M is for Mediation

23 04 2017
county agent 02

The county agent (North Carolina State University)

“Teachers don’t read research.”

This would seem to be a fairly uncontroversial claim, but it generated a fair bit of heat on social media when I made it at last month’s IATEFL conference – see for example, the Dynamite ELT blog.

Why don’t teachers read research, as I claimed? Simon Borg (2009) reports that lack of time and accessibility, combined with a perceived lack of relevance, are often cited as reasons. More to the point, perhaps, is the fact that the researchers don’t write with teachers in mind. As Tom Farrell (2016, p. 352) suggests:

When the word “research” is used in any publication, readers have particular expectations about what they will read in terms of the language that is used in the publication. In most cases, such publications in education are written with a particular audience in mind that for the most part is academics.

Mark Clarke (1994, pp 12-13) goes further, in suggesting that what underpins the ‘dysfunctional discourse’ between researchers and practitioners is an issue of power: ‘Given the hierarchical nature of the profession and the higher status of theorists […] the voices of teachers are subordinated to the voices of others who are less centrally involved in language teaching.’

In sum, as Nat Bartels (2003, p. 737) concludes, ‘studies of teachers’ consumption of and attitudes towards academic research articles show that such articles do not seem to function well as a mechanism for communicating information for teachers.’

So, how is information communicated to teachers?

This is a question that I have been grappling with in preparing the new edition of An A-Z of ELT. I raised it again at the ELTRIA Conference in Barcelona this weekend.

It is the same question that John Carroll engaged with, as long ago as 1966 (p. 98): ‘How is the researcher going to communicate with the consumer of his [sic] research? Does he even know what his audience is and what his audience wants?’

Carroll answers his own question by arguing that the researcher is unqualified to ‘speak’ directly to practitioners. Instead, mediators are needed to ‘translate’ research into its practical applications – or to ‘particularize’ it, in Clarke’s (1994) terms.  Carroll likens this mediating role to that of the ‘county agents’ who functioned in the US as a bridge between agricultural scientists and actual farmers in their fields. ‘There could be an analogue of the county agent in education: the individual who makes a specialty of communicating the findings of research to the potential consumer, the teacher, teacher trainer, educational policymaker, or prepare of instructional material.’ And he adds, ‘the major problem that would be encountered… is the shortage of persons qualified to do this kind of educational liaison.’County-agent-Ammons-Ruth-O-Kelly-1925

What would qualify a person to take on this mediating role? And what qualifies me? In attempting to answer that question, I approached four other ‘county agents’ in our field – i.e. writers of well-known, globally marketed methodology texts —  and asked them a number of questions about the way they achieve ‘educational liaison’, including this one:

  • How important is it, do you think, to link research and classroom practice?

Interestingly, their answers varied considerably:

A. Imperative! Teachers need to ground their teaching in research-based findings and assumptions. And, more importantly, teachers themselves should not shrink from engaging in their own classroom-based “action research.” It’s an all-important interaction.
B. I simply fail to understand people who deny the role of research in helping us understand our practice and improve it. Research is, after all, what all good teachers would do if they had the chance.
C. It’s sometimes a useful support and can provide interesting insights, but it’s certainly possible to write helpful and valid professional guidance for teachers with no research references whatsoever.
D. I’ve never found much formal “research” very helpful to my own classroom work. I am not “anti-research” but I do carry a suspicion of many statistical studies in teaching.  … I more often look at the literature to see if it can help me understand what I have already noticed myself.

Which raises the question: Would you consult/recommend/approve of a methodology text that made little or no reference to published research? And would you expect the writers of such texts to be established researchers in their own right?


Bartels, N. (2003) ‘How teachers and researchers read academic articles’. Teacher & Teacher Education, 19.

Borg, S. (2009) ‘English language teachers conceptions of research.’ Applied Linguistics, 30/3.

Carroll, J. (1966) ‘The contributions of psychological theory and educational research to the teaching of foreign languages.’ In Valdman, A. (ed.) Trends in Language Teaching. New York. McGraw-Hill.

Clarke, M. A. (1994) ‘The dysfunctions of the theory/practice discourse.’ TESOL Quarterly, 28/1.

Farrell, T. (2016) Review of Teacher-Researchers in Action, by Dikilitaş et al (eds.) ELT Journal, 70/3.

