MultiHub Forum

I'm designing my master's thesis in sociology, which will involve conducting interviews and surveys on community responses to a new local policy. While I'm confident in my qualitative interview guide, I'm less sure about constructing a valid and reliable survey instrument that avoids leading questions and captures nuanced attitudes. For those experienced in mixed-methods social science research, what are your go-to strategies for pilot testing a survey and ensuring your quantitative data will meaningfully complement your qualitative findings?

Solid project. Start with cognitive interviews (roughly 5–12 participants) to test wording and concept clarity, then run a small pilot (n ~ 30–60) to check timing and reliability. Keep scales simple (5-point Likert), watch for ceiling/floor effects, and aim for Cronbach’s alpha around .7+ if you have multi-item scales.