2014, 36: 746756

AMEE GUIDE

Quantitative and qualitative methods in medical

education research: AMEE Guide No 90: Part I
MOHSEN TAVAKOL1 & JOHN SANDARS2
1

The University of Nottingham, UK, 2The University of Leeds, UK

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

Abstract
Medical educators need to understand and conduct medical education research in order to make informed decisions based on the
best evidence, rather than rely on their own hunches. The purpose of this Guide is to provide medical educators, especially those
who are new to medical education research, with a basic understanding of how quantitative and qualitative methods contribute to
the medical education evidence base through their different inquiry approaches and also how to select the most appropriate
inquiry approach to answer their research questions.

Introduction

Practice points

Over the past few decades, major advances have occurred in

both the understanding and practice of medical education.
Medical education research has contributed considerably to
these advances by adding reliable new knowledge to an
existing body of educational knowledge to produce best
evidence that can help medical educators to make better
decisions about important areas of medical education, such as
teaching and learning, effective curriculum design and assessment. Through research, data can be collected and analysed to
better understand the teaching and learning process (Norman
2002) and also to inform decision making about how well a
particular programme, practice, procedure or policy is operating (Tavakol & Gruppen 2010). However, there is often little
interest by clinicians in medical education research, possibly as
a result of a lack of training in education research methods,
and with many clinical educators also feeling less confident in
the application of qualitative research approaches (Tavakol
et al. 2008). This could be due to the fact that the nature of
qualitative studies in comparison with quantitative methods
has not been recognised (Morse 2005), especially since
medical educators tend to gather empirical data that are
grounded in objective rather than subjective reality (Buckley
1998). However, the contribution of qualitative studies in
evidence-based practice has increasingly been recognised in
both healthcare systems and educational research (McEwan
et al. 2004; Ong & Richardson 2006; Bower & Scambler 2007).
The purpose of this Guide is to provide medical educators,
especially those who are new to medical education research,
with a basic understanding of how quantitative and qualitative
methods contribute to the medical education evidence base
through their different inquiry approaches. It also provides
readers with the primary steps of the research process and an
understanding of how to select the most appropriate inquiry
approach to answer their research questions

Quantitative and qualitative studies are not contradictory, but complementary. Both develop new
knowledge for solving research problems.
Quantitative research has a positivist paradigm, in
which the world to be researched is viewed as an
objective reality, but qualitative research has a naturalistic paradigm, in which the world to be researched
is viewed as a socially constructed subjective reality.
Qualitative research provides an opportunity to
generate and explain models and theories inductively,
whereas quantitative research provides an opportunity
to test theories deductively.
When there is little knowledge about the phenomenon
of interest, qualitative approaches are suggested to
explore and understand the phenomenon.
In quantitative research, the accuracy of the research
results depends on the validity and reliability of the
measurement tools, whereas in qualitative research
the trustworthiness of the research findings heavily
relies on the researcher as a tool, and hence participants should verify their findings.
Quantitative researchers rely on numerical values
obtained from statistical procedures and their corresponding p values, whereas qualitative researchers rely
on excerpts from the actual voice of participants to
describe and support the identified themes.
All research must consider essential ethical principles
to ensure that participants are not harmed, either in the
process of data collection or by the presentation of
results.

Correspondence: Dr. Mohsen Tavakol, Medical Education Unit, University of Nottingham, Nottingham NG7 2UH, UK. Tel: +44(0)115 823 0014;
E-mail: mohsen.tavakol@nottingham.ac.uk

746

ISSN 0142-159X print/ISSN 1466-187X online/14/9074611 2014 Informa UK Ltd.

DOI: 10.3109/0142159X.2014.915298

Quantitative and qualitative methods in medical education research

What is medical education

research?

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

Research is . . . investigation or experimentation aimed at the

discovery and interpretation of facts, revision of accepted
theories or laws in the light of new facts, or practical
application of such new or revised theories or laws
(Merriam-Webster 2013). The ultimate goal of research is to
gain new knowledge that can then be added to a body of
existing knowledge in order to develop new insights and
create more useful knowledge to solve a problem. Medical
education research is a careful or systematic study designed
to answer the fundamental questions raised by medical
educators in order to make educational decisions that can
be based on rigorous research-based findings rather than
personal experiences.

Understanding of the knowledge

construction process
Paradigms
Medical educators are always faced with questions in the real
world, such as why do students struggle to learn genetics or
does using a video improve learning? They use specific
methods based on their own views of the world in order
to find out the best answers to these questions. In the
terminology of research, a paradigm is a comprehensive
belief system or a worldview that provides a general perspective or framework to guide an understanding of the phenomenon under investigation. Hence, paradigms direct medical
education researchers to employ the optimal methodological
techniques given the nature of the phenomenon under
study. According to Guba (1990), the paradigms that are
adopted by educators respond to three questions: (a) What is
the nature of reality (known as ontology or metaphysics)?
(b) What is the nature of knowledge, its limitations and its
relationship to the researcher (known as epistemology)?
(c) How should the researcher go about finding out knowledge (known as methodology, by which the researcher

chooses to conduct the investigation of the phenomenon)? For

example, researchers may employ a cross-sectional design
with one of the quantitative research traditions to answer their
questions. The ontological and the epistemological questions
focus on philosophical issues underlying research paradigms.
In medical education research, there are a variety of research
questions that may be raised by medical educators. Answering
these questions requires the use of different types of research
paradigms. Two main paradigms that guide disciplined inquiry
in medical education, the positivist paradigm and the naturalistic paradigm, are discussed below.

The positivist paradigm

Epistemologically, over the last few centuries, the positivist
paradigm was dominant to create new knowledge. It
was introduced by Descartes in 1637. For Descartes, there
is an objective reality that is directly observable and this can
be measured using mathematical models that can predict
future events. He believed that the researcher must distance
him/herself from the participant to avoid any distortion of
the interpretation of the findings from the study. Positivists
believe that objective collection of data and its analysis
must be independent of the opinions of the researcher. In
Table 1, the ontological, epistemological and methodological
assumptions of the positivist approach to research are
summarised.
Within positivism, a hypothesis is derived from a theory
and then empirically tested and replicated by a neutral
researcher. Based on the result of a statistical hypothesis test,
the researcher identifies the relationship between cause and
effect within a value-free inquiry (Denzin & Lincoln 2011).
According to positivism, there is an objective knowledge that is
to be discovered and human beings cannot socially construct
this knowledge. The generalisation of the study results to
another situation is possible within a positivism paradigm; the
nature of reality is fixed, it is single (i.e. the study results either
support or reject a particular hypothesis), and it is measurable
(Denzin & Lincoln 2011; Rubin & Rubin 2012). A positivist
researcher will neutrally rely on statistical inferences, and if the

Table 1. Some assumptions of the positivist and constructivist approaches.

