The Research Proposal:Research Design

Introduction


Crookes and Davis describe the research design as ‘the overall plan of how the researcher intends to implement the project in practice’ (2004: 73). In the same vein, Parahoo (1997) talks about research design as a plan within which the researcher details the how, the when and the where of the way data will be collected and analysed during the study.


This chapter introduces, and discusses, the elements that come together to produce a well-designed research proposal, followed by the research study. It is linked to the study design section of the web program.


A proposed research study must be structured in such a way that the results and conclusions are reliable and can be seen to be rigorous. In a well-designed study, the various elements follow one another seamlessly, but the starting point for developing any research design is the selection of a topic and a paradigm. A paradigm is a method, or pattern, by which the topic of the research is explored (this is discussed later in this chapter). Other elements in a research design include:



  • the use of previous literature on the topic (chapter 4);
  • the purpose statement, which will include questions, objectives and/or hypotheses (chapter 3);
  • the ethical issues (chapter 6);
  • the proposed participants in the research – the sample (chapter 7);
  • the methods of data collection (chapter 8);
  • how the data are going to be analysed (chapter 9);
  • how the results will be disseminated (chapter 10).

This chapter encompasses and discusses all the elements that make up a good research design.


The focus of a research study


‘The heart of every research project is the problem. It is paramount to the success of the research effort. To see the problem with unwavering clarity and to state it in precise and unmistakable terms is the first requirement in the research process’ (Leedy & Ormrod 2005: 43).


Before you can start doing any research, you need to select a problem to investigate: this is the focus of your research study – the central concept that you will be examining (Creswell 1994). There is frequently some confusion as to what is a suitable focus for a research study. It is often thought of as solving a problem. However, this is by no means the case. The focus of your research study may well be solving an existing problem, but the focus could also be identifying whether or not there is a problem or, if you know that a problem exists, what it actually comprises.


How do you select a problem to be the focus of your study? This involves asking the right questions in the right way and/or communicating with others who are familiar with the area that you wish to explore. It is highly unlikely that the problem you select as a potential focus for your research will be totally unique. Usually, some work will have been carried out on the problem you are interested in, although the aspect of the problem that has been previously investigated may be different. For example, one of the authors undertook a research study into the early hospitalisation of children following bone marrow transplantation for severe combined immune deficiency and the effect of that and of isolation on them and their families. These topics had already been explored in relation to cancer, for example, but this was the first time anyone had looked at these concepts in relation to severe combined immune deficiency. As a result discoveries were made, which were also found to relate the children with cancer and others in a similar situation.


Returning to your need to select a focus for your investigation, this might come about due to:



  • something that you have read in a professional journal or book;
  • something that you have seen in your professional practice, wherever that might be sited;
  • something that somebody has mentioned as being of concern and/or interest – this could be another professional (healthcare or other), a patient/client, a family member of a patient/client or someone totally unconnected with your work (e.g. a friend or even a chance meeting with a stranger);
  • your general interest in the area;
  • an inspirational thought – never discount inspiration.

Once you have selected your general focus, you have to narrow it down to something much more specific so that you arrive at a particular problem that you wish to investigate (this is the same process that you came across in chapter 3). To help you narrow down the parameters (limits or boundaries) of your focus, you need to specify what that problem is. Research problems are usually stated in the form of a question, although they can also be stated in the form of a hypothesis (a proposition that is assumed to be fact for the sake of an argument), for example:


Questions



  • Why do some patients improve on a particular drug regime, whilst others do not?
  • Do families who have a child who has undergone a bone marrow transplant have any psychosocial problems?
  • How can we prevent cross-infection in a hospital ward?
  • What are the needs of adults dying of cancer?
  • What is the safest way of x-raying an infant?
  • How best can we improve the quality of life of someone in a wheelchair?
  • What is the safest way to transport a neonate with meningitis to a specialist hospital?

Hypothesis



  • Smoking causes lung cancer.

Producing a hypothesis/research question


Above, we have considered problems which provide a possible focus for several different research studies. However, these are still quite broad in scope. In order to design a research study to look at a particular focus, we need to refine our question. This we can do by means of a hypothesis or a specific research question.


