Reflections on Belonging in Access and Participation Plans
Jim Dickinson’s recent analysis in WONKHE of belonging interventions in approved Access and Participation Plans raises questions that many of us working in this space have felt but perhaps not articulated. His central argument is uncomfortable: that the sector’s consensus approach to addressing belonging gaps may inadvertently reinforce the hierarchies we seek to dismantle. Having spent considerable time developing APPs and working with institutions on implementation, I find myself both recognising the pattern he describes and understanding why it exists.
The question warrants careful consideration. Not because APP professionals are getting it wrong. They are working exceptionally hard within significant constraints, but a structural tension appears to exist between what regulatory frameworks require and what research suggests actually works.
What the Wisconsin research revealed
The evidence Dickinson cites comes from work by Markus Brauer at the University of Wisconsin-Madison. About a decade ago, the institution faced persistent achievement gaps between marginalised groups and privileged students across STEM courses. Brauer’s intervention was neither remedial support nor special programmes for disadvantaged groups. Instead, he tested social norms messaging.
One group of students watched a standard explanation of bias and microaggressions. Another saw students describing the benefits of diversity, revealing that 87 per cent actively supported inclusion. The latter approach had significant effects. Students from marginalised backgrounds reported more inclusive treatment. However, more strikingly, the achievement gap had disappeared entirely by the end of the semester.
The intervention altered what students believed their peers valued. It transformed inclusive behaviour from something exceptional into something normal. This matters because the research suggests that most students already hold pro-social attitudes; they simply underestimate how widely their peers share these values. When you correct that misperception, behaviour shifts.
The pattern across current APPs
When you read through approved Access and Participation Plans, and I have read rather too many of them, a remarkably consistent pattern emerges. Nearly every institution identifies that specific disadvantaged groups report lower belonging scores. Black students, mature students, care-experienced students, students with disabilities and commuter students. The list varies slightly, but the framing remains constant.
Universities then design targeted interventions. Black Unity Events providing spaces for Black students to “authentically be themselves”. Mature student networks and social “mingles”. Care-experienced student buddy schemes. Disability-specific groups. Commuter-specific transition support. The language across institutions is strikingly similar: disadvantaged groups “may struggle to feel they fit in”, “can lack a sense of belonging”, “feel disconnected” or “isolated”.
I should clarify that I’m not suggesting these interventions are not developed thoughtfully. They are often delivered in consultation with students by skilled professionals who genuinely care about student success. The programmes reflect genuine understanding of the challenges students face. Yet the structure remains consistent: belonging is framed as something specific groups lack, requiring special intervention to remedy.
Consider how this differs from the Wisconsin approach. Brauer’s intervention said, “Most students already value diversity. Here is proof”. The typical APP approach says, “We will create spaces where Black students can feel they belong”. Wisconsin said, “Inclusive behaviour is normal here”. APPs say “we will train mature students how to access support networks”. Wisconsin changed what everyone believed about everyone else. APPs provide disadvantaged groups with resources they apparently lack. The Wisconsin research explicitly warns against this. Empowering marginalised groups through special initiatives can paradoxically highlight their different status. What is intended as support becomes a marker of otherness.
Why this pattern exists
The question becomes: why are institutions doing this? The answer is not that they have misunderstood the problem or ignored evidence. Rather, the APP framework itself appears to push institutions toward particular types of intervention.
When developing an APP, you must demonstrate spending on disadvantaged groups. Your interventions must be documentable, measurable by the group, and defensible to regulators. You need to show which target group received which intervention, at what cost per head, with what expected outcomes. This architecture makes perfect sense from a regulatory accountability perspective. It becomes problematic when the evidence suggests that effective belonging interventions work differently.
How would you report “we surveyed students and publicised that 78 per cent actively welcome international students” in an APP? Which target group received that intervention? What was the spend per head? How do you attribute outcomes when you have changed perceived social norms for everyone rather than providing a service to a specific demographic? The intervention that research suggests is effective does not easily map onto the APP template.
Similarly, you cannot write “we made peer support universal and student-led because that is simply how induction works here” because that reads as mainstream provision, not an access and participation intervention. You cannot write “we redesigned tutorial systems so they work properly for everyone by default” because that sounds like you are spending APP resources on privileged students.
The result is that institutions, acting entirely reasonably within the constraints they face, design interventions that fit the framework. Targeted programmes. Characteristic-specific support. Documentable spending on identified groups. The APP architecture rewards this approach whilst making alternatives difficult to propose or evaluate.
