Application of Mixed

Method Research
Methodology
By :
Dr. Noor Dayana Abd Halim
17th November 2017
 Steps for Conducting a Mixed Methods Study
Determine if a mixed Identify a rationale for a
methods study is feasible mixed methods study

Step 1 Step 2

Priority Identify the data collection

Write the report as one- or strategy and type of
two-phase study Sequence design

Step 7 Visualization Step 3

Develop quantitative,
Analyze data separately or Collect quantitative and qualitative and mixed
concurrently qualitative data methods questions

Step 6 Step 5 Step 4

 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.
 Pragmatism – practicality; multiple view points; biased
and unbiased; subjective and objective
 When to Use Mixed Methods (cont)

 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 detailed information
 Mixed Methods Research Methodology

 Utilizes both quantitative and qualitative data collection

methodologies
 Examples
- Interviews and Questionnaires
- Performance Tests and Observation
- Questionnaires and follow up Focus groups
- Document analysis, Performance Test,
Questionnaire and Interviews
 What is Mixed Methods Research?

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.
Key Characteristics of Mixed Methods Design: Rationale
 Rationale
- Test findings of first phase
- Explain results of first phase in more details
-provide a more complete understanding than either
quantitative or qualitative alone
 Collecting both quantitative and qualitative data
- Numeric data
- Text data
Key Characteristics In Choosing Mixed Methods
Study

 The level of interaction between the quantitative and

qualitative strands
 The priority of the strands
 The timing of the strands
 Where and how to mix the strands
 Mixed Methods Legend
Notation Defined
QUAN Quantitatively driven study
QUAL Qualitatively driven study
quan Quantitative data is secondary to qualitative
data
qual Qualitative data is secondary to quantitative
data
Indicates that quantitative and qualitative data
are collected concurrently
+ Indicates that quantitative and qualitative data
are collected sequentially
 Key Characteristics of Mixed Methods
Design: Priority and Sequence
 Priority
- Equal weight
- QUAN carries more weight than qual
- QUAL carries more weight than quan
 Sequence
- collect both quantitative and qualitative data at the same
time
- Collect quantitative data first followed by qualitative data
- Collect qualitative data first followed by quantitative data
The Major Mixed
Methods Designs
 The Convergent Parallel Design

Quantitative
Data Collection
and Analysis
Compare or
relate Intepretation

Qualitative Data
Collection and
Analysis
 The Convergent Design
 The researcher:
 Collects quantitative and qualitative
concurrently
 Analyse the two data sets separately
 Mixes the two databases by merging the results
during interpretation (and sometimes during
data analysis)
 Purposes for the Convergent Design
 Obtain a more complete understanding from two
databases
 Corroborate results from different methods
 Compare multiple levels within a system
 When to use the Convergent Design

 Choose this design if:

 Need to collect both types of data in one visit to
the field
 Both types of data have equal value for
understanding the research problem
 Have quantitative and qualitative research skills
 Can manage extensive data collection activities
individually or with a team
 Convergent Design

 Philosophical assumptions:
 Best suited to an “umbrella” paradigm such as
pragmatism

 Common variants:
 Parallel-databases variant
 Data-transformation variant
 Data-validation variant
 Strengths: Convergent Design

 Intuitive
 Efficient
 Lends itself to teams
 Strengths: Convergent Design
 Requires substantial effort and expertise
 Issues related to the samples and sample sizes
 Difficult to converge two sets of different data
 How to resolve discrepant results
 The Explanatory Sequential Design

Quantitative Data Qualitative Data

Collection and Follow up Collection and Intepretation
Analysis with Analysis
 The Explanatory Design
 The researcher:
 Starts by collecting and analyzing quantitative
data
 Collects and analyzes qualitative data in a
second phase as follow-up to the quantitative
results
 Connects the phases by using the quantitative
results to shape the qualitative research
questions, sampling and data collection
 Purposes for The Explanatory Design

 To use qualitative data to help explain quantitative

results that need further exploration
 To use quantitative results to purposefully select
best participants for qualitative study
 When to use the Explanatory Design

 Choose this design if:

 Researcher and research problem are
quantitatively oriented
 Know important variables and instruments are
available
 Participants available for second data collection
 Have time to conduct two phases
 Have limited resources and need to collect and
analyze one data type at a time
 New questions emerge from quantitative results
 Explanatory Design