A hypothesis is basically a question that is turned on its head so that it becomes a statement (see also chapter 3). The purpose of a hypothesis is to stimulate experimental research so that you can either prove or disprove the statement that makes up the hypothesis. Thus, if we look at the first four research questions listed above, we can develop the following four hypotheses from each of the questions that describe the particular problem that we wish to investigate. The hypotheses for each of these four problems can now be written as:



  • Patients with liver disease who are in hospital will improve when prescribed a particular drug, whilst those patients who are at home and are prescribed the same drug do not improve.
  • Families with a child who has undergone a bone marrow transplant suffer from severe psychosocial problems.
  • Cross-infection in a hospital ward can be prevented by rigorous hand washing.
  • Adults who are terminally ill with cancer need the support of specially trained health professionals.

The researcher then has to come up with an experiment to test these hypotheses and either prove or disprove them.


You can now see that each of these problems has a much narrower focus. This makes it easier for us to design a research study to tackle that particular problem.



To Do

Take the next three research questions in the list of questions above and develop hypotheses (the plural of hypothesis) from them.

Then take the hypotheses in the list above and make a research question out of them.

Null hypothesis


Some researchers are concerned that if we use a hypothesis, we will be predisposed to be favourable to it and that might bias our research study. Consequently, some prefer to use a null hypothesis. In effect, this is a negative statement, such as:



  • Smoking does not cause cancer.

To ensure that there is no positive bias in the research study. It is thought that by introducing a possible negative bias by using a null hypothesis, then if the research does produce a positive result, it is seen as stronger, more rigorous and valid research.



To Do

Now take all the hypotheses that you have come across so far in this chapter and turn them into null hypotheses.

Research question


Not all research is experimental. Non-experimental research does not require a hypothesis, because in non-experimental research you are not trying to prove or disprove something; rather, you are exploring a problem. This is usually the case in qualitative research as opposed to experimental quantitative research that does require a hypothesis, but it has to be noted that a lot of quantitative research is also non-experimental – for example, research within the category of descriptive quantitative research.


With non-experimental research, whether quantitative or qualitative, we do not use a hypothesis because we are not trying to test something but are investigating a problem. When working within non-experimental research paradigms we often use research questions rather than hypotheses to underpin the research. Research questions are usually more general than hypotheses and so allow for modification and expansion of the elements that make up the research question as the research study develops.


To return to our first four research problems, possible research questions that can be developed from these problems include the following:



  • Why do patients in hospital improve when taking drug A, whilst patients at home do not?
  • What are the psychosocial results for a family whose child has undergone a bone marrow transplant?
  • Will hand washing on its own prevent cross-infection in a hospital ward?
  • What do adults terminally ill with cancer perceive the needs to be that can be met by the palliative care services?

In each case, we have narrowed down the initial problem to a question that can provide a specific focus for a research study, and by so doing we have helped to clarify not only what we will be investigating, but how we will be investigating it.



To Do

Now take the remaining research questions from the list and try to clarify and narrow their focus so that you begin to have questions that can be more easily investigated.

So far in this chapter we have discussed the focus of the research study because this is probably the single most important aspect of research design and of research itself. A poor hypothesis/research question can lead to a poor research design and consequently a poor research study, whilst a good hypothesis/research question can lead to good research design and a good research study because all the other elements of the research design and study are dictated by, and come from, the initial hypothesis/research question. As Leedy & Ormrod (2005: 43) state:


‘the problem or question is the axis around which the whole research effort revolves. The statement of the problem must first be expressed with the utmost precision; it should then be divided into more manageable subproblems. Such an approach clarifies the goals and directions of the entire research effort.’


Research paradigms


‘The design of a study begins with the selection of the topic and paradigm. Paradigms in the human and social sciences help us to understand phenomena’ (Creswell 1994: 1).


In research, paradigms are composed of various sets of beliefs and practices that are shared by researchers. These paradigms provide frameworks and processes through which research and investigation can be carried out (Weaver & Olson 2006). Within each paradigm you will find a combination of vocabularies, theories, principles, presuppositions and values related to a particular research inquiry (Bunkerset al. 1996, Weaver & Olson 2006) . Weaver & Olson ( 2006) simplify matters by defining paradigms as being sets of philosophical underpinnings from which specific research approaches flow – for example, in nursing we tend to be concerned with quantitative (or positivist) and qualitative (or naturalistic) methods and approaches to research inquiries. (Research philosophies are explored more fully in chapter 2 so we will not consider them in this chapter.)


