SWOT Analysis and Business Plan Proposal

2089255 - Elsevier Health Sciences ©

• If you were a person from a remote part of the world with no exposure to such bank notes, you might see it as a decorative object and value it as a curio. (Qualitative.)

• If you were a person having financial difficulties, ≤$100 would be seen as a means to buy food or cover bills. (Quantitative.)

• If you were an accountant, you would focus on the number printed on the bill for making fiscal calculations. (Quantitative.)

• If you were a counterfeiter (we hope not!), you would be intensely interested in the physical properties of the object, such as its length, width, colour and texture: details required for successful forgery. (Quantitative.)

Money is very much a social construct. It is ‘real’ enough; having it or not having it can influence all aspects of life, including the level of health services a person can access. At the same time, money is an abstraction; it can take virtually any physical form (e.g. coins, notes, credit cards or a set of numbers representing your bank balance). The reality of money is constructed through the actions of a number of institutions (e.g. banks, state treasuries, etc.) which determine its value for buying goods and services. The point here is that health researchers can look at a problem from different perspectives and select the appropriate method to solve it depending on the questions they are asking and the nature of the information they wish to acquire.

The importance of integrated qualitative and quantitative research approaches is reflected in the strong emphasis that is now given to patient-/person-centred care in contemporary health systems and services. There are many policy and scholarly papers about patient-centred care; however, a good definition is offered by the Institute of Medicine: ‘Providing care that is respectful of, and responsive to, individual patient preferences, needs and values, and ensuring that patient values guide all clinical decisions’. Various international health agencies such as the NHS (United Kingdom: https://www.england.nhs.uk/integrated-care- pioneers/resources/patient-care/ ) also provide extensive discussions about the desirability of patient-centred care in contemporary health care. This contemporary emphasis requires that health research methods can accommodate the rigorous study of patient values, preferences and needs, and hence we are obligated to use both qualitative and quantitative approaches.

2089255 - Elsevier Health Sciences ©

Contrasting Qualitative and Quantitative Methods In this section, we will identify and discuss the fundamental differences between

quantitative and qualitative methods ( Table 2.1 ). One key difference is that these two methods have different historical and disciplinary backgrounds. Quantitative methods are based on the traditions of the natural sciences (physics, biology, etc.), whereas qualitative methods emerged from disciplines such as philosophy, anthropology and other social sciences (see Ch. 09 ).

TABLE 2.1 Contrast between Quantitative and Qualitative Methods

Quantitative Qualitative

Perception of subject matter

Reductionistic: identification and operational definition of specific variables

Holistic: persons in the context of their social environments

Positioning of researcher

Objective: detached observation and precise measurement of variables

Subjective: close personal interaction with participants

Database Quantitative: interrelationships among specific variables

Qualitative: descriptions of actions and related personal meanings in context

Theories Normative: general propositions explaining causal relationships among variables

Interpretive: providing insights into the nature and social contexts of personal meanings

Theory testing Controlled: empirically supporting or falsifying hypotheses deduced from theories

Consensual: matching researcher’s interpretations with those of participants and other observers

Applications Prediction and control of health-related factors in applied settings

Interacting with persons in a consensual, value- consonant fashion in health care settings

Source: Adapted from McGartland & Polgar (1994). Copyright (1994) The Australian Psychological Society Ltd. Reproduced by permission.

Perception of Subject Matter

2089255 - Elsevier Health Sciences ©

As discussed in Chapter 1 , quantitative researchers favour a ‘naive’ realistic view of knowledge. The associated research strategies focus on the discovery of novel facts about the world. Reductionism is central to conducting quantitative research. Reductionism refers to the process of reducing complex phenomena to simpler, more fundamental elements. For example, the advancement of the discipline of chemistry has been greatly enhanced by reducing the immense variety of materials to the interaction and combination of elements, represented in the periodic table. Also, the enormous array of physical and mental symptoms has been compiled into taxonomies of illnesses. Good health can be differentiated from ill health along with a number of clearly defined factors, such as blood pressure, cholesterol levels or the presence or absence of infectious agents. For example, ischaemic heart disease can be attributed to diminished blood flow in the coronary arteries. This in turn can be linked to risk factors such as high cholesterol or high blood pressure.

Understanding the risk factors and causes of heart disease enables the implementation of interventions to reduce the burden of this condition on individuals and in the general community. To identify causal associations, clearly stated hypotheses need to be tested in research studies (see Ch. 1 ). An important example of this is the relationship between factors or variables representing changes in signs and symptoms of a disease. For example, Lipitor is effective in lowering blood cholesterol, thus reducing the incidence of heart attack.

However, by reducing health problems to specific disease processes and associated signs and symptoms, quantitative researchers may lose track of the individual with the health problem. A holistic patient-centred perspective focuses on the person with the medical problem in the context of the individual’s physical and social environment and their values, needs and preferences. Qualitative researchers aim to understand how individuals experience health problems and the reasons for their actions to cope with these problems. For example, qualitative researchers investigating heart disease would consider the participants’ experiences and understanding of their health issue. Therefore the person at risk of a heart attack is no longer reduced to a failing mechanism, but rather is perceived as an active agent for achieving positive changes.

Positioning of Researcher Quantitative research is conducted to discover valid and reliable facts about the

world, as discussed in Chapter 1 . The fundamental positioning of the researcher is ‘ objective ’, aiming to perceive and record events dispassionately, without any personal bias or distortion. However, there is recognition of the possibility of error

2089255 - Elsevier Health Sciences ©

and bias entailed in collecting data. Therefore quantitative health research has invested much effort in devising and evaluating accurate measuring instruments and tests (see Ch. 14 ). Also, research designs such as ‘double-blind randomized controlled trials’ (see Ch. 7 ) have been implemented to reduce the effects of both observer and participant bias.

The problem with highly structured data collection is that participants’ true responses may be restricted or hidden because of the method employed to uncover them. Qualitative researchers find standardized instruments intrusive, as they may restrict self-expression and they may impede the understanding of the actual experiences, ideas and emotional responses of the participants. To understand personal meanings and subjective experiences, one has to become involved with the lives of the participants being studied. In this way, the researcher takes a subjective position, a degree of empathy may develop between the researcher and the participant and this may enhance consideration of the patient’s values, needs and preferences as required by patient-centred care principles. By empathy we mean the ability to ‘put ourselves in the other person’s shoes’, that is, to see things from their perspective(s).

Data and Evidence By ‘data’ we mean the results of a study which are collected and analysed to

provide the evidence required to solve a research problem. The data obtained in quantitative research consist of sets of observations and measurements, descriptions of physical and behavioural events, health and illness. These are summarized and analysed in accordance with statistical principles outlined at an introductory level in Sections V and VI.

In qualitative research, the database is essentially an analysis of the narrative (or a story, if you like) that reports what has happened to people, what they did or said in specific situations. This narrative should be adequately detailed so as to illuminate for the reader the personal meanings that the health-related events had for the informants.

