Here at Evidence Based Education, we make sure that evidence is at the heart of everything we do. As well as providing teachers with evidence-backed professional learning resources, we allow teachers to gather their own evidence via student surveys, peer surveys, and self-reflections, to make informed decisions about their development and understand their impact.  

What many may not know is that these survey responses are just as useful for us as they are for our users. As EBE's Research Statistician, one of the most interesting bits of my job is data analysis — getting to dive into response data from student surveys, self-reflections and peer surveys completed using the Great Teaching Toolkit. By pooling together anonymised responses from such a large and diverse sample, we can better understand areas where teachers are doing well, where they might need more support, and how we can improve the GTT to provide them with the tools they need to develop and succeed in their professional learning goals.  

Analysing the survey data

We know schools have been particularly interested recently in exploring inclusion and belonging, so this summer I took a look at response data from inclusion-related GTT survey items from over 127,000 student surveys, nearly 600 peer surveys, and some 2,000 teacher self-reflections. When I approach a dataset like this, I often find that the simplest, first steps of data analysis — descriptive statistics — can tell us some really interesting things. I always start by looking at the means for each statement — on average, do respondents agree or disagree, and how much? This gives me a quick pulse on what people are thinking, which is important in terms of knowing broadly where our sample of teachers generally are doing well or need more development. 

However, important insights can be lost in averages. A mean that falls in the middle could reflect an item where respondents just don't have much of an opinion one way or another… But could also reflect a really polarising item with large groups of respondents holding opposing views. That's why for me, the real story lies in the distribution, or what proportion of respondents chose each response choice. I examine these carefully for each item to understand the spread of the data, knowing that very similar means can obscure very different response patterns. 

With our survey data, it's important to understand not only the patterns among the individual groups — students, teachers, and peers — but also how (or even if) they align. How a teacher views their own practice is undoubtedly important, but it's when we combine that with students' own experience of the teachers' efforts, and what their peers may see as they observe, that we come to a true understanding of where a teachers might want to direct their professional learning.  

What we've learned about belonging and inclusion

Our descriptive data analysis of Great Teaching Toolkit survey belonging and inclusion items shows that there are lots of things where all three groups are aligned in their views. For example, when we looked at the distribution of student responses, there was high agreement that teachers respect their students' views and who they are as people, and that they care about all students in their class — not just the clever or well-behaved ones. We see a slightly different story around the topic of bullying in the classroom. Teachers and students agree at a similar rate that aggressive, bullying or disrespectful behaviours occur between students in their class, but peers noticed this less, perhaps because these more relational aspects of the classroom environment require more immersion to notice.  

Unsurprisingly, there are also areas where ratings are quite different between the groups. For example, most teachers believe that they know when something is annoying or upsetting their students, but most students don't agree that this is the case. It's not all doom and gloom, however. I was especially interested to see there are some areas, such as teachers being ambitious about what their students can achieve, where teachers simply aren't giving themselves credit for how positively students view them in this area!  

These are just a few of the countless insights we've been able to draw from our GTT survey dataset. In the next blog, we'll take a look at the specific findings around the classroom environment in greater depth. What do we know? What does it mean? And most importantly, what does it mean for teacher professional learning? 

If you're interested to find out more about belonging and developing an evidence-informed inclusion strategy, you can read more from EBE's Stuart Kime.