Interview (conducting interviews)
Interviewing is systematically gathering data from the empery by asking people what they think or have experienced. These data are used for the analysis to answer the research question.
Compared to observing, the information is not withdrawn from the empery directly. Other people’s senses are used to find out what happened. Moreover, the object of study is often what people think about the empery (the reality) and not what happened in the empery. Market and marketing research are usually focused on the opinion of customers. It is not about whether products are of good quality, but about the perceived quality. It is the opinion of a customer that matters.
A proper way to get the opinion of people is simple: ask for it. But hold on a moment, if it is that easy then why a whole library is filled with books on interviewing? Well, there are several things to take care of.
First of all you have to be sure the individual to whom a question is addressed, understands the question. Besides that the interviewer who writes down the answer should know what the respondent wants to say. In short, they should understand each other. In science this problem is known as the problem of validity. To be sure the interviewer understands the respondent well, he should regularly summarize the given answers and ask for an explication if a technical term is used.
After the interview the interviewer goes home (or to his office) and has to work out the information in a suitable data base. For most interviews this might be text. It will help if the interview is recorded, so the interview can be re-listened as often as necessary. Take care not to misinterpret the words of the interviewed person. Mistakes are called interviewer bias. To be sure no mistakes were made, the text can be sent to the interviewed person so he can make corrections. This is called an authorised interview report. The corrections are not necessarily due to mistakes of the interviewer. The interviewed persons might change their of opinion or withdraw his opinion because he said things in an impelling moment.
In the second place, the answer do not always correspond with the reality. Respondents may give answers which will show that he is a better human being than he really is. People do not always tell the truth. This problem is known as the problem of reliability. There are several ways to find out if an interviewer gets reliable answers. In short, always try to check the answer with information from elsewhere, or doublecheck the information by asking a similar question (without annoying the respondent).
The interviewer must pay attention to disturbing influences in the room where the interview is conducted. Are only the interviewer and interviewed people/person in the room? Is it quiet there? Are children or relatives of the interviewed person in the room? Are these persons listening to the answers? Are they interfering the interview? All these aspects might influence the results.
3 types of interview
In textbooks 3 types of interviewing are distinguished: completely open (an unstructured interview), a complete closed (a structured interview) and semi open (semi structured interview).
A complete open interview is used when the interviewer doesn’t know much about the topic. For instance, if you want to investigate the healthcare in Finland, you might go to the ministry of healthcare and ask: “please tell me all about it”. You might be surprised about the information you get, maybe even information about things you didn’t even suspect would be nice to know.
A complete closed interview is used when you need specific information. For example, if you want to know whether an elderly person can still live on his own or needs to be admitted to a retirement home. Now the interviewer asks questions and has to fill in a prescribed form.
Most commonly used are the semi open interviews. In this type of interview the research question is well described and has a subset of questions. The interview only has the object to gather information about these topics. Any other topic can be disregarded. If the investigation is about the healthcare of elderly people in Finland, it is okay to have subjects like education and training for the co-workers, social views on how to take care of elderly people, costs and so on. If a respondent wants to talk about demographic changes, this might be interesting too. But if isn’t relevant, gently return to the topics of your investigation, because that’s what you came for and that is what you have to spent your time and effort on.
A group interview
So far we spoke about interviewing on a one to one basis. However, interviews can be done with two persons or even a group. There are even separate terms for this type of interview because it has special techniques, for example focus groups. Sometimes the interview is not even seen as a kind of research, such as brainstorming, and sometimes this is better, because research is more than just collecting data. These types should be mentioned here, because these technique can be used in research as well.
Interviewing is one of the five methods for gathering data to fill a database. The other four are observing, measuring with an instrument, copying and transferring data, and generating data. Every method is described separately. On this page interviewing has been described briefly. As mentioned before a whole library can be filled about this topic. The best tip however is to attend a course to prevent you from making big mistakes and then get started. Practice, practice and practice, till you become a very good interviewer.
And yes, I almost forgot to mention this, but scientists do not have to do the interview themselves. Sometimes its better to hire a well-trained interviewer. Such a person might do the interview all alone, but it is okay to watch through a one way screen, or even sit in the room too (but then don’t interfere the interview, because it can disturb the gathered data).
Related topics to Interview:
- Gathering data
- Transferring data
- Generating data
Deepen your knowledge and read our manual about ...
- How to analyse data from interviews