The data in qualitative research are descriptive, a detailed or thorough description of what people said, their actions and activities, and interactions with other people. An important aspect of qualitative research is keeping thorough, up-to-date field notes. These should be recorded as closely as possible to the time of occurrence of the phenomena under study. The field notes should contain direct quotations from the participants and the settings in which the statements and actions were recorded. Where possible (where it is appropriate and not overly intrusive), the researcher may

2089255 - Elsevier Health Sciences ©

use audio and video recordings. This helps record interviews, and it improves accuracy in conveying what was said and done in a given setting, since it is possible to review the obtained information.

Theories Theories represent our current state of knowledge about the state of the world.

Theories are abstract, coherent explanatory systems that integrate and accommodate a broad range of research findings. Theories may be constituted of premises stated in everyday language, with particular attention paid to the appropriate use of concepts and the logical development of the premises.

Theories based on quantitative evidence integrate patterns of findings concerning the interrelationships among variables. Such theories often contain ‘models’, which may be mathematical and/or systems representations of the patterns of findings. Models of anatomic and physiologic processes, such as those of the circulatory or nervous systems, are good examples of successful quantitative models. Quantitative theories are ‘normative’ in the sense that they aim to describe and explain, as closely as possible, how things ideally work. They represent what is assumed to be stable and universally applicable. Conversely, theories integrating evidence from qualitative research do not address facts about how objects are constituted and interact, but rather are the overall interpretations of personal meanings emerging in specific social settings.

Qualitative health researchers approach theory formation by attempting to identify common ‘themes’ or categories of meanings emerging from the data. The important point here is that the theoretical categories are developed from evidence expressing personal meanings, rather than ‘facts’ derived from the statistical treatment of objective measurements of specific variables. In this way, theory is said to be ‘grounded’ in the narratives of a group of individuals.

Theory Testing Theories based on quantitative evidence are tested through precisely stated

predictions or hypotheses logically deduced from the theories (see Ch. 1 ). The accuracy of a quantitative theory is judged by the extent to which the predictions generated by the theory match the evidence produced by methodologically rigorous research. However, as discussed in Chapter 1 , theories are not easily discarded on the basis of preliminary, insufficient evidence.

Testing qualitative theories is somewhat different, as no causal mechanisms are included in the theoretical framework. The simplest verification of qualitative

2089255 - Elsevier Health Sciences ©

interpretations is to go to the participants themselves to establish if the researcher’s interpretations make sense to them. The extent to which a consensus develops between researchers and their participants is one of the important indications of the truth of qualitative theories.

Applications in Health Care Delivery The applications of quantitative evidence and theories are essentially technical,

providing mechanisms in terms of which we can predict and control specific health- related variables. That is, we apply quantitative approaches for (1) discovering the causes of diseases and disabilities, (2) predicting the burden of diseases and disabilities on individuals or populations, (3) developing and validating assessment procedures and (4) evaluating the effectiveness of interventions.

In contrast, qualitative research provides evidence and theories that enable us to better understand our participants as human beings. This research discloses how illnesses, disability and health care delivery affect people’s lives interpreted from their points of view.

Summary Qualitative research is based on the assumption that reality is socially constructed.

It entails disciplined enquiry examining the personal meanings of individuals’ experiences and actions in the context of their social environments. By qualitative we mean that the data consists of detailed descriptions based on language or pictures recorded by the investigator. The term ‘ discipline ’ indicates that the enquiry is guided by explicit methodological principles for defining problems, collecting and analysing the evidence, and formulating and evaluating theories. Personal meaning refers to the way in which individuals subjectively perceive and explain their experiences, actions and social environments.

In contrast, quantitative research holds the view that reality is independent of personal views and social contexts. The role of the researcher is to discover the objects and processes as they exist in the world. Quantitative research also involves disciplined enquiry based on reducing issues and problems to defined variables. Data collection involves measurement and observation under controlled conditions, enabling researchers to generate mechanistic or systems theories for explaining how variables are interrelated. The data are presented in a numerical form and are analysed using statistical techniques.

Different research designs may be used to generate evidence of the same processes, although from different perspectives. For instance, any complex clinical

2089255 - Elsevier Health Sciences ©

phenomenon, such as schizophrenia, may be studied using any of the research strategies outlined in this text. To understand the scope of the problems and the effectiveness of the appropriate treatments, it is desirable to use a variety of research strategies. Conversely, a comprehensive theory of a clinical problem should generate any number of hypotheses within the realm of the research strategies discussed in this book. We examine the analysis of qualitative data in Chapter X.

References

1. Doyle C, Lennox L, Bell DA. Systematic review of evidence on the links between patient experience and clinical safety and effectiveness. BMJ Open. 2013;3(1):e001570. doi:10.1136/bmjopen-2012-001570.

2. Roddis JK, Liversedge HL, Ryder I, Woodhouse M. Incorporating the patient experience into clinical guidelines: recommendations for researchers and guideline developers. BMJ Evid Based Med. Published Online: September 18, 2018 (doi:10.1136/bmjebm-2018-111015).

2089255 - Elsevier Health Sciences ©

3 The Research Process

Outline

Introduction, 12 Sequential Steps of the Research Process, 12

1 Research Problems and Questions, 13 2 Research Planning, 13 3 Design, 13 4 Data Collection, 13 5 Organization and Presentation of the Data, 13 6 Data Analysis, 14 7 Interpretation of the Evidence, 14 8 Evaluation and Dissemination of the Results, 14

The Structure of a Research Paper, 14 Title and Abstract, 15 Introduction, 15 Method, 15 Results, 16 Discussion, 16 References and Appendices, 16

The Publication Process, 16 Research and Evidence-Based Health Care, 16 Summary, 17

Introduction The previous two chapters were concerned with conceptual and philosophic

foundations of the scientific method. In this chapter, we will examine the method in a

2089255 - Elsevier Health Sciences ©

more practical way, as it is applied to collecting data and generating research evidence. Health and medical research is conducted as a stepwise, problem-solving approach. The steps for conducting research are reflected in the way in which research is reported in professional journals (which you will be reading throughout your career as part of the process of keeping up-to-date). Almost all clinical disciplines now have on-going mandatory requirements for continuing professional development (CPD), also known as continuing medical education (CME), that involve

reading and understanding published health and medical research. 1 This is a fundamental component of quality improvement activity to ensure that clinicians have a contemporary knowledge of the best clinical practices. The central aim of this chapter is to enable you to understand the rationale for conducting original research, which is the foundation for your clinical knowledge and clinical practice. The specific aims of the chapter are to:

1 Outline the steps of the research process. 2 Describe the conventions for presenting published research. 3 Briefly outline the ways in which research publications are essential for

evidence-based practice.

