The Students Who Don't Feel at Home

Our data from 14,000 residents across 191 properties shows a belonging gap that the student accommodation sector has barely begun to address. Transgender students are 16 points less likely to feel at home. Non-binary students score 11 points lower on satisfaction. And the gap isn’t closing.

There is a question in the Investor in Students survey that asks residents whether they feel “at accommodation,” the phrase we use instead of “at home” because we want to understand whether the property has become a place of belonging rather than merely a place of sleep. It is, in my experience, the most revealing question in the entire survey, because belonging is the thing that separates a building from a community, and a community from a place that students actively want to return to.

When you split the Spring 2026 responses, 14,189 of them, across 191 properties and 56 cities, by gender identity, the results are not subtle.

Male students scored a net promoter score of +51.3, with 81% reporting they had made close friendships and 54.6% saying they felt at accommodation. Female students scored +49.2, with 82.9% reporting close friendships and 53% feeling at accommodation. These numbers are broadly consistent with what we see year on year: a healthy, engaged resident population that, on the whole, is having a positive experience.

Transgender students scored +44.2. Non-binary students scored +40.4.

The gap between male students and non-binary students is 11 points of NPS. The gap in feeling “at accommodation” is 9 points. Among transgender students, only 38.2% say they feel at accommodation, compared to 54.6% of male students. That is a 16-point gap in belonging, across a dataset large enough to be statistically meaningful and consistent enough to be operationally urgent.

These are not marginal differences. They describe a fundamentally different experience of the same buildings, the same staff, the same events, and the same service model.

The Friendship Collapse

The belonging gap is driven, in large part, by a friendship gap, and the friendship data is where the numbers become genuinely uncomfortable.

Among cisgender students, 81-83% reported making close friendships during their time in accommodation. Among transgender students, that figure drops to 69.7%. Among non-binary students, it’s 72.2%.

This is not a small difference. A 12-14 percentage point drop in friendship formation represents hundreds of students, across our portfolio, who are living in the same buildings as their peers but failing to form the social connections that drive satisfaction, retention, and wellbeing. The IIS data has consistently shown that making close friendships at accommodation is the single strongest predictor of resident satisfaction, with a 30-point NPS swing between students who made friends and those who didn’t. The students who don’t make friends are not just less happy. They’re in a fundamentally different category of risk.

When transgender and non-binary students are less likely to form friendships, they’re not just missing out on social connection. They’re missing out on the primary mechanism through which accommodation satisfaction is produced. The friendship deficit compounds into a belonging deficit, which compounds into a satisfaction deficit, which compounds into a retention and reputation problem that operators can measure but often can’t explain.

Community Scores Tell the Same Story

The IIS survey asks residents to rate their sense of community on a scale of 1 to 10. This is one of the two metrics used for GOLD certification, alongside Value for Money, and it is the metric we consider the leading indicator of property health. Properties that score below 7.0 on Community almost invariably sit in negative NPS territory.

Male students rated Community at 8.1. Female students at 8.0. Both comfortably above the GOLD threshold of 8.0.

Transgender students rated Community at 7.4. Non-binary students at 7.6.

Half a point may not sound like much on a 10-point scale, but in the context of Community scores, it’s the difference between a property that’s thriving and one that’s struggling. We know from the broader IIS data that when Community drops below 7.0, NPS tends to go negative regardless of operational quality. The transgender average of 7.4 sits in a zone where the margin for error is thin, and for individual properties with smaller transgender and non-binary populations, the scores will be lower still.

The community data tells us something that the friendship data reinforces: transgender and non-binary students are not integrating into the social fabric of their properties at the same rate as their cisgender peers. They’re living in the same buildings but experiencing a different community, one that is less connected, less belonging, and less supportive.

What the Research Says

The IIS data is not an outlier. It sits within a body of research that has been building for years and that the PBSA sector has been slow to absorb.

A 2023 Stonewall survey found that 34% of LGBTQ+ students in the UK had experienced discrimination in university accommodation. That’s one in three, and the figure includes both PBSA and university-managed halls. A study published in the Journal of LGBT Youth found that 66.6% of transgender and non-binary students with access to gender-inclusive housing reported dissatisfaction with the options available to them. The issue wasn’t the absence of gender-inclusive housing. It was that the housing was gender-blind rather than gender-affirming, a space that didn’t exclude them but didn’t actively include them either.

Research from AKT, in collaboration with the Universities of Kent, Bristol, and Southampton, found that LGBTQ+ young people are twice as likely to experience hidden homelessness, the kind where you’re technically housed but functionally displaced, sleeping on sofas, staying with friends, moving between temporary arrangements because the place you’re supposed to call home doesn’t feel safe. For transgender students, the accommodation experience is not just a matter of preference. It’s a matter of safety, both physical and psychological.

