Research - Jonathan Osborne
School of Education, Stanford University and King's College London
I started my career as a teacher of physics and science in the then Inner London Education Authority in 1973. I taught physics in two schools for 9 years and then took a secondment as an advisory teacher for 3 years. When that finished the job at King’s College appeared in 1985 and I was fortunate enough to be offered the post. It took me 10 years from then to complete my PhD and to produce any publications of note. However, it was a good apprenticeship and after that my career took off with a number of publications – some of which are listed below. In 1996 I was promoted to a senior lecturer and in 2001 to a professor. In 2003, I was appointed to the Chair in Science Education. From 2005-2008 I was Head of Department of Education and Professional Studies at King’s and then in 2009 I moved to Stanford University in California to take up the California Chair of Science in the School of Education.
The article that follows may help others to start their research in science education (and get published!)
1. How do you define/identify a research area?
2. The Importance of Reading
3. How do you identify participants?
4. How do you decide on a methodology?
5. Disseminating your Work
6. Seminal influences on my thinking and work
7. Key publications
Becoming a Researcher
Jonathan Osborne, School of Education, Stanford University and King's College London
The simple answer to this question is that a good research area must be one that you are interested in. After all, you are going to spend many lonely hours pursuing this area of research and it is only your interest in or passion for the topic that will sustain you through the long, dark hours! That said - a passion for the history of science education between 1860 and 1870 is unlikely to interest many others. Thus an issue in considering which research area to pursue is what is the relevance of this research? Who will want to know about its findings? Will it, for instance, be existing researchers because your work will have added to our understanding of topic X? If so, you need to consider whether your work will simply add to an existing body of work or research program or, alternatively, whether it will be developing a new area of research. The latter is more of a risk as, whilst it might be highly original, there is the possibility that others will not see it as being of much value and you may have difficulty publishing your work. And, if you are working in an academic environment, your future career will be dependent on getting your work published.
What this means is that it is wise not to rush into the first thing that interests you. Rather spend some time digging up the research in the area that could possibly be a focus for your research attempting to answer two questions: What has been done? How extensive is the work? Is it sufficiently valid and reliable? If you were to do a similar study what would it add? However, perhaps the more important question to ask is¬ - ‘What has not been done?' A good way to do this is to try and write a review of the research in the area which should generate (or not) an argument for more research of the kind that you are interested in. If you are finding it difficult, it might be because your idea does not stand up. The other way of testing your idea is to talk it through with colleagues. First of all can you explain to them why it might be needed and valuable research? Secondly, what is their response to, and feedback about, your idea? It is well worth considering visiting somebody else in another university who you think has some knowledge of the area and discussing it with them. Most people will be prepared to give you an hour of their time to help somebody who is starting out in research and the time spent will be very valuable in developing your thinking.
Yet another way of developing your understanding of research is to join an existing research group. Participating in their workshops and seminars will start to give you a better idea of how experienced researchers think and develop your knowledge of research methods. My own career benefitted by being invited to work with Paul Black and Wynne Harlen on the Science Concept and Exploration (SPACE) Project. Working out what needed to be done in order to collect and analyse the data was a very valuable apprenticeship into the challenges of research.
Alternatively, you might be concerned to do work which will have practical value in the classroom. The common response here is to work on curriculum materials in the belief that these define what happens in the classroom. The reality is that what happens in the classroom is much more dependent on the pedagogy that the teacher deploys and the assessment framework. When it comes to the teaching of science there is a much smaller body of research in these two areas. For instance, the study of assessment that might assess the students' ability to synthesise, evaluate and analyse is still very poorly researched and is crying out for more work. Thus, working on curriculum materials is probably not advisable unless you have a specific idea such as producing sets of materials to promote critical thinking in science.
