The Association for Science Education
  1. Home
  2. Resources
  3. Welcome to Sci-Tutors
  4. Research
  5. Research - Jerry Wellington

Research - Jerry Wellington

Abstract

Jerry Wellington – pen portrait 

I did a degree in Physics and Philosophy at Bristol University before travelling up the A4 to do my PGCE at the Institute of Education in London, focusing on ‘Integrated Science’ as it was called then. After my teaching practices in London, I was offered a job in Bethnal Green, Tower Hamlets. I enjoyed my time there, teaching Physics and ‘Combined Science’ at various places, moving on to become head of department. Whilst teaching full time, I did a Masters at the Institute taking 4 years to complete as a part-time student.

I then moved from Tower Hamlets to the University of Sheffield, School of Education, to teach on the PGCE there and also some of the higher degrees. Over twenty years after my move North, I am now a professor in Education in Sheffield and something of an adopted Northerner. My research and publishing has spanned a number of areas in science education and more recently on research methodology. A few of my publications are shown below, starting with the first refereed article I ever had published, which gave rise to a lot of glee and air punching ( I still do this).

The main article gives advice to those new (or not so new) to Science Educational research - to get you started and get you going.

Contents

1. Defining a research area, choosing a topic and formulating research questions
2. The linear model of research as shown in some textbooks .... And the reality 
3. Achieving quality by locating research in ‘what's been done before' 
4. Deciding on methodology and methods 
5. How do you identify and treat participants? The key issues of ethics, sampling and access 
6. Going public: disseminating for different audiences 
7. Further Reading 

DOING EDUCATIONAL RESEARCH
Jerry Wellington - University of Sheffield, School of Education

1. Defining a research area, choosing a topic and formulating research questions

I always think this is the hardest part. There are so many areas of science education that are interesting that it is almost impossible to choose one and then to narrow it down until the research is ‘do-able' - for me, that it is the main criterion for a piece of research: it has to be do-able or manageable within the limiting constraints of time, energy and money that we all face.

I would suggest that the main questions you should ask when choosing a topic would include:

  • Does it really interest you? Why? e.g. it is inherently interesting; it relates closely to you current job? You have always wanted to look into it?
  • What do you know about what has been published already about this topic?
  • What really needs exploring next / what ‘gaps' are there in our existing knowledge base?
  • What is the historical perspective on it? Is it a recurring debate or issue?
  • What might you contribute to this area by doing your own piece of research (your brick in the wall)? Will it be worthwhile? 
  • Who might use or be influenced by your research e.g. practitioners, policy makers, theorists? 
  • Have you discussed your ideas with colleagues?

The starting point for a research project may be a question, or questions, that the researcher would like to address. It may be an idea or a hypothesis to be tested. A slightly less focused start might be an issue to be explored or, more ambitiously, a problem to be solved.

Stating or formulating your purposes under one (or perhaps more) of these categories can help at all stages of a research enquiry, especially at the outset, i.e.

  • What question(s) are you asking/ addressing?
  • What hypothesis are you testing? (if any)
  • What problem(s) are you trying to solve/alleviate?

If you can formulate the questions you wish to ask, it then helps to decide whether they are

  • 'what, which or where' questions;
  • 'how' questions; or
  • 'Why' questions.

'What, which or where' questions often involve descriptive research, sometimes a fairly straightforward collection of information. For example, what computer programs are most commonly used in year 7 of science education? Which children in a certain school have parents who participate in the daily reading scheme? Where are the multi-media stations in secondary schools located?

The 'why and how' questions (i.e. exploratory and explanatory) are usually the more interesting but invariably the most complex and intractable. A search for ‘How' certainly implies a ‘hunt' for some sort of mechanism; the search for ‘why' implies looking for a reason. Both seem to be searches for different types of causal explanations. Finding causal explanations (cause-effect relationships) can be acknowledged as the very core of science, part of any process really worthy of the label of ‘science'. Though, naturally, there are complications. We may have to acknowledge a chasm between explaining (neutral, 3rd person, nomothetic accounts of cause and effect) and understanding (1st person, ideographic accounts).

