Thematic Analysis in Mixed Methods Research

Yes, thematic analysis can be effectively incorporated into mixed methods research.

Mixed methods research combines qualitative and quantitative data. It offers a fuller understanding than either approach alone.

Thematic analysis (TA) is a flexible method for identifying and analyzing patterns in qualitative data. It can be used with a range of theoretical and philosophical commitments, making it adaptable to different research designs.

TA is valuable in mixed methods research because it can reveal the “why” behind quantitative results. It provides a deeper understanding of the quantitative findings.

Researchers can use qualitative data to explain statistical relationships, contextualize findings, and explore the underlying mechanisms of observed phenomena.

For example, a survey might reveal a correlation between a certain personality trait and career success. Thematic analysis of interview data can then show how that trait manifests in different work environments and contributes to that success.

Real studies show this integration in practice. King et al. (2015) combined a large patient survey with thematic analysis of interviews to find which quality indicators shape patients’ choice of provider.

Stewart et al. (2021) paired quantitative outcome measures with thematic analysis of interviews to study how young people, parents, and clinicians experienced online eating-disorder treatment during COVID-19.

Integrating Thematic Analysis in Mixed Methods Research

Integration, a key aspect of mixed methods research, involves intentionally combining quantitative and qualitative research to create interdependence and synergy between the two approaches.

There are multiple potential levels of integration, including at the design, methods, and representation levels.

1. Design-Level Integration:

Thematic analysis is compatible with a range of mixed methods designs, including convergent parallel, exploratory sequential, and explanatory sequential designs.

The choice of design depends on the research questions and the function of qualitative data within the study.

• Convergent parallel design involves simultaneously collecting quantitative and qualitative data, analyzing these datasets separately, and then merging the results for interpretation.
• Explanatory sequential design is used in research when you want to use one type of data to help explain the findings of another type.
• Exploratory sequential design is a type of research that involves two phases of data collection and analysis, with the qualitative phase coming first.

2. Method-Level Integration:

All methods used must work in synergy with each other and support the research goals.

Method level integration combines data collection approaches in mixed methods research by using insights from one method to shape data collection in the other.

Approaches to integration:

• Shaping Tools: Thematic analysis of qualitative data can help shape quantitative data collection tools or provide context for interpreting quantitative findings.
• Qualitative to Quantitative: Researchers use interview themes and participant language to develop targeted survey items and measurement scales that better capture the phenomena studied.
• Quantitative to Qualitative: Quantitative findings can guide qualitative data collection by helping researchers refine interview questions, focus on significant variables, or probe unexpected results.
• Concurrent Integration: Researchers simultaneously collect and analyze both data types, allowing real-time adjustments to sampling and measurement based on emerging patterns and feedback.

Examples of integration:

• Alongside an RCT: Thematic analysis of qualitative data collected alongside an RCT can illuminate participants’ experiences, perceptions of the intervention, and factors influencing outcomes.
• Informing Sampling: Themes from initial interviews can guide the sampling strategy for a later quantitative survey, ensuring diverse perspectives are represented.

Importance of methodological rigor and transparency:

• Rationale and Purpose: Researchers should clearly articulate their rationale for using thematic analysis and how it contributes to achieving the study’s objectives.
• Transparency: Transparency in the analytical process, including coding procedures, theme development, and integration strategies, is crucial. This transparency enables other researchers to understand and evaluate the findings.
• Established Guidelines: Using established guidelines for thematic analysis, such as those outlined by Braun and Clarke (2006), enhances the rigor and trustworthiness of the findings.

3. Representation-Level Integration:

Qualitative data can illuminate the mechanisms behind observed statistical relationships, providing insights into how and why certain variables are associated with particular outcomes.

It can also provide context-specific information that helps to interpret the meaning of quantitative findings. Thematic analysis helps to explain and contextualize quantitative results.

Qualitative themes derived from thematic analysis can be woven into the narrative presentation of quantitative findings, offering deeper insights and understanding.

Several methods can be used to integrate qualitative and quantitative data:

• Data Transformation: Converting one data type into the other, for example turning qualitative data into numerical counts that can be integrated with quantitative data for analysis.
• Pattern Matching: Connecting qualitative themes to quantitative data points to reveal how they relate.
• Meta-Inferences: Comparing or relating identified themes to quantitative data generates new insights known as meta-inferences.

Examples of these methods in practice:

• Stress and Burnout: Participants who frequently discussed workplace stress in interviews also scored higher on standardized burnout assessment scales, an example of pattern matching.
• Comparing Across Groups: Themes can be compared across levels of a quantitative variable, such as contrasting interview themes from individuals with high versus low grit scores.

Visual representations, such as tables or matrices, can be used to showcase the relationships between quantitative results and qualitative themes from thematic analysis, facilitating the understanding of complex relationships.

For example, thematic quote excerpts alongside relevant statistical graphs.