high linguistic standards when these are not necessary for the task advantage those who are studying in their first language, and disadvantage those who are studying in an additional language. It can also disadvantage those who have disabilities that interfere with rapid reading or processing of texts. 7. Inflexible deadlines for assessment must always be adhered to.
Timing in assessment design manages a range of competing tensions between the practical resourcing issues of grading, returning comments on work to students prior to the next assessment, and results submission requirements for graduation and admission to subsequent programmes. Rigidity Rigid timelines advantages those who have no other commitments. They disadvantage students who have work or carer commitments in addition to study, and those who have fluctuating conditions who cannot predict when they may be well, or ill, or not travelling. Special consideration requirements can involve substantial paperwork which adds to the time (Continued) with the central components of the task, in its format and its content. Problematising commonly implemented adjustments and their relationship to assumptions about assessment is therefore also important for inclusion. Moreover, only taking an adjustments-focused approach implies that the core requirements of the assessment are unquestionable and untouchable. In the COVID-19 pandemic, much of what was previously considered 'untouchable' (in assessment, and more generally in higher education) has indeed changed when really required (Bartolic et al., 2021;Watermeyer et al., 2020). This suggests different processes could be designed to support assessment for inclusion. Whilst the inclusive education literature is sizeable, with many conceptual articles about what should be done in assessment, there is less empirical work on assessment (Lawrie et al., 2017;Roberts et al., 2015;Tai et al., 2021), and what exists is focused mainly on students' perceptions of assessment. Interventions to promote inclusion in assessment have focused on students with disabilities, especially those with dyslexia (Matheson & Sutcliffe, 2017;O'Neill, 2017), and where the language of instruction is an additional language, for instance, international students and students in countries with several recognised languages (Hurst & Mona, 2017;Kaur et al., 2017;Ragpot, 2011). Most of these studies focus on choices in assessment: again, a tinkering around the edges rather than questioning the central tenets of the assessments. To our knowledge, only one has explored the impact on student performance through comparison of grades (O'Neill, 2017). There is little focused discussion regarding broader scale implementation of inclusive assessment within higher education, and how inclusion might mesh with common assessment dilemmas. It is important to engage with these dilemmas, and so the following section identifies three substantial concerns in assessment design. Assumptions about easy access and operation of assessment technologies tend to advantage students who have up-to-date hardware and software. Students with older computers or limited access to technology and internet may be disadvantaged. Without appropriate testing, students who use assistive technology may be disadvantaged. 9. Teachers always set tasks, outcomes, and criteria.
In promoting student capability to continue to produce high quality work, students themselves may need to own the assessment process. Thus, joint setting of criteria and, on occasion, outcomes for a unit can be a way for students to sustain their learning into the future.
All people, including teachers, hold individual perspectives. Assessments that allow others to express their own view of quality, suitably moderated, can include diverse ways of knowing and help reduce problematic oversights. Thus, the academy shows it is open to students' valuable perspectives and experiences which can enrich understanding.

Three concerns to attend to when thinking about assessment for inclusion
Research on assessment design has noted that educators commonly externalise barriers preventing change, referring to policies, procedures, or professional accreditation bodies which limit alterations to assessment (Bearman et al., 2017). These often are not actual barriers but beliefs, which may perpetuate the practices outlined in Table 1 and hamper learning through assessment. These are important concerns to attend to, and so in this section, we probe some key arguments, which can be phrased simply as: 'this is a necessary part of how we do things here'; 'we need to assess these learning outcomes'; and 'students will cheat if we don't assess in this way'. For each we offer some possible solutions.