Framing research questions should always be the first step in the research process. It should always be a case of questions first, methods later. For example, it makes no sense to decide: 'I am going to use questionnaires / interviews / observations' before clarifying the questions which you wish to address or shed light upon. To sum up, they may be what, which, where, how or why questions. The former may imply a straightforward collection of information, perhaps a survey approach. But the latter, i.e. the how and why questions which seek explanations will demand more in-depth exploratory approaches.

2. The linear model of research as shown in some textbooks .... And the reality

The research process has often been depicted as a linear, logical sequence starting with the formulation of aims, then planning, collecting, analysing and interpreting data, and ending with conclusions and writing up. One example of the 'ideal' research sequence is shown in Figure 1.



As the biologist Sir Peter Medawar famously pointed out, this after-the-event portrayal is a fraud, and many others since have admitted that it does not happen like this. Medawar often called scientific research a mixture of 'guesswork and checkwork'. A more realistic approach (see Figure 2) is to admit that the process is cyclical or iterative in that people go back and replan/refocus their research, collect and analyse their data and realize that they need more, or different, data; start to write up and realize that they are addressing the wrong questions; find that the targets or samples they have set themselves are too low /high, and so on. Even the cyclical version in Figure 2 is a cleaned up and idealized version of what really happens.

 

In summary, real life research is a messy business and it would be wrong to pretend otherwise in a report, article, thesis or a book. One of the most common activities in real research is compromise. We compromise over time spent, distance travelled, methods used, samples chosen, literature reviewed, words written and money spent.

3. Achieving quality by locating research in ‘what's been done before'

It is essential, before beginning one's own research endeavour, to explore what has already been done - one does not want to appear like a ‘jack in the box' who has suddenly appeared on the scene and who neglects the fact that a whole raft of research may be out there already.

The importance of finding out what is already ‘known' in an area of research, what's been done before, and (just as importantly) how it's been done is an aspect of research which many of us, including physical scientists at times, are apt to neglect. One of the dangers of such neglect is the old cliché of re-inventing the wheel.

There are many methods of finding what is already known, e.g.

  • Start by reading a textbook or a review paper that gives a broad overview of the area we are interested in, and then following selected references.
  • Start from a seminal or much-cited paper and go from there Each paper will have references at the end which will lead to other references, and so on. This method can be called "snowball searching". The problem is that the process is rather like a chain reaction and the list of publications one "should" read grows exponentially!
  • Ask the experts in the field for a list of key authors or references
  • Use an online search engine (e.g. Google or Google Scholar) to get you started.

These simple search techniques are invaluable and generally sufficient to get a broad overview of the problem.

3.1 Reading and Judging Other People's Research

One of the difficult knacks of reviewing the literature is knowing when to stop and where to draw the boundaries. An equally difficult skill is to be able to judge - critically but fairly - the research reported by others. We tread on dangerous ground if we examine too negatively the work of others - ancient slogans about not throwing stones if we live in glasshouses spring to mind. But there are certain areas, and within them certain criteria, which can be applied when critically examining research reports. Here is a list ofeleven, though there are probably more:

