Are all classrooms unique?

The following two statements have become something of an orthodoxy in education in recent years.

(View 1) Pupils are unique and classroom contexts are unique – this means that large-scale research projects (such as randomised controlled trials) are of limited use in education.

(View 2) Teachers require educating – preferably with university input – because there is more to teaching than just ‘craft’.

It is not at all uncommon for some people to hold both of these views. The problem with this is that, when stripped away of the fluff, these two statements are ultimately mutually contradictory.

It certainly is the case that all pupils are unique and all classroom contexts are unique. It should also be noted, however, that the very notion of teacher education is based on the idea that what is learnt in one classroom is transferable to another. The similarities that exist between children and classes are pretty self-evident to anyone who has taught for more than a couple of years. Pupils frequently come to lessons with common misconceptions (e.g. regarding negative numbers, or the association between class and wealth). Pupils frequently exhibit the same kinds of behaviour and respond to behaviour management in a similar number of ways. If I had to re-learn everything every time I changed school (or indeed every time I got a new class) then there would be absolutely no point in building up any kind of experience (through professional practice, small-scale research, or anything else for that matter) as all of this knowledge and experience would become redundant when I encountered my next bunch of unique children.

This principle underpins a number of approaches to research in education, particularly any form of research that aims to produce a generalisation as its outcome. The whole basis of research is that there is enough commonality between people in order to reach generalisations. If you think teachers have something to learn from Dweck’s research into mind-sets, then you are accepting that pupils are ultimately quite similar in how they think about things. If you think that women are often discouraged from doing physics or maths (for a range of factors) then you are again accepting that there is commonality between what women experience in different contexts.

And yet those who would be happy to argue that teachers need to spend time learning about teaching, and who would enthusiastically accept research into pupil mindsets, can at times be heard criticising research into cognitive psychology, particularly any research that involves a randomised controlled trial (RCT). Now I recognise that there is a debate to be had about whether RCTs in education can ever work well in practice, but the principle underpinning them is the same: pupils are more alike than they are different, and this means that it is useful for teachers to be taught theories (such as those from cognitive psychology) because these can be meaningfully applied in the classroom.

So here is my plea: I do not mind if you hold View 1 or View 2 above (I think that View 1 is false and View 2 is true), but to hold both is a logical contradiction. This contradiction is conveniently overlooked by those who want to reject some things (such as Willingham’s work on cognitive psychology or Lemov’s collation of effective practice) because it does not fit with their own ideological position. We need some common starting points if we are going to have meaningful discussions about education, but one of these has to be that there are considerable similarities in terms of how pupils learn and respond to their education. If this is not true, then there is no point in us having any conversation about teaching at all.

Advertisements

7 Comments on Are all classrooms unique?

  1. An interesting read, thank you.
    It’s true that we are encouraged to view every child as a unique individual, for example through Scotland’s GIRFEC approach, however I agree with what you say about common similarities that cannot be ignored. I think that when planning and implementing learning, educators need to be aware of the similarities between pupils, but also must consider individuality when deciding why something has/hasn’t worked as predicted.

  2. I am not sure I agree that these two views contradict each other. I believe that it is true that each student is necessarily unique because of the unique nature of their personal experience and the way this shapes how they perceive the world and therefore each classroom or combination of students is also unique. As a result, there are some limits to what large scale research projects can tell us about learning. This is reflected in my own experiences as an experienced teacher when I find that particular activities or explanations are more or less effective in different contexts.

    However, I also believe it is true there is a great deal of commonality in how students learn and therefore a lot can learned from studies into things like the testing effect. I read some educational research and my teaching has been informed by people like Willingham and Bjork. It has not, however, been determined by it. As an example of what this means practically, in almost every lesson that I teach there will be some form of retrieval practice (ideally using the idea of interleaving). However, the form that the retrieval practice takes will vary from class to class and sometimes from person to person.

    This position seems to me to be coherent.

  3. I would echo the above. I think it is possible to say that a narrow trial in a couple of contexts is unlikely to be transferable as so many of the variables change. This can be seen in the application of all sorts of “initiatives” in schools which are copied into other contexts. That said, I do agree that there are larger and more general aspects of teaching which may well cross contexts. The key is not rejecting all research outright but equally not accepting all at face value. In essence we come back to the critical approach – something (hopefully) instilled in a good PGCE.

  4. Hamish Chalmers // 19 October 2015 at 12:14 // Reply

    Great post. If teachers truly did believe that all children are unique (or perhaps, more accurately, sufficiently different that no generalisations about how to teach them can be made) then every time they opened the classroom door to start a lesson they would have to wipe completely from their minds any assumptions about what is likely to work for that lesson and completely re-invent teaching from scratch. Everyone uses research to inform what they do in the classroom, even if that research is merely ‘what worked for me the last time I did this’. Teachers must assume that there is a starting point which says ‘on average, this is likely to get the results I am hoping for’.

    Your post also alludes to a common misconception about randomised trials. A randomised trial is a research design, no more, no less. It is a good research design if you are interested in trying to determine casual relationships. This is because it is the best way that we know of creating unbiased comparison groups. Because, in a randomised trial you have randomly allocated your cases to comparison groups you can be quite confident that any differences between your comparison groups are due to the play of chance rather than some known or unknown systematic difference between groups. Therefore, any differences you observe between outcomes in each group can be more confidently attributed to the intervention. It is important to note here that this is nothing to do with the dreaded Q words. You can just as easily use random allocation to create unbiased groups from which you collect qualitative data as you can quantitative data.

    The research design makes no claims about generalisability – at least none that can be bettered by other research designs.

    I really enjoyed reading you post. Thank you.

  5. Hamish Chalmers // 19 October 2015 at 13:00 // Reply

    In the second paragraph of my comment I wrote ‘casual’. I, of course, meant ‘causal’. Oops!

  6. We do need to make a distinction between different kinds of research. Classroom based research is of little value, because there are too many variables that cannot be controlled. Cognitive psychology, on the other hand, creates artificial experiments in which specific aspects of human cognition can be isolated. Therefore we are far more likely to gain widely applicable information from cognitive science. Skepticism about research partly derives from the huge quantity of flaky research which is out there, in which every single initiative claims to have a positive impact.

You can offer your thoughts here:

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: