This Wednesday (20/08/14) Episode 2 of the new BBC Radio 4 series The Educators aired and is now available on iPlayer. In this episode Sarah Montague interviews John Hattie, Professor of Education at the University of Melbourne about his work analysing a wide range of educational research.
John talks about his work analysing over 60000 studies (encompassing a quarter of a billion students) to find out what has the largest impact in improving the educational outcomes of students. This work took him over 15 years and is probably familiar to many through his books on Visible Learning.
In the programme, which is well worth a listen, following interviews of parents discussing what they look for in a school, Hattie makes the comment that they focus on what they can observe; things like class sizes, leadership of the school, Ofsted reports and whether children appear happy in the school. Hattie points out that these are poor proxies for what is really important – the quality of teaching. I, as a teacher (and I’m sure mos teachers too) agree that it is the the teaching that makes the difference between success and a pupil not achieving their potential.
Hattie lists many things that actually have zero or little effect such as class size, homework (greater effect in secondary schools than in primary) and the type of school. All things commonly held to be important. This is all discussed in more detail in his books, with proper references to the original studies and comments on the validity and quality of the studies looked at.
On the subject of the quality of the study he made one comment that I found fairly strange:
If the effect is quite large, then the quality of the study is not that important
I really don’t agree with this, I could make up any old junk and get a large effect size, but surely the validity of the method used in a study is important to whether the study and it’s conclusions have any value! Not being a statistician, I don’t feel qualified to critique Hattie’s methodology, though I would be amiss to not point out that such criticisms do exist. Some of these, including the use of the concept of effect size are discussed in the blog of @OllieOrange2. As I say, I am not a statistician and am also a big fan of Hattie’s books and research, but some points raised in this blog do raise concerns that at some point I should look into in more detail.
Hattie makes two points that struck me in this programme: 1) The biggest predictor of future health, wealth and happiness in the future is the number of years in schooling, not achievement at school. 2) In excellent schools there is a dialogue about the teaching and the impact that teaching is having.
I would recommend this radio programme and John Hattie’s books to all teachers, wishing to investigate what makes a difference in schools. However I would echo Hattie’s remark that he is not giving school leaders a recipe for what they should do, rather he is giving them a way to think – to investigate the impact teaching and decisions have in their school.