A question of style

Pioneering work in participant observation with Glasgow gangs, a study of unemployed young adults in the North East of England, and a survey of vandalism and graffiti… Frank Coffield has contributed to knowledge throughout a distinguished research career stretching back more than 30 years. Perhaps appropriate then, I thought as I went to meet him, that he also has a long-standing interest in vocational education and what is now termed 'lifelong learning'.

Frank taught in comprehensive schools and at Jordanhill College of Education before moving to Keele University and then holding chairs at Durham and Newcastle. From 1994 to 2000 he was director of the ESRC's Learning Society Programme, during which time he edited four reports and two books of findings. He is now at the Institute of Education, London.
At a time when it seems difficult to move in any phase of education without stumbling across a learning style questionnaire, Frank and colleagues based in Newcastle have just carried out the most comprehensive, systematic (and, it must be said, critical) international review of learning styles so far undertaken.

You have just led a major two-year project of research into learning styles. How did this come about?
We entered a peer-reviewed competition that was set up by the Learning and Skills Development Agency [www.lsda.org.uk]. We were interviewed in London and were subsequently awarded the contract.

We had four main research questions: What models of learning styles are influential or potentially influential? What empirical evidence is there to support the claims made for these models? What are the broad implications for pedagogy, i.e. how are we to teach students? What empirical evidence is there that models of learning styles have an impact on students' learning?

What stands out for you as being the most significant outcome of the project?
The very marked variability in the quality of learning style instruments or questionnaires. It matters fundamentally which instrument is chosen. In all, we found 71 different varieties or models and we evaluated in detail 13 of the most well known and of the most promising. Most, though, had problems with reliability, validity or evidence of their impact on pedagogy. Indeed, some of the best known and commercially successful instruments have such low reliability and negligible impact on pedagogy that we recommend that their use in practice and research should be discontinued.

What instruments, if any, emerged as the most robust?
Out of the 13 instruments we evaluated, Allinson and Hayes' Cognitive Style Index – the CSI – emerged as the most robust. We found that it had the best evidence for reliability and validity, although the pedagogical implications of the model have not been fully explored yet. The CSI was designed for use in adult organisational contexts and as a research tool. It is well regarded as a means of asking pertinent questions about how adults think, behave and learn in the world of work, but it is not suitable for use with students in education.
 
Your focus was post-16 learning, but a lot of schools have become enthusiastic about learning styles and have invested in questionnaires and staff development in the belief that it will enable them to add more value to pupil learning. Does your research imply that schools have been wasting their time and money?
I'd like to respond more positively. A learning style questionnaire can be used in classrooms to begin a dialogue between students and teachers about how people learn and fail to learn, about how people learn in different ways in different contexts, and how teachers can help or hinder these processes. But the teachers need to be knowledgeable about the strengths and limitations of the model they are using; and they also need to be aware of the dangers of labelling students 'globalists' or 'pragmatists'. Far better if they discuss 'deep', 'surface' and 'strategic' approaches to learning, which don't suggest that individuals can have only one learning style.

Many readers of The Psychologist teach in further and higher education. What are the implications for them?
Interest in, and use of, learning styles questionnaires is now very widespread within the education system as a whole and the language of learning styles has entered into the vocabulary of teachers and tutors in all phases. Indeed some of the better instruments have been developed for university students; for example, Entwistle's Approaches and Study Skills Inventory for Students, and Vermunt's Inventory of Learning Styles. What we would now like to see is teachers and students in all phases having a deeper understanding of the strengths and weaknesses of the learning styles approach and of the particular inventory they are using.

What about the reaction to your research? I imagine that some people in the field will not have welcomed it.
It is still a little too early to say – our two reports were published at the end of June last year and no reviews have appeared so far. On the other hand, I am receiving a daily stream of enquiries from practitioners and academics, and the authors of some of the instruments we evaluated have been very complimentary about our reports.

Most research contains surprises of one kind or another. At this point, what have you found most surprising about this research?
The size of the research literatures involved. For example, David Kolb's wife, Alice, drew up a bibliography on all the research studies on his experiential learning theory and his Learning Style Inventory: it runs to 1004 items up to January 2000. Corresponding figures for the Dunn and Dunn model are 1140, and for the Myers-Briggs model for the 10-year period between 1985 and 1995, over 2000.

So it's a potentially overwhelming task for anyone to evaluate the evidence for one model of learning style rather than another. As an applied psychologist working for a local education authority, I get the impression that there is considerable interest in learning styles. But surely that in itself can't account for the size of the literature?
The research literatures are now so huge because the topic has been seized upon by a large number of professional groups as a possible easy answer to very complex problems. It is no longer just psychologists and educationalists that are researching the notion of learning styles, but also business studies lecturers, geographers, lawyers, engineers, physicists and medics. Learning style obviously has an intuitive appeal to many academics, but that needs to be balanced by a careful and dispassionate analysis of the research record.

What thread runs through your diverse research activity?
From my first research project, on Glasgow gangs, to my last, on three groups of marginalised learners in the post-16 sector, the common thread has been a concern with those who have been badly treated by the educational system. Only last month I received an e-mail from an education tutor who was visiting a student in a primary classroom on the south coast. All the pupils had labels on their desks indicating whether they were a 'visual learner', a 'tactile learner' or a 'kinaesthetic learner'. This is nonsense, and dangerous nonsense at that. My aim has been to change policy and practice to improve their lot.

The learning styles reports, one aimed at practitioners and one at academic researchers, are published by the Learning and Skills Research Centre. They can be downloaded from www.LSRC.ac.uk, or hard copies can be obtained from Sally Farady, Research Manager, LSDA, Regent Arcade House, 19–25 Argyll Street, London W1F 7LS.