The OIA steps in to fill gaps in harassment regulation

Its proposed new section of the Good Practice Framework is being positioned as consolidating existing guidance in one place.

In practice, it goes much further – setting out detailed expectations on everything from how panels should evaluate witness credibility to whether mediation is ever appropriate in sexual misconduct cases.

There’s a good reason for that. In its February 2025 casework note, the adjudicator noted that while harassment and sexual misconduct cases remain under five per cent of its overall caseload, it upholds or settles a higher proportion of these cases than any other complaint category.

When students complain about how their harassment report was handled, they’re more likely to win than in any other area.

The Good Practice Framework has historically been principles-based guidance. But as the GPF has grown, it has started to read differently. Look at paragraph 189 on disciplinary hearings:

For the responding student, it is essential that they can challenge and question the evidence against them. But this does not mean that they must be allowed to put questions directly to any witness, including the reporting student. It will usually be appropriate for responding students to put their questions to the Chair of the panel. The Chair is responsible for deciding whether and when to ask the witness to answer the question.

That’s less a principle, more a procedural requirement – and one that many universities don’t currently follow. The draft goes on to say that Chairs “should document a brief rationale if they require questions to be modified before asking them or if they decide not to ask them at all.”

Paragraphs 202-207 cover evaluating witness credibility with detailed guidance on trauma-informed approaches, training requirements for decision-makers, and warnings about relying on behavioural cues. Panel members will “benefit from cultural awareness training” and should understand why “a reporting student may not come forward for some time” or why “they may have continued to interact with the responding person.”

The level of detail reflects what the OIA is seeing in its casework. The consultation notes that “panel members who must test evidence appear to feel constrained” – a diplomatic way of saying panels aren’t always doing their job properly. And providers show “less certainty and consistency of approach” when supporting reporting students than when running disciplinary processes.

The OIA has found that providers are generally more confident handling the disciplinary mechanics – investigating the allegation, running the panel, reaching a decision – than supporting the person who made the report.

So the draft sets detailed expectations on initial responses (don’t test the veracity of the account at the first meeting), ongoing support (designated points of contact who don’t take on investigatory functions), and informing students of outcomes. On that last point, the OIA notes that:

…in several cases, providers haven’t understood that informing a reporting student that a disciplinary process has taken place is not a complete outcome.

For providers in England, the relationship between OIA guidance and OfS regulation matters. Condition E6 sets out what providers must do – the OIA reviews whether they did it fairly. But E6 is notably light on procedural detail. It says providers must have “credible and fair investigations” without defining what that means.

The OIA is stepping into that gap. Its guidance on burden of proof, panel composition, witness questioning, outcome notification, and appeal rights will become the de facto standard against which “credible and fair” is measured. When a student complains that their harassment report wasn’t handled properly, the OIA won’t be comparing the provider’s process to E6’s broad requirements – it’ll be comparing it to paragraphs 189-201 of the Good Practice Framework.

The two frameworks don’t conflict, but they don’t feel especially coordinated either. E6 emphasises understanding why students don’t report and closing the gap between prevalence and reporting rates. The OIA’s draft lists barriers to reporting – uncertainty about whether behaviour is “worth” reporting, fear of retaliation, embarrassment – but can only address these procedurally.

The draft guidance on “Report and Support” systems says providers should explain that anonymous reports help identify “trends and patterns” and may encourage confidence in reporting. Meanwhile OfS’ free speech guidance seems to present anonymous reporting portals – particularly those allowing free-text identification of individuals – as potentially chilling lawful speech. The OIA can’t resolve that tension – it can only set expectations for how providers handle the reports they receive.

Similarly, the draft includes guidance on freedom of speech considerations – when controversial views cross into creating a hostile environment, how context and manner of expression matter, and whether views are directed at specific individuals. It’s navigating genuinely difficult territory, not least because students’ free speech complaints will come to the OIA rather than OfS’ own impending scheme.

There’s also the question of Wales. E6 only applies to OfS-registered providers in England, but the OIA covers both nations. Welsh providers will be subject to different regulatory requirements – Medr’s condition on staff and learner welfare, currently being consulted on – but the same OIA Good Practice Framework.

For some students, the timelines remain a problem. A taught postgraduate who reports harassment in January may have finished their course before the investigation concludes. International students on Graduate visas have 90 days to find skilled worker sponsorship – they’re not hanging around to see how their complaint is resolved. The draft doesn’t grapple with what meaningful support or effective process looks like when the student is about to leave, or has already left, or is in another country entirely.

There are some other gaps that the OIA can’t fill.

The casework note acknowledges that “students are sometimes put off making a complaint because they are worried that they may be subject to fitness to practise or disciplinary proceedings if the complaint isn’t upheld.” BMA research last week found that two in five female medical students have experienced sexual harassment or assault, with 20 per cent harassed specifically on clinical placement – by senior doctors, consultants, NHS staff and patients. Students described being passed between institutions, each claiming it wasn’t their responsibility.

The OIA’s rules are clear about its jurisdictional limits. It cannot look at complaints about placement providers such as schools or hospitals unless they are OIA members. It can only review whether the university fulfilled its obligations in the arrangements it makes with those outside bodies.

So if a student complains that their university failed to support them after they were harassed by a consultant, the OIA can consider whether the university’s response was reasonable. But it can’t consider whether the trust should have disciplined the consultant, because the trust isn’t an OIA member.

The new draft says partners should “agree” their respective responsibilities. But it doesn’t grapple with the fundamental jurisdictional problem. Given the scale of harassment in professional programmes, this is an area that needs more detailed guidance on what “fulfilling obligations to students in arrangements with outside bodies” actually means – and what universities should be requiring of placement providers in those arrangements. Maybe that’s OfS’ job.

There’s also a capacity problem the draft doesn’t really address. At large universities, separating investigatory and support functions is straightforward – different teams, different buildings and so on. At small and specialist providers, the student services team might be three people, the SU a lone officer. The person a student discloses to may also be the person gathering evidence, or sitting on the panel, or managing the responding student’s support needs.

The draft says reporting students should have access to “independent” support, but the sector is still struggling with what independence means when everyone knows everyone, or when the SU is a single sabbatical officer who received the original disclosure and now sits on the panel. For students at these providers, the perception of independence may matter as much as the reality – and the draft offers little on how to achieve either. Maybe that’s OfS’ job too.

The OIA has long walked an awkward line. It reviews complaints after internal processes are exhausted, which means it often sees what goes wrong rather than what works. Its natural instinct is procedural – if something failed, fix the process.

But some problems can’t be fixed procedurally. Someone reading the section on “Establishing an appropriate environment for study” might expect it to say something about what kind of learning environment students should actually enjoy – the culture they should experience, how harassment should be prevented, what a safe and respectful academic community looks like. Instead, it’s almost entirely procedural – setting behavioural expectations in codes of conduct, defining what constitutes harassment, guidance on neutral language.

This isn’t a criticism of the OIA – it’s doing what a complaints adjudicator can do. But as it fills more gaps, the limits of what procedural guidance can achieve become clearer. The OIA can ensure that when students do report, the process is fair. It can’t make students feel safe enough to report in the first place.

The deadline for responses is Friday 6 February 2026.