Sampling

Even though it may look like random sampling, asking people in the streets is just an arbitrary selection of respondents and very prone to being biased. You shouldn’t pick anyone passing by, but select the respondent deliberately.

In fact, true random sampling with personal interviewers is not very common in market research, as it would be too expensive in the vast majority of cases. Simply imagine the travel expenses of the required interviewers to distribute the sample geographically. In addition, some groups of people would be really hard to reach: telephone or online samples have more advantages in this regard.

However, personal interviewing has clear benefits for many research questions, especially product and concept tests. In these cases, a test studio allows you to prepare the product samples for the participants of your study in the most efficient way, especially if you require refrigeration, heating or any other stationary equipment.

As said, if respondents are sourced from intercept samples in shopping malls or pedestrian zones, the quality of the sample is defined by how active field control is exerted. As random sampling is not possible, it mainly depends on the interviewer to select the right mix of people and engage with them for the study. Therefore, sample quality in personal interviews has a lot to do with selecting and training the interviewers.

Selecting the right interviewers

One of the first textbooks about market research is Lyndon O. Brown’s “Marketing and Distribution Research” from 1937. In this book, he describes the main principles of selecting the right interviewers and most of these recommendations are still valid today. In general, an ideal interviewer has to meet two requirements.

Appearance

Research has shown that answers are less biased, if the social difference between the interviewer and the participant is small. This is why you should avoid deploying interviewers who belong to very special sociodemographic groups.

“Experience shows that one who is rather plain looking and not to well-dressed is much more satisfactory. The investigator’s voice and manner of speaking should be pleasant, clear and firm. Some researchers prefer that investigators do not appear too intelligent or sophisticated, as these qualities may put the person interviewed on guard and make it difficult to obtain the desired information” (Lyndon O. Brown)

The relevant factors of social distance may have changed over the decades, but the general principle is still true: interviewers need to engage with all kind of people and must not attract merely some and repel other target groups.

This is not just a matter of the outer appearance, but depends on the interviewer’s personality as well. Does the interviewer dare to engage with all groups of people? Imagine a male interviewer who is too shy to talk to women (or too persistent). The outcome of this research would be massively biased. Neutrality is also a matter of the personal attitude and this brings us to the next point.

Attitude

“There are many opportunities for dishonesty in spite of the care with which supervision and systems for checking the work of field interviewers are developed. Personal reliability is one of the most single qualification for field investigators” (Lyndon O. Brown)

These words are still true. Interviewers will work independently during most of their time, as it is simply impossible to supervise them permanently. And working as an interviewer can be really hard. It includes adverse experiences like standing in the cold or being rebuffed. At the same time, you usually have very strict instructions on whom to contact (i.e. quotas) and how to exactly conduct the fieldwork. Interviewers who don’t have the right attitude may take shortcuts and violate the quality standards of a study. Hence, quality mainly depends on interviewers having the right mindset and discipline when being alone in the field.

Obviously, it is not always easy to find such interviewers. This is why we think that fieldwork agencies should try to tie in the most experienced interviewers and maintain long-term relationships with them.

Basic training of the interviewers

Once we have found suitable interviewers, we give them a basic training on how to professionally conduct interviews. As we have seen in the previous paragraphs, interviewers can put the overall quality at risk if they are not conscious about the pitfalls of fieldwork and have learned the right techniques to deal with them.

In order to reduce personal variation and standardize the interviews of different persons, we train our interviewers in communicating the right way. Among other things, this includes how to approach and talk to people, how to present oneself, what basic information should be provided (e.g. duration, privacy policy) and how to deal with common situations.

As long-term relationships are so important when striving for a superior quality, we believe in a constant exchange between the supervisors and our interviewers. If interviewers get feedback about the quality of their job, they will become even better over the course of time. So, as mentioned before, we rather don’t want to work temporarily or even sub-contract external interviewers, but always prefer to have our own staff on the job.

Project briefing

Finally, we brief our interviewers about the specifications of a given project, for example the right target group, the screening questionnaire, the process of the study or how the technical equipment should be used. For many clients, the project briefing seems to be the main step towards data quality, but probably it’s just the most visible one.

So much more needs to be done in order to get a good quality when doing personal interviews. Many things can go terribly wrong, if you don’t have trained and experienced interviewers on the job. Therefore, data quality has a lot to do with developing the skills and maintaining the personal relations with your interviewers.

Coming back to our initial question, is personal interviewing an easy and available way of doing fieldwork on one’s own? It definitely isn’t. Whenever you ask people, you will get some kind of answers, – that’s for sure. However, that doesn’t mean that your insights are solid and can withstand critical questions. Only if you do your research with scientific rigor and operational excellence, you can really depend on the results.