Always remember that ideas are fragile things and exposing your own ideas to criticism is a difficult and challenging experience. However, it is absolutely essential to do this and listen to the comments that come back. You can start by making presentations to colleagues and then try a small conference for young researchers where people are possibly more sympathetic. The key thing is to evaluate whether the criticisms are peripheral or core. The latter kinds of criticism are much more serious. More fundamentally, remember that the criticisms are not criticisms of you as a person, but of your ideas. They are not one and the same thing and, whilst it is impossible to separate them totally, it is important to keep that separation.
The great privilege of doing research is that you are being given time to be reflective about the practice of science teaching. Most teachers of science are so busy doing it they will tell you that they have no time to think about their work. Given that there is a lot that could be improved in the teaching of science, you should be able to find a plethora of problems which are worth pursuing. Working as a researcher in science education, I tend to see myself as a jack of all trades - a smidgeon of philosophy, a dash of psychology, mixed with a little bit of history and sociology which is all rooted in a foundation of science. Our job is to draw on the work of these more fundamental disciplines to develop the ideas about effective science education and/or to explore new ways of learning and teaching. Working at King's I have always been imbued with the concept that somehow what I do and research should inform our understanding of what constitutes good practice. In short, that it helps not to lose sight of the classroom. This is not to say that more abstract, laboratory based work that psychologists might undertake is not valuable - that simply that is not the basis on which I can work.
My own research career started with in a post at King's with one publication to my name, no PhD and only a science Masters. I began by reading a lot, conscious of my own ignorance of the field and regard this as an essential aspect for anybody starting out to be a researcher. Only by opening your self to a wide range of ideas, both within science education and outside of it, can you begin to develop a sense of the lay of the landscape and identify what ideas matter and how they are related.
It is absolutely essential to keep good notes of your reading in some form of computer-based database such as Endnote or Reference Manager. Whilst this makes writing much easier in terms of citing references, its real function is to keep a record of what you have considered important in your reading. Everything that I read that I consider to be of any significance I add to this database that now contains over 3,500 references. When I read something, I am always reading it reflectively trying to distill what are the major points in this article or book. Whenever something particularly salient stands out I mark it with a pencil or a post-it note and, before returning the book to the library, putting it on the shelf or filing the paper, I will literally type the material into the database. Both the citation and the pdf file can often be downloaded from a website if your institution has an Athens subscription. So there is no need to type in the details of the reference which can just be imported. Typing in any quotes is made easier by the fact that I taught myself to type at age 19 as I had nothing very much to do that year. This has proved to be an invaluable skill which simply requires ten minutes of practice a day for 3 months.
The pdf file itself can be linked to the Endnote database so that you can instantly retrieve the article. Whilst keeping such a database might seem to be time consuming it is an invaluable resource. First, you have all your references to hand. Second, and more importantly you have a record of why you considered it important. Third, if a student needs references for their reading you can instantly retrieve the ones you think important and send them the reference, the pdf or both.
Finding people to undertake your research with is always challenging and depends on the exact nature of the research. Most researchers recruit people opportunistically from people they know or have met. The latter may well be teachers who have attended in-service courses or Master's students. Persuading teachers to participate in research is not straightforward. Teachers are busy people and your research is not their priority. Thus before asking somebody to participate in your research you need to think through your answer to their question which is likely to be ‘What is in it for me? There is a range of answers to this - it might be an opportunity to get better data about their students; to contribute to the development of a research idea themselves; to try something new or to develop themselves professionally. The important thing is that you do have an answer to this question.
Another consideration that you will be faced with is that of gaining ethical approval for your research. In most cases, if your work is judged to be low or medium risk, this is relatively unproblematic but you will need consent from the participants. For these reasons, when starting out with research it is often best to do something exploratory which involves only one teacher or one class to minimize the administrative problems.
Many universities have their own research start-up funds which are designed to support such studies giving you a small amount of financial assistance for either administrative support or buy out of teaching for things such as critical days when you have to get into a school. Do ask and apply for such funds as you will need all the support that you can get.
Finally, look to see if you can find a collaborator to work with - either in your own institution or another. When working with someone else, your ideas benefit from your mutual but often differing perspectives on the problem, the critiques you offer of each other's ideas and your mutual enthusiasm. It is really very hard to be a lone researcher and collaboration really helps to sustain your work.
There is no easy answer to this question. The methods you choose are highly dependent on your research question. The important thing is to be open to different methodologies rather than carrying prejudices about any particular methods. It would be fair to say that a lot of educational research is dominated by qualitative research - some of which does not have a good reputation because of its lack of rigour. The place to start is by undertaking a self-audit of what research skills you have. If you really are a beginner, it is a good idea to attend the courses that are offered in your own institution. You can learn much more from a well-taught course than you ever will from reading some standard tome on research methods as these are usually taught by experienced researcher.
It is important to realise that there are software tools for undertaking research and that you need to acquire the skills of how to use these. For quantitative research the standard tool is SPSS. Although this has its own strange quirks, it really does much more than an Excel spreadsheet. It is best learnt when used in conjunction with a course on either exploratory or confirmatory data analysis - preferably both. And the earlier you get to grips with such tools, the longer you will have to become really competent.
For qualitative research, you will need to learn how to interview people, design an interview schedule and then analyse the data. When it comes to interviewing people, this is not a simple straightforward task but does require training. See if you can find a course locally you can attend. It will improve your confidence considerably and you will at least have a good idea of what you are supposed to do. Conducting interviews requires equipment - either tape recorders or video recorders - and there are many lessons to be learnt about ensuring that these work, that the batteries are fresh, and that you have spares.
As for analysing qualitative data, generally you have to transcribe the interview. This is one of the more tedious tasks involved in research. It has been considerably simplified by the introduction of digital technology and software that will enable you to repeat the segment of the interview you were just listening to so that you can transcribe it. Attempting to do it without such technology will double the time required so do go out and ask somebody about the technology, get yourself a copy, and work out how to use it. The basic formula that all researchers work with is that one hour of interview requires three hours of time to transcribe it. The only way of avoiding this commitment is to pay somebody to do it and most transcribers now charge £12-15 an hour.
When you have transcribed your interviews you are now faced with analysing them. The basic approach in qualitative data analysis is one of data reduction by labelling sections of the text with codes. These are then grouped into themes and hypotheses that you have about the data are then tested by looking to see if you can find sufficient codes to support your hypothesis. However, you do need to learn the principles of qualitative data analysis which can only be really done by attending a course. There are standard texts but learning from a course is a much better approach and more interesting.
However, the rigour of qualitative data analysis has been much improved in the past decade by the development of software tools such as NVivo and Atlas. Both of these also enable you to analyze video. Not only do these force you to be much more rigorous and systematic, they can retrieve all the text coded with a specific code virtually instantaneously - something which manual methods are very slow at doing. Moreover, they enable you to search all the texts for key words or phrases. However, perhaps most importantly, they provide a summary of your coding at the end and allow you to test hypotheses by looking for specific combinations of codes or just to look at what the major themes are in the data. It is rigorous because the computer will count the instance of use of a given code, identify combinations and do it very quickly - all tasks which would take so long manually that very few people do it.
So how do you choose between qualitative and quantitative approaches? The basic answer to this question is that quantitative methods are generally good at answering the question of what exists. For instance, how many of the population of this age have this particular type of view? They are weak at explaining why they might hold that particular view. This is where qualitative methods are much better as the interviews enable you to ask questions focussing on why the subject thinks what they do think. However, interviews are time consuming and the size of your sample is often limited by pragmatic constraints. The problem then is how representative and general is the sample you have interviewed? Many qualitative studies select subject purposively for this reason - not claiming that they are necessarily representative but that such individuals give some insight into the views of the specific groups that have been selected.
Quantitative methods are often very valuable for surveys and the data can easily be represented in charts and tables which makes it easy to tell a story about what the data show in a relatively short space. Qualitative research relies on many extracts from the interviews and it is difficult to condense this into the length of article that many journals will publish. So the answer to the question about what methods are most appropriate is dependent on the nature of the question you are asking.
We now live in a world where we are surrounded by sources of information which are all competing for our attention. If you don't either write about your work or talk about it, nobody is going to know about it. Many people are daunted by the idea of writing so beneath are some thoughts about how to overcome this fear.
5.1 Learning to Write
Many of us entered science because it required us to write less than that required by our peers studying humanities based subjects. Taking up an academic post in a university requires you not only to do research but also to write articles for publication. For many young researchers this is one of the biggest hurdles that can make or break their careers . There is no quick route to becoming a skilled and fluent writer of English. There are some things which can be done from which you can learn though. In no particular order, these are:
1. Academic writing is a genre of writing. It has rules. Like all rules, these can be broken but you must know the rules first of all. There are range of style manuals - one that comes highly recommended is:Joseph M. Williams. (1995). Style: Toward Clarity and Grace. Chicago: University of Chicago Press. No style manual is going to be a magic book of tricks but it will stop you making basic mistakes such as ending sentences with prepositions or using that ghastly phrase ‘in relation to' which is always a sign of the lazy writer who cannot be bothered to explain what the relation is.
2. Proof read what you write by reading it out aloud to yourself. If you cannot read it easily it is extremely likely that nobody else will be able to make sense of it. Reading it silently is not good enough as you tend to skip words so easily and miss poor grammatical structures or errors in the English.
3. When you have done this, ask a colleague to read it and comment and see what they think about it. The feedback is always difficult. What you thought was your perfect prose comes back littered with edits and questions, but it is an important lesson.
4. When you read something that is well-written take a moment to ask yourself why? What is it about the use of language that makes the ideas so elegantly expressed.
5. The most common error we make in writing is to repeat words either in the same paragraph or even in the same sentence. All computers now come with a Thesaurus which is the first port of call to find a duplicate word. If this fails, ask yourself if the sentence can be expressed in a different manner.
6. Identify the audience you are writing for. Writing for academics is different from writing for teachers or writing for publication in a popular magazine. Focus on the kind of person you think you are writing for and ask yourself whether the words you are using are too academic or not academic enough.
7. Learn to love language. Words are the writer's tool and although writing is hard, there is a lot of pleasure from writing a good paragraph or a good article which says what you want it to say in a graceful and pleasing manner.
Finally, practice makes perfect. Only by working at your writing, and thinking about the feedback you are given, will it improve significantly.
5.2 Talking about your work
The key method for talking about your work is to make presentations at conferences or seminars. How hard it is to obtain a slot at a conference varies. Some conferences are just happy to have people volunteer to come along and present a issue relevant to the conference. The ASE annual conference is like that and is a good venue for presenting ideas. Presenting your ideas at a more academic conference usually requires that you submit a 1-4 page outline of your work which covers aspects such as the research question(s), background literature, methods and findings. Getting your paper accepted for presentation requires that you pay careful attention to the rubric in the goal about the nature of the information required in the submission and you are strongly advised to use their headings to frame your writing. Most of these submissions have to be put in 9 months before the conference. A good idea is to do one jointly with a more experienced colleague from whom you can learn what is the appropriate style and how the key points of the work have to be summarised.
If you get it rejected, do not be put off. The review process is generally done blind and the reviewers do not know you. Under this process, even the most senior researchers sometimes go wrong and get rejected - something that has happened to me several times in my career. If it is accepted, then you need to start thinking ahead about whether you have done all the work necessary, what needs to be done before the conference, and how you will get the paper written before then. Do not attempt to write a 30 page, single-spaced paper for the conference. Nobody will read it. Rather, think of getting the main ideas about your work in something like 8 pages with clear, structured headings. This will force you to organise your thinking, work out your main arguments and people can always ask for more information if they are that interested.