1. The Title: is it descriptive i.e. does it accurately reflect what is in the article? Is it attractive?
2. The Abstract: does it tell you: why the research was done and why it's important; how it was done, with whom, with what; and what were the key findings? does it provide a map for the reader?
3. The Literature Review: does it give an overview of the range of literature related to the research? is it new or creative in suggesting other, alternative areas of literature which might be "laterally applied" to this area? does it explain where and why boundaries were drawn?
4. Theoretical Framework: does it start from a theoretical framework which might help to inform it or structure it? does it shed light on any existing theories or models? does it lead (inductively) to new theories or models?
5. The Aims of the Research: is it addressing a significant problem or issue? are its aims and purposes clear? is it reasonably well-focused?
6. Methodology: is it adequately described, so that the reader could replicate the study, should he wish? Do the methods chosen match the purposes/aims? why were these methods chosen and not others? did they prove to be appropriate and productive?
7. Sampling: is the sample described in sufficient detail? on what grounds was the sample chosen? e.g. why these schools/colleges; why these classes? why these documents or reports?
8. Data analysis: is the way of analysing the data appropriate for the problem? Is it described in sufficient detail?
9. Evaluation, self-evaluation and reflexivity: is the research evaluated, both for ‘content' and method/methodology i.e. is it a reflective account? do the researchers evaluate their own role, their own position, and their effects on what is being researched i.e. are positionality and reflexivity included?
10. Drawing out conclusions and implications: have the data been ‘milked' for all they are worth? are the conclusions related back to the literature review - are the two woven together? do the researchers stick their necks out too far ..... or not far enough?
11. Presentation: is it clearly-written and well-structured - or is it turgid and verbose? could Ockham's razor have been applied in places? Does it have at least some ‘story-line'?

These points together make a hard act to live up to - in a way they are counselling perfection and they do not all apply to every piece of published research. But they can be a useful checklist in examining and reading research in a critical way (and in reflecting on your own writing and reporting -see section 6).

4. Deciding on methodology and methods

There is a range of methods available to practitioners in educational research - they can be divided up crudely into those acting as primary sources of data and those acting as secondary sources. This crude distinction is shown in the tree diagram in Figure 3 Primary sources would include observation, interviews, questionnaires, focus groups and so on. The secondary sources can be put together and called ' documents'.
Figure 3 A way of crudely classifying methods in educational research


{N.B: I do not have the space here to discuss each method in detail but a full guide can be found in many of the books listed under Further Reading.}

4.1 What is methodology?

Methodology is defined by the Shorter Oxford English Dictionary as the ‘science of method' or more historically as ‘treatise on method'. Methodology is: the activity or business of choosing, reflecting upon, evaluating and justifying the methods you use. Indeed, the latter is an essential feature of any research report or thesis i.e. justifying the decisions that have been made on methods. No one can assess or judge the value of a piece of research without knowing its methodology.

Methodology is a vital part of any research project, small or large. The following table summarises some of the key aspects of methodology in research.

Table 1 Methodology: questions to consider
1 How was the study designed?
2 Was the design appropriate?
3 Why were particular methods of data collection used, and not others? Could, or should, other methods have been used? Why?
4 How could the sample have been better?
5 What was the quality of the data?
6 Why were the data analyzed the data the way they were? Could, or should, other methods have been used? Why? 
7 Can one ‘generalise' from the data?
 8 How did the researcher affect the data collected?
4.2 Integrating methods: triangulation

Even in a small scale study, a mixture of methods can often be adopted; qualitative and quantitative methods can exist side by side in an enquiry.
The concept of using a multi-method approach in collecting data, information or evidence can be called ‘triangulation'. The origin of this term lies in the use by navigators, surveyors, astronomers, artillerymen and others of several observational positions to ‘survey' or pinpoint a location.
A typology of triangulation has been suggested by various authors ( most prominently Denzin,1970) The types can be summarised briefly as follows:

1. Data Triangulation which is subdivided into:
(a) time triangulation: the researcher attempts to consider the influence of time using cross-sectional and longitudinal research designs
(b) space triangulation: researchers engage in some form of comparative study e.g. of different regions, different countries

2. Investigator Triangulation: more than one person examines the same situation.

3. Theory Triangulation: alternative or competing theories are used in any one situation.

4. Methodological Triangulation, which involves "within method" triangulation, that is the same method used on different occasions, and "between method" triangulation when different methods are used in relation to the same object of study.

5. How do you identify and treat participants? The key issues of ethics, sampling and access

5.1 Ethics: The Researcher's Responsibilities

Ethics are important in the physical sciences (which investigate inanimate objects); they figure even more prominently in the biological sciences where plants and animals are the objects of study. But this factor is multiplied in educational research, where people are studying people. Every researcher to place it foremost in the planning, conduct and presentation of his/her research. Ethical considerations override all others.

