home  news  search  vre  contact  sitemap
 Capacity building resources

Educational Research and policy: epistemological perspectives


Aim of this resource
'What works'
Systematic reviews (web resources)
Generic issues
Large scale population studies
Case study
Stories and narratives
Action research
Philosophical enquiry
Re-imagining educational research


Action research

By Lorraine Foreman-Peck and Jane Murray, University of Northampton

[Note: the full version of the paper which is summarised and referred to here will be published as a special issue of the Journal of Philosophy of Education and subsequently as a book edited by David Bridges, Paul Smeyers and Richard Smith and published by Blackwell in 2008]

This paper is addressed to the epistemological question of what kind of justification or warrants can be given for knowledge claims made by action researchers. Claims made by action researchers often have an ambiguous epistemological status. Firstly, the claims are not typically warranted by conventional standards for research reporting in the social sciences, rather they often resemble personal learning accounts. Secondly, there is an influential body of theorising that would deny that action research belongs to social science and is actually a form of practical philosophy. The question that the paper seeks to address is what kind of confidence policy makers at various levels can have in action researchers claims which do not fully meet social scientific standards of warranting or, alternatively, are the outcome of practical philosophising . Action research is not informed by a common set of epistemological assumptions, nor is it to be identified with a particular set of methods. Given the national policy for evidence based teaching however and the proliferation of teacher research accounts now available, it is a question worth asking. The approach of this paper consists of a theoretical discussion of epistemological considerations in evaluating action research reports and a discussion of two examples, one in the positivist tradition, and one in the critical theory tradition associated with Carr and Kemmis (1986). While a privileged account of knowledge appropriate to teaching is rejected, the paper outlines the way in which teacher researcher accounts in both traditions may be warranted.

2. Action research may be characterised as a form of practitioner research carried out by teachers into a practice problem that they are in some way responsible for. The intention is to improve a situation. Beyond this every characterisation of action research is contested and theorising is continuously evolving as action researchers bring new theoretical perspectives to bear. Writers are divided between those who think that action research is a ‘broad church' and those who see social scientific or positivist forms as a subversion or as an error (e.g Carr 2006). More generally action research is conceived of as a cycle or cycles of action and thinking involving problem formulation, planning, intervention and evaluation. There is no common underpinning epistemology of action research. An influential typology introduced by Carr and Kemmis (1986), distinguished three kinds of knowledge interest: technical, practical and critical. These were developed as distinct categories, however in practice knowledge interests may not be so easily divisible. Potential knowledge claims deriving from action research may encompass any of the following: the efficacy of teaching methods (technical) the development of theory (practical/critical) the efficacy of policy or managerial practice (technical/practical/critical) the realisation of values in practice and the development of personal and professional theorising.

3. Politicians are probably more interested in predictive generalisations, to guide policy making. However teachers, managers, parents, are more likely to be interested in how action research findings can be related to their own experience and contexts. Practitioners at different levels (below the level of politician), therefore have an interest in the degree of confidence they can have in knowledge claims made by action researchers. Confidence is based on a critical evaluation of the basis or warrant of a knowledge claim.

4. Warrants are assessed on the quality of the evidence and the claims made on the basis of that evidence (Gorard 2002). Researchers can make warranted claims on the basis of relatively poor evidence, if the weakness of the claim is acknowledged. Knowledge claims can be invalidated where the claims made go beyond what the evidence will bear, even where the quality of the evidence is good. It may be objected that claims to know are however not the same as claims that something is true. A strongly warranted claim may still be false. It is the case that all knowledge claims are fallible (Hamlyn 1970). It is enough to say that the justification given is in line with the standards operating at the time, for this kind of claim, (i.e all alternative accounts are less plausible than the justification given). In reading research reports, we assess whether the evidence and the arguments are strong or weak. On this account then teachers' professional learning accounts are ‘research products' but because they do not meet the strongest requirements of social science in terms of the presentation of evidence or the explicit processes of reasoning, may be termed weak claims to knowledge. Critical action research which aims at emancipatory knowledge also makes claims, but these are based on a consensus view of the truth. It is beyond the scope of this paper to examine this in detail but it should be obvious that a consensus does not guarantee infallibility anymore than empirical claims can be thought of as certain. Action research which takes the form of practical philosophy, aims at the realisation of values in practice. This form of action research should demonstrate the link between inadequate values in practice, intervention and more adequate principles. The claim will be philosophical and therefore may be judged by standards of valid and coherent reasoning.

