1. S N
D G S:
O SI
TH E
E D
M H W
ED ARC VIE
IX E R
M ES VE
R O
A N
DR. GUSTAVO DANIEL CONSTANTINO
CIAFIC-CONICET
ARGENTINA
2. MMR: THE NAMES
Multitrait/multimethod research (Campbell &
Fiske, 1959)
Integrated/combined research (Steckler et al.,1992;
Creswell, 1994)
“Quantitative & Qualitative Methods” (Fielding &
Fielding, 1986)
Hybrids ( Ragin, Nagel & White, 2004 )
Methodological Triangulation (Morse, 1991)
Mixed Methods Research (Tashakkori & Teddlie,
2003, 2010; Cresswell & Plano Clark, 2007;
Tedlie & Tashakkori, 2009)
3. MIXED METHODS RESEARCH: A DEFINITION
MMR is a research design with
philosophical assumptions
(pragmatism) as well as methods of
inquiry.
As a methodology, it involves
philosophical assumptions that guide
the direction of the collection and
analysis of data and the mixture of
qualitative and quantitative
approaches in many phases in the
research process.
4. MIXED METHODS RESEARCH: A DEFINITION (CONT.)
As a method, it focused on collecting,
analyzing, and mixing both quantitative
and qualitative data in a single study
or series of studies.
Its central premise is that the use of
quantitative and qualitative
approaches in combination provides a
better understanding of research
problems than either approach alone.
5. MMR PRAGMATISM
Consequences of actions
Problem centered
Pluralistic
Real-world practice
oriented
Ontology: singular and multiple realities
Epistemology: Practicality (what works)
Axiology: Multiple stances (biased and
unbiased perspectives)
Methodology: combining
Rhetoric: formal (quan) or informal (qual)
6. MMR: CENTRAL PREMISE
The combination of QUAN and QUAL approaches
provides a better understanding of research
problems than either approach alone.
But, in what way is it better?
3. MMR provides strengths that offset the weaknesses of
both approaches
4. MMR provides more comprehensive evidence because
there isn’t data restriction.
5. MMR can help to answer complex questions that cannot
be answered by QUAN or QUAL approaches alone.
6. MMR encourages researchers to collaborate in despite of
their paradigmatic posture
7. MMR encourages the use of multiple worldviews or
paradigms
8. MMR is “practical”: free to use any research method and
any type of thinking (inductive – deductive)
7. TYPES OF RESEARCH PROBLEMS AND MATCHING
METHODS OR DESIGNS (CRESSWELL & PLANO CLARK, 2007)
Type of Research Problem Type of Methods (Designs)
suited to study the problem
To see if a treatment is effective Experimental design
To see what factors influence an Correlation design
outcome
To identify broad trends in a Survey design
population
To describe a culture-sharing group Ethnography design
To generate a theory of a process Grounded theory
design
To tell the story of an individual Narrative Research
8. MMR: THE FOUR MAJOR TYPES
Triangulation Design
Useful when a researcher needs to directly compare and contrast
quan statistical results with qual findings or to validate or expand
quan results with qual data.
Embedded Design
Useful when a researcher needs to embed a qualitative component
within a quantitative design (correlational or experimental design)
Explanatory Design
Two different (QUAN-QUAL) sequential phases for Follow-up
Explanations (QUAN emphasized) or Participant Selection (QUAL
emphasized)
Exploratory Design
Two different (QUAN-QUAL) sequential phases for Instrument
Development (QUAN emphasized) or Taxonomy Development (QUAL
emphasized)
9. MMR: THE TRIANGULATION DESIGN
Purpose: to obtain different but complementary
data on the same topic
Rationale: to bring together the differing
strengths and non-overlapping weaknesses
of QUAN methods (large sample size, trends,
generalization) with those of QUAL methods
(small N, details, in depth).
10. MMR: VARIANTS OF THE TRIANGULATION
DESIGN
The convergence model ( traditional; to end
up with well-substantiated conclusions about
a single phenomenon)
Data transformation model (to quantify
Qual data or to qualify Quan data)
Validating quantitative data model
( including qual data to validate main quan
data)
Multilevel model / multilevel research
(different methods (quan/qual) are used to
address different levels within a system)
11. MMR: TRIANGULATION DESIGN: CONVERGENCE MODEL
QUAN QUAN QUAN
Data Results
Data
Analysis
collection
Compare
Interpretation
And
QUAN+QUAL
Contrast
QUAL
QUAL QUAL
Data
Data Results
collection Analysis
12. MMR: TRIANGULATION DESIGN: DATA TRANSFORMATION
MODEL
QUAN QUAN
Data
Data
Analysis
collection
Compare and
Interpretation
Interrelate two
QUAN+QUAL
quan data sets
QUAL
QUAL Transform
Data
Data QUAL into
collection Analysis quan Data
13. MMR: TRIANGULATION DESIGN: VALIDATING QUANTITATIVE DATA
MODEL
QUAN QUAN QUAN
Data Data Results
Collection: Analysis
Survey
Validate
Interpretation
QUAN
QUAN + qual
results
with qual
results
Qual data
qual qual
Collection:
data results
open-ended
analysis
survey
items
14. MMR: TRIANGULATION DESIGN: STRENGTHS AND
CHALLENGES
The design makes intuitive sense.
It is an efficient design in both types of data.
Much effort and expertise (QUAL and QUAN)
is required.
To converge two sets of very different data
and their results in a meaningful way.
Researchers need to develop procedures for
transforming data and make decision about
how the data will be transformed.
What to do if the quantitative and qualitative
results do not agree?
15. MMR: INCONSISTENCIES AND CRITIQUES
(BRYMAN, 2007; DENSCOMBE, 2008; CONSTANTINO, 2008)
The dividing lines are much fuzzier that typically
suggested in the literature (for example, from post-
positivism or interpretative research paradigms)
Positions are not nearly as “logical” and as distinct as
is frequently suggested in the literature (idem)
The problem of “commensurability” between
quantitative and qualitative methodologies
Incompatibility of philosophical premises leads to use
QUAN e QUAN “in parallel”, each playing to its
respective strengths (as Bryman has demonstrated in
his study)
The 4 senses of “pragmatism”: fusion of approaches;
a third alternative; a new orthodoxy; expedient (lack
of principles underlying a course of action).
The retrospective framing of past studies is not strong
evidence for validate mixed method models
16.
17. MMR: THE QUESTIONS TO MY OWN RESEARCH
• Which of the major paradigms (QUAN,
QUAL, MMR) frames your study?
• What challenges are associated with
your design choice?
• Do you think that relevant data could
be collected and analyzed if you
choose an alternative research
design?
• Are you really satisfied with the design
model that you have drawn?
• Are you doing your research with a
pragmatic (naïve or commonsense)
point of view?
18. MMR: CONTENT ANALYSIS VS.(?) DISCOURSE ANALYSIS
Consider a study in which only one type
of data is collected (QUAL data - texts)
The researcher would analyze the data
both qualitatively (developing themes
using discourse analysis)
and quantitatively (counting words or
rating responses on predetermined
scales, using content analysis).
Are these mixed methods data analysis a
MMR?