A research project could be unethical in at least 3 ways:

1. The design or planning of the research. 
Testing hypotheses about cause and effect often requires experimental design: random assignment of participants to groups which are then treated differently., Any design that deprives some participants of resources or goods they would normally enjoy (books, practical work, privacy) or, worse still, exposes them to some known risks, is likely to be unethical.

2. The methods and procedures employed
Failure to obtain informed consent from participants is nearly always unethical; so is deception or manipulation to gain access (e.g. requesting a 10-minute interview that lasts half an hour). Lack of respect and putting participants in situations that is likely to make them feel uncomfortable or inadequate are other examples.

3. The presentation or reporting of the research
e.g. Revealing personal information about participants without their consent, or describing them using derogatory or biased language; using sexist or racist language.

In summary, research might be unethical in its design, its methods, its data analysis or its presentation (see the table below). Incidentally, research in the sciences could equally well be unethical in any of these areas, despite the myth that it is value-free.

Table 2 A summary of ethically difficult research situations
1. Research without participants' prior knowledge, permission and informed consent. Participants are unaware of what they are letting themselves in for and where the findings might be publicized.
2. Asking people to do anything unsafe, or forcing them to do something
3. Withholding relevant information about the nature and purpose of the research.
4. Deceiving the participants.
5. Invading participants' privacy or taking too much of their time.
6. Withholding benefits from some participants (e.g. in a control group) or imposing disadvantages upon others (e.g. in a control or experimental group).
7. Lack of consideration, respect and honesty.
8. Breaching confidentiality and anonymity, especially in publication.

Ethics Review committees

In recent years, most professional bodies and all universities formalized the process of ethical consideration by requesting that all research involving human participants is approved by a formal ethics committee. Researchers often bemoan the excessive paperwork involved in this process - but no one should criticize being requested to reflect on the ethics of her or his work. It is now almost certain that anyone carrying out research within an organization such as a college or a school (private or public) will be subject to detailed ethical review procedures, laid down and monitored by an ethics review committee. This process will involve the researcher going through a rigorous set of questions, likely to include:

  • What Procedures, methods, measurements and tools will be used?
  • What amount of time will be required for participants in the study?
  • What methods will be used for obtaining informed written consent?
  • How will participants' anonymity and confidentiality be protected?
  • How will participants be fully informed about the study including that they may withdraw from it at any time?
  • Where will data be stored and how will results be published?
  • Are there any possible physical or psychological risks and how will these be addressed?
5.2 Sampling: an issue of Choice and Compromise for the Researcher

Samples and Populations
A sample is a small part of anything which is intended to stand for, or represent, the whole. Thus we can smell a sample of perfume, drink a sample from a glass of wine or sample a small piece of chocolate before we buy the whole bar. In each case, we are not interested in the sample per se, but in the whole thing. We use samples because we believe that they represent the whole thing - that is, the population, as it is formally called.

Sampling inevitably poses a challenge: Can I be reasonably sure that my sample is representative of the population? These questions cannot be answered unless a more basic one is answered first: What is my population? If my ‘population' is a single bottle of wine then it is safe to assume that a single sip represents it well. However, we could not safely extrapolate that that sip represents the entire vintage (all bottles of wine of the same make and same year). The same applies to research: when we choose a sample {which we must), e.g. an entire class of Year 8 children, then we can (perhaps) treat it as representative of every year 8 pupil in that school. It would be problematic, however, to assume it represents all year 8 pupils in that city, or in the UK, and preposterous to claim it represents every pupil of that age band in Europe, or the northern hemisphere, or the world.

The first challenge of good sampling, then, is to decide what my population of interest is. Only then can the next two questions about sampling be answered adequately: How should I select my sample, and how large does it need to be?