Sequential Steps of the Research Process Although there are differences in how health scientists approach specific

problems, Fig. 3.1 shows the steps commonly followed in applied health research. You might recognize this as an ‘algorithm’, a general programmed approach to solving problems. To illustrate the sequential steps, we will refer to a published research paper titled ‘A qualitative study of GPs’ views of treating obesity’ by Epstein

& Ogden. 2 You can access this publication from the British Journal of General Practice to enable you to follow the steps described here in more detail.

2089255 - Elsevier Health Sciences ©

FIG. 3.1 The research process.

Follow for extended description

2089255 - Elsevier Health Sciences ©

1 Research Problems and Questions The first step in the problem-solving process is to state clearly and unambiguously

the problem that you are intending to solve. Research problems reflect both the unmet health care needs of patients and the opportunities created by new theoretical and technological advances for improving health care (see Chs 4 and 24 ). The problem must be realistic, one that can be solved with the resources at the researchers’ disposal. It is pointless and unethical to initiate projects that cannot be completed (see Ch. 6 ).

Having identified a problem, the next step is to ask the ‘right’ answerable research question. A well-formulated research question will guide the research project to produce the evidence required to answer the question and (if all goes well) to contribute to the solution of the research problem. Chapter 4 will discuss in more detail the way researchers formulate questions.

In the case of the paper by Epstein & Ogden 2 , the research problem was identified as the lack of evidence concerning general practitioners’ (GPs’) views of their treatment approaches to obesity. The research question was: ‘How do general practitioners view the treatment of obesity?’ Therefore the aim of the researchers was to obtain evidence to describe and understand GPs’ views using a qualitative approach. It is widely recognized that clinicians have a pivotal role in promoting the

health of their patients and preventing the onset of serious avoidable diseases. 3

2 Research Planning Note that the researchers may have used a structured quantitative method to

obtain similar, pertinent evidence to address the research question. Of course, one may use both questionnaires and interviews in the same study. Obesity in the

community is a serious global health problem. 4 It is understood that the research problem should lead to evidence which will improve practices and health outcomes.

Research planning (see Section II ) involves selecting appropriate strategies and data collection techniques to answer research questions and to test the research hypotheses. (We explain what these are in Chapter 4 .) Research planning relies on detailed knowledge of previous research summarized in a literature review. Also, the plan should take into consideration ethical and economic factors before the appropriate data collection strategies are implemented and the precise research aims are stated. Additionally, the planning process will take into account the target population and formulation of a sampling strategy to select the required sample.

2089255 - Elsevier Health Sciences ©

In the present example, the population was defined as GPs working in London and the sample consisted of 21 GPs who consented to being interviewed for the study. The ethical requirements for this study were minimal, as there were no patients involved and the participants were commenting on their everyday professional experiences relevant to the research question.

3 Design Research designs are clear statements of how the data are to be collected in the

study. Appropriate research designs will lead to obtaining the data suitable for answering the research question (see Ch. 7–10 ). In the study we are discussing, the design is described as qualitative research using semi-structured interviews. This design was appropriate for obtaining data relevant to describing how GPs view the treatment of obesity.

4 Data Collection The next step in the research process is collection of data. We will examine data

collection methods employed in health research, including measurement, interviews and focus groups. In this study, the design that is the plan for data collection depended on semi-structured interviews with the doctors.

5 Organization and Presentation of the Data We will examine basic concepts in this area, outlining how graphs and various

descriptive statistics are used to condense and communicate research and clinical findings for qualitative research designs. When employing qualitative research, the results often are presented in the form of direct quotations of what was expressed. In the present study, the researchers provided a series of direct quotations representing GPs’ understanding and management of their patients’ weight problems. For example, ‘Dr 15’ (Doctor 15) described an unhealthy diet of a patient as: ‘Because of his lifestyle – he was working in the city – his lunch was fast food and so it was a difficult issue for me’.

6 Data Analysis The analysis of quantitative data involves applying the principles of probability for

calculating confidence intervals and testing the research hypotheses. Inferential statistics and decision making is outlined in Chapters 18 to 19 . The presentation and analysis of qualitative data involves identifying key themes which provide insights into participants’ personal meanings of events and experiences. The way in which GPs conceptualized their responsibility in the management of obese patients and

2089255 - Elsevier Health Sciences ©

how they saw the role of the patients in this process were the two key themes which emerged from the interviews.

7 Interpretation of the Evidence The next step in a research project is the interpretation of results (see Ch. 21 ). The

evidence is used to answer the research question and may either support or falsify existing theories or practices. Also, for various reasons, a research project may fail to produce sufficient evidence for confidently answering the research question. It is rare that the findings from any single research project are completely definitive, and often the results may suggest the need for further investigation in related subject areas or contexts. In the discussion section, the authors reported that the GPs conceptualized obesity in terms of the patients’ responsibility. It was suggested that future research was required to resolve an apparent conflict between the understanding of GPs’ and patients’ views of the management of obesity. Further research is required to provide evidence for improving communication between GPs and their patients.

8 Evaluation and Dissemination of the Results For research to be meaningful, investigators must publish their results in

professional journals, present them at conferences or both. Research findings become part of scientific knowledge only if the studies stand up to methodological critique and replication. We will outline steps for critically evaluating published research. Also, we note that evidence from various related publications can be synthesized for producing systematic reviews guiding evidence-based practices in Chapter 23 . Epstein and Ogden disseminated their results by publishing their findings in the British Journal of General Practice. In doing so, the results of the study may generate further discussion and research regarding the role of GPs in helping people manage weight problems. At the time of writing, the paper was cited by 208 other researchers, so it had a significant effect on thinking in this area. The research effect of papers can be measured by the various databases and tools designed for this purpose, such as Google Scholar, Research Gate and Thompson citation tools. The translation effect of such studies also can be measured by the effect on clinical practice and clinical standards and guidelines.

The Structure of a Research Paper The format of a professional publication reporting empirical studies generally

reflects the stages of the research process discussed in this book. Table 3.1 represents the relationship between the stages of research and the commonly used

2089255 - Elsevier Health Sciences ©

publication format for empirical studies. This form is not necessarily followed for all types of scholarly communications, such as for some qualitative studies, theoretical papers and literature reviews. In the subsections that follow, we examine in more detail each of the components of a research report shown in Table 3.1 .

TABLE 3.1 Format of Research Publications and the Research Process

Publication format

Research process

Title Explains the focus and content of the study

Abstract Provides a brief informative outline of what was done and what was found

Introduction Research planning. Answers the question, “why was this work necessary?”

Method Design. Explains how the research was conducted

Participants Explains who participated in the study

Apparatus Measurement and data collection

Procedure Provides a step-by-step description of how the research was conducted

Results Descriptive statistics or narratives and inferential statistics or thematical analysis describe the study’s outcomes

Discussion Interpretation of the evidence and recommendations for clinical actions and future research

References A list of the work used to inform the study design and results

Appendices Provided if necessary to explain the study and its results

Title and Abstract The title is a descriptive sentence stating the exact topic of the report. Many titles

of quantitative research reports take one of two forms:

• y as a function of x

• the effect of x upon y

In causal research, such as experiments, y refers to the dependent variable being measured and x refers to the independent variable being manipulated (see Ch. 7 ).