 Philosophical assumptions:
 Begin from post positivism for the quantitative
phase
 Shift to constructivism for the qualitative phase

 Common variants:
 Follow-up explanation variants
 Participant-selection variant
 Strengths: Explanatory Design

 Appealing to quantitative researchers

 Straightforward to implement two phases
 Final report can be written in two phases
 Lends itself to emergent approaches
 Challenges: Explanatory Design

 Two phases require lengthy time to implement

 Difficult to secure IRB approval when second
phase cannot be specified before first phase
complete
 Need to decide what results to follow up
 Must decide criteria for selecting participants
 Need to contact participants for a second round of
data collection
 The Exploratory Sequential Design

Qualitative Data Quantitative Data

Collection and Follow up Collection and Intepretation
Analysis with Analysis
 The Exploratory Design

 The researcher:
 Collects and analyzes qualitative data first
followed by quantitative data
 Analyze the qualitative data and uses results to
build to the subsequent quantitative phase
 Connects the phases by using the qualitative
results to shape the quantitative phase by
specifying research questions and variables,
developing an instrument, and/or generating a
typology
 Purposes for the Exploratory Design

 To first explore because variables, theories,

hypotheses not known
 To develop an instrument or typology that is not
available
 To assess whether qualitative themes generalize
to a population
 When to use the Exploratory Design
 Choose this design if:
 Researcher and research problem are
qualitatively oriented
 Important variables not known and instruments
not available
 Have time to conduct two phases
 Have limited resources and need to collect and
analyze one data type at a time
 New questions have emerged from qualitative
results
 Exploratory Design

 Philosophical assumptions:
 Begin from constructivism for the qualitative
phase
 Shift to post positivism for the quantitative
phase
 Common variants:
 Theory-development variant
 Instrument-development variant
 Strengths: Exploratory Design

 Straightforward to design, implement, and report

 Quantitative component can make the qualitative
approach more acceptable to quantitative-biased
audiences
 Researcher produces a product, such as an
instrument
 Lends itself to emergent approaches
 Challenges: Exploratory Design

 Two phases require lengthy time to implement

 Difficult to specify quantitative procedures when
applying for initial IRB approval; may have to
apply twice
 Deciding the qualitative findings to use for
quantitative phase
 Procedure for developing a valid and reliable
instrument
 The Embedded Design

Quantitative (or Qualitative)

Design

Quantitative (or Qualitative)

Data Coolection and Analysis
Interpretation

Quantitative (or Qualitative)

Data Collection and Analysis
(before, during, or after)
 Embedded Approach
Collect qual Experimental Design (Can use quasi-experimental
design)

Interpretation
QUAN QUAN
Intervention based on
(pretest) (posttest)
QUAN (qual)

Correlational14 Experimental

QUAN designs to be used within Explanatory design Between-subjects

an Embedded Approach - Pre- and posttest design
Predictive design Within-subjects
- Cross-over design
- Factorial design

Collect qual

QUAN QUAN Interpretation

(IV or predictor (DV or criterion based on
variable) variable) QUAN (qual) Correlational Design
 Purposes for the Embedded Design
 To address different questions that call for different
methods
 To enhance an experiment such as by
- improving recruitment procedures
- examining the intervention process
- explaining reactions to participation
 When to Use the Embedded Design

 Choose thus design if:

- Have expertise with the primary design
- Are comfortable with the primary orientation
- Have little prior experience with the supplemental
method
- Resources limit placing equal priority on both
methods
- Need for a secondary data set emerges
 Embedded Design
 Philosophical assumptions:
- Worldview may reflect the primary approach, use
pragmatism for a concurrent approach, or shift in a
sequential approach
 Common variants:
- Embedded experiment
- Embedded correlational
- Embedded instrument development and validation
- Mixed methods case study
- Mixed methods narrative research
- Mixed method ethnography
 Strengths: Embedded Design

 May require less time and fewer resources

 Improve the larger design with supplemental data
 Fits team approach well
 May be able to publish result separately
 Appealing to those accustomed to traditional
designs
 Challenges: Embedded Design

 Need expertise in primary design and mixed

methods
 Must specify purpose for collecting the
supplemental data
 Must decide when to collect supplemental data
 Result are difficult to integrate
 Must consider treatment bias if qualitative data
collected during experiment
THANK YOU