Type of assumption
Epistemology (what is the relationship
between the researcher and
knowledge)

Ontology (what is the nature of reality)

Methodology (What is the research
process?)

Positivist approach (quantitative)

Constructivist approach (qualitative)

Knowledge is uncovered by detached scientific obser- Knowledge is socially constructed through interaction of
the researcher with research participants. The values
vations. The reality is independent of any opinions of
of both the researcher and the research participants
the researcher. The researcher tries to minimise
contribute to knowledge, with there is a lack of
subjectivity and to maximise objectivity
neutrality and objectivity.
The reality is singular. Reality is constructed based on Multiple realities exist. Each study participant has a
cause and effect inferences.
different view on the phenomenon being studied.
Deductive reasoning: Statistical hypothesis testing
Inductive reasoning: theory or hypothesis construction
Objective and measurable
Subjective and non-measurable
Validation of theories
Explore participants experiences
Prediction and estimation
Provide rich description of the phenomenon being
Identifying associations between variables
investigated
Generalization from samples to population
Generate hypothesis or theory
Rule-bound
Generalisation does not matter
Statistical analyses
Context-bound
Internal and external validity
Sample size is small
Sample is large or random

747

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

M. Tavakol & J. Sandars

study results do not fit a theory/model, then the theory/model

can be modified and subsequently tested using statistical
procedures.
Positivism was criticised by post-positivists and in the
last quarter of the twentieth century was rapidly deflated
(Alvesson & Skoldberg 2009). Although proponents of the
positivist approach believe that there is a fixed and objective
reality that can be investigated, post-positivists argue that
the absolute reality of knowledge can never be obtained, it
is only estimated. Post-positivism considers that the reality
is captured based on multiple methods and researchers seek
to test, verify and refine theories to understand the
world (Creswell 2014). Post-positivist researchers begin
with a theory, then collect data in order to either support or
refute the theory, and then make necessary changes and
collect further data to check on whether the theory is
supported or refuted (Denzin & Lincoln 2011; Creswell
2014). Thus the assumptions of post-positivist mostly support
quantitative inquiry approaches rather than qualitative inquiry
approaches.

The naturalistic paradigm

A different perspective to understand the world is the
naturalistic view, otherwise known as constructivism, and is
associated with qualitative inquiry approaches. This perspective is typically considered as an approach to qualitative
inquiry (Lincoln & Guba 1985; Creswell 2014). The constructivist paradigm, a social movement opposed to positivism,
began with writers such as Weber and Kant (Polit & Beck
2014). According to the constructivist epistemology,
knowledge is the result of a dialogical process between the
self-understanding person and that which is encountered,
whether a text, a work of art, or the meaningful expression of
another person (Smith 1990). For constructivist researchers,
individuals do not passively receive knowledge, but they
actively construct knowledge through engagement with each
other and the social world they are living in. Uncovering and
understanding the social world of individuals thus leads to the
production of meaningful knowledge. The epistemological
perspective of constructivism has a focus on the meaningmaking activity of an individuals mind (Crotty 1998; Ritchie
et al. 2013). In Table 1, the ontological, epistemological and
methodological assumptions of the constructivist paradigm to
research are summarised.
Ontologically, from the constructionist perspective, reality
is not a single (one knowable) reality, but there are multiple
realties which are constructed by those who participate in the
study and this reality is negotiated with the research participants. As previously mentioned, positivist researchers believe
that the individual parts of reality are not interrelated and can
be separated into separate dimensions. However, naturalists
believe that the reality is a whole and is not divided into its
parts. For example, the parts of a whole cloth (as a reality) are
interrelated and inseparable. By removing part of the cloth, we
actually destroy the cloth (its meaning) (Erlandson et al. 1993).
From an epistemological perspective, constructivist researchers collect data subjectively to explore a single overall
dimension of a phenomenon so that its true meaning is
748

captured rather than reducing it in a number of different

individual dimensions of the phenomenon. Additionally,
constructivist researchers consider that inquiry and knowledge
are value-laden since the researchers beliefs highly influence the interpretation of knowledge (Tashakkori & Teddlie
1998; Griffin & Museus 2011). In contrast to positivists, which
have a deductive approach to analysis (i.e. they choose a
theory first and based on that theory they formulate
hypotheses to test), the approach of constructivist researchers
is inductive (i.e. the researcher begins with the participants
perspective and then a hypothesis or theory is created which
is grounded in the real-life experiences of participants).
This emergent theory illuminates the phenomenon under
investigation.

Inductive and deductive

approaches
Knowledge is created, based on either the inductive or
deductive approach. As previously pointed out, qualitative
researchers use the inductive approach to generate knowledge
whereas quantitative researchers use the deductive approach
to generate knowledge. The process of the inductive approach
(a bottom-up method of analysis) begins with exploring the
specific details of participants experience and then gradually
moves to more general principles of the phenomenon being
investigated (Liehr & Smith 2002). For example, suppose a
clinical educator is interested in exploring the experiences of
medical students in problem-based learning (PBL). The clinical
educator can use an inductive approach and by using
qualitative research methods, such as interviewing students
who have experienced the PBL approach, he/she can generate
new understanding and theory about the PBL experience.
Each student has their own specific experiences but interviewing several students will identify several common themes
across the students. The process of the deductive approach,
on the other hand, begins with formulating a research
hypothesis about the phenomenon of interest (which is
usually based on a theory). The hypothesis is then tested
using statistical procedures to support or refute this hypothesis.
Causal explanations, generalisation and prediction may be
made based on these statistical procedures. For example,
suppose the same clinical educator has formulated a hypothesis entitled the PBL group will score higher than the non-PBL
group in the communication skills course. The clinical
educator needs to consider theories that have been previously
developed by qualitative researchers to develop the assessment questions and then use quantitative research methods to
collect and analyse data. Therefore, quantitative researchers
test theories in order to generalise the study results to the target
population. Both approaches are important for generating
knowledge and the choice is based on the question being
investigated.

