A false dichotomy is sometimes drawn between the need to structure a research project around either qualitative or quantitative techniques. In practice, most research studies will utilise elements of measurement (quantitative) as well as understanding (qualitative). A semi-structured survey will seek to mix elements of both.
When conducting a semi-structured interview, the interviewer will usually follow a questionnaire that is a hybrid of a qualitative “topic guide” and a quantitative questionnaire. As such, a semi-structured interview will include a mix of both closed and open questions.
One benefit of working from a discussion guide with some structure is that it ensures there are set questions asked of all respondents, allowing the researcher to make comparisons between the different responses.
In addition to this, semi-structured surveys also offer the interviewer some flexibility. This means that they can adapt the discussion as they go along, to suit the individual respondent and the answers being given. The interviewer has the ability to ask follow-up questions, and encourage the respondent to speak in more depth about particular topics of interest.
Interviewing in this manner allows for themes and ideas to emerge, which may not have been known or predefined prior to the interview taking place.
Semi-structured surveys are ideal for use in exploratory research projects – such as trying to gain understanding of a particular market. They can help the researcher to develop knowledge and understanding, which may help them to inform later stages of the research project.
In business-to-business research, semi-structured surveys are useful for gathering a large range of views and responses from very diverse targets. Semi-structured interviews are also adaptable, meaning they can be administered over the telephone or face-to-face, with individuals or with groups.
A semi-structured approach also works well where:
Time or budget is limited: Classically, qualitative research would precede quantitative measurement. However, adding a discrete qualitative stage to a research design can add considerable time and expense to a study. A single, semi-structured stage is can often be a helpful compromise.
The universe of target respondents is limited: This is especially observed in b2b markets. In these cases, each interview must work hard in delivering a range of information about the target audience. Typically this includes both deeper understanding and measures of key issues.
Semi-structured surveys are best administered by experienced interviewers, who have the ability to think on their feet and ask relevant, spontaneous questions to further the discussion. Where necessary, the questionnaire should also provide a clear prompt to the respondent where a more structured answer is required, as this might not be clear from an interview that is otherwise quite discursive.
Interviews conducted in this format should ideally be audio recorded to aid the accurate capture of open-ended responses as part of the written interview transcript.
A close review of the first few interviews completed as part of a semi-structured survey is also advisable: This “pilot” phase allows the researcher to identify areas where open questions can be adapted to closed ones (and vice versa).