2089255 - Elsevier Health Sciences ©

For example:

• The incidence of alcoholism in health professionals as a function of work-related stress

• The effect of major tranquillizers on the cognitive functioning of persons with schizophrenia

For descriptive quantitative or qualitative research, the title should inform the reader about the groups being studied and the characteristics being reported, for example, ‘The experience of chronic pain in injured workers’. In general, titles should be concise and informative, enabling a prospective reader to identify the nature of the investigation. Immediately below the title will usually be the name(s) of the investigator(s) and affiliation (i.e. place of work).

The abstract is a short (approximately 250 words) description of the entire report. The purpose of this section is to provide the reader with a general overview of the communication. It should provide enough detail to enable the reader to decide whether the article is of interest. Writing an abstract is a challenging task because it should be precise and informative but at the same time very brief. When writing an abstract you should include:

1 A statement about previous findings that led to this research project. 2 The hypothesis and/or aims. 3 Methods, including participants, the research design and procedure. 4 A short description of what was found and how the data were analysed. 5 What was concluded and the nature of further research need.

Introduction The introduction describes the planning stages of research, as discussed

previously. A good introduction sets the stage for the research question by discussing the theoretical background of the problem under consideration and describing and evaluating the relevant research previously completed. The introduction thus serves as a link between past work and the present knowledge.

Generally, all aspects of the previous research literature cannot be covered in a relatively brief introduction required in a research paper; therefore the review of past research is done with a bias towards only those aspects of the problem that are of direct relevance to the report. In this way the research question can be derived in a logical manner. For this reason, the introduction generally starts by making a few

2089255 - Elsevier Health Sciences ©

general statements about the field of research, then focusing on specific research questions, aims and/or hypotheses.

Method The purpose of the methods section is to inform the reader of exactly how the

study was carried out. It is important to remember that the methods section should contain enough detail to enable another researcher to replicate the study. (Of course, replications may not be feasible for a unique event, such as a case study of a specific individual.) Conventionally, three subsections are reported: participants (or subjects), apparatus/tools and procedure.

Research Participants

Three questions must be answered concerning the participants: (1) who were they, (2) how many were there and (3) how were they selected? Specific information must be given concerning the participants, as results may vary from one sample to another. Ethical considerations also should be discussed here. Medical researchers generally refer to people in the sample as the ‘subjects’ and social scientists and qualitative researchers as ‘participants’.

Apparatus/Tools

A description of all equipment, questionnaires and standardized tests used in the research must be provided. If it is commercially available, the reader should be provided with the manufacturer’s name and the commercial identification of the equipment. Alternatively, if the equipment was privately made, the reader should be given enough information to allow replication. Measurements and perhaps a diagram will be necessary. In qualitative research, equipment often includes devices used to record interviews.

Procedure

Once again, this section should provide enough information for other researchers to replicate the study. Details of how the research was carried out should include how research participants were assigned to groups, how many research participants were in each group, the experimental procedure used and a description of how data were collected.

Results The results section presents the findings of the investigation and draws attention

to points of interest. Raw data and statistical calculations are not presented in this

2089255 - Elsevier Health Sciences ©

section. Rather, the principles of descriptive and inferential statistics should be used to present the summarized and analysed data: graphs, tables and the outcomes of statistical tests are presented in this section. It is essential that all the findings are presented and that the graphs and tables are correctly identified. Qualitative researchers will present original quotes from participants and describe how analysis was conducted.

Discussion The discussion section restates the aim(s) of the research study and discusses the

results with reference to the aims or research hypothesis stated in the introduction. Did the researchers obtain the results expected? How do the present results relate to previous results and theories?

It is important to remember that the results of one study in isolation cannot make or break a theory or establish the efficacy of a novel practice. Thus the discussion should connect the findings with similar studies and especially with the theory underlying such studies. If unexpected results were obtained, possible reasons for the outcome (such as faulty design and controls) should be discussed. By doing this, the discussion will point the way to further problems that remain to be solved. Unconstructive, negative or unimportant criticism should be avoided so that the report does not end with an unreasonably long discussion of possible reasons for the outcome. A brief, concise discussion is more appropriate.

In the conclusion, which is usually the last paragraph of the discussion section, the main findings are summarized and suggestions made for further research. For example, you may have demonstrated certain phenomena that may have implications for explaining broader concepts which can be empirically tested. You are therefore taking your findings and generalizing them to phenomena not directly tested in the present research.

References and Appendices It is expected that all the literature discussed in the paper be listed in the

references section. This enables the reader to evaluate the sources. Information on how references should be listed can be found in appropriate style manuals. Sufficient information must be provided for an interested reader to be able to identify and retrieve the sources. Additionally, a report may include labelled appendices. These might include a full description of questionnaires or other measuring instruments, raw data or statistical calculations if required.

The Publication Process

2089255 - Elsevier Health Sciences ©

The formal knowledge representing the empirical and professional basis for your professional practice is in large part stored in journals, books and conference reports. Journals are published by appropriate professional associations, government departments or private publishing corporations. Having completed a research project, how does one publish it in a professional journal? After all, the value of research is negligible if it is not made public.

In general, the prospective author will:

1 Select a professional or scientific journal appropriate for the material. 2 Present the research report in a format required by the journal. 3 Send the completed manuscript to the journal editor.

The editor is generally a person of high standing in a given scientific or professional area. If the article is judged as being appropriate for the journal, the editor will send the article to two or more peer reviewers and, on the basis of the reviewers’ reports, publish or reject the manuscript. Sometimes the reviewers recommend certain additions or changes which have to be made by the author before the manuscript is judged to be publishable.

Therefore when you read research publications in peer reviewed journals, you can be confident that experts have scrutinized the articles. However, discussed in Chapter 22 , the fact that a research study has been published in a reputable journal does not necessarily guarantee the truth of either the evidence or the conclusions.

Research and Evidence-Based Health Care Research methods cover a wide variety of skills and techniques aimed at the

methodologically valid investigation of questions of interest to the researcher. These methods of enquiry are not restricted to research laboratories, nor need they involve expensive equipment or large research teams. Rather, these methods imply an approach to stating and answering questions in any setting.

Research methods interact with health practices in multiple and mutually productive ways. An important general aim of this textbook is to discuss the relationships between theories, practices and the ways in which research methods contribute to improvements in health care. The term ‘ evidence-based medicine ’ has been defined by Sackett et al as ‘… the integration of best research evidence with

clinical expertise and patient values’. 5 Evidence-based health care is a contemporary movement aimed at ensuring that health services are based on the best available scientific and clinical evidence. The fundamental question that is explored in this book is ‘what constitutes best research evidence to inform clinical practice?’ The

2089255 - Elsevier Health Sciences ©

evidence is, of course, produced by research conducted in accordance with the principles of the scientific method. Clearly, if you expect to participate in the delivery of evidence-based health care you will need to understand research methods sufficiently for making informed and critical judgements concerning the quality of the available evidence to drive high-quality and effective clinical practice. As outlined in Chapter 23 , systematic reviews of available evidence are very helpful to clinicians in keeping up-to-date in their clinical practices.