Probability and Non-Probability Sampling
Probability sampling includes various types of drawing by chance. Probability sampling is most suited for larger populations and for studies where the researcher is interested in the typical trends and patterns occurring in those populations (and not in the exceptions to those patterns). Probability sampling is used most often in the context of quantitative research. The main advantage of random sampling lies in the fact that the probability of a sample being unrepresentative of the population can be calculated precisely using statistical techniques, and taken into account when interpreting the results.
Non-probability sampling includes methods of participant selection that do not rely on chance , but on your own judgement. Its main types are listed in table 3

Table 3: Some types of 'purposive' sampling
1. Convenience: Accessible, easy-to-contact, well-known (to the researcher) people or settings are selected
2. Opportunistic: Selecting people or settings present themselves during fieldwork
3. Typical: Selecting people or setting that the researcher believes to be most typical for the target population
4. Maximum variation: Deliberately selecting people or settings that represent the greatest differences or extremes in the target population
5. Atypical: Cases clearly outside the norm are chosen deliberately e.g. exceptional students or teachers
6. Criterion: A more generic label for samples chosen according to predetermined criteria
7. Snowball: One case suggests another who suggests another. ..(also called 'ancestry' or 'recommendation' sampling)
8. Critical: Choosing special cases e.g. those with a reputation for 'good practice' in a certain field
9. Guided (directed): An informant, a knowledgeable guide, an 'expert' in a field suggests particular settings, and may even help with access

Non-probability sampling is generally associated with qualitative research methods and are often inevitable in studying very small, or very hard to access populations

It should be pointed out that probability and non-probability sampling may sometimes be combined, e.g. a school, a group, organization or other unit may be selected for its convenience or its special features and then random samples may be taken within it.

Sampling always involves a compromise between the desire to make the sample representative and the practical constraints of time, money and access opportunities. Making such compromises is an inevitable (if frustrating) part of being a researcher. Social research is always the art of the practical or the 'art of the possible' (Medawar, 1979).

The key responsibility for the researcher is to describe, clearly and candidly, all relevant characteristics of the samples studied, including all limitations of the sampling process the researcher is aware of. Transparency is the key in reporting and disseminating (see section 6).

5.3 The Researcher's role in ‘gaining Access‘

The Problem of Access
Whatever plans we might make in our research, they are almost certain to be compromised by the problem of gaining access to what we want. This might involve access to schools, students, teachers, policy makers, places, organizations or documents. Let's start with some extreme examples: in the UK many of us would like to interview the Prime Minister and a range of his minions in various Departments; some researchers might like to interview a range of 'captains of industry' to ascertain their views on science education.

In such extreme cases, access is likely to be impossible and would therefore force the unrealistic researcher to return to their drawing-board. But there are far less fantastic examples where access may well be a problem; for example, interviewing all the pupils who have been excluded from a given school in, say, the last three years; interviewing all the head teachers from a cluster of schools; observing the lectures, or the lessons, of every lecturer, or teacher, in a given department; interviewing a random sample of parents and/ or observing them in their own home.

In all conceivable cases, unrestricted access and a 100 per cent success rate are likely to be difficult if not impossible to achieve, often for purely practical reasons (and sometimes for ethical or safety reasons). The business of access can therefore seriously affect the design, planning, sampling and carrying out of social research. But we have to do something, and a compromise is always involved. This is why opportunistic or convenience sampling feature so commonly in real research into natural settings such as schools.

Guidelines in Gaining Access
Access is difficult; it requires time, effort and perseverance. But there are certain guidelines which can be followed in improving it. These may help to avoid upsetting people, 'getting their backs up' and falling foul of any of the ethical issues discussed already.

1. First, remember that a researcher may be viewed in a selection of different ways:as an academic whose feet are 'off the ground', as a suspicious stranger, as a knight to the rescue, as a friend or confidante, as a trusted colleague, as an expert, or as a puppet or instrument of the funding body, principal or the managing director. Attitudes towards the researcher are likely to vary from suspicion, mistrust or cynicism, to awe, trust or friendship. It is to be hoped that any negative viewpoints and attitudes at the outset would give way to positive attitudes and dispositions towards the end of the research.

