School belonging: are we measuring what matters?

Director Phil Stock considers what the latest DfE research tells us about understanding pupils' experiences.

Why belonging matters
Most school leaders would agree that feeling a sense of belonging matters. But how confident are we that we understand what belonging actually means to our pupils, how their experiences differ, or whether our efforts to strengthen belonging are making a difference?
Last month, the Department for Education published a timely report drawing on 16 qualitative school case studies exploring how schools understand, develop and assess belonging, while maintaining high standards of attendance and attainment.
The schools were selected to reflect a range of contexts, with evidence of strong or improving attendance and attainment alongside approaches intended to strengthen pupils' sense of belonging.
The findings reveal just how complex belonging can be and raise important questions about whether schools really understand the different ways pupils experience it. The different means by which schools attempt to measure belonging partly reflect this complexity, but they also raise questions about how confidently they can judge whether their efforts to strengthen it are making a difference.
The findings reveal just how complex belonging can be and raise important questions about whether schools really understand the different ways pupils experience it.
The findings will inform the DfE's forthcoming Pupil Engagement Framework, as part of the wider reforms around inclusion, further indicating that belonging is becoming an increasingly prominent consideration for leaders.
But as attention turns towards measuring belonging more systematically, we need to think carefully about what we are measuring, what we can reasonably infer from the findings and, perhaps most importantly, how we use them to inform our decisions.
As attention turns towards measuring belonging more systematically, we need to think carefully about what we are measuring, what we can infer from the findings and how we use them to inform our decisions
Belonging is more complicated than we think
Across the 16 primary and secondary schools involved in the study, belonging was understood through the lens of five interconnected dimensions:
Safety and security: both physical and psychological safety as foundational
Inclusion: feeling accepted and able to participate.
Collective identity: feeling part of the wider school community.
Individual recognition: feeling known and valued as an individual.
Positive relationships: experiencing supportive connections with others.
What I find particularly interesting isn't the identification of these five dimensions - I suspect most school leaders would recognise their importance, albeit express them in other terms - but the possibility that pupils experience them very differently, both from one another and over time.
A pupil's sense of belonging is not necessarily consistent across different aspects of school life, nor does it remain unchanged over time.
For example, a pupil might feel safe, valued by their teachers and proud to be part of their school community, but still struggle to feel accepted by their peers.
As we explored in the first blog in this series, feeling connected to particular individuals is not necessarily the same as experiencing a wider sense of belonging.
A single headline measure of belonging, particularly when averaged across a whole school, could easily conceal these differences.
A pupil's sense of belonging is not necessarily consistent across different aspects of school life, nor does it remain unchanged over time.
Are we measuring what matters?
One of the things that struck me most in the DfE report was the vastly different ways that schools measure belonging.
Many leaders appeared to struggle with how to measure belonging consistently and meaningfully, drawing instead on a mixture of indirect indicators, such as attendance and attainment, alongside pupil surveys, pupil voice and participation in school life.
The report itself openly acknowledges the measurement challenges that school face:
‘Senior leaders, staff and parent governors agree that belonging is not necessarily a single, measurable outcome; rather, it is a coincidental and organic result of many different supportive actions across the school, making it hard to isolate and quantify and therefore difficult to measure.’
This problem of measuring belonging in a consistent way is not just restricted to school leaders. As John Jerrim suggests in his analysis of PISA, TIMSS and PIRLS surveys to investigate how school belonging relates to school outcomes, these ‘studies do not capture this construct in the same way.’
Jerrim refers to these different interpretations of belonging as a ‘jingle-jangle’ problem – ‘where different things are given the same name (jingle) or different names are given to the same thing (jangle)’.
These concerns are well founded. As we have seen, belonging is multifaceted, and different measures may capture different aspects of pupils' experiences. However, I think we need to distinguish between the difficulties of defining belonging consistently and our ability to measure particular aspects of it reliably.
We need to distinguish between the difficulties of defining belonging consistently and our ability to measure particular aspects of it reliably.
In psychology, belonging has a substantial research history, with established measures such as Goodenow's Psychological Sense of School Membership scale, providing a systematic way of assessing pupils' perceptions of acceptance, inclusion and support within school and seeing how this changes over time and differs for different groups.
It can be tempting for leaders to try and write their own questions to measure belonging in their schools, or perhaps to cherry pick items from different sources to create their own belonging surveys.