The genuine dilemma
This creates a real dilemma for institutional leaders developing APPs. You are responding to regulatory guidance that explicitly asks you to consider belonging. You are addressing genuine evidence that disadvantaged groups report lower scores on belonging. You have gaps in continuation, completion, and attainment that demand attention. The instinct to create support structures for specific groups feels responsible and caring.
Moreover, some targeted support remains entirely necessary. Financial hardship funds. Specialist mental health provision. Disability services. These are distinct from belonging interventions and should continue. The question specifically concerns how we structure the social and academic architecture that enables students to develop a sense of belonging.
There is also the matter of what institutions can reasonably accomplish. Research on social norms messaging suggests that its effectiveness partly depends on revealing what is actually true about student attitudes. If you survey your students and discover that only 45 per cent actively support diversity, you cannot simply assert otherwise. The intervention requires truth-telling. This means institutions need to know what their students actually believe, which requires asking rather than assuming.
Then there is implementation. Even if an institution wanted to shift from targeted belonging interventions to universal design approaches, how would this appear in an APP? Would it be approved? Would it satisfy monitoring requirements? These are not rhetorical questions but genuine uncertainties about what the regulatory framework will accommodate.
There is also a curious tension between regulatory frameworks themselves. Previous TEF guidance has asked institutions to demonstrate holistic approaches to student success, showing how teaching, assessment, support and enhancement work together for all students. Yet APP frameworks require demonstrating targeted interventions for specific disadvantaged groups. At once, we’re asked to consider the entire cohort and focus on specific characteristics. Both demands are legitimate. Both reflect important priorities. Yet they can pull institutional strategy in different directions, making it harder to adopt the universal design approaches that research on belonging suggests may prove most effective.
Questions, not prescriptions
I am conscious that analysing the problem proves considerably easier than proposing solutions. Indeed, I am not certain there are straightforward solutions when regulatory architecture and implementation realities pull in particular directions. Yet several questions seem worth considering.
Does the APP framework itself constrain institutions from adopting approaches that research suggests may prove more effective? The structure of required reporting, target-setting, and evaluation may inadvertently favour interventions that sound documentable over those that evidence suggests are effective. This is not a criticism of the OfS, which has legitimate accountability concerns, but an observation about unintended consequences.
How might regulatory frameworks create space for universal design approaches? If the most effective belonging interventions focus on what all students believe about each other rather than providing specific groups with additional resources, how should this be reflected in planning and monitoring? What would success look like, and how would we measure it?
What would it mean to work with advantaged groups as the primary intervention? The Wisconsin research suggests that closing the achievement gap requires changing what white students believe and how they behave. Yet APP spend focused on white students to benefit Black students would likely appear problematic in current frameworks. This raises difficult questions about where intervention should be directed.
Can we measure changing social norms rather than counting participants? Traditional APP evaluation counts the number of students served, resources provided, and programme completion rates. If effectiveness depends on shifting what communities believe about themselves, our evaluation approaches may need to be rethought. This poses real challenges for demonstrating impact and value for money.
Perhaps most fundamentally, how do we balance genuine support needs with evidence on what creates a sense of belonging? Students from disadvantaged backgrounds do face additional challenges. They do need support. But if providing that support through characteristic-specific programmes reinforces their sense of difference, we face a genuine paradox with no obvious resolution.
Where this leaves us
Every current APP includes detailed commitments to targeted belonging interventions, complete with evaluation frameworks and expected outcomes. Institutions have allocated budgets, designed programmes, appointed staff and set objectives. Rowing back is not straightforward, even if the evidence increasingly suggests the approach may be counterproductive.
Yet the evidence does warrant attention. If belonging depends more on universal structures and perceived norms than on programmes for specific groups, this has significant implications for how we approach access and participation work. The challenge for both institutions and regulators is determining how to accommodate this within frameworks designed with different assumptions.
Dickinson suggests that what actually works – changing social norms, creating universal structures, enabling student leadership – may be easier and cheaper than current approaches. Perhaps so. However, it requires both institutions and their regulator to acknowledge that the problem may have been framed in ways that make effective solutions more difficult to implement.
I do not have definitive answers to these questions. I have learned that quality enhancement often works quite differently in practice than in policy frameworks. The tension between institutional autonomy and system-level planning remains a real concern. What I can say with some confidence is that when regulatory architecture and research evidence point in different directions, we need to at least ask why, and whether frameworks might evolve to accommodate what we are learning.
The sector has developed considerable expertise in access and participation work. We have skilled professionals, robust evidence bases and genuine commitment to reducing gaps. The question is whether our regulatory frameworks enable or constrain that expertise from being deployed in ways that research suggests will prove most effective. That seems worth examining, even if the examination proves uncomfortable.