The role of theory in research

A theory is a set of interrelated constructs (concepts),
definitions, and propositions that present a systematic view

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

Quantitative and qualitative methods in medical education research

of phenomena by specifying relations among variables, with

the purpose of explaining and predicting the phenomena
(Kerlinger 1970). Fundamentally, a theory is an idea, a guess,
or a speculation, which may account for reality. Theories
should guide the research process both in qualitative and
quantitative research methods (Morse & Field 1995).
Qualitative researchers, use the inductive approach to
research and explore the observed data for the patterns
and relationships and then develops and tests hypotheses
to generate theory or uses developed theories to explain
the data (Morse & Field 1995). In this approach, research
questions are created by the qualitative researcher, and
then data are collected in the participants setting. Data
analysis is inductively built from specific (particulars) to
general themes (generating categories and themes). The
themes are finally interpreted by the researcher (Creswell
2014). However, sometimes qualitative researchers use a
deductive approach in the initial stage of qualitative
data analysis. They can develop a template (codebook) that
uses a theoretical framework in order to organise the qualitative dataset for interpretation. However, these categories may
not accurately reflect the participants views of the phenomenon under investigation. An interesting example of a hybrid
approach of inductive and deductive approach was well
illustrated by Fereday & Muir-Cochrane (2006).
On the other hand, quantitative researchers formulate a
research hypothesis deductively from an existing theory, and
then the research hypothesis is tested by gathering data. Based
on the statistical procedures used for the hypothesis testing,
the existing theory is either revised or supported. For
example, based on the humanistic education theory (by Carl
Rogers), a researcher may hypothesise that small group
teaching is more effective than large group teaching and this
hypothesis can be tested using statistical procedure (Lodico
et al. 2010).
Quantitative researchers who do not employ a theoretical framework for their own research study, particularly
those who wish to establish cause-effect relationships,
may struggle to explain why some independent variables
influence the dependent variables (Kawulich 2009). In quantitative studies, theory-driven investigations are essential for
the generalisation of the study results. Unfortunately, in many
medical educational papers, the research question or hypothesis is not connected to a theoretical framework.

Concepts, constructs and

variables
In qualitative studies, the building blocks of a theory are called
concepts (Brown 2010; Polit & Beck 2014). Concepts are
abstractions of particular characteristics of human behaviour,
such as empathy, motivation and pain (Polit & Beck 2014).
Researchers are unable to directly observe concepts in the real
world but can measure them indirectly as a construct. For
empathy to be measured as a construct, researchers need to
identify the behavioural manifestations of empathy that can be
considered as proxies of empathy, such as summarising the
feelings that are expressed by participants. The terms of

concept and construct are often used interchangeably in

research.
A variable is a concept, and as its name suggests, is
something that is likely to vary. From a quantitative point of
view, a concept is observable and measurable and takes
different values. For example, age, gender, teaching
methods all are variables as they vary from one individual
to another. Quantitative researchers are interested in investigating how or why phenomena vary, and also how the
variation in a variable is explained by the variation in
another variable. As an example, consider the variable of
learning and a research study that wishes to investigate
what factors can affect student learning. Motivation as a
variable may be investigated as a learning factor by the
researchers. Quantitative researchers quantify student performance, for example, ranging from 0 to 100. It is noteworthy to mention that if every student obtained a mark of 60,
student performance would not be a variable, it would be a
constant. Qualitative researchers do not quantify a variable.
For example, student performance could be reported using
qualitative words, such as inferior, poor, borderline, satisfactory, good and excellent.

Dependent and independent variables

Quantitative researchers make a link between the basic
building blocks of theory and the basic unit of scientific
studies in order to establish the cause and effect relationships
between variables. For example, does an educational
intervention produce improvement in the reliability of
OSCEs? In this example, researchers face a cause and effect
relationship between educational intervention and the
improvement of OSCEs. The presumed cause is the independent variable (sometimes called the exposure or predictor)
whereas the presumed effect is the dependent variable
(sometimes called the response or outcome). Quantitative
researchers are interested in knowing how the independent
variable causes the change in the dependent variable, especially if the independent variable predicts the dependent
variable (Brown 2010). Sometimes it is very difficult to decide
which of two variables in a study is the independent variable
and which is the dependent variable (McBurney & White
2010). As an example, consider there is an association
between drug education programmes and medication compliance. It is very difficult to conclude whether drug education
programmes cause medication adherence or whether a
predisposition to medication adherence causes people to
adhere to a medication regimen. In medical education
research, there are many confounding factors that can
influence the dependent variable (outcome). Sometimes
researchers are unable to manipulate independent variables
in order to see its effect on the dependent variable. Examples
include age, gender and year on a medical school programme.
Qualitative researchers, on the other hand, are not interested
in quantifying associations and relationships, or in seeking
cause and effect connections. They are interested in similarities
and differences in patterns of association in order to
explore the underlying meanings of the phenomena under
investigation.

749

M. Tavakol & J. Sandars

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

Debate over the quality of

qualitative and quantitative
research
At the beginning of this Guide, we discussed philosophical
perspectives of qualitative and quantitative research methods.
When we are speaking of research methods, most medical
educators and clinicians think of research studies that have a
large sample size, and are randomly taken from the population
of interest. They think how to randomly assign their study
participants to groups (intervention and non-intervention
group). They also think of gathering numerical data in order
to use statistical procedures to produce study results. Although
research studies that follow these steps in the research process
can be useful, it is not enough to produce knowledge about
reality, especially where situations are examined through the
eyes of the participants (Cohen et al. 2008). For example, how
clerkship students interact with the parents of unconscious
children in hospital, what are the processes and strategies of
clinical reasoning used by the students to produce treatment?
(Khatami et al. 2012) or medical students understanding of
empathy (Tavakol et al. 2012). Such social situations can be
explored best by a researcher who integrates him/herself in
the situation and obtains ideas, feelings, expectations,
perceptions, experiences and behaviour patterns from the
participants point of view (Brown 2014). Perhaps more
importantly, when researchers have little knowledge about
new phenomena or new meanings of phenomena, qualitative
inquiry methods are the best for gaining a deeper understanding of the phenomenon from the participants perspective
(Trice & Bloom 2014). In addition, when a theory is missing,
qualitative studies can be used to generate theory (Leedy &
Ormrod 2005). Quantitative researchers need to be guided by
theories that are developed by the qualitative researchers.
Qualitative study results can shed light on phenomena that are
not accurately understood in teaching and practice. In
addition, qualitative research methods are the most humanistic and person-centred way of discovering and uncovering
thoughts an action of human beings (Halloway & Biley 2011).
The underlying distinctions between quantitative and
qualitative methods and their epistemological and ontological
considerations have contributed to a better understanding of
research issues. However, qualitative inquiry approaches have
often been criticised by quantitative researchers (who view
phenomena independent of the behaviour of the researcher),
who consider quantitative research results to be more
objective and value-free. In medicine, qualitative research
continues to be devalued, and is considered to be subjective,
biased, and opinion based (Morse 2006, 2011). Quantitative
researchers also argue that qualitative research does not have a
strong design, and hence they do not recommend it for
funding (Morse 2006). There are arguments against the quality
of the knowledge produced by qualitative inquiry: quality in
qualitative research is a mystery to many health services
researchers (Dingwall et al. 1998). Perhaps more importantly,
the highest level of evidence has been awarded to quantitative
research and the importance of qualitative research has been
undermined and ranked at the lowest level of evidence
750