When it comes to the presentation - look at how much time you will have. Some conferences pack in presentations so that the presenter only get 15 minutes. You cannot present all of the detail of your research in that time. You have to look on your slot as a marketing opportunity which explains why your research is important, what its research questions are, what are the bare bones of the methods, and then presents the major findings. Essentially you are trying to whet people's appetite and interest them in your work. The motto to be adopted here is less is more. Any PowerPoint presentation should restrict itself to no more than 1 slide per minute. People simply will not absorb anything which is more dense. Also, make sure that you leave some time for people to ask you questions at the end. This is important as the questions they ask enable you to judge the level of interest in what you are presenting and get some critical feedback about the work. It is considered very bad style to use up all of your allotted time and not permit any questions. It is even worse to attempt to overrun your time and any hardened chair like myself would simply stop you as you are eating into somebody else's time.
I have been fortunate in that I was recruited to work in an institution where so many bright, able and thoughtful people were working. More importantly, these people were not afraid to say what they thought and to challenge my thinking. Finding such an environment is very important to anybody's intellectual development as it is very difficult to develop your thinking on your own. In short, the message is look for a collaborator - somebody you can work with.
In no particular order, these are the people that have influenced by thinking.
Michael Shayer and Philip Adey. These two are some of the very few people who have established a research program based on a well-defined theoretical frame - that of Piaget and persisted with working through the implications by conducting an extensive body of empirical work. In the 1970s they conducted a nationwide survey to establish the level of cognitive development and its progression from age 9 to 16 showing why most science curricula were too demanding at that age. In the 1980s, they developed an intervention program to accelerate students' cognitive development and demonstrated empirically that it worked and in the 1990s, Philip Adey established an extensive program of professional development around this work which has become self-sustaining. This work is captured in three books:
Shayer, M., & Adey, P. (1981). Towards a Science of Science Teaching. London: Heinemann.
Adey, P., & Shayer, M. (1994). Really Raising Standards. London: Routledge.
Adey, P. S., Landau, N., Hewitt, G., & Hewitt, J. (2003). The Professional Development of Teachers: Practice and Theory. Dordrecht: Kluwer.
Rosalind Driver: Ros was a colleague for only two years before she died. Her work contributed enormously to framing the primary science project that I worked on with Paul Black and Wynne Harlen - the Science Process and Concept Exploration (SPACE) project that I worked on from 1987-1994. She and her collaborators were responsible for establishing the significance of the constructivist view of learning which is perhaps most easily comprehended from:
Driver, R. (1983). The Pupil as Scientist? Milton Keynes: Open University Press.
During the 1980s there was fairly intense debate about the limits of constructivism as this group did not think the work of Piaget, and by implication that of Shayer and Adey, was as important in explaining the difficulties of student learning as their own. This debate is important and some of it is summarised in another important book by Michael Matthews:
Matthews, M. R. (1994). Science Teaching: The Role of History and Philosophy of Science. New York: Routledge.
This book is also important in outlining why teaching about the history and philosophy of science matters and needs to be read by everybody working in science education.
My own view about this debate and its resolution is to be found in:
Osborne, J. F. (1996). Beyond Constructivism. Science Education, 80(1), 53-82.
Ros Driver and her colleagues articulated a more sophisticated or subtle version of their position in their article:
Driver, R., Asoko, H., Leach, J., Mortimer, E., & Scott, P. (1994). Constructing Scientific Knowledge in the Classroom. Educational Researcher, 23(7), 5-12.
Ros Driver was also important to my work in showing how important it is to support and encourage younger people in their work. Her enthusiasm for things coupled with a critical mind was infectious. Together we worked on the proposal to Nuffield which led, after her death, to the publication of the document Beyond 2000: Science Education for the Future which led to a greater emphasis on ‘How Science Works' in the curriculum. In particular, she led me into a line of work on argumentation in school science which helped me to bring together an interest in learning about science and the history and philosophy of science and the thinking behind that work can be found in:
Driver, R., Newton, P., & Osborne, J. (2000). Establishing the norms of scientific argumentation in classrooms. [Paper presented at the ESERA conference, Rome 1997]. Science Education, 84(3), 287-312.
