Because you do things in a sensible order, you have your research question, right? Good. It’s very important to have that first. The method (or methods) you choose should be the one (or ones) most likely to help you answer your question. You can’t figure out which methods are most likely to help if you don’t yet know what your question is. So if you’re actually not sure of your question, stop reading this RIGHT NOW and go settle your question, then come back and carry on reading.
OK, now you definitely have a question. You’ll probably have an idea of what kind of research method may be most help. (If you don’t, I recommend you get into the research methods literature, such as this book.) Let’s say you think your question could be answered most usefully by doing a bunch of interviews. Your next step is to think through the pros and cons of interviews as thoroughly as you can. You may find it helps to read relevant excerpts from the research methods literature. For example, here is a breakdown of the pros and cons of interviewing, taken from page 143 of my book Research and Evaluation for Busy Students and Practitioners: A Time-Saving Guide (2nd edn; Policy Press, 2017).
| Pros | Cons |
| Interviews yield rich data | Interviews are time-consuming for researchers and participants |
| Face-to-face interviews let the interviewer include observational elements, e.g. from the participant’s appearance or body language, that are not available with other methods | The researcher’s interpretation of the data from a face-to-face interview may be affected by the quality of the rapport they developed with their participant |
| Interviews can be conducted by telephone, which saves time and costs and increases anonymity | Not everyone is comfortable using the telephone, and it can be harder to create a rapport over the phone than in person |
| An interview equivalent can be conducted by email, which avoids transcription and so saves time and money; this also helps in reaching some groups of people e.g. those with severe hearing impairment | Conducting an ‘interview’ by email can make it more difficult to follow up interesting answers with supplementary questions |
| Interviewers can follow up interesting answers with supplementary questions | Interviewers’ input can influence participants’ answers |
| Unstructured interviews can be particularly useful at the exploratory stage of a research project | Unstructured interviews run the risks of missing important issues or degenerating into a general chat |
| Semi-structured interviews allow participants to participate in setting the research agenda, which may be more politically acceptable, lead to more useful data, or both | Semi-structured interviews make it harder to compare data from different individuals or groups |
| Structured interviews enable clearer comparison of data from different individuals or groups | Structured interviews require the question designer to be able to consider all the issues that are relevant to the participants |
| Recording data enables exact reproduction of someone’s words and pauses | Transcribing interview data is time-consuming and expensive |
This kind of thinking will help you to decide on your research method. Also, you will need to be pragmatic. For example, if you have a very tight deadline and no time or budget for transcription, then interviewing is not a good idea however much it might fit the research question. In such a case you would need to consider other methods. My book contains similar ‘pros and cons’ tables for using secondary data, questionnaires, focus groups, documents as data, observational data, visual data, and collecting data online. Of course this is not an exhaustive list, and if you’re considering using, say, mobile methods, soundscapes, or ethnography, you might need to construct a ‘pros and cons’ list of your own. To do this, you would need to read, watch videos, and talk to people with more knowledge about the method of interest.
Once you have established the pros and cons of the method, these need to be weighed against pragmatic considerations of available time, money, and other resources. This assessment will be different for each research project, in its own context; there are no hard-and-fast rules. But however you do an assessment like this, your results will always be better than those of someone who uses the method that first springs to mind. For sure you may end up using the method you first thought of, in which case you might say to me, Helen, what is the point of doing all that thinking? The point is you’ll be making a considered and informed decision to use the method. That means you’ll be able to justify your decision to readers, reviewers, tutors, supervisors, managers, examiners, or whoever else has an interest in your work.
Helen Kara 
Pingback: Developing questions for a semi-structured interview – Anuja Cabraal