Last year I was sent an email by my line manager about a little bit of research called ‘The shuffling of mathematics problems improves learning,’ by Rohrer and Taylor, that took place in America. The two different theories in this research really hit home with me as it challenged the idea of teaching maths in neat, topic based sequences. I would strongly recommend this to all teachers across all curriculum areas, not just maths. Having spent today on an SSAT course I have had the chance to reflect upon how important the context of a school is and I am fortunate enough to work in a supportive school that is driven by research, making the implementation of and enquiry driven by my Rohrer & Taylor reading all the more feasible.
Sharing my vision
To undertake the research I wanted to take two classes and make completely parallel sets and was particularly keen to focus on middle attainers, this being a focus of our departmental improvement plan.
How would it all work?
One class would learn in the traditional blocks of learning that takes place in our education system, e.g. A block of learning on Pythagoras’ theorem, then straight line graphs etc.
The other class would take the first 3 topics of learning in a scheme of work and split these up lesson by lesson in a week. In this way lesson 1 of the week would be topic 1 – Number, lesson 2 in the week would be topic 2 – Fractions and decimals and so on. We’re fortunate to have 4 hours of maths with our year 9 students each week and so the final lesson of the week was dedicated to DIRT (dedicated improvement and reflection time) and homework. The department responded brilliantly; eager and keen to see what impact this might have. One concern was raised and this was how some of the ‘weaker’ students in the group would cope, something I am now reflecting upon after the first term. Luckily SLT were fully on board with my ideas and helped to facilitate the enquiry.
Ready, set …Go!
So at the start of the year two class sets had been created to run parallel. We chose year 9 as the second cohort to run through the new 1-9 curriculum. We assessed students at the end of their previous year and from one assessment ranked the students and paired them up.
The first issue we had was convincing students they were all now 9 set 3 (how we ‘name’ sets is another blog altogether!) and we changed their class code. For this blog we’ll call the shuffled class ‘Alpha’ and the other ‘Beta’ and the only way I felt I would be able to evaluate/share/promote this idea was if I took on the shuffled class.
And so it began… in their first week I was amazed how students just accepted a different topic each lesson! I briefly described what we were trying but I didn’t want them to panic or have predetermined ideas of how it would work. The biggest challenge was for me, I had to stay on top of how the class were progressing and write a weekly plan for the following week of where the students had got up to, ensuring I looked at feedback and AFL. I did also hit some speed bumps when students missed a lesson for absence, this would sometimes mean it would be 2 weeks before they looked back on a topic.
Beta lesson structure
Wednesday: Fractions and Decimals
Pausing to reflect
As I’m writing this we have now completed our first shuffled term. The mathematician in me has bullet pointed keys aspects that we have found in carrying out the trial:
• Student voice – In the early stages I asked students for verbal feedback of how they felt the shuffled teaching was going and two points stood out for me ‘it’s not boring, if you don’t like a topic you can change next lesson’ and ‘just when you start to get the hang of it the lesson finishes and you don’t get chance until the next week’. This is not a fair questionnaire as mathematicians will know. I was the one asking the questions but I’m confident students felt able to be honest. We’ll definitely conduct a student’s survey with anonymity.
• Time to align – to ensure we are conducting as fair a research as possible time to align between two staff has been vital and has sometimes taken up more time in an already time conscious job. We use the same starters, lessons and homework so as to ensure equipoise as far as possible.
• Speed and time to teach – strangely and something I didn’t anticipate was how quickly I moved through the curriculum compared to the other class. The other teacher said ‘I know I’ve allowed time for the students to ‘get it” and I felt that possibly I’d rushed through a few things in order to get things done in a lesson. Of course with the traditional teaching you’re able to adapt your lesson for the following day depending on how the students responded but one thing that concerns me is whether the shuffled class have the same amount of time.
• One thought we did have was ‘we’re bringing up the middle students, but are we pushing the high flyers in the groups?’
Where’s the evidence?
At this point I know so many of you will be eager to hear the data!!! We have created our own assessments from an exam boards GCSE bank of questions and used common GCSE questions from the first 3 topics we’ve taught. This is new in itself as we move away from ‘topic test’ to interleaved assessments.
With regards to hard evidence to say which way is best, I don’t actually think we’ll get this (I hope that we do) until the end of the year. It might be that within our experiment we have no impact! And that the parallel sets stay parallel as it were. This is just a trial! I’ve bullet pointed some findings from our first assessment. We ranked out students in order of score and this is what we found:
• Top 25% – 8 students Alpha, 4 Students Beta
• Middle 50% – 11 students Alpha, 11 students Beta.
• Bottom 25% – 5 students Alpha, 6 students Beta
(Some students were absent and unable to take the assessment)
Where to now?
Well, we’re going to carry on in the same way with the same structure and hope that the next round of results reflect any further changes. I also think student voice here is key, how do they feel about this? Can they see an improvement? We’ll wait and see.