(Cochrane 1972, 1989; Morse 2011). It should be noted that,

however, chapter 20 of the Cochrane Intervention handbook
has outlined how qualitative studies can contribute to
Cochrane Intervention reviews. The handbook stated there
are many methods of qualitative evidence synthesis that are
appropriate to the aims and scope of Cochrane Intervention
reviews. For example, qualitative researchers should systematically review related individual papers in order to address
important outcomes and questions directly related to the
effectiveness review. For more details, see the handbook
(Higgins & Green 2011). Although the number of papers in
relation to qualitative evidence syntheses is growing (McInnes
et al. 2011; Marshall et al. 2012), systematic reviews of
qualitative studies are not well established in comparison
with quantitative studies.
Although the issue of the nature of the knowledge
produced by qualitative researchers has been raised by
quantitative researchers, the discrimination against qualitative
methods continues (Morse 2006), qualitative researchers can
address the issue of quality in their research. . . its methods,
can and do, enrich our knowledge of health and health care
(Mays & Pope 2000). Criticisms about qualitative research
methods are based on a lack of understanding of what is the
purpose of qualitative research studies to produce knowledge
for medical educators and clinicians (Brown 2014). According
to Morse the fact that so little is generally known about what
qualitative inquiry is appalling (Morse 2005, 2006). While
qualitative inquirers gain knowledge of social reality, which is
completely different from quantitative inquiries, both
approaches can produce a wide range of knowledge about
the phenomenon under investigation. There is a growing body
of evidence that suggests the combination of qualitative and
quantitative methods are important (Ashley & Boyd 2006),
although it takes time and effort to understand both styles and
see how they can be complementary (Neuman 2003).
Although quantitative and qualitative inquiry methods each
have different underlying epistemological and ontological
assumptions about the generation of knowledge and reality,
their differences do not make one better or worse than the
other. They are complementary rather than contradictory.
Complex research questions require complex answers which
can be achieved through the integration of qualitative and
quantitative approaches. Qualitative researchers have now
recognised that the perspectives of participants are not enough
per se and multiple forms of evidence are essential. Likewise,
quantitative researchers have realised that the perspectives of
participants can play an essential role in quantitative results
(Taylor 2013). It is perhaps for that reason that mixed methods
research is grown increasingly popular among researchers. We
need to think first that research is about inquiry and that
distinctions between quantitative and qualitative approaches
are arguably reified more by a need to label approaches than
by true differences in in purpose (Newman & Hitchcock
2011). Approaches should act as servants rather than as
rulers and can be considered as different tools to be used
when most appropriate to answer a research question
(Silverman 2010). From a learning perspective, therefore,
medical educators should learn techniques that are used in
both quantitative and qualitative methods.

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

Quantitative and qualitative methods in medical education research

Steps of the research process

Literature review

The quantitative research processes have a linear sequence,

and consists of different steps, beginning with the identification of research questions and ending with a statement
answering those questions (Nieswiadomy 1998; Polit & Beck
2014). The qualitative research process, on the other hand,
tends to have a nonlinear sequence (or an iterative, repeating
or recursive process). For example, qualitative researchers
collect data and analyse them concurrently. They immediately
begin data collection and analysis with the first interview.
The next interview will be planned based on this interview.
This process (i.e. data collection and analysis) continues until
they sense that data saturation is achieved, that is until no new
themes are identified. Unlike qualitative researchers, quantitative researchers analyse their data after all of the data are
gathered. Due to the nature of qualitative studies, they may
have different research process flow diagrams. Figure 1
illustrates the main steps in a quantitative research study.

Reviewing the related literature is an important step of the

research process and the research report. Literature review
informs us about the feasibility of researching the study topic
before proper research begins (Hart 2005). Table 2 shows how
a literature review can contribute to the study topic.
Quantitative researchers conduct a literature review to
gather information on what is already known about the topic
and the methods that have been used to study the topic before
any data are collected. This will enable the researchers to
provide the rationale for the study that they are planning.
Consulting the literature can be useful for both quantitative
and qualitative studies without considering the researchers
paradigm (Mertens 2010). Although some qualitative researchers acknowledge the importance of doing a review literature
prior to commencing their research, some believe that
qualitative researchers should not review the literature
Table 2. The contribution of the literature review to the study
topic (Gillis & Jackson 2002; Hart 2005; Aveyard 2010).

Defining the problem

As seen in Figure1, researchers first need to identify the
problem under investigation (Ary et al. 2006). Researchers
should clearly state why they want to conduct a particular
study. What is the knowledge gap in the field of study that
needs to be closed? What is the importance of the problem?
Researchers need to provide a rational for the study that they
intend to undertake. The research problem should be logically
developed into a discussion of the reasoning behind the study
of interest, and end with a statement of the research question.
Carefully constructed research questions will facilitate the
search for a solution. It is worth mentioning that some
researchers state the purpose of the study instead of the
research question, with the aim of the study at the end of the
introduction.

Figure 1.











Ensures a comprehensive, relevant, clear picture of the studies

available on the study topic;
Identifies the main issues related to the study topic and hence
establishes the importance of the study topic;
Identifies areas where there are consistency or inconsistency in
research results;
Determines what is known and unknown about the study topic (the
knowledge gap);
Identifies experts in the related fields;
Helps construct a theoretical/conceptual framework for the study
topic;
Helps the researcher to plan methods;
Identifies how other researchers have measured and analysed their
data;
Discovers instruments or tools that can be used to measure the study
variables;

The main steps in a quantitative research study.

751

M. Tavakol & J. Sandars

before proceeding with data collection. Opponents argue that

by reviewing the literature, the conceptualisation of the
phenomena under investigation might be contaminated.
They believe that this should be explored based on participants perspectives rather than prior findings (Morse & Field
1995; Polit & Beck 2014). However, others believe that
qualitative researchers should be initially aware of what is
already known about the phenomena under investigation,
but that an additional literature review is also required during
the research to make sense of the data (Marshall & Rossman
2006).

