AI12 min read11 May 2026

What 14,000 Students Reveal About the Psychology of Where They Live

The largest student satisfaction dataset in the UK has a psychological dimension that nobody is talking about.

David Chadderton
David Chadderton
Behavioural scientist · Creator of the STAR Framework

The largest student satisfaction dataset in the UK has a psychological dimension that nobody is talking about.

I’ve just analysed 14,131 responses from the Global Student Living Index 2026, cross-referenced against the STAR Framework’s four psychological types. The dataset covers satisfaction, community, wellbeing, AI usage, booking behaviour, premium features, extracurricular engagement, and more. And the patterns that emerge are not just statistically interesting. They are operationally significant for anyone running, designing, or marketing student accommodation.

All satisfaction scores are on the GSLi scale (0-100). All recommend scores are NPS (-100 to +100).

Here’s what the data says.

The Type Distribution

The STAR Framework identifies four primary motivational orientations: Adventurers (driven by autonomy), Thinkers (driven by competence), Socialisers (driven by relatedness), and Realists (driven by security). Among the 13,414 respondents with a valid STAR profile:

Thinkers are the largest group, followed closely by Socialisers. This matters because the two types have fundamentally different expectations of their environment, and they experience the same accommodation in fundamentally different ways.

The Satisfaction Gap

Here’s the finding that should reshape how the sector thinks about satisfaction measurement.

Overall satisfaction by STAR type (GSLi 0-100):

The gap between Socialiser and Thinker is 7.7 GSLi points. That might sound small. It isn’t. On a 100-point scale, 7.7 points is the difference between a product that delights and a product that merely satisfies. And the gap is consistent across every single dimension measured.

Socialisers rate every aspect higher than all other types. Every one. Internet, communication, student care, condition, location, design, value for money, bedroom space, study environment. Not by a little. By 5 to 10 points on every metric.

This is not because Socialisers are easier to please. It’s because the accommodation is delivering on their primary psychological need: relatedness. The social infrastructure, the communal spaces, the sense of belonging, the staff interactions, these are the things that make a Socialiser feel at home. And when that need is met, everything else looks better through that lens.

Appraisal Theory of Emotion, one of the seven pillars of the STAR Framework, explains the mechanism. We don’t evaluate our environment objectively. We evaluate it through the lens of our goals and needs. When the environment meets our needs, our appraisal of every dimension shifts upward. The Socialiser who feels connected rates the internet higher, not because the internet is better for them, but because their overall emotional appraisal of the environment is more positive.

The Thinker, by contrast, is driven by competence. They want an environment that works efficiently, that is logically structured, and that doesn’t waste their time. When the accommodation delivers on those terms, the Thinker is satisfied. But the bar is higher, because the Thinker’s evaluation is more analytical, more granular, and more resistant to the halo effect that lifts the Socialiser’s scores.

Self-Determination Theory, another STAR pillar, explains why. The Thinker’s core need is competence, the sense that things work as they should, that standards are maintained, and that their time is respected. When that need is unmet, the Thinker doesn’t just feel disappointed. They feel disrespected. And that feeling colours their entire experience.

The Recommend Gap

The NPS scores tell the same story.

Recommend likelihood by STAR type (NPS):

Socialisers are 20.5 NPS points ahead of Thinkers. In commercial terms, that’s the difference between a brand with momentum and one running to stand still.

Social Identity Theory, another STAR pillar, explains why. We derive part of our self-concept from the groups we belong to. When the accommodation feels like a community, the Socialiser doesn’t just enjoy living there. They identify with it. It becomes part of who they are. And when something is part of who you are, you recommend it to others, not as a product, but as an extension of your own identity.

The Thinker doesn’t have that relationship with their accommodation. The accommodation is a service. It either meets the specification or it doesn’t. The Thinker will recommend if the evidence supports it, but they won’t evangelise. Evangelism is a social behaviour, not an analytical one.

The Community Effect

Sense of community by STAR type (GSLi 0-100):

The gap between Socialiser and Thinker on community is 12.4 points. This is the largest gap in the dataset, and it has direct implications for retention.

Socialisers feel the strongest sense of belonging. This is not surprising. What is surprising is the magnitude of the gap. The Socialiser’s experience of community is not just slightly better. It is qualitatively different. They experience the accommodation as a social environment. The Thinker experiences it as a functional one.