However, within these two paradigms, there is a ‘rich array of approaches and methods available’ (Polit & Hungler 1999. 14), and it is up to the researcher to decide which paradigm, and which approach and method, to use in order to explore the problem. This is where the importance of the research question/hypothesis comes in. The paradigm, including the approach and methodology, is dictated by the initial research question/hypothesis. If the research question/hypothesis does not properly reflect the problem/inquiry, then the methodology could be inadequate or even completely wrong for the research. As a result, either the research will fail or it will be very poor research indeed. In other words, it is important that the research question/hypothesis is first framed and agreed by you, the researcher (and your research team if you are part of one), and then the most appropriate method is chosen by which the question can be answered or the hypothesis proven or not proven. The researcher should ‘choose the correct tool to do the job rather than just using the tool and then asking “now what job can I do?” ’ (Crookes & Davies 2004: 74). In other words, select your topic and then determine your hypothesis or research question. Only then should you decide on the actual research paradigm that will allow you to prepare, and successfully complete, your research study.


We shall explore below the research design in terms of the two paradigms mentioned above, but you need to be aware that there are certain aspects of any research study that are common to all research. These include:



  • ethical issues;
  • the rigour of the research;
  • ensuring that you have the correct sample for your particular research study;
  • writing the research report;
  • the dissemination of results from your research.

Samples


As you prepare your research proposal, you will need to consider who is going to participate in the study. This is called the sample. A sample is a group of people who have been selected as representatives of a population as a whole. A ‘population’ (in terms of a research study) consists of the people who may be affected by the phenomenon(a) you are investigating. The sample must be large enough to allow you to investigate fully the phenomenon(a) using the research methodology that you think will best answer the research question/problem/ hypothesis.


Different types of sample can be used, according to your research methodology, research question/problems and the aims of your research. These different types of research samples include:



  • random samples;
  • theoretical samples;
  • purposive samples;
  • convenience samples;
  • snowball samples;
  • volunteer samples.

These different types, and how we decide which one to use, are discussed in chapter 7 as well as in the web program.


Next we consider the research design in terms of the two major paradigms used in nursing research – quantitative and qualitative paradigms.


Quantitative research


Quantitative research is the principal method we use when considering scientific investigation in nursing (Burns & Grove 2005). According to Porter & Carter (2000: 19), quantitative research is ‘a formal, objective, systematic process for obtaining quantifiable information about the world, presented in numerical form and analysed through the use of statistics’.


We use a quantitative research paradigm to test relationships between phenomena and it is particularly brought into use to examine cause-and-effect relationships (e.g. smoking and cancer, or obesity and type-2 diabetes).


Norbeck (1987) notes that there is a belief among researchers that quantitative research provides a much sounder and more reliable knowledge base for the purposes of guiding nursing practice than does qualitative research because quantitative research is thought to produce ‘hard’ science involving:



  • rigour;
  • objectivity;
  • control.

Unlike researchers who use qualitative methodologies, quantitative researchers believe that truth is absolute and that there is a single reality that can be defined by careful measurement. Therefore, in order to find this ‘objective truth’, the researcher must be completely objective. In other words, the researcher’s values, feelings and personal perceptions are not allowed to be brought into the measurement of ‘reality’.


Consequently, quantitative researchers hold that all human behaviour is:



  • objective;
  • purposeful;
  • measurable.

Within the quantitative research paradigm, there are several methodologies that the researcher can use, namely:



  • experimental research;
  • quasi-experimental research;
  • non-experimental research (including descriptive research and correlational research);
  • survey research;
  • evaluation research.

Experimental research


Experimental research is the type of research that most people think of whenever quantitative research is mentioned (Polit & Hungler 1999). It is the most appropriate and powerful quantitative method for testing cause-and-effect relationships because of its rigorous control of variables and it is considered the gold standard for demonstrating something in a rigorously scientific manner.