Robin Millar: Robin has been a major collaborator both on the project Beyond 2000: Science Education for the Future and the Evidence-Based Science Education Project which was funded by the ESRC. He is a very good critical thinker and inquisitor with a detailed knowledge and understanding of science. Important elements of his work can be found in:
Millar, R., & Driver, R. (1987). Beyond Processes. Studies in Science Education, 14, 33-62.This article is still very much the defining contribution on the interrelationship between content and process in science.
The work he conducted with Ros Driver and others on Young People's images of science remains one of the most important insights into the picture of science that school science develops in young people:
Driver, R., Leach, J., Millar, R., & Scott, P. (1996). Young People's Images of Science. Buckingham: Open University Press.
Stephen Norris and Linda Phillips: I had the privilege of meeting these two when they spent a year in London. Stephen is a philosopher by background and Linda Phillips is an expert on literacy and the teaching of literacy. I can only point to three pieces of his work which I think are essential reading:
Norris, S. (1997). Intellectual Independence for Nonscientists and Other Content-Transcendent Goals of Science Education. Science Education, 81(2), 239-258.
Norris, S., & Phillips, L. (1994). Interpreting Pragmatic Meaning When Reading Popular Reports of Science. Journal of Research in Science Teaching, 31(9), 947-967.
Norris, S., & Phillips, L. (2003). How literacy in its fundamental sense is central to scientific literacy. Science Education, 87, 224-240.
Finally, let me say that you should never be afraid to write to any author of a paper with a critical comment. Firstly, most of them will be (a) flattered that somebody bothered to read the paper and (b) even more flattered that you were sufficiently bothered to write. Most will respond and the dialogue can be very productive and may lead to collaboration in the future. Also, writing such e-mails forces you to think critically about their ideas and express yourself coherently. And remember, as Bachelard argued ‘Two people must first contradict each other if they really wish to understand each other. Truth is the child of argument, not of fond affinity.'
Bachelard, G. (1940). The Philosophy of No. Paris
Osborne, J. F. (1993). Alternatives to Practical Work. School Science Review, 75(271), 117-123.
Osborne, J. F., & Simon, S. (1996). Primary Science: Past and Future Directions. Studies in Science Education, 27, 99-147.
Osborne, J. F. (1996). Beyond Constructivism. Science Education, 80(1), 53-82.
Millar, R., & Osborne, J. F. (Eds.). (1998). Beyond 2000: Science Education for the Future. London: King's College London.
Newton, P., Driver, R., & Osborne, J. (1999). The Place of Argumentation in the Pedagogy of School Science. International Journal of Science Education, 21(5), 553-576.
Driver, R., Newton, P., & Osborne, J. (2000). Establishing the norms of scientific argumentation in classrooms. [Paper presented at the ESERA conference, Rome 1997]. Science Education, 84(3), 287-312.
Osborne, J. F., & Collins, S. (2001). Pupils' views of the role and value of the science curriculum: a focus-group study. International Journal of Science Education, 23(5), 441-468.
Wellington, J., & Osborne, J. F. (2001). Language and Literacy in Science Education. Buckingham: Open University Press.
Osborne, J. F., Ratcliffe, M., Collins, S., Millar, R., & Duschl, R. (2003). What 'ideas-about-science' should be taught in school science? A Delphi Study of the 'Expert' Community. Journal of Research in Science Teaching, 40(7), 692-720.
Osborne, J. F., Simon, S., & Collins, S. (2003). Attitudes towards Science: A Review of the Literature and its Implications. International Journal of Science Education, 25(9), 1049-1079.
Osborne, J. F., Erduran, S., & Simon, S. (2004). Enhancing the Quality of Argument in School Science. Journal of Research in Science Teaching, 41(10), 994-1020.
Osborne, J. F., & Dillon, J. (2008). Science Education in Europe: Critical Reflections. London: Nuffield Foundation.