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

Develop a theoretical framework

Another step in the research process (Figure 1) is to develop a
theoretical framework to the research study. Theories operate
as a lens through which to view the phenomena of interest
(Sclater 2012). Theories are generated to describe, predict and
understand the relation between two or more different
concepts in order to construct universal laws. A theoretical
framework (sometimes called a conceptual framework) is a
part or a brief explanation of a theory that researchers can
verify by hypothesis testing or seek answers to research
questions that are driven from theory. A theoretical framework
underpins the research problem under investigation, formulates the research questions or hypotheses, guides the data
collection process, explains and predicts the underlying
cause the phenomena under study (Reeves et al. 2008;
Creswell 2014). Moan and Rise, for example, tested the use
of the Theory of Planned Behaviour (Ajzen 2005) for explaining and predicting students intentions to quit smoking and
their subsequent behaviour six months later (Moan & Rise
2005).
In quantitative studies, the theoretical framework is deductively established before data are collected. Quantitative
researchers report explicitly the theoretical framework of
their studies in the introduction section, immediately after
addressing the research questions or hypotheses. Qualitative
researchers, on the other hand, generate, explain and understand a theory inductively during the research study. Theories
and hypotheses are inductively generated after data analysis
has begun (Morse & Field 1995). Carefully designed studies
use a theoretical framework in order to guide the phenomenon
to be studied. Quantitative researchers often do not explicitly
discuss the theoretical framework in their reports in comparison to qualitative researchers. However, there are several
educational theories that are relevant for medical education
research, and these can be explored further through the work
by (Kaufim 2003).

Constructing hypotheses
A hypothesis predicts the relationship between the independent variable and the dependent variable. Some quantitative
studies explicitly address one or more research hypotheses but
qualitative studies, on the other hand, do not have research
hypotheses. This is because qualitative researchers want the
inquiry to be guided by participants viewpoints rather than by
their own hunches (Polit & Beck 2014). As previously stated,
hypotheses are sometimes formulated from theories and often
these are formulated from a large body of evidence. For
example, a study hypothesised that women will show higher
levels of empathy than will men as this hypothesis is
consistent with the previous studies (Toussaint & Webb
2005). Descriptive studies do not have a hypothesis.
A hypothesis contains the population, the independent
variable, the dependent variable and a predicted relationship
between them. Hypotheses are dichotomised into two groups:
directional or non-directional. In a directional hypothesis,
researchers can predict the direction of the association, either
positively or negatively. In a non-directional hypothesis,
researchers do not specify the direction of the association.
Table 3 shows some examples of directional and non-direction
hypotheses.
You may recall from statistics courses that there are two
types of hypotheses: null hypotheses and alternative hypotheses (sometimes called research hypothesis). Researchers want
to know whether or not their theories can be supported
when subjected to the rigors of scientific investigations
(Daniel 2005). The null hypothesis is a hypothesis of no
difference (i.e. there is no difference between the independent
and the dependent variables). The null hypothesis is either
rejected or accepted by statistical procedures. If the null
hypothesis is rejected, the alternative hypothesis is supported
as the available data are incompatible with the null hypothesis
(Daniel 2005). Hypotheses neither are proved nor disproved,
but they are either supported (accepted) or rejected.
Study results are not always definite and researchers maybe
unable to prove or disprove research hypotheses (Polit & Beck
2014). For example, consider the hypothesis that tall medical
students show more empathy than shorter students. If a
sample of students shows that tall medical students have
higher levels of empathy than short ones, we cannot conclude
that height is related to a students empathy since in realty
there is no relationship between height and empathy with
patients. There are also other influences, including sources of
measurement error that can influence statistical inferences,
such as the accuracy of measures and factors that are not
under the control of the researchers.

Table 3. Directional and non-directional hypotheses.

Type of hypothesis
Directional
Non-directional
Directional
Non-direction
Directional
Non-directional

752

Hypothesis
PBL students are better able than non-PBL students in disclosing bad news to patients with life-threatening illness
There is an association between PBL student and non-PBL students in disclosing bad news to patients with life-threatening illness
OSCEs better measure medical students clinical performance than do mini-CEXs
There is a relationship between OSCEs and mini-CEXs with respect to measuring medical students clinical performance.
Female medical students have more positive attitudes towards epidemiology training than male medical students.
Female medical students differ from male medical students with respect to epidemiology training.

Quantitative and qualitative methods in medical education research

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

Quantitative research designs

Quantitative research designs differ from qualitative research.
Quantitative research designs are classified into three groups:
experimental designs, quasi-experimental designs and surveys
(Creswell 2013). Experimental designs explain the cause and
effect relationship between the independent and dependent
variables. Three important features of experimental designs
(or Randomised Controlled Trials, RCTs) are: control, manipulation or intervention, and randomisation. These features help
quantitative researchers to ensure that the study outcome is
caused by a particular intervention rather than by other
variables. Researchers control variables (major factors) which
may influence the study outcome as they want to ensure that
the study outcome is caused by the intervention rather than
other variables (sometimes called extraneous or confounding
variables). Quantitative researchers also manipulate the independent variable (cause) and then they measure its consequence on the dependent variable (effect). Another feature of
experimental designs is randomisation. Researchers randomly
assign study participants into experimental or control groups.
The experimental group is exposed to a treatment, but the
control group does not receive any treatment. Randomisation
means that each participant has an equal chance of being
selected to either group. By taking these features into account,
researchers are able to generalise the study outcome to the
population of interest. To illustrate, consider a medical
educator investigating the effect of high-fidelity simulation
(independent variable) on medical students knowledge and
clinical performance (dependent variable). In this example,
simulation may influence knowledge and clinical performance. The medical educator can assign students to two groups
randomly (i.e. each student has an equal chance of getting the
experiment or control condition). Before the intervention was
experienced, the educator can assess (pre-test) the knowledge
and clinical performance of students in each group.
Subsequently, students in the experimental group can be
exposed to the simulators whereas students in the control
group are exposed to a traditional intervention which was
similar to the information covered in the simulator. The
educator then can reassess the knowledge acquisition and
clinical performance in both groups (post-test). Finally, the
educator can compare the differences between the post-test
scores of the two groups in order to identify the effect of the
simulator on the knowledge acquisition and clinical performance, while the educator considers other variables (Figure 2).