Experimental research involves observation, usually to test cause-and-effect relationships between variables under conditions which, as far as possible, are controlled by the researcher. It involves not only scientific observation, but also the manipulation and control of phenomena. According to Porter & Carter (2000), a classic experimental research design involves the selection of subjects (a sample) who are randomly allocated (see below) to either an experimental group, who are exposed to the variable that is the purpose of the study, or a control group, who do not come into contact with the variable, and hence act, as the name implies, as a control within the experiment.


You will come across these types of research studies particularly in relation to trials for new drugs.


Double-blind tests are typically used in drug trials. ‘Blinding’ means that either the patient or those providing care to the patient do not know whether the patient is in the experimental group or the control group. ‘Double-blinding’ means that neither the patient nor the caregivers are aware of the group assignment of the patient. The purpose of double-blinding is to ensure that the risk of bias (particularly subconscious bias) from either the patients or the caregivers is avoided.


Randomisation


Note that in quantitative research we think of randomisation when we are looking at the make-up of our sample. Randomisation means that we choose the subjects of the sample randomly. As a result, every individual in a population has an equal chance of being selected for the sample. One of the reasons why we would opt for randomisation (random sampling) is that it helps to eliminate any bias that might distort our findings. We have already mentioned that, in experimental studies, we may use an experimental group and a control group (e.g. when testing a new drug). In this case, subjects would be randomly selected for the study, and then further selection would take place when the subjects selected for the study are randomly assigned to the experimental group or the control group.


Quasi-experimental research


Quasi-experimental research may look very much like true experimental research in that it involves the manipulation of an independent variable, but it is not the same, because quasi-experimental research studies lack one or both of the essential experimental research properties of randomisation and a control group.


The main drawback with quasi-experimental research is that, compared to experimental research, it has a weakness in that it is not possible to deliver cause-and-effect results (Polit & Hungler 1999). In other words, we cannot infer from quasi-experimental research that, for example, doing one thing causes a particular phenomenon.


Descriptive research


This type of research, as its name suggests, describes what exists, but it may also uncover new facts and meanings that were previously not known or apparent.


The purpose of descriptive research is to observe, describe or document aspects of a situation as it naturally occurs (Polit & Hungler 1999). This involves the collection of data that will provide an account or description of individuals, groups or situations. Examples of the types of instruments that we use to obtain data in descriptive studies include questionnaires, interviews (closed questions) and observation (e.g. checklists). There is no experimental manipulation, or indeed any random selection to groups, as there is in experimental research.


The characteristics of individuals and groups such as nurses, patients and families may be the focus of descriptive research. It can provide a knowledge base which can act as a springboard for other types of quantitative research methods.


Correlational research


Correlational research within a quantitative paradigm aims to investigate systematically and explain the nature of the relationship between variables in the real world. Often the quantifiable data (data that we can count) from descriptive studies are analysed in this way. Correlation means to co-relate; in other words, correlational research studies go beyond simply describing what exists and are concerned with systematically investigating relationships between two or more variables of interest – they co-relate (Porter & Carter 2000). Such studies only describe and attempt to explain the nature of relationships that exist; they do not examine causality (i.e. whether one variable causes the other), as is the case in experimental research.


Survey research


According to Polit & Hungler (1999), a survey is used to obtain information from groups (populations). The information so obtained may be concerned with the prevalence, distribution and/or interrelationships between variables within these groups. For example, the UK census comes under the heading of ‘survey research’. In this type of research, data collection tools include:



  • personal interviews;
  • telephone interviews;
  • questionnaires.

It should be pointed out, however, that, depending on the problem under investigation, this type of research may be better explored by means of a qualitative paradigm, although there is still a place for quantitative research methodology in survey research. For example, the UK census return report is based mainly on the quantitative paradigm, but more recently it has included elements from the qualitative paradigm, such as detailed, face-to-face surveys that also take place as part of the census data collection.


Evaluation research


This is an ‘applied form of research that involves finding out how well a programme, practice, procedure or policy is working’ (Polit & Hungler 1999: 201). The aim is to assess/evaluate the success of a particular practice or policy (see also chapter 1).


Examples of this type of research can be seen in various types of analysis/evaluation, including:



  • process/implementation analysis;
  • outcome analysis;
  • impact analysis;
  • cost–benefit analysis.

