Mixed-methods research combines the depth of qualitative inquiry with the breadth and generalisability of quantitative analysis. It is particularly well-suited to research questions where the ‘what’ (how common is this phenomenon?) must be understood alongside the ‘why’ (why does this happen, and what does it mean to those experiencing it?).
The most common mixed-methods design in thesis research is the explanatory sequential design: quantitative data is collected and analysed first, and qualitative data is then collected to explain or elaborate on unexpected or interesting quantitative findings. The reverse — qualitative first, then quantitative — is the exploratory sequential design, used when the concepts are insufficiently understood to design a reliable questionnaire.
SPSS handles the quantitative component. Use it for descriptive statistics, inferential tests, and regression modelling on your survey or experimental data. Clean and code the dataset before analysis, and run assumption checks before each test. Export output tables as images or formatted tables for inclusion in the results chapter.
NVivo handles the qualitative component. Import interview transcripts, focus group recordings (transcribed), field notes, or documents. Code the data using nodes — first descriptive codes that stay close to the data, then interpretive codes that begin to identify patterns and themes. Use matrix coding queries to explore relationships between themes and participant characteristics.
The integration point is where mixed-methods theses most often struggle. It is not enough to present quantitative results in one chapter and qualitative findings in another. Show how one strand of data explains, confirms, or challenges the other. A joint display — a table or figure that presents both strands side by side — is an effective way to make the integration explicit and examinable.
