Notes from Week 4 discussion on a Mixed Methods approach to research.
From Mixed Methods group precis
Creswell asserts that a researcher should formulate a framework when
preparing a research proposal so that research methods can be developed
that are justifiable to the author and audience. Three frameworks to
support the development of an appropriate proposal are described
(‘approaches to research’): quantitative, qualitative and mixed methods,
as well as three framework elements (‘elements of inquiry’),
alternative knowledge claims, strategies of inquiry and methods.
Three Elements of Inquiry
Alternative Knowledge Claims
Knowledge
claims are the theoretical assumptions made by researchers about how
and what they will learn during their research. These claims can be
described as philosophical assumptions, epistemologies, ontologies or
research methodologies. Knowledge claims can relate to what constitutes
knowledge (ontology), how we know if something is knowledge
(epistemology), what values go into knowledge (axiology), how we write
about knowledge (rhetoric) and the process of studying knowledge
(methodology).
Creswell proceeds to discuss four ‘schools of
thought’ about knowledge claims and aims to reduce these to a practical
context. He outlines four primary alternative knowledge claim positions,
postpositivism, constructivism, advocacy/participatory and pragmatism.
a. Postpositive Knowledge Claims
These
assumptions are primarily associated with scientific research and the
concept of ‘cause and effect’.They are also referred to as quantitative
research, positivist/post positivist research, and empirical science.
Postpositivism means challenging the traditional notion of the absolute
truth and understands that we cannot be “positive” about our claims of
knowledge when studying the behaviour and actions of humans. The
approach adopted by postpositivist researchers is to develop a theory,
collect data to prove/disprove the theory and then make any required
alterations before conducting any additional tests.
b. Social Constructivism
The
assumption that governs social constructivism is that the environment
where the research is conducted gives new meaning and interpretation to
the research based on the historical and social perspective. Human
beings are stakeholders in this research position as they contribute to
constructing the knowledge by their own social experiences and
backgrounds. It recognises that the subjectivity of individuals’
experiences and involves exploration the complexity that results from
this. Researchers use open-ended questioning to capture the
participants’ views and interpret what they say in the relevant social
and historical context to generate meaning. Unlike the postpositive
approach where a researcher starts with a theory which is then tested,
the process of qualitative research involves the inquirer developing a
theory or meaning based on the data collected in the field.
c. Advocacy/participatory
As
a result of the limitations of postpositivist and social constructivist
assumptions, advocacy/participatory approach evolved during 1980s and
1990s. This was based on the opinion that the other theories did not
fully address issues of ‘social justice’ and the marginalised in
society. This approach suggests that research should be linked with
politics, and address contemporary social issues in a collaborative way
to give participants a ‘voice’ and improve their lives. The researcher
advocates for reform, considering social issues like ‘empowerment,
inequality, oppression, domination, suppression and alienation’.
d. Pragmatism
In
this approach the focus is on “problem” and “what works” to solve the
given problem instead of considering the methods used. Researchers using
this approach do not subscribe to a specific philosophy or approach to
research. Therefore researchers have a freedom of choice and employ
elements of qualitative and quantitative approaches to collect data.
They accept that research takes place in a social, historical and
political context and believe that ‘truth is what works at the time’.
Strategies of Inquiry
The Strategies of Inquiry introduced in this paper are 1. Quantitative 2. Qualitative and 3. Mixed Methods.
1.
Quantitative
approach has long been established using data gathered through
experimental practices. Creswell notes two quantitative strategies of
inquiry;
- Experiments: True – with random subject assignment. Quasi-experiments – with nonrandomized design.
- Surveys: Cross-sectional and longitudinal studies. Surveys aim to generalise from a sample to a population.
2. Qualitative approach
methods have gathered momentum over the last 20 years, by incorporating
several approaches such as ethnographies; grounded theory; case
studies, phenomenological and narrative research.
Cresswell outlines the qualitative approaches as follows:-
- Ethnographies;
Researchers aim to explore cultural phenomena by gathering
observational data over time from a specific cultural group. This
involve a flexible process that adapts to the ongoing researched
situation.
- Grounded Theory; Researchers
aim to generate or discover a theory from data systematically obtained
from social research. It draws abstract theory from participants’ views
in a study. The theory involves several stages and collection of data
from different groups with evolving comparisons between data categories.
- Case
Studies; Researchers analyse an actual event to gain detailed
information, through various data collection techniques. There is a
detailed study of a program, process, activity or individuals.
- Phenomenological
Research; Researchers relate their own experiences with how those are
perceived by the participants in a given study. It aims to understand
the “lived experience” of a small number of participants regarding a
phenomenon. This has philosophical and methodological features.
- Narrative
Research; Researchers and collects real-life stories to create a
narrative involving both researcher’s and participants’ views.
3. Mixed Methods
approach. These approaches include elements of qualitative and
quantitative methods e.g combining ‘field’ methods such as interviews
(qualitative) with traditional surveys (quantitative). The premise is
that biases in one method may offset biases in another to produce more
reliable data. This approach led to triangulation methods, which aim to
seek convergence across research methods.
