Things you learn from students - why small class size does matter.
Last week, during a 1:1 tutorial with an A level Biology student, I heard a way of describing something quite tricky from the syllabus that just cut straight to the heart of the idea. Pow! A perfect metaphor. I’ve probably taught that idea over 150 times, and I've taught it to undergraduates, A level students, GCSE students, and KS3 students. They always struggle to grasp the central core of the idea.
I heard ‘the perfect metaphor’ from a student who had all but given up on A level Biology because she felt like there was no time to understand anything and that the juggernaut just kept on rolling along regardless of how much she struggled. In that instant she showed such insight because we had time to draw out the idea in an unlimited and unscripted way, and because she’s smart and funny!
Checking for understanding
A central challenge of teaching, and I’d say even more so for tutoring, is understanding what the student’s don't understand. You can use formalised tests, but really nothing substitutes talking in an open-ended way and just proding around at the limits of the students' understanding until you find a loose edge. I say ‘understanding’ here, because finding failures to fully absorb Gradgind’s curriculum of facts is easy, verging on mechanical.
Effect of class size on student academic performance
Does class size make a difference? According to Meta-analyses - a statistical method that allows one to pool results from multiple academic studies - hardly at all.
Meta analyses are much in favour at the moment and are digested, I’d argue uncritically, by managers keen to find the most effective tools to increase student success. If you work in education you will hear all about them pretty much every day. They are the golden panacea of our problems. There is an organisation called the Education Endowment Foundation which has done an astonishing job of funding and publicising research into which interventions work and which don’t.
It sounds like an attractive idea, and like any evidence based approach it has to be understood in context. If you take the time to look behind the ‘toolkits’ and ‘dashboards’ and ‘value added scores’ there are some nasty problems with meta analysis. The most pressing are ‘lumping apples with oranges’’, publication bias, and reductive thinking.
The lumping problem
Let’s start with the lumping problem. Feedback is widely considered to be among the most useful techniques to improve student performance. The trouble is, there’s no clear definition of what ‘feedback’ is. If you look in detail at the research used in the meta-analyses you find feedback can encompass anything from signs in factories saying ‘don't forget to wear your safety gear’, peer review of essay questions, and everything in between.
Publication bias
If you dig a bit deeper and ask where the evidence in the meta-analyses has come from, you discover that there is a huge publication bias. Studies which find a positive effect of feedback, and end up being used in the meta-analyses, are 100’s of times more likely to be published than studies that don’t.
But who has time to investigate the bias? Who has time to question the evidence they are shown when in 8 minutes class: 11c needs to go over the mock, and you need the loo, and the flush doesn't fill the cistern quickly enough, and the vets rang to ask you to collect the tablets and…
Reductive thinking
The reductive thinking aspect is a much wider philosophical problem. We assume that technical thinking, by which I mean thinking where the outcome is clearly understood and defined in a quantifiable form ahead of activity, is the best way to understand a situation or problem. But, is it? Is there not room for a different way of thinking which explores meaning and defines its parameters as it progresses? An unfolding understanding reached by doing? There’s a vast literature on the difference and utility of technical thinking vs practical wisdom which goes back to the time of the Ancient Greeks. If you're interested in this sort of thing I’d heartily recommend the fabulous ‘Back to the rough ground’ by Joseph Dunne.
Technical solutions to improve educational outcomes are mighty attractive because they give the promise of being scalable to however many pupils we might have in the class. Faced with diminishing budgets, a recruitment and retention crisis, and the ever-increasing number of students with mental health difficulties, a technique which promises to fix all is obviously appealing.
Don’t get me wrong here, my academic training is in Ecology which is a highly abstracted, mathematical, and reductionist discipline. Obviously there’s a place for this sort of thinking, and a really important place, but we shouldn't let it fool us into believing it's the only way to think.
Pedagogy is suffering with ‘physics envy’
A pattern I've observed in my own career is that the greatest enthusiasts for teaching by ‘meta-analysis’ are those who have the least formal training in statistical inference. I’m going to stick my neck out further here: I’d argue pedagogy is suffering from ‘physics envy’. The jealous regard of the certainty of physical sciences, and the misguided adoption of methods from the physical sciences into all aspects of human endeavour. We want children to be electrons; immutable, wholly defined, equivalent units of analysis. It’s a mistake. To get back to the Greeks again, Aristotle once wrote that "it is the mark of an educated man to look for precision in each class of things just so as the nature of the subject admits".
I’ve come quite a way from ‘Does class size matter?’, but I think it’s important to think deeply about which answers we give credence to.
In the Education Endowment Foundation dashboard, the evidence for the effectiveness of reducing class size is described as ‘Low impact for very high cost based on very limited evidence’’, and I would argue it will remain that way because who would fund a rigorous analysis of the effects of class size? I’d argue no one would. Why? Well, imagine the cost implications if the result was, as pretty much every teacher I’ve ever talked to would agree, that class size is critical.
Why class size matters
Class size does matter and it matters because individual attention matters. Individual attention matters because feedback is, by definition, a two way exchange of information. You need to be able to understand the student’s understanding, and in the ‘one teacher to 25+ students’ model, which every year creeps towards 30, even in the A level sector, there is no way to do that. If we spent every lesson with 20% of the time in students talking 1:1 with the teacher then, in an average of five hours contact time per week, each student in a 25 student class would get 0.2 x 5 / 25 hours of talking = just shy of 2 ½ minutes per week. There is no way that in a subject as information heavy as A level Biology that 20% of the lesson time could be given to exploratory talk between student and teacher.
You see the problem? Some students can easily spend an entire term without really having an open-ended conversation with the teacher, indeed I’d suggest it's not unusual for some to never, in 2 years of A level, have the time to talk like this.
The things you learn from students
Which gets us back to where we came from - things you learn from students. Last week I heard the perfect metaphor for the difference between density dependent and density independent factors in population regulation. Over 25 years of teaching, where I have always made it a priority to hear the students’ ideas and explore their thinking with them, I’ve gained hundreds of little metaphors, jokes, silly rhymes and all sorts of insights from students on how to make things make sense. My recent move to working as a full time A level Biology tutor has brought me much more opportunity to talk and listen with students. Yesterday another student explained energy to me through the metaphor of money. Tutoring is so much more an exchange than teaching ever was and so much more effective a way of finding the limits of what they understand. And don’t even get me started on the use of tech to automate teaching…
Education faces several modern challenges, but at its heart teaching is a thoughtful exchange that nurtures the unique potential within each student. It relies on using human connection to reach mutual understanding, and I worry that if we lose touch with that, we’ll lose one of the most powerful tools in our possession. As always, I’d love to hear your thoughts on this topic, so please feel free to get in touch by emailing tim@timdickinsoneducation.uk.