The Realist’s Bedroom

Here’s a finding that most operators won’t expect. Realists rate nearly every bedroom dimension highest.

Bedroom ratings (GSLi 0-100):

Aspect Realist Socialiser Gap
Bed comfort 77.7 71.7 +6.0
Bedroom size 81.0 75.0 +6.0
Shelving 80.4 68.4 +12.0
Lighting 75.5 63.2 +12.3
Design 77.2 68.4 +8.8

The Realist’s need for a secure, well-appointed personal space is being met. Socialisers, by contrast, rate bedrooms lowest across most dimensions. They don’t care about the room. They care about the people.

The one exception is desk chairs. Socialisers rate desk chairs highest (65.6), while Realists rate them lowest (56.5). And desk chairs are the lowest-rated bedroom item across the entire dataset (60.0). Every type agrees: the desk chair needs work.

The AI Adoption Hierarchy

Now to the finding that surprised me most.

ChatGPT usage by STAR type:

“AI usage increased a lot” in the past year:

Socialisers are the heaviest AI users and the fastest adopters. Thinkers, the type you’d expect to lead a technology revolution, are the slowest.

This is counter-intuitive until you remember what AI actually is. Conversational AI is a social technology. It talks to you. It responds to your questions. It adapts to your style. It feels like a conversation, even though it isn’t one. For the Socialiser, who is driven by relatedness, AI is a natural fit.

The Thinker, by contrast, evaluates AI as a tool. Does it work? Is it accurate? Can I trust its outputs? Dual Process Theory, another STAR pillar, explains the difference. The Thinker defaults to System 2, the slow, analytical mode. They evaluate before they adopt. The Socialiser defaults to System 1. They adopt because it feels right, because their peers are using it, because the social signal outweighs the analytical one.

AI Tracks with Satisfaction

The most important AI finding is not about usage. It’s about what usage predicts.

AI Increase Satisfaction (GSLi) NPS Community
Increased a lot 84.1 +37.8 76.3
Increased some 81.4 +29.7 72.0
Not changed 78.5 +22.2 67.2
Decreased 78.0 +20.1 66.6

Students whose AI usage increased a lot score 6.1 GSLi points higher on satisfaction and 17.7 NPS points higher on recommend likelihood than those whose usage decreased. AI engagement is not just a technology metric. It’s a proxy for overall engagement. The students who are most engaged with new tools are the ones who are most engaged with their environment.

The International AI Divide

International students are dramatically ahead of domestic UK students on AI adoption.

ChatGPT usage among Socialisers:

“Usage increased a lot” among Socialisers:

Chinese students are adopting AI at twice the rate of UK students. International students arrive with higher AI literacy, higher AI expectations, and higher AI usage. If you’re designing for the UK domestic student, you’re designing for the segment that is least prepared for an AI-mediated world.

The City Divide

Not all cities deliver the same experience.

City Satisfaction (GSLi) NPS
Birmingham 84.5 +43.4
Glasgow 83.5 +33.8
Sheffield 83.3 +36.9
Nottingham 83.2 +32.5
Coventry 82.7 +40.8
Edinburgh 82.1 +28.7
Exeter 81.6 +30.5
Leeds 79.6 +21.4
Leicester 76.7 +17.7
London 74.0 +7.2

London is the clear underperformer: 10.5 GSLi points below Birmingham and 36 NPS points behind. Birmingham and Coventry are the standout performers, suggesting the Midlands market is delivering strongly. London’s low scores warrant investigation: is it pricing, expectation mismatch, or product quality?

The International Satisfaction Gap

International students are consistently more satisfied than domestic ones.

Nationality Satisfaction (GSLi) NPS Community
India 84.8 +45.3 77.2
China 84.4 +34.6 79.3
Nigeria 82.8 +36.7 75.1
Pakistan 81.7 +32.8 74.1
United Kingdom 80.3 +26.5 69.3
United States 76.7 +16.5 60.0
Hong Kong 71.3 -2.1 63.8

India leads on satisfaction (84.8) and NPS (+45.3). Hong Kong is the only nationality with a negative NPS (-2.1). That’s a significant outlier, and it suggests something is systematically wrong with the experience for Hong Kong students. It warrants urgent investigation.

