- ✓Data collection should be guided by the research questions and designed to generate data that directly addresses them.
- ✓Survey instruments should be piloted before full deployment to identify ambiguous questions and technical problems.
- ✓Thematic analysis is one of the most widely used methods for analysing qualitative data, involving systematic identification of patterns and themes.
- ✓Statistical analysis methods must match the type and distribution of data collected; using the wrong test can produce misleading conclusions.
- ✓Maintaining a detailed research diary throughout data collection supports later reflection and strengthens the transparency of the process.
Listen to the full episode inside the course. Enrol to access all 80 episodes, plus assignments, tutor support and Student Finance funding.
Start learning →Alex: Today we're looking at how to actually conduct research and analyse the data you collect. Sam, once you've chosen your methodology, where do you start?
Sam: You start by designing your data collection instruments. If you're running a survey, you need to write the questions. If you're conducting interviews, you need to design an interview guide. If you're running an experiment, you need to design the experimental conditions and the measurement approach. Good instrument design is critical: poorly designed data collection produces data that can't answer your research questions, no matter how sophisticated your analysis.
Alex: What makes a good survey question?
Sam: Clarity, above everything else. Each question should ask about one thing and one thing only; double-barrelled questions that ask about two things simultaneously produce uninterpretable responses. Questions should avoid leading the respondent toward a particular answer. Response options should be exhaustive and mutually exclusive. And the overall survey should be as short as possible while still covering what you need, because longer surveys get lower completion rates and more fatigued, less reliable responses.
Alex: And for interviews?
Sam: A semi-structured interview guide lists the main topics and some suggested questions, but leaves the interviewer free to follow up on interesting responses and explore themes that emerge organically. Open questions beginning with 'how', 'what', and 'why' generate richer responses than closed questions with yes or no answers. Probing questions like 'can you tell me more about that?' and 'what did you mean when you said...?' help you understand what the participant is really saying.
Alex: Once you have data, how do you analyse it?
Sam: The approach depends on your methodology. For quantitative data, statistical analysis is central. Descriptive statistics summarise the data: means, medians, standard deviations, frequency distributions. Inferential statistics allow you to draw conclusions about a population from a sample or to test whether differences between groups are statistically significant. Choosing the right statistical test depends on the type of data and the research question.
Alex: And for qualitative data?
Sam: Thematic analysis is one of the most widely used approaches. You read through your data, identify patterns and recurring themes, code those themes systematically, and then organise and interpret them to answer your research questions. The process involves moving back and forth between the data and your emerging interpretations, refining your understanding as you go. It's iterative and interpretive, which is why qualitative researchers need to be transparent about their process.
Alex: Any advice on maintaining rigour throughout the process?
Sam: Keep detailed records of every decision you make and why. Record and transcribe interviews where possible rather than relying on notes. In quantitative research, define your analysis approach before looking at the data to avoid unconsciously cherry-picking the analysis that supports your preferred conclusion. In qualitative research, seek disconfirming evidence: actively look for data that challenges your emerging interpretations rather than just confirming them.
Alex: Brilliant. Thanks Sam. Next we look at how to communicate your research outcomes.