2. Secondly, the important first task is to establish individual contacts who can act as a link, i.e. names with direct phone numbers or e-mail addresses. These 'contact points' will help with the next task which is to ascertain which people, or gatekeepers , and channels need to be gone through in order to gain permission and consent. This involves understanding the structure and hierarchies in an organization. Insider knowledge needs to be tapped in order to follow the correct protocol and to not leave anyone out (especially those who might take offence).

3. This links to the next task which is to make clear to all concerned the extent of the study, the demands it will make, the reasons for doing it and the likely forms of publication. This will involve telling people exactly what will be expected of them (e.g. a 30- minute interview, being a member of a focus group, and filling in a two-page questionnaire) and telling them what you plan to do with it. This applies to informants of any age or status.

4. Fourthly, the researcher needs to become aware, as early as possible, of any sensitive or controversial issues which might arise - for an individual or for an organization. As mentioned in the first point, subjects of research may feel threatened or intimidated by a newcomer -a researcher or even by an insider adopting the role of researcher.

These are just a few of the points needing consideration in gaining access. The important general point is that it would be foolish to pretend that a project could be designed and planned, or sampling established, before access had actually been arranged; hence the portrayal of research shown earlier and the unrealistic idea that a research project proceeds along a straightforward linear pathway.

6. Going public: disseminating for different audiences

A piece of research cannot be considered to be complete unless it "enters the public domain" - its findings are disseminated. This is true of both ‘applied' research (which tries to ‘make a difference' by influencing practice or policy in an area) and more ‘theoretical' research (which aims to influence thinking and theorising in a field of study). 

Ways Of Reporting And Disseminating
There is a range of possibilities for disseminating and presenting research findings: some will involve spoken presentations (typically using visual aids); some will involve writing for conference proceedings, journals or book publishers; or a combination of spoken and written forms.
Source: Wellington, J and Szczerbinski, M ( 2008)

Table 4Main types of research publication
Articles ("papers") in peer-reviewed journals With respect to their content, papers can be divided into:
Primary literature: reports of new, previously unpublished data, 
Narrative literature reviews: critical summaries of a current state of knowledge on a given topic
Quantitative literature reviews/metaanalyses: ‘pulling together' and statistical reanalysis of results of all (quantitative) studies on one particular topic, in order to draw a general conclusion about their outcome.
Books  With respect to authorship and editorial process, academic books may be divided into:
Scholarly monographs: books addressing a single topic, written by one or few authors
Edited books: books where each chapter is written by different authors. Chapters are revised by editors, who take responsibility for overall consistency, coherence and cohesion. They usually (though not exclusively) address a single topic. 
With respect to content, academic books are typically narrative literature reviews, though they may also report primary findings or quantitative literature review/metanalyses.
Presentations at conferences (organised by learned societies or professional organisations)

Different forms are possible:
Oral presentations: oral accounts (typically using visual aids) of research in front of the peer audience
Posters: a ‘single page' summary of research, presented during "poster sessions" that are part of most conferences
Conference proceedings: Printed summaries of research presented during conferences. May be very brief (abstracts) or more substantial (resembling research papers)

Commissioned reports The commissioning body may be the government, a charity, a quango, a commercial company, etc.
Other  Technical reports:Typically prepared for internal distribution (e.g. for a sponsor of the research project)
Working papers: Reports of work in progress, ahead of more formal peer-reviewed publication. They are often made available online
Blogs: Blogs as the means of disseminating research findings have been increasingly adopted e.g. by some scientists researching online communities.

Different types of research publication serve different purposes. For example, conference presentations (oral or posters) are usually used to present ‘work in progress'. This gives the researcher not only the opportunity to disseminate the ‘fresh' findings, but also to get the valuable feedback from the audience (regarding data analysis, interpretation and presentation), which they can take on board in preparing a subsequent written publication (a peer-reviewed paper or a book).