This experimental design is called a randomised control-group

pre-test-post-test design (sometimes called an RCT design).
Sometimes in medical education, educators are unable to
assign participants randomly to experimental and control
groups, but they want to assess a particular intervention. If this
is the case, educators need to follow a quasi-experimental
design. Sometimes this design is called trials without randomisation. There are different types of quasi-experimental designs
(Shadish et al. 2002; Harris et al. 2006) but two important
designs will be discussed here: the non-equivalent
control group pre-test-post-test design and one group pretest-post-test design.
The non-equivalent control group pre-test-post-test design
(sometimes called a comparison study) compares two or more
groups of participants before and after a particular intervention
without assigning participants to the experimental and control
groups. This design is the same as the pre-test-post-test
experimental design (see above), except participants are not
grouped (the experimental and control groups) randomly
(Polit & Beck 2014). As an educational researcher, you can
compare students at medical school A as the experimental
group with students at medical school B as the control group.
The experimental group attend clinical simulation (intervention) activities whereas the control group attend the normal
clinical programme for three weeks. Before the intervention is
implemented, the clinical performance of students in two
medical schools is assessed as a baseline. Data on clinical
performance in both medical schools after 3 weeks when the
intervention is made are collected to see the effect of the
simulation on clinical performance.
The second quasi-experimental design is the one group
pre-test-post-test design. Here, as in the previous example, the
educator assesses the knowledge and clinical performance of
all students before simulation training. This time, however, all
students are exposed to simulation training. The difference
between the pre-post test scores may be an indication of the
change in the use of simulators.
As groups (the experimental and control/comparison
groups) are not randomly allocated, it is impossible to say
the groups are equivalent at the beginning of the study. Hence,
the study findings of quasi-experimental designs are less
decisive in comparison with experimental design. In the other
words, if the clinical performance of students in groups is not
equivalent, the effects of the intervention will not be clear.
Sometimes researchers are not able to randomise participants into groups or they cannot manipulate the independent

Figure 2. Randomised control-group pre-testpost-test design.

753

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

M. Tavakol & J. Sandars

variable in order to observe its effect on the dependent

variable. This could be due to ethical considerations or to
factors out of the control of the researcher. For example,
consider gender as an independent variable; if researchers
want to compare female and male students about a particular
phenomenon, they cannot manipulate gender and they cannot
randomly assign students to be either female or male. When
the researcher has no control on the independent variable, the
study is non-experimental (sometimes called an observation
design). This design is widely used in medical education
research. There are different non-experimental designs, but
three study designs which are commonly used in medical
education research will be discussed: correlational studies,
cross-sectional studies and longitudinal studies.
Sometimes when researchers wish to establish a relationship between the variables in their study, but they cannot
design an experimental or quasi-experimental study, they plan
for correlational studies. In correlational studies, researchers
make a claim about the relationship between variables in
theories or models. For example, what is the relationship
between student ability and their score on the UK Clinical
Aptitude Test (UKCAT)? By calculating the correlation between
students marks and UKCAT scores, researchers can address
the association between student ability and UKCAT.
In cross-sectional studies, researchers collect data at one
point in time or over a short period (Kevin 2006). In this
design, independent and dependent variables are identified in
a given population and then the associations between them
are determined. For example, medical educators may be
interested to determine the association between medical
student year and empathy. If the researchers have knowledge
of the medical student year and the empathy scores of
students, they will be able to identify the relationship using
statistical procedures. In epidemiological studies, retrospective
studies or prevalence studies are usually cross-sectional. Data
on the independent and dependent (outcome) variables are
collected simultaneously (Polit & Beck 2014).
Sometimes study participants are followed over time and
data are collected at multiple follow-up times. This is called a
longitudinal study. In other words, the same participants are
measured at each point of the time scale. Such studies are
always concerned with individual change (Goldstein 1968). As
an example, a longitudinal study was designed to examine
changes in medical students empathy during medical school
and to determine when the most significant changes occurs
(Hojat et al. 2009). In epidemiological studies, prospective
studies or incidence studies are usually longitudinal. Cohort
studies are longitudinal studies which involve a large sample
size. The interested reader is referred to the book A study
guide to epidemiology and biostatistics for a discussion of the
epidemiological studies (Hebel & McCarter 2012).

Conclusions
This part of the Guide has explained how quantitative and
qualitative methods can be used in medical education research
to produce new knowledge. From a learning perspective,
therefore, medical educators should learn techniques that are
754

used in both quantitative and qualitative methods. Although

the philosophical assumptions of quantitative research differ
from qualitative research, they certainly do not contradict each
other, but they are complementary. From a quantitative
perspective, a concept is observable and measurable, and
is analysed using statistical procedures. From a qualitative
point of view, the phenomenon of interest does not quantify,
but the qualitative researcher provides a rich description of
the phenomenon to be studied. In quantitative studies, the
research process is linear and deductive, whereas in qualitative
studies the research process is recursive and inductive. In this
part, we have also explained some essential steps in the
research process and quantitative methods.

Notes on Contributors
MOHSEN TAVAKOL, PhD, MClinEd, is a Lecturer in Psychometrics. His
main interests are in medical education assessment, psychometric analysis
(Classical Test Theory, Generalisability theory, Item Response Theory
Models), robust statistical methods, multivariate statistics, quantitative and
qualitative research methods and communication skills.
JOHN SANDARS, MD, MSc, FRCGP, FAcadMEd, MRCP, Cert Ed., is an
Associate Professor at the Leeds Institute of Medical Education, University
of Leeds, UK. His main interest is in developing teaching and learning by a
scholarship approach, in which real life problems are investigated by both
quantitative and qualitative research.

Declaration of interest: The authors report no conflicts of

interest. The authors alone are responsible for the content and
writing of the article.

