Ch 18 - Mixed Methods and Other Special Types of Research Flashcards
Advantages of mixed methods
- Complementarity
- Practicality- use whichever works best in the current research situation
- Incrementality- build off each other
- Enhanced Validity- supported by multiple types of data, triangulation
Pragmatism
provides basis for a position that has been stated as the “dictatorship of the research question”
researchers consider that it is the research question that should drive the inquire and it’s design/methods, don’t want to force a specific type of research on a study
Aspects of MM research questions
- exploratory questions (qualitative)
- —-combined with——- - confirmatory questions (quantitative)
Examine
- causal effects: quantitative
- causal mechanisms: qualitative
What situations are good for MM?
- concepts that are poorly understood
- findings that can be enhanced by another source of data
- when picking one approach wouldn’t be sufficient
- when quantitative results are difficult to interpret
Situation for MM: Developmental Work
NEW construct - qualitative research helps to capture full complexity
-qualitative research can help form research questions for quantitative research
Situation for MM: Hypothesis Generating and Testing
qualitative studies may bring to light insights into constructs/relationships and then an be studying/tested with larger quantitative samples
Situation for MM: Explication
qualitative provides meaning behind quantitative descriptions/relationships (WHY - clarification)
Situation for MM: Theory Building, Testing, and Refinement
an ambitious application of MM
–>if a theory can survive both qualitative and quantitative study then it will provide stronger contest for the clinical work
Decisions in MM: Design Decisions/Notation
- Sequencing
-qual first, quant first, both simultaneously (concurrent)
QUAN –> qual, QUAN + qual, or QUAN (qual) - Prioritization
–>deciding which approach to emphasize
(equal status vs. dominant status)
ex. QUAL/quant, QUANT/qual, or QUANT/QUAL
Decisions in MM: Specific MM Designs
- convergent parallel design
- embedded design
- explanatory designs
- exploratory designs
Convergent Parallel (Triangulation) Design
- ->seek to obtain different, but complementary data bayou the central phenomenon
- goal to converge truth about a problem/phenomenon by allowing the limitations of one approach to be offset by the strengths of the other
(QUAL + QUAN) = data collected simultaneously and with equal priority
Embedded Design
oen type of data is used in a supportive capacity in a study based primarily on the other data type
QUAL(quan) or QUAN (qual)
Explanatory Design
sequential designs with quantitative data collected int he first phase, followed by qualitative data collected in second phase
–>qualitative builds on the quantitative
QUAN –> qual
Exploratory Design
qualitative data collected first (explore in-depth a poorly understood topic) and then quantitive data to focus on measuring/classifying it
(QUAL –> quan) or (QUAL –> QUAN)
Decisions in MM: Sampling and Data Collection
Quant: rely on sampling strategy that enhances researchers ability to generalize from sample to broader population
Qual: adopt purposeful sampling method to select information-rich cases who are good informants
Sample size: different in qual vs. quan, issue: people can be in both strands (useful in nested approach)
Data Collection:
- group/individual interviews
- psychosocial scales
- observations
- biophysiological measures, etc
- ->intramethod mixing