As with the survey research methods, this type of research may best be carried out as a qualitative piece of research, depending on the original research question.



Scenario

You are a researcher who has been commissioned to investigate the incidence of leg ulcers in the under 50 year olds in the catchment area of a Primary Care Trust situated within a city. Because you are going to be investigating the incidence of something – not the causes or experiences – you will base your study within a quantitative paradigm. The question that you have to answer is: which type of quantitative research will you use for your study, and what will be your rationale for choosing this type of research?

Possible suggestions are given on page 107.

Data collection


There is a variety of techniques that can be used to collect data in a quantitative research study (see chapter 8). However, all of them are geared towards numerical collection (i.e. the data are in the form of numbers rather than words or any other media).


Numerical data can be collected by means of:



  • observation;
  • interviews;
  • questionnaires;
  • scales;
  • physiological measurements.

In quantitative research, the data are collected and recorded systematically, and are then organised so that they can be entered into a computer database (Burns & Grove 2005). However, some of us are old enough to remember doing this using only pen and paper and our brains (later, we were able to turn to the help of very basic calculators). Computers and computer databases are comparative newcomers in the recording and analysis of numerical data, but they are now so sophisticated that if you enter the correct data, the computer will analyse it in seconds (and even represent the findings graphically).


Piloting and pilot s tudies


If you are planning to undertake quantitative (or even qualitative) research, you really need to consider undertaking a pilot study before attempting the main study. A pilot study is usually preliminary to a main study and as such should follow the design of the main study as closely as possible – or rather, the main study (amended as necessary) should fully follow the pilot study. In addition, the sample should consist of subjects who resemble, as closely as possible, those who will be used in the main study. Another criterion that a pilot study should meet is the extent to which the areas covered or the questions asked by interview or questionnaire measure what they are supposed to measure.


It is not unusual to exclude participants in a pilot study from the main study as they will have been exposed to the intervention or interview protocol and respond differently from those who have not experienced the procedure (Peat et al. 2002). However, in some cases it may not be possible to exclude the pilot participants as the main sample may be too small, particularly where the research has to reach groups such as prisoners and homeless individuals. Holloway (2005) believes that pilot data are of less concern in qualitative research than in quantitative research because in the former, analysis is often ongoing and so including participants from different stages of the study is less crucial, whereas in the latter, the data could be flawed or inaccurate because all data must be rigorously collected from people with the same experiences of the phenomenon, with no extra variances (Peat et al., 2002). The important point is to justify whichever method you choose to adopt.


The reason for running a pilot study is that it serves as a testing ground for your data collection instruments, sample and method of analysis so that if mistakes are made in any part of your research design, it will not be unduly costly in terms of money or time, because you can sort out the problems before you move on to the main study. The intention to undertake a pilot study should be stated in your research proposal because both the ethics and the research and development committees will expect this.


Qualitative research


The second of the two research paradigms that we are going to discuss is qualitative research. Qualitative research is an umbrella term that covers several styles of psychosocial research and draws on a variety of disciplines, among them sociology, anthropology and psychology. However, there are two common elements to these approaches that begin to give some sense to the term ‘qualitative research’. These are:



  • A concern with meanings and the way people understand things. Human activity is seen as a product of symbols and meanings that are used by members of the social group to make sense of things. One of the symbols and meanings that can be analysed is ‘text’, i.e. the written or spoken word. Other symbols that we can use in qualitative research include drawings and play.
  • A concern with patterns of behaviour. The focus in terms of patterns of behaviour is on regularities and irregularities in the activities of a social group, such as rituals, traditions and relationships, and the way that these are expressed.

Qualitative data


Qualitative data, whether they are depicted in words, images or any other medium (including, occasionally, numbers), are the product of a process of interpretation. The data only become data when they are used as such. Data do not exist in their own right waiting to be discovered, but are produced by the way they are interpreted and used by researchers. Qualitative research can be part of an information-gathering exercise and useful in its own right, or it can be used as the basis for generating theories. In neither case, however, are its descriptions ever ‘pure’ ; they are always the outcome of researchers’ interpretations. Nevertheless, even quantitative research can never be ‘pure’, because the numerical data have to be interpreted (even if they have been analysed by computer), and all humans are fallible.