Reaching Different Audiences
The main purpose of writing up social research is to communicate with other people (although this is not always obvious when reading some of it!). It is hardly worth doing research if it is not disseminated. Communication can, and should, take place with a number of different audiences in mind: one's peers and fellow- researchers, practitioners, policy-makers, curriculum planners and developers, teachers or lecturers, parents or the general public. Once again the ground rule is horses for courses: 'Different purposes and different audiences require different styles of writing' (Woods, 1999, p. 48). In addition, different aims and audiences require different lengths of writing.

Writing For Journals: Tips For Improving Your Acceptance Chances
I suggest the following tips for submitting a piece of writing to an academic journal:

  1. Decide on the journal you want to submit to before you write anything. The choice should depend on your intended audience (e.g. fellow researchers? Practitioners? policy makers?) and, naturally, the perceived quality of your work (better you consider your work to be, higher impact factor journal you. Remember that submitting the same paper to several journals at the same time is considered unethical. To make the choice wisely, leaf through a good number of back issues of potential journals (identifying recurring topics, debates or themes), and talk to more experienced colleagues.
  2. Read and the journal's guidelines to authors (regarding manuscript structure, format, referencing, word length, a number of copies to be submitted, etc.) and follow them. This is essential: you don't want to write a 10000 words long paper, only to discover that a word limit in your chosen journal is 6000 words!
  3. Look for key traits / characteristics in your chosen journal and attempt to model them in your writing.
  4. Prepare your manuscript meticulously. Remember it must be a finished product, not a draft. Ask a critical friend to proofread it and comment on it before sending it off. 
  5. Accompany your submission with an appropriate letter to the editor.
  6. Then, prepare to wait (don't hold your breath).

The next step is to deal with referees' comments. The outcomes of the refereeing process will be one of the following:

  • Accept without revisions. In modern academic publishing this rarely happens. Action: celebrate.
  • Accept with minor revisions. Action: celebrate and just do them
  • Accept with major revisions. Action: strong drink, then take each point one by one
  • Reject. Action: it depends. You may change the manuscript (taking reviewers' comments on board) and then send it to another journal ASAP. Or you may agree with the reviewers that your study is not good enough - and start working on a better one. Most reviewers try to be constructive, so a letter of rejection typically contains plenty of useful feedback that allows you to make your work better.

Finally, be patient: the whole process takes time. A two year interval from the initial submission to the actual publication is not uncommon. Journals often provide information regarding average ‘turnover' time - check it.

Good luck.
Jerry Wellington

7. FURTHER READING

Becker, H (1986) Writing for social scientists: how to start and finish your thesis, book or article. Chicago: University of Chicago Press.( my favourite book on writing and how to do it - a classic)

Bell, Judith (first published in 1993) Doing Your Research Project: A Guide for First-time Researchers in Education and Social Science. Buckingham: Open University Press. (A concise, valuable guide for new researchers, now in its 5TH edition.)

Cohen, L. and Manion, L. ( first out in 1994) Research Methods in Education (now in later editions). London: Routledge. ( definitely not bedtime reading, but everything you need to know or to look up is there)

Denzin, N. (1970) The Research Act. Chicago: Aldine.

Denzin, N. and Lincoln, Y. (1994) Handbook of Qualitative Research. London: Sage ( the handbook on everything to do with real life research)

Two classics from Sir Peter Medawar:
Medawar, P. (1963) 'Is the scientific paper a fraud?'. The Listener, September. 
Medawar, P. (1979) Advice to a Young Scientist. New York: Harper and Row.

Robson, C. (first out in 1993) Real World Research: A Resource for Social Scientists and Practitioner-Researchers. Oxford: Basil Blackwell. (A 510-page resource covering almost everything from design to data collection and 'making an impact'.)

Wellington, J (2003) Getting Published, London: Routledge Falmer

Wellington, J and Szczerbinski, M ( 2008) Research Methods for the Social Sciences, London: Continuum (a full version of many of the points made here, with descriptions of all the main research methods)

Woods, P. (1999) Successful Writing for Qualitative Researchers. London: Routledge ( new editions now available - a superb discussion of writing for different audiences)

Published: 19 Jan 2009