Creswell notes three mixed methods strategies of inquiry:
(i) Sequential procedures: The researcher begins with either a
qualitative or quantitative method, and then develops findings with the
other.
(ii) Concurrent procedures: Quantitative and qualitative data
are collected at the same time to develop a comprehensive picture. The
researcher may embed one data form in a larger data collection.
(iii) Transformative procedures: The design involves quantitative and
qualitative data analysed through a theoretical lens. The lens gives a
focus for factors such as outcomes or data collection. The data
collection procedures may involve a sequential or concurrent approach.
Research Methods
The researcher must specify the
methods
of data collection and analysis. The choice of methods used by a
researcher depends on whether the intent is to specify the type of
information to be collected in advance of the study or to allow it to
emerge from participants in the project. The knowledge claims,
strategies and the method all contribute to a research approach.
A
quantitive approach is one in which the investigator primarily uses
post positivist claims for developing knowledge and employs strategies
of inquiry such as experiments and surveys.
A qualitative approach
is one in which the inquirer often makes knowledge claims based
primarily on constructivist perspectives or advocacy/participatory
perspectives . It uses strategies of inquiry such as narratives.
A mixed methods approach is one in which the researcher tends to based knowledge claims on pragmatic grounds.
Three criteria to consider when selecting a research method are;
1. The research problem
2. Personal experiences of the researcher
3. The audience to whom the report will be written
1. The research problem:
A quantitative approach is best to test a theory or explanation. A qualitative approach is best if there is concept or phenomenon that
needs to be understood because little research has been done on it. When
the researcher does not know the important variables to examine, a
qualitative approach is most appropriate. In general, this approach is
ideal for exploring new topics.
Mixed methods can be used to capture
the best of both quantitative and qualitative approaches. The
researcher may initially adopt a qualitative approach to learn about
what variables to study and then uses a quantitative approach to study
the variables within a large sample. Alternatively, the researcher may
initially use a quantitative approach on a large sample, then use a
quantitative approach to go in depth with a few individuals from that
sample.
2. Personal experiences of the researcher
Personal
training and experiences should be considered when deciding on a method.
A mixed methods researcher needs to be familiar with both quantitative
and qualitative research, and also needs to understand the rationales
for combining both forms of data.
Quantitative research is most common. There is therefore more support available for those choosing this approach.
Mixed methods research takes the longest because of the need to collect and analyse both qualitative and quantitative data.
3. Audience
When deciding on an approach, the experience and preferences of those
who will support and read your research should be considered. For
students, preferences of their supervisors need to be considered.
Summary
Creswell outlines how when research philosophy, strategies and methods
are combined, it is possible to generate different frameworks under
which to conduct research. Researchers must choose the most appropriate
approach based on the research question, their own experiences of
research and the potential audience. The research approaches that are
available to researchers are qualitative, quantitative and mixed
methods. Each is underpinned by assumptions about knowledge claims,
strategies of inquiry and specific research methods.
From Don Passey
The important element for me when I'm setting up a study is to match
what I want to study with appropriate methods that will allow me to do
that. Planning then becomes important - identifying the how, the when,
and the extent of data gathering that can be done within time periods
and still enable adequate data analysis and outcomes to be drawn out.
From Seema Pillai
Mayer, R. E. (2003), the research should be “issue-driven rather than
method-driven”. If the chosen research problem demands a mix of
quantitative and qualitative approach to better address the issue then
definitely researcher should carry out a detailed study on this
innovative research method to minimize the chances of a pitfall.
The typology suggested by Creswell and Plano Clark
(2007) provides a clear role for each component in terms of timing,
weighting, and mixing and is thus very useful for novice researchers.
"Creswell and Plano Clark (2007) proposed four main design types, with
multiple variants:
(1) the triangulated design, to obtain different but complementary data;
(2) the embedded design, in which one data set provides a supportive secondary role;
(3) the two-phase explanatory design, which builds or explains quantitative results; and
(4) the exploratory design, which is also two-phased but led by the qualitative."
The results of Jerome (2011)
study using triangulation mixed method and embedded design in two
different educational contexts will be useful to learn more about the
benefits and challenges of mixed method research.
From Mary Hamilton
The reality is that people use different starting points. The "ideal" (expressed carefully by Don in his post above) is to
identify the research problem or question(s) you want to investigate and
design your research strategy around this.
Commonly, however, what is practically possible in terms of
access to different kinds of data, researchers predilections, training
and beliefs act as filters to the kinds of methods you may consider in a
given study.
Bob and I like to talk in terms of a "chain of evidence" that a
researcher has to build in order to present a convincing story of
their project from start to finish. The idea is that the questions,
framework, design and data have to link in a logical way in order to
produce "knowledge claims" at the end of the day that other people will
(hopefully!) accept as valid.