Finally, we need to keep in mind that the research is produced by the hard work of men and women who were undergraduate students, just like you. Ultimately, it will be up to you (as nurses, podiatrists, physiotherapists, speech pathologists and occupational therapists) to carry out the research that underpins the effectiveness, advancement and prestige of your profession. If not you, then who?

Summary In this chapter we discussed the ways in which the scientific method is directly

applicable to conducting research in the health sciences. The sequential steps of the research process follow a general problem-solving algorithm. Identifying these steps enables you to understand the rationale underlying the planning, conduct and interpretation of research. We outlined how the general format of a published research paper is associated with these steps. We also examined the style and structure of a research publication. The style involves clarity, accuracy and sufficient completeness for colleagues to understand or replicate the research project. Research is published in journals, which are generally edited by persons of high standing in the field. Every effort is made by editors to ensure the validity of the research published in their journals. The individual researcher also is bound ethically to report findings in an unbiased and truthful way. Although the format and style outlined in this chapter might seem rather arduous, poor presentation may destroy the intrinsic value of a research project.

References

1. World Health Organization. Regional Guidelines for Continuing Medical Education (CME)/Continuing Professional Development (CPD) Activities. New Delhi: World Health Organization Regional Office for South East Asia; 2010.

2. Epstein L, Ogden J. A qualitative study of GPs’ views of treating obesity. Br J Gen Pract. 2005;55(519):750-754.

2089255 - Elsevier Health Sciences ©

3. Beaglehole R, Epping-Jordan J, Patel V, et al. Improving the prevention and management of chronic disease in low-income and middle-income countries: a priority for primary health care. Lancet. 2008;372(9642):940- 949. doi:10.1016/s0140-6736(08)61404-x.

4. Ng M, Fleming T, Robinson M, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2014;384(9945):766-781.

5. Sackett DL, Strauss SE, Richardson WS, Rosenberg W, Haynes RB. Evidence-Based Medicine: How to Practice and Teach EBM. 2nd ed. Edinburgh: Churchill Livingstone; 2000.

2089255 - Elsevier Health Sciences ©

Introduction

The first stage of research involves developing the detailed plan for conducting the project. The project plan is documented in the research protocol. Before the research project is initiated, usually one or more research ethics committees (Medical Research Council 2018; National Health and Medical Research Council 2018) examine the protocol and funding bodies to ensure that the project plan conforms with general methodological and ethical principles. The three chapters in Section 2 aim to outline the basic considerations for the successful preparation of a research protocol. This section will help you build a picture of the overall research process.

The primary reason for proposing a research project is to obtain evidence that will, we hope, advance theory and practice in the health sciences and the efficacious patient-centred treatment of people with health problems (see Ch. 4 ). Before all else, we must be sure that we are asking the right research questions; that is, raising issues and problems which are crucial to progress to the development of high quality and effective health care. We must convince the experts who evaluate our proposal that the questions we are asking are of central importance. Asking the right research questions depends on being creative, for example, identifying unanticipated patterns in data, or the construction of novel theories that predict new, as yet unobserved facts regarding health care.

To justify the research proposal it is essential to write a literature review. The literature review is a summary and critical evaluation of previous research and theory relevant to the problem under investigation. The literature review provides both a conceptual background for our proposal and justifies the need for further evidence by identifying ‘gaps’ in our knowledge essential to solving the research problem. In contemporary health research it is now standard practice to use a systematic review approach. The reporting of reviews is now a highly structured and systematic process. Systems such as Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA; http://www.prisma-statement.org/ ) are used to report what is known about particular research issues and questions. In a contemporary

2089255 - Elsevier Health Sciences ©

health research paper, if the paper does not itself use a systematic review, it is typical to report such reviews in the paper rationale and introduction. Systematic reviews may be found in many academic journals but there are also databases that are specifically focussed on systematic reviews. The most famous of these is the Cochrane Database of Systematic Reviews ( http://www.cochranelibrary.com/ ). This database contains many thousands of systematic reviews about an exceptionally broad span of topics. The Cochrane library is actually mandatory reading for clinicians because it is a repository of the best evidence concerning clinical interventions. It is notable that the Cochrane Database includes a wide range of qualitative, quantitative and mixed reviews. This reflects the development of health research in including patient and clinician perspectives as valuable contributions to our knowledge base.

The way in which research study samples are selected is discussed in Chapter 5 . Selection of an appropriate study sample is crucial for the generalizability (or external validity) of your findings. The aim of sampling strategies is to select a representative sample of the study population. Also, we need to select a sample size which is sufficiently large to enable accurate inferences on the bases of sample data.

A project should not proceed unless it is judged to be ethically sound by an appropriate ethics committee (see Ch. 6 ). A research proposal is judged to be ethical if it conforms to the rules of moral conduct and values of the community and the evaluation criteria used by the committee to consider this issue. These rules and values are made explicit in documents and research publications representing the standards of professional groups and of institutions (e.g. hospitals, universities and research councils). The research protocol must describe the research in sufficient detail the degree of burdens imposed on the participants.

2089255 - Elsevier Health Sciences ©

4 The Formulation of Research Questions

Outline

Introduction, 20 Identifying Research Problems, 20

Environmental and Social Changes, 20 Demographic Changes, 21 Scientific and Technological Advances, 22 The Cost-Effectiveness of Health Services, 22

Literature Reviews and Research, 22 Formulating the Research Question, 23 Methodological Considerations, 23

Deciding Between Using Qualitative and Quantitative Methods in Your Research, 23 Research Participants and Sampling, 23 Study Design, 23

Ethical Issues and Procedures, 24 Aims and Hypotheses, 24 Research Proposals, 24 Summary, 24

Introduction The planning of a research study begins with the identification of a research

problem, followed by the formulation of the research question(s). In the ‘Introduction’ section of a research paper you will find the research question guiding the

2089255 - Elsevier Health Sciences ©

investigation. The formulation of a research question is the outcome of a complex, iterative process. Researchers refine and focus the question as they plan the stages of the research project, as shown in Fig. 4.1 . The aims of this chapter are to:

1 Describe the process for identifying research problems. 2 Discuss the use of literature reviews for developing research questions. 3 Describe the relationship between formulating research questions and

methodology. 4 Explain the relevance of ethics and resources to the statement of specific aims

and hypotheses. 5 Describe what constitutes a research proposal.

2089255 - Elsevier Health Sciences ©

2089255 - Elsevier Health Sciences ©

2089255 - Elsevier Health Sciences ©

FIG. 4.1 Reiterative process for formulating research questions.

