Evidence-Based Innovation Blog

Business School Qualitative Research: How many people should you interview?

Posted on Feb 9, 2011 9:15:00 AM

Find me on:

How many people talk to

Core to the success of a higher education market research study is method of sampling -- a sample is a group of people drawn at random from all those people who are in your chosen market. If a researcher does not design the right sampling plan, then the research results will likely be inaccurate; thus, the right design is a key, first critical step in defining a research study.

The sampling design for qualitative research (such as incoming student focus groups, MBA alumni online chats, etc.) is not a pre-defined process like quantitative survey research design. The latter is often based on certain sample sizes needed to provide statistically significant differences between key target groups or topic areas.

Qualitative research design, on the other hand, is not such an exact science. It is often rooted in setting up the right number of conversations to get a strong read on why, or how, people are talking about certain topics. The determination of how many people are required to provide this information is often based more on instinct and subject matter, than on statistical probabilities.

However, there are some solid rules of thumb that can be used to determine the right number of people per session to target in business school qualitative research, depending on the subject matter and study goals.

Use of 1:1 interviews

Higher education qualitative research aimed at understanding “expert” views, such as input from chief learning officers into an employer outcomes assessment, are often best done on a one-on-one basis due to the depth of inquiry often required, as well as the difficulty in trying to schedule a group of experts at the same time due to their busy schedules. In-depth interviews are also best used when a study is tackling some very personal or provocative subjects, such as in-depth follow-up of negative views expressed in an MBA student satisfaction study.

Use of mini-groups: 3 to 5 per session

In those cases where a study is aimed at very personal or provocative subjects, but there is a desire to understand the “social” aspect to these subjects, a mini-group is best employed to minimize potentially embarrassing situations which could impact study accuracy. In these instances, mini-groups often utilize groups of friends, who are often more willing to discuss sensitive topics with each other versus among a group of strangers.

Use of medium-sized groups: 6 to 8 per session

Medium-sized groups are best for MBA qualitative research where there is a fair amount of “brainstorming” or “conceptual” thinking required. Often, this size is best for research that is truly exploratory in nature, where a social aspect to the research is helpful in stimulating ideas, but where discussion will be stunted if there are too many people involved in the session. A student loyalty online bulletin board study could be one that fits into this category, especially if this study focused on generating the key factors that really engage students with a particular business school program.

Use of large-sized groups: 9 to 12 per session

In any other instance, large-sized groups are the norm; they are often best for studies aimed at understanding a wide variety of subjects, or connections between topics, without a lot of depth. An MBA market online chat study could fall into this arena, as the pros and cons of various graduate business school programs, and how they connect to a prospective student’s choice of a particular curriculum, could be discussed in an efficient manner. These large groups are also very helpful when a study’s timing constraints require talking to a lot of people in a short amount of time. Note that a typical maximum is twelve people per session; otherwise, groups prove too unwieldy to be conducted in an accurate way.

Other considerations

Beyond defining how many people to talk to per session, qualitative research design needs to also account for how many sessions are required to meet a research study’s objectives. This type of consideration is often more linked to the budget and time allowances provided by the study, as well as the demographics targeted (e.g., age, work experience, geographic location) than it is about subject matter. But there are some basic design parameters to consider:

1)    A study should allow for as many sessions as possible in total, to make sure that the largest number of people and viewpoints are included in the study.

2)   The number of sessions organized around demographic targets should be as balanced as possible. So, for instance, if planning MBA graduate online bulletin board study sessions across a wide range of alumni ages, there should be the same number of sessions planned among younger alumni, middle-aged alumni, and older alumni to ensure that each groups is properly represented.

3)   This balance should incorporate both census targets as well as any oversampling required by the goals of a study. For instance, normally males and females are evenly distributed within U.S. groups due to the similarities in size of both groups nationally. However, if an MBA alumni online chat study is focused on understanding the female alumni response to a specific topic, females would be the majority within each group while males would be a minority.

Another consideration is whether these sessions should be executed in-person or virtually -- discussed in a recent blog article.

Incorporating the parameters listed here will help to ensure that business school qualitative research design includes the right number of participants in the sample to facilitate accurate results.



Dinesh Mathew, the guest writer for this article, is a marketing consumer research expert and qualitative research consultant. Dinesh welcomes your questions and comments.


Topics: Dinesh Mathew, Qualitative Research, In-depth Interviews, Sample Size