Lecture 3
Mixed Methods
A mixed methods research design is a
procedure for collecting, analyzing, and
“mixing” both quantitative and
qualitative research and methods in a
single study to understand a research
problem.
To utilize this design effectively, you
must understand both quantitative and
qualitative research.
Mixed methods research combines
elements of quantitative
research and qualitative research in
order to answer your research
question. Mixed methods can help
you gain a more complete picture
than a standalone quantitative or
qualitative study, as it integrates
benefits of both methods.
Mixed methods research is often
used in the behavioral, health, and
social sciences, especially in
multidisciplinary settings and
complex situational or societal
research.
Mixed methods research question examples
1. To what extent does the frequency of
traffic accidents (quantitative) reflect cyclist
perceptions of road safety (qualitative) in
Karachi?
2. How do student perceptions of their
universityenvironment (qualitative) relate
to differences in test scores (quantitative)?
How do interviews about job satisfaction at
Company X (qualitative) help explain year-over-
year sales performance and other
KPIs (quantitative)?
How can voter and non-voter beliefs about
democracy (qualitative) help explain election
turnout patterns (quantitative) in Town X?
How do average hospital salary measurements
over time (quantitative) help to explain nurse
testimonials about job satisfaction (qualitative)?
When to Use Mixed Methods Designs
When both quantitative and qualitative
data, together, provide a better
understanding of your research problem
than either type by itself.
When one type of research (qualitative or
quantitative) is not enough to address the
research problem or answer the research
questions.
To incorporate a qualitative
component into an otherwise
quantitative study
To build from one phase of a study
to another
Explore qualitatively then develop
an instrument
Follow-up a quantitative study
qualitatively to obtain more
As you formulate your research
question, try to directly address how
qualitative and quantitative methods
will be combined in your study. If
your research question can be
sufficiently answered via standalone
quantitative or qualitative analysis, a
mixed methods approach may not be
the right fit.
Research example
You want to research cycling safety in
high-traffic areas of Karachi. If you’re
interested in the frequency of accidents
and where they occur, this could be a
straightforward quantitative analysis. If
you’re interested in the nature of
complaints submitted by cyclists, or their
perceptions about cycling in particular
areas, then a qualitative approach may fit
best.
But mixed methods might be a good choice
if you want to meaningfully integrate both
of these questions in one research study.
For example, you could use a mixed
methods design to investigate whether
areas perceived as dangerous have high
accident rates, or to explore why specific
areas are more dangerous for cyclists than
others.
Keep in mind that mixed methods
research doesn’t just mean
collecting both types of data; you
need to carefully consider the
relationship between the two and
how you’ll integrate them into
coherent conclusions.
Mixed methods can be very
challenging to put into practice,
and comes with the same risk
of research biases as standalone
studies, so it’s a less common
choice than standalone qualitative
or qualitative research.
Mixed methods research designs
There are different types of mixed
methods research designs. The differences
between them relate to the aim of the research,
the timing of the data collection, and the
importance given to each data type.
*As you design your mixed methods study, also
keep in mind:
*Your research approach(inductive vs deductive)
*Your research questions
*What kindof data is already available for you to
use
*What kind of data you’re able tocollect yourself
Convergent parallel
In a convergent parallel design, you
collect quantitative and qualitative
data at the same time and analyze
them separately. After both
analyses are complete, compare
your results to draw overall
conclusions.
Example: Convergent parallel design
In your research on cycling safety in Amsterdam, you
undertake both sides of your research simultaneously:
On the qualitative side, you analyze cyclist complaints
via the city’s database and on social media to find out
which areas are perceived as dangerous and why.
On the quantitative side, you analyze accident reports
in the city’s database to find out how frequently
accidents occur in different areas of the city.
When you finish your data collection and analysis,
you then compare results and tie your findings
together.
Embedded
In an embedded design, you collect and analyze
both types of data at the same time, but within
a larger quantitative or qualitative design. One
type of data is secondary to the other.
This is a good approach to take if you have
limited time or resources. You can use an
embedded design to strengthen or supplement
your conclusions from the primary type of
research design.
Example: Embedded designAs part of a
quantitative study testing whether the
number of cyclist complaints about an
area correlates with the number of
accidents, you could “embed” a series
of qualitative interviews with cyclists
who submitted complaints to further
strengthen your argument. The bulk of
your research remains quantitative.
Explanatory sequential
In an explanatory sequential design,
your quantitative data collection and
analysis occurs first, followed
by qualitative data collection and
analysis.
You should use this design if you think
your qualitative data will explain and
contextualize your quantitative findings.
Example: Explanatory sequentialYou
analyze the accident statistics first and
draw preliminary conclusions about which
areas are most dangerous. Based on these
findings, you conduct interviews with
cyclists in high-accident areas and analyze
complaints qualitatively.
You can utilize the qualitative data to
explain why accidents occur on specific
roads, and take a deep dive into particular
problem areas.
Exploratory sequential
In an exploratory sequential
design, qualitative data collection and
analysis occurs first, followed
by quantitative data collection and
analysis.
You can use this design to
first explore initial questions and
develop hypotheses. Then you can use the
quantitative data to test or confirm your
qualitative findings.
Example: Exploratory sequential
designYou first interview cyclists to
develop an initial understanding of
problem areas, and draw
preliminary conclusions. Then you
analyze accident statistics to test
whether cyclist perceptions line up
with where accidents occur.