R is for Research

24 10 2010

“Where’s your evidence?!”  Given the vehemence with which some methodological matters are argued (especially on blogs), it’s surprising that this question doesn’t come up more often. Well, a good place to start looking for evidence might be The British Council Directory of UK ELT Research, compiled by Shelagh Rixon and Richard Smith,  the primary aim of which is “to disseminate and share information generally in the area of UK-based ELT research”. (You can access it via the TeachingEnglish website here).

It makes a fascinating read. I was particularly interested to find out what people have been researching at doctoral level. (Where else can you find this information, after all?)  These are just some of the topics that have recently been investigated, and whose findings I’d love to get my hands on:

  • The lexis and grammar of English as a Lingua Franca
  • The use of interactive whiteboard technology
  • ‘Nativespeakerism’ and the status of non-native teachers
  • Formulaic language and SLL
  • A systemic view of emergent course design
  • Collaborative learning via e-mail discussion
  • Group influences on individual learner’s motivation

And this one, not least because it is research about the effects of research:

Andon, N. 2008. What roles do theory and research play in language teaching? A case study on the task-based approach in language teaching.

The researcher’s aim was “to examine the ways that language teachers make use of theory and research presented to them in the professional literature and on training courses”.

As both a writer of ‘professional literature’ and a teacher educator, this goes to the heart of what I do. I’m often accused (and probably guilty) of selecting research evidence to support my own point of view, and ignoring that which doesn’t;  or, worse, of not having any evidence at all. This is particularly the case with the Dogme ELT philosophy:  it’s not enough to wheel out a supportive bibliography in order to situate Dogme on a  firm theoretical base (as I did on Jeremy Harmer’s blog recently). Nor will anecdotal evidence do: the Dogme discussion list is strewn with feel-good accounts of  ‘successful’ materials-light, talk-driven classes. But people want concrete proof. They want research evidence.

Fair enough. But what kind of evidence would that be, and what’s the guarantee, anyway, that this evidence would satisfy the sceptics?

Let’s take Dogme: how could you provide convincing evidence that it works? Here are some possible lines of attack:

1. Measure the outcomes of teaching two matched groups, one taught with coursebooks, one taught without. Problems: too many variables (teacher, students, context factors…); what outcomes do you meaure (fluency? accuracy?) and how do you ensure your assessment criteria don’t automatically favour one approach over another? Also, it would probably need to be done over an extended period to produce significant findings.

2. Record and transcribe a sequences of ‘Dogme-style’ lessons, and track ’emergent language’, i.e. language that learners have seemingly appropriated and then re-used subsequently, thereby showing that learning can take place without a pre-selected syllabus and solely through interaction. Problems: an enormous amount of work (all that transcription); you would also need to pre-test – but what would you be pre-testing for if the learnt language is not pre-selected? Also, without a control group, there’s no way of knowing if that same language would also have emerged in a more orthodox setting.

3. Ethnographic case study of a Dogme class over an extended period, using observations, interviews, questionnaires, etc, to gather a ‘thick description’ from the point of view of the participants. Problems: nothing to compare it with; too context specific, hence ungeneralisable; ‘Hawthorne effect’, i.e. subjects out-perform when they know they are being experimented on; attitudinal questionnaires are unreliable – subjects say what they think you want to hear.

4. Fine-grained, micro-analysis of classroom interactions in a Dogme class compared to a ‘traditional’ class, to demonstrate, for example, a greater quantity of and/or better quality of communicative, scaffolded, authentic, creative, etc language use in the former. Problems: again, assuming you could control the variables, the specificity of the findings is unlikely to satisfy the sceptics; also, because the findings are evaluated through the lens of a specific theoretical model – e.g. sociocultural learning theory – your conclusions depend on this theory being generally accepted – which it isn’t.

Have I missed anything out?

In the end, though, there’s probably nothing you can do to convince the doubters (let alone the cynics). Which makes one wonder: why do research at all?  One way of answering this question might be to re-assess what research is capable of achieving. Nunan (1992) distinguishes between two alternative conceptions of research: “The first view is that external truths exist ‘out there’ somewhere.  According to this view, the function of research is to uncover these truths.  The second view is that truth is a negotiable commodity contingent upon the historical context within which phenomena are observed and interpreted”  (p. xi-xii).   Researchers in the second tradition are interested less in proving a theory than in deepening their understanding of their own situated practices. This understanding may, in turn, influence the way these practices evolve.

But isn’t this a cop-out? Is there no way my research can be generalised to your context?


Nunan, D. (1992). Research Methods in Language Learning. Cambridge: Cambridge University Press.