Follow for extended description

Identifying Research Problems Research in health and medical science often is focused on identifying effective

interventions for preventing or treating health problems. Many research problems

are derived from improving current practices, as discussed in Sackett et al. 1

Economic considerations, that is, the availability of resources, also can have a strong influence on research planning. For example, you may have designed a qualitative research project that involves 100 ‘in-depth’ interviews with persons suffering from a disorder. Say that you are an ‘honors’ student and have only 1 year and a limited amount of funds to complete the project. You would be advised to reduce your sample size or, if this is not possible, change your topic. Ethics or funding bodies will only approve projects which are realistic – that is, those which can be completed feasibly with the available resources.

Thus research planning is a process through which we transform our initial ideas into well-planned, ethical and realistic research projects ( Table 4.1 ). Experts judge the feasibility of the project before permission is given to proceed.

TABLE 4.1 Essential Research Questions for Evidence-Based Medicine and Health Care

Categories of practice Typical research questions

Improving diagnosis and screening

Does this new diagnostic test accurately discriminate between people who suffer from Alzheimer’s disease and multistroke dementia?

Does this new genetic test accurately identify people who are likely to suffer from epilepsy?

Producing and evaluating strategies for predicting health outcomes

Does the immunization of children result in increased incidence of autism?

What is the fatality rate among people who were exposed to high levels of mercury?

2089255 - Elsevier Health Sciences ©

Devising and evaluating treatments

What are the comparative benefits of hydrotherapy and nicotine replacement therapy for smoking cessation?

Does aspirin reduce the incidence of stroke?

Harm minimization What are the comparative infection rates after surgery across a group of metropolitan hospitals?

Do hypertension medications increase the risk of kidney failure?

However, there are numerous other research problems, such as gaps in our knowledge of the causes and consequences of health problems and opportunity to improve the quality of our practices. Let us look at examples of some research problems.

Environmental and Social Changes As the biological and social environment changes, new health problems may

emerge. For example, abundant food supplies and increasingly sedentary lifestyles have led to a high prevalence of obesity in many communities. The Global Burden of Disease studies ( http://www.healthdata.org/gbd ) conducted by the international Global Burden of Disease consortium is an exceptionally comprehensive epidemiologic study global program in which obesity and other disease-burden risk factors have been studied. Many publications in the Lancet and other journals report this work. We consider a good knowledge of this work as a valuable tool in the clinician’s clinical toolkit. It is well worth a comprehensive read.

Obesity is a serious risk factor for numerous health problems, including type 2 diabetes and coronary heart disease. The research problem of interest to a researcher may be the lack of effective interventions for preventing and reducing obesity in the community.

Another example is the health effect of climate change, a process which is predicted to affect population health. The environmental changes (in particular the prediction of global warming) have led to identifying new research problems and questions about the effect of climate change. Many countries have research priorities based on the prevalence of diseases as identified by epidemiologic studies (see Ch. 8 ). Public and political concern about the ‘burden of disease’ for a community results in funding bodies preferentially channelling resources to researchers in these areas.

2089255 - Elsevier Health Sciences ©

Demographic Changes Changes in population characteristics, such as the increasing number of elderly

people, have shifted the health needs of communities in many countries. There is increased research activity concerning the health problems associated with ageing and the ways of designing and delivering effective and efficient health services for elderly patients. Also, some countries have an increasing number of immigrants and asylum-seekers who have health needs different than the rest of the population. Relevant research questions have been asked regarding language or sociocultural barriers which prevent migrants from accessing appropriate health services. In general, research problems arise from the limitations of the current health services being offered to a population with changing needs.

Scientific and Technological Advances There are continuing and remarkable discoveries in the biological, behavioural,

social and information sciences that can be applied to improving the assessment, treatment and prevention of illnesses. Advances in electronics and information technology have created opportunities in a variety of health professions, devising new diagnostic techniques or data collection and management strategies. The research problems arise from the need to translate pure research into practical solutions. An example of this is stem cell research, which has led to spectacular discoveries in the biological sciences. Translational research in this field of regenerative medicine aims to develop cellular therapies for the treatment of a variety of disorders such as heart disease, diabetes and Parkinson’s disease. Chapter 24 examines the rationale for translational research.

The Cost-Effectiveness of Health Services Economic pressures from increasing health costs require that service providers are

accountable for the way in which they use available resources. Program evaluations for services are now routinely incorporated into service delivery. Evidence from this research is necessary to demonstrate that the resources are being used in the best way and that the services are meeting the needs of patients and the community. Although often a source of stress for health professionals and administrators who participate in them, well-designed evaluation programs can contribute to the development of cost-effective, ‘value-for-money’ services to the community. For example, if you can demonstrate that program A and program B provide the same benefits but A costs half as much as B to deliver, then you will have made a

2089255 - Elsevier Health Sciences ©

significant contribution to health care. The money saved can be used to provide better services to other patients with other problems.

We hope this discussion demonstrates that there are innumerable research problems waiting for you when you graduate! Perhaps you will take up the challenge and by asking the right questions will produce research which will contribute to the advancement of health care for the community.

Literature Reviews and Research Before embarking on the design and conduct of a research project, the investigator

must review previous work and publications relevant to the aims of the intended project. (The process of literature review is described in detail in Ch. 23 ). This process is essential, both for providing the appropriate background and context for the investigation and for justifying the investigation in contributing to existing knowledge. It is potentially a waste of money to inadvertently duplicate very similar research (although deliberate replication of previous studies may be a legitimate activity if there are uncertainties about the validity of a previous research). Another reason for reviewing the literature is to examine how previous researchers have approached sampling, design, data collection and analysis decisions in their research. There are benefits in reviewing how previous researchers succeeded or failed in achieving their research and clinical objectives.

There are several standard approaches to identifying research papers of interest to a specific research study or problem. An example is PICO (see http://pubmedhh.nlm.nih.gov/ ), an acronym for:

P = Target P opulation/ P atients I = I ntervention (or type of exposure) C = C omparison (alternative intervention) O = The O utcome measures

By examining the keywords, title and abstract of a paper, you can see how these criteria apply. Imagine that you are interested in identifying papers that investigate the efficacy of neural transplantation for treating Parkinson’s disease. PICO has been used for a long time. For example, you should be able to find the paper, ‘A double- blind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease’ by

Olanow et al. 2 Here:

P represents patients with Parkinson’s disease. I refers to the bilateral foetal nigral transplants.

2089255 - Elsevier Health Sciences ©

C refers to the placebo surgery group. O refers to outcomes on standardized tests.