Entrenched perspectives and practices
Across disciplines and professions, academics have worldviews or perspectives, and have their own epistemic bases for conducting assessments in particular ways (Ashworth et al., 2010). Additionally, local institution specific assessment traditions may exist, and students will also bring their pre-existing experiences and assumptions to assessment. These perspectives contribute to establishing what types of assessment are valid, and who should succeed in those assessments. Students can be influenced by socially constructed norms or stereotypes, for example, gender, race, or class assumptions which can impact perceptions of competency, and therefore motivation (R. M. Ryan & Deci, 2000).
Students might interpret messages about who belongs in higher education through particular assessment activities, such as examinations or oral presentations. Solutions commonly employed to redress student deficitssuch as writing workshops, elocution lessons, dressing differentlyor in some cases, recommending a different course of study, encourage conformity. Engaging staff in discussion to facilitate reflection and identify effects of these taken-for-granted assumptions may create opportunities for change. Low-stakes early assessment might also build student confidence through providing opportunities to practice and develop capabilities.

What success is judged against in assessment
In assessment, work is always judged against some concept of quality. These notions of quality may be explicit standards or criteria, or be held implicitly by a group of professionals (Bloxham et al., 2016). Present standards-based quality frameworks require learning outcomes to be explicitly identified, and for assessment to assure that students meet them. In such a context, the learning outcomes that assessment is seeking to judge should be examined in addition to the assessment itself. There is a risk when operating in such a framework that learning outcomes become rigid and inflexible, cementing assessment practices in place when it is the learning outcomes themselves that need to be challenged. An unreflexive interpretation of outcomes and the way they should be assessed can lead to inadvertent exclusion. Students might very well meet requirements if they were defined in other equally legitimate ways. Furthermore, assessments do involve compromises between competing interests and ideals to meet stakeholders' needs in the real world (Bearman et al., 2017). Instead of buying into the belief that a particular stakeholder only accepts one assessment configuration, discussions should be convened with those stakeholders (e.g., accreditation bodies and industry representatives) to consider how outcomes and the assessment of them are defined, and therefore what different assessment designs might be acceptable.

Preventing cheating on assessment
The spectre of academic dishonesty shapes much of what is done in assessment. The traditional, closed-book exam is often employed out of a fear that without invigilation students will cheat, presenting a threat to the validity of assessor judgements, whilst satisfying equality (but not equity) since conditions are the same for all. Approaches to address cheating often come with consequences for inclusion. To stop students sharing their answers with peers who are yet to sit the exam, educators often elect to have only one exam sitting, but the inability to choose an exam sitting is exclusionary towards those who may be unavailable at that time, e.g., due to caring responsibilities. The exclusionary effects of exams are so potent there is even a research literature around test anxiety (Woldeab & Brothen, 2019), a trait which is associated with poor exam performance. Remote proctored exams have been criticised as ableist due to features like eye tracking that expect to see unobstructed neurotypical eye movements (Logan, 2020).
Addressing the problems of cheating while being inclusive is a worthwhile challenge. The positive, values-based mission of academic integrity, which aims to develop students' ability to work with integrity at university and beyond (Fishman, 2014), is consistent with values of inclusion. However, approaches such as integrity education and honour codes do not on their own sufficiently address the problem of cheating. As a result, institutions deploy assessment security (Dawson, 2021) approaches such as invigilated exams or textmatching software to enforce anti-cheating rules. Such approaches, which do not guarantee integrity, act to enforce a limited set of normative behaviours in assessment which can operate to reinforce exclusion. The upholding of integrity is undoubtedly important, but it must be enacted in ways that embrace inclusion. Any proposed integrity or security practice must be judged against its effects on the most diverse students.

Repurposing existing ideas in assessment to effect inclusion
There is a substantial overlap between what might be considered good assessment design practice to promote assessment for learning, and assessment for inclusion. Contemporary approaches like authentic or programmatic assessment were not originally developed for reasons of inclusion. In this section we illustrate how three assessment design approaches can readily be harnessed to better serve an inclusion agenda across students' background characteristics, present capabilities, and to support their future goals. We encourage readers to consider how these ideas might be adopted within their own local context using the prompting questions that conclude each section.