References
Ajzen I. 2005. Attitudes, personality, and behaviour. New York: Open
University Press/MaGraw Hill.
Alvesson M, Skoldberg K. 2009. Reflexive methodology: New vistas for
qualitative research. Thousand Okas: CA, SAGE.
Ary D, Jacobs L, Sorensen C, Razavieh A. 2006. Introduction to research in
education. Belmont, CA: Wasdsworth.
Ashley P, Boyd B. 2006. Quantitative and qualitative approaches to
research in environmental management. Australasian J Environ Manage
13:7078.
Atkinson P, Pugsley L. 2005. Making sense of ethnography and medical
education. Med Educ 39:228234.
Aveyard H. 2010. Doing a literature review in health and social science
care: A practical guide. Glasgow: The McGrawHill Companies.
Bower E, Scambler S. 2007. The contributions of qualitative research
towards dental public health practice. Community Dentistry and Oral
Epidemiology 35:1619.
Brown C. 2010. Conceptualizing research. In: Aparasus R, editor. Research
methods for pharmaceutical practice and policy. London:
Pharmaceutical Press. pp 1735.
Brown S. 2014. Evidence-based nursing. Burlington, MA: Jones & Bartlett
Learning.
Buckley G. 1998. Partial truths-research papers in medical education. Med
Edu 32:12.
Cochrane A. 1972. Effectiveness and efficiency: Random reflection on
health services. London: Nuffield Prvincial Hospitals Trus (original
publication, 1972).
Cohen L, Manion L, Morrison K. 2008. Research methods in education.
London: Routledge.
Creswell J. 2013. Research design: Qualitative, quantitative, and mixed
methods approaches. Thousands Okas: SAGE.

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

Quantitative and qualitative methods in medical education research

Creswell J. 2014. Research design: qualitative, quantitative, and mixed

methods approaches. Thousand Oaks: SAGE.
Crotty M. 1998. The foundations of social research: meaning and
perspective in the research process. London: SAGE.
Daniel W. 2005. Biostatistics. Hoboken, NJ: Wiley.
Denzin N, Lincoln Y. 2011. The discipline and practice of qualitative
research. In: Denzin N, Lincoln Y, editors. The SAGE handbook of
qualitative research. Thousands Oaks: SAGE. pp 119.
Dingwall R, Murphy E, Watson P, Greathbatch D, Parker S. 1998. Catching
gold fish: quality in qualitative research. J Health Serv Res Policy 3:
167172.
Erlandson D, Harris E, Skipper B, Allen S. 1993. Doing naturalistic inquiry.
London: SAGE.
Fereday J, Muir-Chocrane E. 2006. Demonstrating rigour using thematic
analysis: Hybrid approach of inductive and deductive coding and
theme development [Online]. [Accessed July 2012] Available from
http://www.ualberta.ca/iiqm/backissues/5_1/pdf/fereday.pdf.
Gillis A, Jackson W. 2002. Research for nurses: Methods and interpretation.
Philadelphia: F.A. Davis Company.
Goldstein H. 1968. Longitudinal studies and the measurement of change.
Statistician 18:93117.
Griffin K, Museus S. 2011. Application of mixed-methods approaches to
higher education and intersectional analyses. In: Griffin K, Museus S,
editors. Using mixed -methods approaches to study intersecrionality in
higher education. Ann Arbor: Wiley. pp 1526.
Guba E. 1990. The alternative paradigm dialog. In: Guba E, editor. The
Paradigm dialoge. London: Sage Publication. pp 1527.
Halloway I, Biley F. 2011. Being a qualitative researcher. Qual Health Res
21:968975.
Harris A, Mcgregor J, Perencevich E, Furuno J, Zhu J, Peterson D,
Finkeistein J. 2006. The use and interpretation of quasi-experimental
studies in medical informatics. J Am Med Inform 13:1623.
Hart C. 2005. Doing a literature review. London. SAGE.
Hebel J, McCarter R. 2012. A study guide to epidemiology and biostatistics.
Bulington, MA: Jones and Bartlett.
Higgins J, Green S. 2011. Cochrane handbook for systematic reviews
of interventions [Online]. The Cochrane Collaboration. [Accessed
November 2013] Available from http://handbook.cochrane.org/
index.htm#chapter_20/20_qualitative_research_and_cochrane_
reviews.htm.
Hojat M, Maxwell K, Brainard G, Herrine S, Isenberg G, Veloski J, Gonnella
J. 2009. The devil is the third year: A longitudinal study of erosion of
empathy in medical school. Acad Med 84:118291.
Kaufim D. 2003. Applying educational theories in practice. BMJ 25:213216.
Kawulich B. 2009. The role of theory in research. In: Garner M, Wagner C,
Kawulich B, editors. Teaching research methods in social sciences.
Surrey: Aghgate Publishing Limited.
Kerlinger F. 1970. Foundations of behavioural research. New Yourk: Holt,
Rinehart & Winston.
Kevin K. 2006. Study design III: Cross-sectional studies. Evidence-based
Dentistry 7:2425.
Khatami S, Macentee M, Pratt D, Collins J. 2012. Clinical reasoning in
dentistry: A conceptual framework for dental education. J Dent Educ
76:11161128.
Leedy P, Ormrod J. 2005. Practical research planning and design. New
Jersey: Pearson Merrill Prentice Hall.
Liehr P, Smith M. 2002. Theoretical frameworks. In: Lobiondo-Wood G,
Haber J, editors. Nursing research: Methods, critical appraisal and
utilization. St. Louis: Mosby. pp 107120.
Lincoln Y, Guba E. 1985. Naturalistic inquiry. Beverly Hills, CA: SAGE.
Lodico M, Spaulding D, Voegtle K. 2010. Methods in educational research:
from theory to practice. San Faransico: John Wiley & Sons.
Marshall C, Rossman G. 2006. Designing qualitative research. Thousand
Oaks: Sage Publication.
Marshall I, Wolfe C, Mckevitt C. 2012. Lay perspectives on hypertension and
drug adherence: systematic review of qualitative research. BMJ 344:
e3953.
Mays N, Pope C. 2000. Assessing quality in qualitative research. BMJ 320:
5052.
McBurney D, White T. 2010. Research methods. Bamont, CA: Wadsworth.

