Mixed Methods Flashcards
Mixed methods
Qualitative and quantitative research
Quantitative Approach
- Research makes decisions before study
- Researcher reduces the inquiry to a small number of variables to study and a large number of people
- Researcher ask specific closed-ended questions
- Analyzes numbers
- Importance placed on reliability, validity, generalizability, replicability, control, lack of bias
Qualitative approachs
- Researcher makes decisions based on views of participants
- Researcher open the inquiry up to understand the complexity of the situation
- Asks open-ended questions
- Analyses words and images
- Researcher places emphasis on indivdual meaning, context, and self-reflexivity
Why use mixed methods?
Triangulation
Complementarity
Development
Initiation
Extension
Triangulation
To find convergence and agreement across different methods (increases convergent and divergent validity)
Complementarity
To develop an enriched, elaborated explanation of a given phenomena
Development
Triangulation is used to develop complementarity
Initiation
Look for inconsistencies, contradictions and paradox between methods (challenge existing paradigms and promote theory development)
Extension
Extend the range of inquiry by using different methods for different inquiry components (ex. qualitative for processes; quantitative for outcome)
Philosophical approaches
- Post-positivism
- Social constructionism
- Advocacy/transformative
- Pragmatism
Post-positivism
ACCEPTS an objective reality outside of your senses that we can study
REJECTS the idea of incommensurability (assumes that there is only one way of seeing the world)
All types of measurement are fallible and subject to error
We will never achieve scientific “truth”
Social constructionism
REJECTS the notion of an objective reality that can be studied
ACCEPTS the idea of incommensurability.
Assumes there are multiple equally valid ways of understanding the world
Science should focus on how arguments are constructed and why (i.e., political, social biases)
Advocacy-Transformative
Research is politically motivated
Focuses on power differentials and tried to empower and give voice to stigmatized groups (women, ethnic minorities)
Pragmatism
Only thing that matters is practical, real-world utility of research findings
“the instrumental truth”
The purpose of science is to be useful/practical in day-to-day life
Three steps in designing a mixed method study
- Choose a theoretical lens
- How will data be collected
- When will data integration occur
Step 1: What is the theoretical lens?
Choose one or more of the four approaches
Step 2: How will data collection be implemented and prioritized?
- Sequential Explanatory
- Sequential Exploratory
- Concurrent Designs
- Transformative Designs
Step 3: When will data integration occur?
- Analyze quantitative/qualitative data separately
- Convert qualitative data into quantitative codes, and analyze both together in a quantitative analysis
- Connect the two types of analyses together in separate studies (e.g., qualitative study which informs
Four types of mixed method designs
- Sequential explanatory
- Sequential exploratory
- Concurrent designs
- Transformative designs
Sequential explanatory design
QUAN –> qual
Qualitative portion is a selected subsample of the quantitative data
Qualitative data used to explain quantitative results
RCT to test a new therapy (quan); follow up on unsuccessful patients with qualitative questions
Sequential exploratory design
qual –> QUAN
Usually involves the same people for both quan/qual
used to explore a poorly measured construct, create a new test, to increase the external validity of qual results
Ex. thematic analysis then quantitative analysis on the results
Concurrent designs
QUAN + QUAL
Analyze both sets of data and compare and contrast (to get the same results)
Hill et al., concurrent triangulation
14 participants with PTSD
Dream therapy and loss therapy; 8-11 sessions
Quantitative questionnaires and thematic analysis to measure distress and therapy effectiveness
Quan/qual similar; gain of insight, but no decrease in symptomology
Dream therapy participants had better engagement and a better working alliance
Concurrent nested design
QUAN + qual or quan + QUAL
Like triangulation design
Data usually transformed into the prioritized type