Literature searching may be carried out at appropriate research centres or libraries where scientific and professional journals are stored and, of course, by using the Internet. In health research, as opposed to historical research, almost all searches use Internet search engines and electronic journals databases. There are powerful web-based search methods that can simplify the search. The most widely used search tool for health research publications is probably PubMed, the web-based Public Medline (see http://www.ncbi.nlm.nih.gov/entrez/ ). Professional library staff can help you locate the relevant literature. However, the critical evaluation of the literature depends on the application of research methods for the identification of controversies or ‘gaps’ in the available evidence (see Ch. 23 ). A detailed and critical review of the literature enables us to ask the questions most relevant to planning the kind of research that will enable us to address the research problem.

Formulating the Research Question To formulate a research question, we need to investigate what is and what is not

known in the area (i.e. to identify what we do and do not know when we understand the existing published literature). By stating the research problem and reviewing the literature, we are able to devise a research question. The research question will guide the investigation and guide us in producing the evidence required to solve the research problem. Inspiration for research questions may stem from direct personal experience of working with patients and literature review and analysis. The critical evaluations to an issue or problems may raise questions suitable for a research study. The identification of research questions is a creative act. It draws upon consideration of previous relevant work in the area under study. This work may be reviewed and the review may give rise to questions that have not been answered by previous work.

Methodological Considerations The precisely stated research question requires an examination of the proposed

methodology of the project. To illustrate the methodological issues, let us look at an example based on a publication titled ‘Ribavirin for Crimean-Congo hemorrhagic

fever: systematic review and meta-analysis’. 3 The health problem is the Crimean- Congo haemorrhagic fever (CCHF), which is a potentially lethal viral disease. There is a drug called ribavirin which has been approved by the World Health Organization

2089255 - Elsevier Health Sciences ©

(WHO) and has been used with apparent success to reduce the mortality rate

associated with CCHF. A review by Soares-Weiser et al, 3 identifies a research problem and a possible research question regarding the efficacy of ribavirin:

‘No clear message of benefit is available from the current data on ribavirin as observational data are heavily confounded, and the one trial carried out has limited power. However ribavirin could potentially have benefits in this condition …’

Let us imagine that we are scientists working in public health and that we want to contribute to solving this problem. The first thing to do is to formulate a precise research question and, if appropriate, a research hypothesis derived from the research question. To do this, we need to plan the data collection process for conducting the investigation.

Deciding between Using Qualitative and Quantitative Methods in Your Research

An essential consideration in framing the research question is whether you will use a qualitative or a quantitative research method. Quantitative research is normally structured so as to test a research hypothesis. Hypotheses are propositions about relationships between variables or differences between groups that are to be tested. Qualitative research does not usually test hypotheses. It is concerned with understanding the meaning of how people experience specific issues. However, most qualitative researchers work with clear aims and questions to guide their investigations. Some research questions require both qualitative and quantitative methods.

For our hypothetical example, we would select a quantitative approach. The reason is that we intend to test the causal hypothesis that ribavirin, an antiviral drug, is effective for treating CCHF. If we were investigating the experiences of patients taking this treatment, we would have selected a qualitative approach. However, we are looking at the efficacy of a treatment. The next step will be to refine our research hypothesis, taking into account the sample and proposed design.

Research Participants and Sampling To focus our hypothesis, we need to define the population we wish to study. In the

ribavirin example, the study population consists of people living in regions where

CCHF is a public health concern. According to Soares-Weiser et al, 3 recent outbreaks have occurred in Turkey and Iran. Because of both clinical and pragmatic reasons (see the next section), you might select people living in Turkey as your population of interest. We will look at details of sampling in Chapter 5 .

2089255 - Elsevier Health Sciences ©

Study Design In Section III we will examine research designs in detail. As discussed previously,

we have selected a quantitative methodological approach for our hypothetical investigation. There are several types of research designs which fall under quantitative methods; the most basic distinction is whether they are experimental or nonexperimental observational designs (see Section III ). In our example, we are investigating a causal hypothesis; therefore we would select an experimental design. As discussed in Chapter 7 , a randomized controlled trial (RCT) would be the preferred design for establishing the efficacy of ribavirin.

In an RCT, the participants are assigned (by chance or random method) to a treatment (active intervention) or control (non-intervention) group, and then given the intervention or, in clinical contexts, the treatment. In our example, the treatment is the quantity and administration of the ribavirin. We also need to make a decision about the type of control group we will use (i.e. no treatment, placebo treatment or current standard treatment other than ribavirin). The intervention and control groups are then observed and compared. At this point, we must consider how we are going to measure outcomes for enabling us to make a comparison. The selection of outcome measures is discussed in detail in Chapters 11 to 14 . In this case, since CCHF is a potentially fatal condition, we should use mortality rates as the indicator of treatment efficacy.

Ethical Issues and Procedures When conducting health care research, a primary concern is the well-being and

dignity of the research participants (see Ch. 6 ). Ethical considerations will influence the sample selected, the design of a project and, ultimately, the research question which will be posed. Additionally, the available resources such as time and money will determine the scope of the research project. In our present hypothetical example, we may have decided to:

• Select the sample only from Turkey for budgetary reasons.

• Use a best-available treatment for the participants in the control group for ethical reasons.

• Use mortality as an outcome measure for treatment efficacy.

All studies need to be ethically cleared by a properly constituted human research ethics committee (see Ch. 6 for a detailed discussion of ethics processes).

Aims and Hypotheses

2089255 - Elsevier Health Sciences ©

Having considered the methodology and ethics, we are able to refine our research question as follows: ‘Is ribavirin more effective than the best-available current treatment for reducing mortality in people with CCHF living in Turkey?’ Our aim is to collect evidence showing the efficacy of CCHF. In general, the aim of a research project is to produce evidence for answering the research question.

Because we have selected a quantitative method to investigate the question, we can also state a research hypothesis: ‘Ribavirin treatment will reduce mortality more than best-available treatment in people with CCHF living in Turkey’. The obtained results will enable us to make a decision as to whether the research hypothesis is supported by the data we collect. This decision will enable us to answer the research question and contribute to solving the original research problem. You will recall that the research problem was an uncertainty regarding the efficacy of ribavirin. Fig. 4.1 represents the steps that are followed to develop detailed and precise aims and hypotheses.

Research Proposals By the time we have stated our precise aims or hypotheses, we should have

decided the way in which the data are to be collected. This information is written up as a research proposal. A research proposal is an explicit statement of how and why you intend to conduct a research project. A research proposal will contain essential information:

• Identification of the research problem

• Literature review

• Research aims and hypotheses, which must take into consideration the: 1 participants to be selected, 2 data to be collected, 3 research design/s, 4 ethical considerations and 5 estimated cost.

The research proposal is circulated to various individuals and committees in order to obtain critical feedback from supervisors, institutions and your fellow students concerning the practicality and quality of your research project. Most importantly, the research proposal is the basis for writing a detailed report to relevant ethics committees (see Ch. 6 ).

It often is desirable to carry out a small-scale preliminary study called a pilot study. This is an economical way of identifying and eliminating potential problems before

2089255 - Elsevier Health Sciences ©

you do the large-scale (and expensive) full study. Case studies and so-called safety and efficacy studies often are employed before RCTs are undertaken (see Ch. 7 ).

