The Research Process – Organising Your Research
- The research process is a way of organising a research project.
- The study aims and objectives guide the study.
- The literature review provides the context for the study and determines the need for it.
- All stages of the process should be clearly described with appropriate rationale.
- Issues of ethics are fundamental to the research process.
- Dissemination and implementation complete the research process and start the research cycle with new questions.
This book is about all aspects of the research process as it relates to healthcare research. This chapter offers an overview of what is involved in that process and can be read in conjunction with other, more specific, chapters. Those chapters will, for example, discuss the differences between qualitative and quantitative approaches to research. Whatever approach is chosen, the same broad process is undertaken and it is this broad process that is under discussion here. Many writers have described the research process, often providing subtly different components. We do not subscribe to any one description of the process rather than another. Instead, we offer below a series of suggestions as to how a research project might be organised, based on the following view of the research cycle: research questions, literature review, selection of sample, selection of measures, data collection, data analysis, report writing and dissemination. Indeed, much of this book is constructed with this cycle in mind.
In a way, research is always a cyclical process. Once findings from a study have been established, they nearly always bring us back to new questions that need answering. It is not a good idea to view research as establishing ‘the truth’. It is merely offering the best view we have on a given topic at a particular time in history.
A research project needs a descriptive title. When other researchers are looking to see what research has been done before, it is the title of the project that will alert them, either to want to read the report of that project or to pass over it. The title, then, should be concise and, as clearly as possible, describe the nature and content of the research project. Thus, a title such as:
‘A qualitative study of student midwives’ views of reflective practice’
is preferable to:
‘The mirror of the mind: students and reflective practice’
The first title states clearly both the nature and the content of the study, whilst the second one is less clear. It is probably best to avoid what might be called ‘romantic’ titles, or titles that use an analogy (such as ‘The mirror of the mind…’), as these may be confusing or may miss the point. The title, wherever possible, should indicate both the research approach used and the topic of the study.
The title should, until the research is finished, be considered a ‘work in progress’. It may be found that the title changes as the study is completed. Considerable time should be given to ensuring that the title is as accurate and as concise as possible. It is usually intimately related to the research questions.
One of the final tasks in designing a research project is to write an abstract for the study. It is discussed here because it is usually one of the first pages in a research report or proposal for funding. An abstract is normally a one-page description of all aspects of the study. Thus, the researcher will report, briefly, the aims of the study, the sample, the data collection and analysis methods used and the findings. In writing proposals for funding, it is usually a requirement that the abstract is written in lay terms. This in itself can be helpful in guiding the researcher towards highly focused research aims and objectives.
Although the abstract is one of the first pages of a research proposal, it is often best written when the proposal is complete. In this way, the abstract more easily becomes a concentrated summary of what the project is about.
Aims and objectives
While the heading of this section is ‘aims and objectives’, in practice, various terms can be used here. A research project has to be clearly focused, because it is this focus which provides the direction for the research process as the project takes shape. The researcher must know, in advance, what he or she is trying to find out. Most beginning researchers attempt to do too much and the process of narrowing down the focus of interest can be a lengthy and painful one. A useful means of knowing that you have finally identified the exact focus for a study is when you can state it in one sentence. For example, the beginning researcher might say, when asked what his or her project is about: ‘I want to look at the things students think about reflection and reflective practice. I want to know what their problems with it are and how it affects their clinical practice. I also want to hear about their teachers’ experiences of teaching reflective practice and see if they feel that it makes a difference to clinical practice’. After discussion with a research supervisor and a reading of the literature, the answer to the question: ‘What is your project about?’ might be: ‘It is a descriptive study of student midwives’ attitudes towards reflective practice’.
Having identified exactly what the project is about, the researcher will write either a set of aims and objectives for the project or a number of research questions. Aims are broad, general statements about what the research will be about, while objectives are smaller statements about the actions to be taken. Thus, an aim might be to ‘Identify the attitudes of student midwives towards reflective practice’, while two objectives might be to ‘(1) identify an appropriate, convenience sample of student midwives; (2) use a validated instrument to identify student midwives’ attitudes towards reflective practice’.
A research question is not dissimilar to a research aim but the focus of the research is demonstrated in a question. For example, ‘What are some student midwives’ attitudes towards reflective practice?’
In the case of quantitative studies, a hypothesis will be stated – a very specific statement that is to be tested by the research and found to be either ‘supported’ or ‘not supported’ at the end of the study. Occasionally, the statistical term, null hypothesis, is used to state that something is not likely to be the case at the end of the study (e.g., from the pharmacy world, drug ‘A’ will not be demonstrated as being more effective in the treatment of hypotension than drug ‘B’). Hypothesis testing is discussed in more detail in Chapter 14.
A review of the literature
As we shall see, all researchers need to do a thorough search of the literature that has preceded the study. This search of the literature is to find out what previous research has been done, what others have written about on the topic, and to place the present study in context. Research projects do not simply arise out of the blue. They are always linked, closely, to a particular time in history, a particular set of beliefs that are current about a subject and must always be clearly linked to what is already known about the topic in question.
The development of computer software that allows huge databases of research reports and publications to be searched means that most researchers can now undertake systematic reviews of the literature. The aim of a systematic review is to attempt to identify almost every paper that relates to the current topic being researched. This is no easy task, even with computerised search engines. The researcher must identify key words in order for the engines to identify appropriate papers. Once a range of papers has been identified, the researcher then needs to filter these down to exactly the papers that are relevant to his or her study. Finally, the papers themselves have to be obtained, read, summarised and their reference formally recorded. Searching the literature is covered in detail in Chapter 4.