The TASO (Transforming Access and Student Outcomes) research programme has documented that while many LGBTQ+ students experience improved wellbeing during their degrees, inequalities persist, and the accommodation environment is a significant variable. Students who feel unsafe or unwelcome in their housing carry that experience into their academic work, their social lives, and their mental health.

And then there is the staff training gap. Multiple studies have identified that resident assistants and accommodation staff often lack the training to effectively support LGBTQ+ students, not through malice but through ignorance, and that the resulting indifference can be as damaging as active hostility. A student who encounters a microaggression from a flatmate and finds no support from staff has learned something about the property that no marketing campaign can undo.

The AI Dimension

There is a dimension to this data that is easy to miss and important not to.

In our AI usage analysis, also drawn from the IIS survey, we found that 52.7% of transgender students avoid AI entirely, compared to 17-20% of male and female students. The group that is most likely to need support navigating a new environment, finding resources, understanding their rights, and connecting with community is the group least likely to use the digital tools that could help them do those things.

This creates a paradox that the sector hasn’t articulated. The transgender students who are most at risk of not feeling at home are also the ones least likely to be reached by digital support channels, AI-powered information systems, or the automated service models that are increasingly replacing human interaction in PBSA. As operators invest in chatbots, AI assistants, and digital-first service delivery, they may be inadvertently building systems that serve cisgender residents well and transgender residents poorly.

If half of your transgender residents aren’t using AI, and your support strategy is moving toward AI-powered delivery, you’re optimising for the population that needs it least and abandoning the population that needs it most. That’s not a technology problem. That’s a design problem, and it’s one that can be solved if you see it.

What the Data Demands

The numbers are clear enough to demand action, and the action needs to be more specific than “be more inclusive.”

Measure belonging by identity, not just overall. The portfolio NPS of +49.5 looks healthy. The transgender NPS of +44.2 and non-binary NPS of +40.4 do not. If you’re only reporting on the aggregate, you’re hiding a problem inside a success story. Every operator should be tracking satisfaction, friendship formation, and belonging by gender identity, not as a diversity initiative but as a core operational metric.

Invest in staff training that goes beyond awareness. The research is consistent: the single most impactful intervention is staff who are trained not just to be “LGBTQ+ friendly” in a general sense but to understand the specific needs, risks, and experiences of transgender and non-binary residents. This means training on pronoun use, on how to respond to disclosure, on how to de-escalate conflict that has a gender identity dimension, and on how to create social spaces where transgender and non-binary students feel safe enough to participate.

Design social infrastructure that doesn’t assume cisgender norms. The friendship gap isn’t inevitable. It’s a product of social environments that are implicitly designed for cisgender residents. Events that centre on drinking culture, flat dynamics that assume binary gender norms, social spaces where transgender students feel visible in ways they don’t want to be, these are design choices, not natural laws. The properties that are actively designing inclusive social infrastructure will see it in their friendship data and, consequently, in their NPS.

Don’t let digital-first become digital-only. The AI avoidance rate among transgender students is a warning. As operators automate service delivery, they need to ensure that human support remains accessible, visible, and actively offered to the residents who are least likely to seek it through digital channels. The cost of maintaining human support infrastructure is real. The cost of losing the residents who need it most is higher.

Publish the data. The IIS survey asks about gender identity because we believe the data should be visible. If you’re collecting it, share it. If you’re not collecting it, start. The sector can’t address a gap it refuses to measure, and the students who are living the gap deserve to know that someone is paying attention.

The Question Behind the Data

The IIS Spring 2026 survey is, as far as I’m aware, the largest dataset on student accommodation satisfaction in the UK. Fourteen thousand responses. One hundred and ninety-one properties. Fifty-six cities. When the data shows a consistent pattern, across that many properties and that many residents, it’s not an anomaly. It’s a signal.

The signal here is that transgender and non-binary students are having a materially worse experience of student accommodation than their cisgender peers. They’re less satisfied. They’re making fewer friends. They feel less at home. They rate community lower. And the transgender students who might benefit most from digital support tools are the ones least likely to use them.

The sector’s response to this data has been, so far, underwhelming. A few operators have introduced gender-inclusive housing options. A few have updated their booking forms to include non-binary gender categories. These are necessary steps, but they are not sufficient. The belonging gap is not a booking form problem. It’s a lived experience problem, and it requires a lived experience solution: trained staff, inclusive social design, visible support, and a willingness to look at the data honestly rather than celebrating the aggregate while the margins suffer.

The students who don’t feel at home are not a niche. They’re a measurable, growing, and underserved population, and the operators that figure out how to serve them will not just improve their satisfaction scores. They’ll build the kind of reputation that no amount of marketing can manufacture, because it’s earned in the moments when a student who expected to be invisible discovers that someone was paying attention all along.