The problem with this is that, without establishing the validity and reliability of the questions, we cannot be confident that we are measuring belonging consistently or accurately.
We may simply be collecting responses to questions about belonging without adequately capturing the underlying construct. It would be a bit like writing a handful of comprehension questions and then claiming that we had developed a reliable measure of pupils' reading ability.
Unlike these kinds of DIY surveys or individual questions about whether pupils feel they belong, established measures such as Goodenow's scale use multiple carefully selected items that have been tested with lots of different students for their validity and reliability. This provides a stronger basis for measuring pupils' sense of belonging consistently and making meaningful comparisons between groups and over time.
Established measures such as Goodenow's scale use multiple carefully selected items that have been tested with lots of different students for their validity and reliability.
In other words, just as a well-designed reading assessment can be said to assess reading ability reliably, so a pre-validated scale such as Goodenow’s can be said to capture belonging reliably, or at least his conceptualisation of it.
And this point is important: whilst Goodenow doesn't necessarily capture all five dimensions of belonging identified in the DfE's research, it does provide an established way of measuring particular aspects of belonging, including pupils' perceptions of acceptance, inclusion and support – many of the things that leaders are super interested to find out.
But measurement is only the starting point. Knowing that some pupils report a weaker sense of belonging than others is useful; understanding what might explain those differences, and what schools should do in response, is even more so.
To continue the reading analogy, most schools use standardised reading assessments to identify pupils who may need additional support. But the results don't necessarily tell us why a pupil is struggling, or what support they need. They provide a starting point for further investigation.
The same is true of established measures of belonging, such as Goodenow's scale. A reliable indicator can tell us which groups of pupils might be experiencing a weaker sense of belonging, but we need to explore why this is the case so we can do something about it.
Knowing that some pupils report a weaker sense of belonging than others is useful; understanding what might explain those differences, and what schools should do in response, is even more so.
From measurement to understanding
This is precisely why we developed Pupil Insights. We wanted to provide schools with a more systematic way of understanding pupils' experiences, drawing on established psychological measures of belonging, connectedness and other aspects of wellbeing.
But just as importantly, we wanted to help leaders move beyond the headline findings to identify patterns and differences that might otherwise remain hidden.
Our surveys provide leaders with a reliable overall picture around a range of indicators of inclusion and wellbeing, and they enable schools to examine how things change over time and whether the changes they make to the environment or culture are making a difference.
In the reports that we produce there are recommendations for leaders about potential next steps that will help them to understand the different experiences of pupils in a bit more detail.
We wanted to provide schools with a more systematic way of understanding pupils' experiences, drawing on established psychological measures of belonging, connectedness and other aspects of wellbeing.
Sometimes, the initial findings raise more questions than they answer. This is where our advanced analysis service can help. Schools can identify a particular research question for more detailed statistical analysis, allowing them to explore patterns and relationships that may not be apparent in the headline findings. For example:
Do pupils in some year groups report a weaker sense of belonging than others?
Which aspects of pupils' experiences are most closely associated with a stronger sense of belonging?
Which aspects of pupils' experiences are most closely associated with anxiety?
Of course, identifying these relationships doesn't explain why they exist, but it can help leaders ask better questions and decide where to focus their attention.
Across this blog series, I've tried to make the case that understanding pupils' experiences requires more than simply asking them questions.
It requires us to think carefully about what we're measuring, use the findings to identify patterns that might otherwise remain hidden, and then draw on other sources of evidence to investigate what lies behind them.
Ultimately, the value of measuring belonging lies not in the data itself, but in what schools learn from it and what they choose to do differently.
The value of measuring belonging lies not in the data itself, but in what schools learn from it and what they choose to do differently.
Perhaps the most important message comes in the concluding words of the DfE report itself:
‘This work reminds us that education is fundamentally about human connection and community. In an era of considerable focus on academic data and accountability measures, these schools demonstrate that investing in belonging is not a distraction from the core business of education but rather part of its essential foundation.’
'Investing in belonging is not a distraction from the core business of education but rather part of its essential foundation.'
References
Department for Education (2026). School belonging case studies report.
Jerrim, J. (2026). The jingle-jangle problem of measuring ‘school belonging’. FFT Education Datalab
Goodenow, C. (1993). ‘The psychological sense of school membership among adolescents: Scale development and educational correlates’. Psychology in the Schools, pp. 79–90.




Comments