In reviewing the literature, in this way, the researcher is also required to read critically. What are the limitations of the study report that is being read? What, if any, mistakes were made in the research process? What else needs to be done? Asking these questions can help further focus on the study being undertaken by the researcher who is doing the review. The critical evaluation of research papers is described near the end of this book in Chapter 22.
It is not always necessary for the researcher to design an entirely new project. There is considerable value in undertaking a replication study. In such a study, the methods used in a previous study are used again. If the replication is very close to the original piece of research, much can be learned about whether or not the original work was sound and whether or not the findings from the original are still valid or current.
There is a certain hierarchy in the value of papers that are reviewed. The soundest evidence comes from original report papers. After that, in degrees of value, come theoretical papers, which summarise the research and develop theoretical perspectives. A long way behind these two come ‘opinion papers’. These are usually short, one-page papers that appear in many healthcare journals and magazines. They may help to illustrate some of the current debates about certain issues but, in the end, usually only reflect the writer’s own views. As a result, they are not particularly useful in generating information for a review.
The most thorough researcher also considers the ‘grey literature’. This is that literature which is not formally published in journals or books. Examples of the grey literature include documents such as college reports, curriculum documents and handouts. Again, their status may not be very high in the evidence hierarchy but they can help to identify current trends in a particular discipline.
Some reviewers of the literature attempt a type of content analysis of that literature. This is to say that they count the occurrence of certain issues that arise across a spread of papers and then tabulate their findings. Such counting needs to be done carefully and is probably only really valid when a researcher has managed to identify all the literature in a given topic.
A systematic review of the literature means being systematic in other ways too. The researcher must set up appropriate and careful systems for managing the literature that they find. He or she must also be careful to record the exact reference to a given piece of work. Such references (e.g. author, date, title of book or paper, publisher of book or paper, page numbers [in the case of a journal paper]) can be stored either in a dedicated, computerised, reference database (such as End Note or Reference Manager) or can be stored on cards or in a notebook. What is essential is that the researcher is very clear about how to properly cite references, using either the Harvard or Vancouver systems of citation, and is also very careful to record all the details of a reference for further use in writing the research proposal or report. This point cannot be repeated too much: you MUST know how to reference properly. In our experience, many students at various levels in their academic careers have difficulty with the particular aspect of scholarship. It is worth learning to reference very early on in that career.
A sample is a slice of any given population. For example, we might consider ‘the population’ to be those who live in London. A ‘sample’ would be a selected group of those people.
In quantitative research, there are careful means for selecting a representative sample: that is to say, a sample that can, statistically, represent the larger population. The findings from such a sample can sometimes be generalised out to that larger population. There are computer programs for generating a power calculation which allows the researcher, in quantitative research, to identify the minimum number required in a sample for representativeness, given the size of the total population. Careful selection of a sample, in quantitative research, is essential for identifying the degree to which findings can or cannot be generalised out to the larger population. These issues are discussed further in Chapters 6 and 13
Generally, in qualitative research, three types of sample are routinely used. Researchers may use a convenience sample, a purposive sample or a snowball sample. A convenience sample is one that is made up of people who happen to be available to take part in the research. For example, in a midwifery department, a particular class or group of classes may constitute a convenience sample. A purposive sample is one made up of people who are reasonably likely to be able to offer information or views on a given topic. For example, it would be pointless asking student midwives who had no experience or knowledge of reflective practice for their views of reflective practice, but much more helpful to ask students who did have such experience or knowledge. We could sample specifically for students who had this experience or knowledge, and such a group would be a purposive sample. Surprisingly, if the whole population of students had such experience, no purposive selection of them according to that criterion would be possible. It would not be appropriate to refer to students from this group who eventually participated in the study as a purposive sample simply because they possessed characteristics we would have wished to select for, because the whole population have this characteristic.
A snowball sample is rather different and is usually used, in qualitative research, when interviews are being conducted. Here, the researcher asks the interviewee, at the end of the interview, to recommend someone else that he or she might interview. Thus, in a study of healthcare education, the interviewer may interview a course leader about the economics of such education. After the interview, the researcher may ask the course leader to recommend another person for interview and the interviewee may suggest the local finance officer. Thus, the pattern of sampling ‘snowballs’. Clearly, the snowball sample is also a purposive one. Issues around sampling in qualitative and quantitative research are covered in detail in Chapters 7 and 13, respectively.
Measures and materials
Measurement is usually a feature of quantitative research and involves the selection of measures and scales appropriate to accurate examination of the issues to be explored during the study. In qualitative research, although formal measurement is less frequent, the researcher will want to create detailed interview schedules or guidelines for the completion of field notes. There may also be stimulus materials to be shown to participants. For example, focus group interviews sometimes ask participants to comment on such things as the appropriateness of patient information booklets. In all cases, a clear rationale for the use of particular measures and materials should be described.
The most common forms of data, from social science and healthcare research, arise in one of two forms: numbers or words. Occasionally, photographs and other forms of data are also collected and the collection of these seems likely to increase in the future as more sophisticated and computerised methods of data storage and analysis become available.
In quantitative research, the data are mostly likely to be numerical. These figures can come from various sources: they may be scores on a questionnaire, arise out of direct counting of things or may come out of other data sets. In qualitative research, data (in the form of words) usually come from transcripts of interviews or from the researchers’ field notes.
A short list of data collection methods to be found in social science and healthcare research might include, at least, the following:
Data collection approaches
- Direct, non-participative, observation by the researcher
- Participative observation (observation made while the researcher is working alongside other people)
- Questionnaires and scales
- Field notes
- Focus groups (or group interviews)
- Historical review of documents
- Meta-analysis of findings from other studies