Authentic assessment
Authentic tasks and authentic assessment have been highlighted as useful to engage students in learning and for the world beyond university (Gulikers et al., 2004). Authentic assessment needs to engage the whole person; to integrate what students know, how they act and who they are (Vu & Dall'Alba, 2014). Whilst what counts for authentic is contested since authenticity is individually perceived, there are several dimensions of assessment design that might contribute to greater authenticity. Dimensions can include attributes of the task such as realism in context and realism to the problems faced in professional settings (Villarroel et al., 2018). Students' perceptions of alignment contribute towards authenticity (Smith & Worsfold, 2015). Even when authenticity is designed into assessment, if the student does not perceive the assessment to fit with their own goals, aspirations and expectations, this can render the assessment inauthentic to the student leading to instrumentalism and cynicism (Ajjawi et al., 2020). Authentic assessment must therefore create space for students to integrate their values, capabilities, and their future aspirations. This could enhance inclusivity, since students are able to explore ways of learning and working which are suitable for them.
Authenticity may also involve topics or tasks which are challenging and provoke anxiety and stress in students (e.g., in health, social work, justice), particularly if students are survivors of past trauma. In these cases, if professional practice requires graduates to engage in such activities, then students must be suitably supported to engage in such activities. Given students may have not disclosed their condition or experience to the university or think it is irrelevant, being proactive in mitigating harms through setting expectations of types of encounters and regular offering of available services is important. For instance, the death of a loved one might impact learning in a hospital placement.
Another dimension of authenticity from an assessment security perspective is 'authentic restrictions' (Dawson, 2021), where authenticity also encompasses societal/community/workplace expectations about acceptable ways in which a task might be completed, and thus support students to be able to negotiate how to demonstrate competence beyond the university as well. For instance, a journalist might have a strict deadline for submitting copy, but be able to access spelling, grammar and thesaurus facilities. There is the usual caveat that assessment tasks which closely parallel what a practitioner might do may not be seen by students to be authentic in their terms, whereas an artificial task with no relation to practice, if dressed suitably, might be seen as authentic to students. Thus, to gauge and maintain authenticity, stakeholders including students should be involved in assessment design to provide feedback and enable refinements to be made.
Considerations regarding authenticity therefore include: under what circumstances and in what ways is it legitimate for an authentic assessment task to not be adjustable to meet the needs of a diverse student population? When is a task necessary and nonnegotiable for professional practice? And how can we support students to participate in and learn in these situations?

Programmatic assessment
Assessment decisions are often made at the level of individual course units. This means that the outcomes of course units may be well addressed, but not those of the overall programme. Assessment for inclusion needs to be considered at the programme level as well as that of course units. Assessments that address programme outcomes need particular attention in the inclusion context, since they can make or break overall student success: they have a gatekeeper function which facilitates access to professional worlds. Individual unit assessments might be able to be modified with little implication for the whole programme but considering programme assessment activities is crucial to maintain the accreditation and/or credibility of a course.
Recognising that any assessment involves compromises to meet varied stakeholder needs, it may help to consider how the outcomes of individual assessments might be synthesised to provide a broader picture of a student's capabilities, and to view learning and demonstration of capabilities across time and space (Schuwirth & Van der Vleuten, 2011). This could account for a constellation of approaches, which when taken separately may be imperfect but serve some student groups better than others. However, when taken together they may provide equitable opportunities to demonstrate capability. Though students are just as concerned as educators about their development of a suite of capabilities to prepare them for the world beyond university (Morris et al., 2019), not all skills need be assessed at once, so long as they are assessed during a programme of study. Considering which combinations should be assessed together may also call into question why there is a focus on given formats of assessment, and what content is included across the programme of assessment. Explicitly focusing on student development and progress over time may also have flow-on effects to support student learning in a more holistic manner.
Considerations with regards to programmatic assessment therefore include: at which point should certain capabilities be assessed and assured? Should significant weighting for accreditation be placed on a few high-stakes tasks? How can a broader assessment landscape be represented to students?