Qualitative research m ethodology


Introduction


We have already mentioned that qualitative research consists of a number of differently developed methods that are best suited to address questions of particular interest. There are, however, some general themes of qualitative research design that apply to all approaches and methodologies. These are:



  • Qualitative design is flexible and elastic – it is capable of adjusting to what is being learned during the collection of the data.
  • Qualitative design usually involves mixing various methodologies in terms of data collection (e.g. interviews and diaries).
  • Qualitative design is focused on understanding a phenomenon or a social setting.
  • Qualitative design tends to be holistic; it involves striving for an understanding of the ‘whole’, rather than just an understanding of a part of the phenomenon that is being studied.
  • Qualitative design requires the researcher to become intensely involved in the research study. This can often be over very long periods of time.
  • Qualitative design requires the researcher to become the research instrument.
  • Qualitative design requires the ongoing analysis of the collected data. This in turn drives both the collection of more data and the formulation of theories as the research progresses.
  • Qualitative design forces the researcher to develop a model based on the data collected as opposed to the quantitative researcher who will develop a theory (possibly as an hypothesis) and then collect the data to support/refute the hypothesis.

As mentioned above, there are many different qualitative research methodologies, but we will consider just three methods that are probably the most common qualitative research methodologies nurses use. These are known as:



  • ethnography;
  • grounded theory;
  • phenomenology.

Ethnography (see also the web program)


The qualitative research methodology known as ‘ethnography’ is built on the social science specialism known as ‘anthropology’ . Anthropology is the study of humankind, especially of its societies and customs, and the study of the structure and evolution of the human being as an animal – particularly as a social animal.


An ethnographic research study is one that studies people in their natural environment. It is a descriptive account of social life and culture within a defined social system, and is often thought of as ‘a portrait of a people’. In effect, it is concerned with a holistic view of a culture, including its shared meanings, patterns and experiences.


The aim of ethnographic research


This involves the description and the interpretation of cultural behaviour. It is culturally specific patterns of behaviour and attitudes that give people a sense of being members of a group and, under certain circumstances, set the guidelines for action.


Ethnography is a description and interpretation of a cultural or social group or system. The aim of the ethnographic researcher is to learn from (rather than study) members of a cultural group. The ethnographic researcher’s intention in relation to the members of a particular cultural group is to understand their worldview as they define it.


The ethnographic researcher examines the group’s:



  • behaviour;
  • customs;
  • way of life.

The researcher also studies:



  • the meanings of behaviour;
  • the meanings of language;
  • interactions of the culture-sharing group.

An ethnographer will try to define a particular culture by asking questions such as:



  • What does it mean to be a member of this group?
  • What makes someone an ‘insider’ and others ‘outsiders’?

The ethnographer also tries to make sense of what people are doing by asking:



  • What’s going on here?
  • How does this work?

And the ethnographic researcher hopes to be told about the way the group does things in their society/culture.


Answering these questions requires openness to learning from those who inhabit that culture and a willingness to see everything and suspend premature judgement on what should be selected as data. This quality of openness lies at the heart of ethnography, in its processes, purposes and ethics. Thus, the whole complexity of cultural categories and assumptions and the variety of relationships among them should be examined.


In ethnographic research studies, the researcher is the main instrument of data collection. The researcher must gain acceptance by the group, but at the same time it is crucial that the researcher does not become so involved in the group and individual members of the group that theoretical distance is lost. In other words, the researcher must not lose objective perspective. This is referred to as the management of marginality.


The ethnographic researcher has to fulfil certain roles ‘in the field’. These are:



  • complete observer;
  • observer as participant (observing while participating in group activities);
  • participant as observer;
  • complete participant.

It is important that the researcher continually undertakes throughout the study:



  • reciprocity (mutual actions, or ‘give-and-take’);
  • reflexivity (the ability to reflect on what they see and also their role in the group/society).

When talking about reflexivity in research, you will realise that this is nothing new for nurses and other healthcare professionals, who should be used to reflecting on their work and relationships with patients.


Mar 13, 2017 | Posted by in NURSING | Comments Off on The Research Proposal:Research Design

Full access? Get Clinical Tree

Get Clinical Tree app for offline access