McEwan M, Espie C, Metcalfe J. 2004. A systematic review of the

contribution of qualitative research to the study of quality of life in
children and adolescents with epilepsy. Seizure 13:314.
McInnes E, Seers K, Tutton L. 2011. Older peoples views in relation to risk
of falling the need for intervention: A metaethnography. J Adv Nurs
67:25252536.
Merriam-Webster. 2013. Research [Online]. [Accessed 1 December 2013]
Available from http://www.merriam-webster.com/dictionary/research.
Mertens D. 2010. Research and evaluation in education and psychology.
London: SAGE.
Moan I, Rise J. 2005. Quitting smoking: Applying and extended version
of the theory of planned behaviour to predict intention and behaviour.
J Appl Biobehav Res 10:3968.
Morse J. 2005. What is qualitative research? Qual Health Res 15:
859860.
Morse J. 2006. Reconceptualising qualitative evidence. Qual Health Res 16:
415422.
Morse J. 2011. What is qualitative health research. In: Denzin N, Lincoln Y,
editors). SAGE handbook of qualitative research. Thousand Oaks:
SAGE. pp 401414.
Morse J, Field P. 1995. Qualitative research methods for health
professionals. Thousand Oaks: SAGE.
Neuman W. 2003. Social research methods: Qualitative and quantitative
approaches. Boston: Allyan and Bacon.
Newman I, Hitchcock J. 2011. Underlying agreements between qualitative
and quantitative research: the art short and tall of it all. Human
Recourse Dev Rev 10:381394.
Nieswiadomy R. Foundations of nursing research. Stamford: Appelton &
Lange.
Norman G. 2002. Research in medical education: three decades of
progress. BMJ 324:15602.
Ong B, Richardson J. 2006. The contribution of qualitative approaches to
musculoskeletal research. Rheumatology 45:36070.
Polit D, Beck C. 2014. Essentials of nursing research. Philadelphia: Wolters
KluwerjLippincott Williams & Wilkins.
Reeves S, Kuper A, Hodges B. 2008. Qualitative research methodologies:
Ethnography. BMJ 337:a1020.
Ritchie J, Lewis J, Cnicholls C, Ormston R. 2013. Qualitative research
practice: A guide for social science students and researchers. London:
SAGE.
Rothstein W, Phuong L. 2007. Ethical attitudes of nurse, physician, and
unaffiliated members of institutional review boards. J Nurs Scholarship
39:7581.
Rubib H, Rubin I. 2012. Qualitative interviewing: the art of hearing data.
London: SAGE.
Sclater M. 2012. Theorizing from bricolage: researching collaboration in art
and design education. In: Adams J, Cochrane M, Dunne L, editors.
Applying theory to educational research: An introductory approaches
with case studies. Oxford: WileyBlackwell. pp 157176.
Shadish E, Cook T, Campbell D. 2002. Experimental and quasiexperimental designs for generalised casual inference. Boston:
Houghoton Mifflin.
Silverman D. 2010. Doing qualitative research. London, SAGE.
Smith K. 1990. Alternative research paradigms and the problem of
criteria. In: Guba E, editor. The paradigms dialogue. Newbury Park,
CA: SAGE. pp 1671187.
Tashakkori A, Teddlie C. 1998. Mixed methodology: Combining qualitative
and quantitative approaches. London: SAGE.
Tavakol M, Gruppen L. 2010. Using evaluation research to improve medical
education. Clin Teach 7:192196.
Tavakol M, Murphy R, Rahemei-Madeseh M, Torabi S. 2008. The
involvement of clinicians in medical education research. Qual Prim
Care 16:33540.
Tavakol S, Dennick R, Tavakol M. 2012. Medical students understanding of
empathy: a phenomenological study. Med Edu 46:306316.
Taylor B. 2013. Mixed method research. In: Taylor B, Francis K, editors.
Qualitative research in the health sciences: Methodologies, methods
and processes. New York: Routledge. pp 162176.
Toussaint L, Webb J. 2005. Gender differences in the relationship between
empathy and forgiveness. J Soc Psychol 14:673685.

755

M. Tavakol & J. Sandars

Trice L, Bloom K. 2014. PICOT, problem statement, research ouestion,

hypothesis. In: Boswell C, Cannon S, editors. Introduction to nursing
research. Burlington, MA: Jones & Bartlett.

Med Teach Downloaded from informahealthcare.com by University of North Carolina on 09/28/14

For personal use only.

Suggested readings
Alasuutari P, Bickman L, Brannen J. 2008. The SAGE handbook
of social research methods. London: SAGE.
Bryant A, Charmaz K. 2010. The SAGE handbook of
grounded theory (Paperback Edition). London: SAGE.
Corbin J, Strauss A. 2008. Basics of qualitative research:
Grounded theory procedures and techniques. Thousand Oaks:
SAGE.
Creswell J, Clark V. 2011. Designing and conducting mixed
methods research. London: SAGE.
Creswell J. 2011. Educational research: Planning, conducting, and evaluating quantitative and qualitative research.
Boston: Pearson.
Denzin N, Lincoln Y. 2011. The SAGE handbook of
qualitative research. London: SAGE.
Flick U. 2013. The SAGE handbook of qualitative data
analysis. London: SAGE
Grove M, Overton T. 2011. Getting started in
pedagogic research within the STEM disciplines [Online].
The University of Birmingham on behalf of the National
HE STEM Programme. [Accessed 1 February 2014]
Available from http://www.birmingham.ac.uk/Documents/
college-eps/college/stem/getting-started-in-stem-pedagogicoptimised.pdf.
Gruppen L. 2008. Is medical education research hard
or soft research? Adv Health Sci Educ Theory Pract 13:12.
Harland N, Holey H. 2011. Including open-ended questions
in quantitative questionnaires. Int J Theory Rehab 18:482486.
Johnson B, Christnsen L. 2010. Educational research:
Quantitative, qualitative, and mixed approaches. London:
SAGE Publications.

756

Morse J. 2004. Qualitative comparison: Appropriateness,

equivalence, and fit Qual Health Res 14:13231325.
Morse J. 2009. Mixed method design: principles and
procedures. Walnut Creek: Left Coast Press.
Morse J, Dimitroff L, Harper R, Koontz A, Kumra S,
Matthew-Maich N, Mihas PCM. 2011. Considering the qualitative-quantitative language divide. Qual Health Res 21:1302
1303.
Norman G, Eva K. 2014. Quantitative research methods in
medical education. In: Swanwick T, editor. Understanding
medical education: Evidence, theory and practice. Sussex:
Wiley-Blackwell. pp 349369.
Sandelowski M. 2004. Counting cats in Zanzibar. Res Nurs
Health 27:215216.
Silverman D. 2006. Interpreting qualitative data:
Methods for analysing talk, text and interaction. London:
SAGE.
Tavakol M, Wilcox R. 2013. Medical education research: the
application of robust statistical methods. Int J Med Educ
4:9395.
Teddile C, Tashakkori A. 2009. Foundations of
mixed methods research: Integrating quantitative and qualitative approaches in the social and behavioural. London:
SAGE.
Thomas G. 2013. How to do your research project: A guide
for students in education and applied social sciences. London:
SAGE
Trochim W, Donnely J. 2006. The research methods
knowledge base. Ohio: Atomic Dog.
Web Centre for social research methods [Online]. [Accessed
1 February 2014]Available from http://socialresearchmethods.net/.
Wilcox R. 2011. Modern statistics for the social and
behavioural sciences: A practical introduction. Boca Raton:
CRC Press.