Summary The planning of a research project requires the transformation of preliminary

thoughts on solving a research problem into clearly stated research questions and aims. To achieve this, the researcher should review the relevant literature and evaluate ethical considerations and economic constraints in conducting the investigation. Next, an appropriate research strategy has to be formulated. Having considered the stages of the research project, the researcher is in a position to state precisely the aims or hypotheses being investigated. Before data collection begins, an ethics committee must scrutinize the research proposal to resolve methodological problems and to ensure the implementation of ethical research.

References

1. Sackett DL, Strauss SE, Richardson WS, Rosenberg W, Haynes RB. Evidence-Based Medicine: How to Practice and Teach EBM. 2nd ed. Edinburgh: Churchill Livingstone; 2000.

2. Olanow CW, Goetz CG, Kordower JH, et al. A double-blind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease. Ann Neurol. 2003;54(3):403-414.

3. Soares-Weiser K, Thomas S, Thomson G, Garner P. Ribavirin for Crimean-Congo hemorrhagic fever: systematic review and meta- analysis. BMC Infect Dis. 2010;10:207.

2089255 - Elsevier Health Sciences ©

5 Sampling Methods and External Validity

Outline

Introduction, 26 Basic Issues in Sampling, 26 Representative Samples, 27

Sampling Methods, 27 Sample Size, 30

Sampling Error, 30 Sampling in Qualitative Research, 31 Purposive Sampling, 31 External Validity and Sampling, 32

Ecological Validity, 32 Summary, 33

Introduction Research in the health sciences usually involves the collection of information from

a sample of participants, rather than the entire population in which the investigator is interested. Studies that involve an entire population or group are called census studies, but these are relatively rare and generally prohibitively expensive to perform. Thus a sample is drawn from the target population and studied because it is usually impossible or too costly to study entire populations. For instance, when individuals who have conditions such as diabetes, cerebral palsy or emphysema are being studied, it is not possible to study everyone with the disease because of the large size of such populations. Also because many people do not seek treatment or

2089255 - Elsevier Health Sciences ©

may have been incorrectly diagnosed, it may not be possible to identify all members of the entire population. Therefore in most research, the researcher studies a subset or sample of the target population and then attempts to generalize the findings to the population from which the participants were drawn. This general principle applies to both qualitative and quantitative studies.

The aim of this chapter is to examine ways in which samples can be drawn to permit the investigator to make valid generalizations from the study sample to the target population.

We will also consider the question of generalizing the findings of an investigation to other samples and situations. This is referred to as the external validity or generalizability of a sample.

The specific aims of this chapter are to:

1 Define what is meant by sampling and representative samples. 2 Outline the relative advantages and disadvantages of commonly used

sampling strategies for both qualitative and quantitative methods. 3 Discuss the relationship between sampling error and sample size. 4 Examine the concept of external validity for generalizing research findings.

Basic Issues in Sampling A population is defined as the entire set of persons, objects or events that the

researcher intends to study. That is, the population is defined in terms of the specific research question being investigated. We must examine very closely the research questions being asked to clarify the population which is being studied in a research project. In other words, the population can be any group of people or set of objects. For example, consider the statement from a textbook for health professionals

working with people with Parkinson’s disease (PD): 1

Levodopa is the most effective medication for PD and almost all patients will need

to be on levodopa during the course of the disease. 1 (p26)

This statement refers to the actual and predicted clinical responses of millions of people with PD to a range of drugs which increase acetylcholine activity in the brain. It must be evident to you that the proposition just given is based on evidence from a minute proportion of patients with PD. You might recall that the logic for generalizing from a limited set of observations is called induction. Induction was defined as the logic of producing general statements on the basis of specific observations. In the context of research, the ‘limited set of observations’ refers to sample data collected

2089255 - Elsevier Health Sciences ©

in the course of the study. The process of generalization entails summarizing sample data and then drawing inferences from sample data about the probable true state of the population. We will also examine this concept in the statistical chapters in Section V and VI.

A sample is defined as a subset of the population. When we talk about a population we are not only referring to the individuals who constitute the population but also the factors or variables which we observe or measure in these people. For the example given here, the sample is not simply the people with PD but also the changes in their symptoms. The data for the symptomatic changes in people under the influence of drugs which cause increased acetylcholine constitutes the actual sample.

In the context of scientific research, we call the process of generalizing from sample data to the population inference . Inference is probabilistic; that is, we are not absolutely certain that what we find in our sample is true for the population. To improve the probable truth of our inferences we must ensure that our samples are representative of the population.

Representative Samples There are a variety of different ways by which one can select the sample from the

population. These are called sampling methods . The ultimate aim of all sampling methods is to draw a representative sample from

the population. The advantage of a representative sample is clear: one can confidently generalize from a representative sample to the rest of the population without having to study the entire population. Jones and Attia (2017) in their aptly named paper ‘Sampling: how you choose people is as important as how you analyse their data’ note that ‘if the research design is poor, no amount of complex statistical

analysis can extract useful information from the data collected’. 2 (p67)

A representative sample is an accurate but reduced representation of the population being studied ( Table 5.1 ). Ideally, a representative sample contains in a miniature form all the relevant factors or variables in the population in the same proportions and magnitudes as in the original population. If a population is not representative then it is biased. Biased samples generate sampling errors, which are defined as the magnitude of the difference between the true state of the population and the state of the sample. So if the average age of the targeted population was 45 years but the study sample had an average age of 50 years, then the sampling error for age would be 5 years.

Fig. 5.1 illustrates a hypothetical population composed of three different shapes.

2089255 - Elsevier Health Sciences ©

Follow for extended description

A representative sample is a precise miniaturized representation of the population. An unrepresentative, or biased, sample does not adequately represent the key groups or characteristics in the population, and this may lead to erroneous conclusions about the state of the population. The first sample shown in Fig. 3.1 is representative because it has the same proportion of the three shapes as the population. The second sample is obviously biased and far too small even to correctly represent the population.

Sampling Methods The selection of the appropriate sampling method depends on the aims and

resources of the researchers. For instance, if someone is designing a very expensive health or social welfare program on the basis of a survey of client needs, it is imperative that a good sampling method is used and a representative sample of the clients be obtained so that appropriate conclusions may be reached about the population. Good sampling methods tend to be more expensive and more difficult to implement than poor methods, but, arguably, they are worth it because of the effects upon the soundness and quality of the research.

Incidental Sampling

Incidental sampling is the cheapest and easiest sampling method for conducting clinical studies. It involves the selection of the most accessible and available members of the target population. For example, a researcher who stands in a street in the business centre of London (the City) and quizzes people about their health

FIG. 5.1 A simple representative sample.

2089255 - Elsevier Health Sciences ©

status is practicing incidental sampling. However, it is quite likely that this sample would not be representative of the target population.