Assessment for distinctiveness
To succeed as graduates, students should not aspire to be just the same as other graduates, rather, they need to understand and portray evidence of achievement that distinguishes them from their peers and predecessors. The preferences of employers and their requirements to fill specific roles are highly variable, and graduates with the same qualifications occupy vastly distinct roles in which different subsets of skills and personal attributes are most valued. As students graduate from increasingly large cohorts holding apparently identical qualifications which are not constructed to allow them to personalise their achievements, there has been growing recognition that students need opportunities to differentiate themselves, and that assessment has a role to play in portraying difference as well as commonality (Jorre de St Jorre et al., 2021;Jorre de St Jorre & Oliver, 2018).
It is naïve to assume that offering the same assessment to all students provides equal opportunities for achievement: the conditions under which assessment takes place are never identical because of students' unique personal histories and lived realities. In requiring that students perform the same tasks and be judged against the same standards, assessment fails to acknowledge the value of different perspectives, skills, personal attributes and experience. This also reduces opportunities for students to build on prior learning and pursue growth relevant to their personal aspirations. Instead, assessment could be designed in more open-ended ways that support students to develop and demonstrate their distinctive capabilities. This aligns with Universal Design for Learning guidelines, which suggests that a variety of means of engagement in and expression of learning should be offered (CAST, 2018). Criteria would then need to be developed to take into account this range of possibilities.
The ways in which universities judge and represent achievement have lagged other curriculum innovations. Therefore, considerations relating to assessment for distinctiveness include: How can students be supported to generate diverse artefacts for assessment purposes, demonstrating their strengths, aspirations and experience? What boundaries might be set for such performances?

Conclusions
Assessment for inclusion is a perpetual endeavour requiring robust conversation. Choices in assessment design are never neutral, as each may promote or constrain inclusion differently, and affect different people. While problems with assessment are easy to identify, as done in this article, what works in certain contexts or circumstances is unlikely to work everywhere, so perfect solutions are unlikely to exist at a global level. Changing assessmentsreimagining and redevelopingwill not be easy for educators. Students might also be used to previous ways of being assessed, and resist change. Students usually advantaged may object to their privileges being interrogated or reduced. Some of the work to be done in pursuit of assessment for inclusion threatens previously comfortable ways of working, since it may disrupt expected practices of teaching, learning, and assessing. Within this, the roles of student and academic in higher education systems must be revisited to better support the expected diversity of future cohorts.
Our account here has several strengths and limitations. Firstly, we have undertaken this work from an assessment perspective, and as such our understanding of perspectives beyond these is tempered by our orientation towards assessment and feedback. Drawing further on socio-political theory will help to interrogate assumptions in assessment (McArthur, 2018;Nieminen & Pesonen, 2021). Secondly, we have not involved the voices of those beyond the academy. Future work should collaborate with a broader audience, drawing in stakeholders from diverse perspectives, involving a deliberately diverse population and hosting discussions in many fora across universities. However, this article has been an important initial step to question assumptions about assessment, towards taking on new and diverse ways of thinking about assessment for inclusion.
There is considerable value in working towards inclusion and socially just forms of assessment which every educator can be part of (Moriña et al., 2020;Nieminen & Pesonen, 2021). The influence of COVID-19 has demonstrated that radical assessment change is possible. It is not yet known if any of these changes will become permanent, but the possibilities of alternatives have been directly experienced through a massive natural experiment. This momentum must not be lost. There is an overlap between inclusive practice and good assessment design, since they both require the exclusion of spurious factors in judging student capability, through questioning our assumptions and who and how they impact. Approaches such as authentic and programmatic assessment design, and a focus on distinctiveness may promote inclusion. Designing assessment for inclusion not only changes assessment practices, but fundamentally reconsiders what discriminations are and are not appropriate within assessment. It is highly important to ensure that diverse students have equitable opportunities to learn from assessment and be equitably judged within assessment.