Research Ed 2015 The Best Bit

Research Ed 2015 was the usual head-mangling mix of some of the more dynamic and incisive voices in UK education. But one moment in this years conference stood out for me because it reminded me of the very best part of engaging with research. This moment was coming across a surprising finding I could trust.

The moment came quietly in an understated talk by Jonathan Sharples from EEF and Alex Quigley about peer tutoring and philosophy for children (P4C). They’d found no effect for peer tutoring but they had found philosophy for primary children improved reading!! who’d have thought it, not me!! WTF! (I know this result has been out for some time but I had not given it much thought until Jonathan explained it).

One of the biggest pratfalls in research is succumbing to the temptation of accepting findings without being sure there isn’t some other explanation for results that’s more likely than the intervention causing the outcome. But listening I couldn’t think of how the results might have come about other than the influence of the P4C.
The study was large and included a mix of kids found in most schools, it compared outcomes to a control of similar kids who probably only differed in that they didn’t receive the philosophy and the data had been handled with due scepticism.

Looking at the study there did not seem any other likely explanation for the results other than Philosophy at primary aids reading.

Dealing with research is almost always a turgid business; wading through difficult concepts and fiddly details. It also usually ends in disappointment because there is almost always a mundane explanation for the results that’s nothing to do with the intervention or the programs don’t live up to their promise. But every-so-often a reliable piece of research throws up finding that you did not expect.
This for me is the very most enjoyable part of research because a surprising new finding means that our view of the truth has widened, slightly more detail has been revealed and we must find a way of assimilating this new knowledge into our world view. We can put away the tables of figures and boringly written papers and just think about what this might mean. It may have implications for our assumptions. It may also give us a better ability to influence outcomes. In this case, how can this finding help improve the teaching of reading?

This study got my mind-a-racing, what on earth is it about philosophy that links to reading?

Philosophy is designed to make you think deeply and we know from Cognitive Psychology that words are remembered better when more thought goes into reading them. Words seem to be stored and retrieved according to their meanings.

But P4C does not involve a huge amount of reading so it seems unlikely deep thinking that gets done on the specific written items in the P4C materials is responsible for an improvement in general reading. This suggests that the deep thought on the P4C materials is generalising to reading outside the items presented in P4C. But how could so little reading lead to better reading?

I wonder if philosophies effect on reading has something to do with philosophies highly provocative subject matter.

Here’s one example of P4C written materials

What would you rather find?

A magic sweety bag that is always full
A magic book that talks to you
A magic carpet that can take you anywhere

Each of these example seem well picked to provoke some pretty wild imaginings.

Perhaps this is how fairy tales work so well. Fairy tales always involve some very wacky business going on. In three little pigs wolves blow down houses and in Hansel and Gretel the witch is cooked in an oven! P4C makes a great deal of use of parables and fables, simple brief stories where people make questionable choices.

But perhaps philosophy does more good than by firing the imagination. To do the task you need to do more than imagine some wild stuff, in order to pick between them you need think of the implications of having each of these magic items. Never ending sweets would be great but imagine what a book that talks would say, what conversations could you have but then again with a flying carpet you could go anywhere, imagine the adventures you could have!. Also the task asks for some mental gymnastics because, not only are you expected to imagine these amazing things but you are asked to hold all of them in mind and compare between them to pick the one you want most.

But why would deep thought and mental gymnastics lead to better reading, there must be some link between these imaginative workouts and deciphering text. One possibility is that philosophy is heavily linked to language structure, it might help by demonstrating common language arrangements in a way that help children interpret sentences more efficiently?

The example above delivers some highly evocative ideas in a few very simple and familiar words. The language structure is also very simple, a [common thing] that does [something bloody amazing]. Maybe these extreme examples in plain language demonstrate to a child that language can go to some pretty interesting places. In this example it seems to say, ‘you’d better pay close attention to the second part of sentences [verb phrases] because it might go to somewhere crazy’.

A final thought is that philosophy might help reading because it make the world of words seem a much more interesting place.

But most importantly, whatever explanation I might come up with we do know one thing, P4C as carried out in those schools studied by EFF seems to improve reading. Consequently, we know that an explanation is out there and there is some causal connection between whatever P4C does and ability to read. Greater thought about what P4C can enable us to be better at teaching children to read. I find that a very exciting and interesting prospect that can only come about through engagement with good research.

4 thoughts on “Research Ed 2015 The Best Bit

  1. “Looking at the study there did not seem any other likely explanation for the results other than Philosophy at primary aids reading.”

    Except for regression to the mean and (given the lack of statistical significance of the result) chance.

    https://teachingbattleground.wordpress.com/2015/07/14/does-teaching-philosophy-to-children-improve-their-reading-writing-and-mathematics-achievement-guest-post-by-mjinglis/

    http://ripe-tomato.org/2015/07/14/teaching-philosophy-in-primary-schools/

    This is not reliable research in anyone’s book, surely?

    Like

    • Regression to the mean is part of the statistical dark arts and is a dirty trick as it can be wheeled out to argue against any change over time without fear of contradiction because you can never know how much it is.

      Reading for FSM improves by a substantial one third a standard deviation, not likely to be by chance. That, I think is a result worth further thought.

      The other results seem too small to be convincing.

      I have done my fair share of arguing with Stephen Gorard that he and EEF should use significance testing so we can contrast the size of results against the background noise. But he takes a very strict
      principled stance which I’ve so far failed to wear down.

      Like

  2. Hmmm…Can’t we have some sympathy for the decisions here, to make it an RCT schools were selected at random, given the small number a sampling bias unluckily crept in. Matched samples design is very different and once P4C had been allocated to schools there was no way back. It was never the plan to match samples because sampling bias wasn’t anticipated understandably because randomisations should have prevented this.

    Like

Leave a comment