5. Action research is not informed by a common set of epistemological assumptions. However all versions seek a solution to a problem and it has been though that pragmatism might be its underpinning philosophy Bridges (2003). Pragmatism is not a unified philosophy and various versions are not unproblematic. Dewey's pragmatism, it has been argued is fatally inconsistent, and it has been persuasively argued that Rorty's is also a failure (Mounce 1997). It is beyond the scope of the paper to engage in scholary exegisis. The basic tenet is that what is true is ‘what works' and all forms of action research are directed towards a particular meaning of what works. It is apparent by now that ‘what works' implies different epistemological claims, i.e. the claim takes on different meanings in different theoretical discussions. Suffice it to say that a claim that something ‘works', on any version of action research is not sufficient to establish a strong truth claim, since it eschews the necessity for e.g. the challenge of contrary evidence, or the consideration of opposing values. It basically relies on testimony and authority. It is also not self contained. It needs a correspondence theory of truth.

6. The full paper illustrates the points made by a discussion of two examples of action research and tries to determine what confidence we can have in the knowledge claims they make. Jennings (2002) –points to be made are that it is practitioner research but some may deny that it is action research, because it is a) technical and b) it does not repeat the cycle. Does this matter? It is has potential for application and testing in Jenning's and other's classrooms. Jennings makes a strong empirical claim, but does not conform to all the criteria for good reporting in the social sciences. It is transparent in that it presents evidence, and we are justified in claiming a high degree of confidence. Is her whole enterprise in error, as some theorists claim? It is not apparent that it is, since it builds praxis. Arguments against the positivistic view depend on a mechanical and implausible idea of the means-end relationship. Dewey's analysis of the means /end relationships provides a much better analysis that fits with our idea of the normative quality of teaching. It is argued that a teacher would be justified in using this research, with the proviso that she held the findings as hypothetically or tentatively, as others have argued (e.g. Bassey 1999).

In the second paper, Melrose and Reid (2000) identify their research as belonging to the critical theoretical model of action research. It is argued that the authors make a weak claim to knowledge because they simply indicate their evidence and do not sufficiently demonstrate their reasoning. It is more like practical philosophy in working out principles for ‘right' action, but it is written up in such a way that it does not look like philosophy- there are no explanations of dilemmas, values, arguments. Would we be justified in using this research? Even so, depending on context and the readers purposes, this research could be useful.

7. Conclusion. Action researchers address questions that are likely to be of interest to the profession, and the move towards an evidence based profession requires that we know how to evaluate knowledge claims made by action researchers whose reports do not fully comply with conventional empirical standards for reporting empirical research or conventional standards governing philosophical theorising about practice. The question that the paper sought to address is what kind of confidence policy makers at various levels can have in action researchers claims which do not fully meet social scientific standards of warranting or, alternatively, are the outcome of practical philosophising.

At a minimum knowledge claims should be clearly identified, the warrant or justification for the claim should be available so that a reader can make a judgement. Depending on our current state of knowledge, our contexts and purposes, weak claims to knowledge can be useful and valuable. However all research findings , even when thought to be strong , should be held hypothetically by teachers applying them in their own practices , since the context in which they were developed is , inevitably a different one.


Bassey, M. (1999) Case Study Research in Educational Settings. Buckingham, Open University Press.

Bridges, D. (2003) Fiction written under oath ? Essays in Philosophy and Educational Research , London , Kluwer Academic publishers.

Carr, W., and Kemmis, S. (1986) Becoming Critical: Education, knowledge and action research. London , Falmer

Carr, W., (2006) Philosophy, Methodology and Action Research. Journal of Philosophy of Education, Special Issue Philosophy, Methodology and Educational Research part 1, 40,2, pp421-437

Gorard, s (2002) Warranting claims from non-experimental evidence . Occasional Paper series 48, ESRC Teaching and Learning Research Programme. Cardiff University School of Social Sciences

Hamlyn, D.W. (1970) The Theory of Knowledge. London, MacMillan Press Ltd.

Jennings , S. (2002) ‘Helping weak readers up the reading ladder' ch.5. in O. McNamara ( Ed ) Becoming an evidence-based practitioner: A framework for teacher–researchers . London and New York , Routledge, Falmer

Melrose , M., and Reid M. ( 2000)The Daisy Model for Collaborative Action Research: application to educational practice . Educational Action Research, 8, no.1,pp151-165

Mounce, H. O. (1997) The Two Pragmatisms: from Peirce to Rorty. London and New York , Routledge.



Creative Commons License TLRP Resources for Research in Education by Teaching and Learning Research Programme is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 2.0 UK: England & Wales License




homepage ESRC