The Disability Gap

Metric With Disability Without Disability Gap
Satisfaction 77.7 81.2 -3.5
NPS +18.3 +29.3 -11.0

Students with disabilities score 3.5 GSLi points lower on satisfaction and 11 NPS points lower on recommend. Among Thinkers with disabilities, NPS drops to just +8.6, the lowest in the entire dataset. The analytical, detail-oriented types notice what’s missing most acutely.

AI has enormous potential as an accessibility tool. It can read text, summarise documents, generate notes, translate languages, and adapt content to different learning styles. But the students who would benefit most from these capabilities are the ones who are least likely to be using them. This is both a social responsibility argument and a market opportunity.

The Gender Question

The data shows a gender gap in AI adoption, but it’s not where you’d expect.

Segment Satisfaction NPS Community
Male Socialiser 87.9 +45.1 81.9
Female Socialiser 85.5 +39.6 78.3
Male Adventurer 80.8 +29.0 70.9
Female Adventurer 80.2 +28.7 70.5
Male Realist 79.2 +23.9 67.8
Female Realist 79.7 +32.0 69.0
Male Thinker 79.7 +21.4 67.7
Female Thinker 78.3 +22.2 67.4

Male Socialisers are the highest-scoring segment in the entire dataset: 87.9 satisfaction, +45.1 NPS, 81.9 community. They are your super-promoters. The personality type matters more than the gender: Male Socialisers outscore Male Thinkers by 23.7 NPS points.

The Socialiser Paradox

Socialisers are most likely to say “I’m just here to study” (51.5%). This sounds wrong. Socialisers are the most social type. Why would they say they’re just here to study?

The answer is that their social needs are already being met by the accommodation itself. The community they feel (79.7 on the GSLi scale) is the accommodation. They don’t need extracurricular activities to feel connected, because the connection is already there. The accommodation is the social life.

This is the highest compliment a PBSA operator can receive. When the Socialiser says “I’m just here to study,” they’re not rejecting social engagement. They’re saying the social engagement is so embedded in their living experience that they don’t need to seek it elsewhere.

What Each Type Would Sacrifice for Lower Rent

Nobody wants to lose bathrooms, bedrooms, or safety. But Thinkers and Realists would happily sacrifice social events for cheaper rent. Socialisers wouldn’t. The social programming has value, but only for one type.

The Hero Nominations

When students nominate a staff member as their “hero,” the reasons diverge by type:

Same staff. Four different reasons for admiring them. The Adventurer values warmth. The Thinker values competence. The Socialiser values community-building. The Realist values reliability.

What This Means for the Sector

The data tells a clear story. Student satisfaction is not uniform. It is filtered through psychological architecture, and that architecture predicts everything: how students rate their experience, whether they recommend, whether they stay, how they use technology, what they value, and what they ignore.

The Socialiser is your promoter. They rate highest, recommend most, and feel the strongest community. Keep them connected and they will sell your accommodation for you.

The Thinker is your critic. They rate lowest, recommend least, and evaluate most harshly. But they are not unsatisfiable. They just need evidence. Show them the data. Prove the system works. They’ll respect you for it.

The Adventurer is your premium customer. They want the lifestyle. Give them something distinctive, something novel, something that feels like an upgrade. They’ll pay for it and they’ll tell their friends.

The Realist is your baseline. They want the basics done well. Clean, responsive, safe, reliable. Get those right and the Realist is satisfied. Get them wrong and the Realist leaves quietly, without telling you why.

The organisations that understand these patterns, that design for four types of human rather than one generic “student,” will own the next decade of student experience. The rest will keep measuring satisfaction without understanding why it varies.

Analysis based on 14,131 respondents from the Global Student Living Index 2026. STAR types assigned via the STAR Cognitive Profile Assessment (CPA). The STAR Framework synthesises seven psychological theories: Self-Determination Theory, the Big Five (OCEAN), Dual Process Theory (System 1 and System 2), Regulatory Focus Theory, Social Identity Theory, Cognitive Bias Theory, and Appraisal Theory of Emotion. David Chadderton is the creator of the STAR Framework and author of The STAR Framework: Rewriting the Rules of Consumer Engagement (NYC Big Book Award, 2025), The STAR Operating System, and Dear Algorithm, It’s Not Me, It’s You. He is CMO at Homes for Students, VervLife, and Orla.

The STAR Framework

If you enjoyed this essay, you'll find the full argument — and the framework behind it — in the book.