Audience Dialogue

Qualitative research methods

The traditional way to do research is to count everything in sight, then to perform elaborate statistical tests on the numbers. With qualitative research, you take a totally different approach: you deal with words, not numbers.

The words can come from various sources. For example:

Consensus groups

Here's a detailed paper on Consensus Groups - a qualitative technique that I (Dennis List) have been developing over the last 10 years or so. My intention was to create a qualitative method which relatively untrained people could use to obtain reasonably accurate qualitative information. The key to the solution was for participants to play a much greater role than with traditional focus groups. Though focus group moderators need both training in psychology and extensive experience, consensus groups can be conducted successfully after only a few days' training.

The co-discovery conference

This is another research method developed by Audience Dialogue. We began with a family of techniques such as the Search conference and Future Search, to create a new method that's not quite research and not quite planning. With traditional research, the process works like this:

  1. Client has problem and calls in researcher.
  2. Researcher converts client problem into questionnaire.
  3. Researcher carries out survey and analyses survey results.
  4. Researcher writes report.
  5. Client reads report (maybe!) to find answer to original problem.

The co-discovery conference, by bringing client and audience together in carefully controlled conditions, bypasses most of those steps. It works like this:

  1. Client has problem and calls in researcher.
  2. Researcher organizes co-discovery conference.
  3. Audience members at co-discovery conference solve client's problem.

Instead of the audience knowledge being passed back by the researcher, the audience members communicate directly with the client. Though the process sounds simple when described like this, it's actually quite radical in terms of research. And (as we've discovered over the last five years) it's not as simple as it seems: if the psychological environment isn't well planned, the knowledge transfer doesn't happen well.

In-depth interviews

When a topic has already been researched in detail, it's normal to commission a survey. You know enough about the background of the situation to ask exactly the right questions, and thus get appropriate answers. But when little or no research has been done previously, a survey will not usually be much help - because some key questions were missed. In this situation, a set of in-depth interviews is more useful than a survey. The sample size will be a lot smaller, but that doesn't matter. The aim is to get understanding, not numbers.

Because of the time required, and the high skill level needed, in-depth interviews aren't cheap. A study with 30 interviews may cost as much as a survey of 500 people. One way to reduce the cost while retaining objectivity is for us to train client staff to do these interviews. You can read more about in-depth interviews in Chapter 10 of Know Your Audience.

Further reading on qualitative research

When I started doing qualitative research around 1980, I could find nothing at all written on this topic. Now there are hundreds of books and journal articles. Unfortunately, most of them are all but unreadable. Sometimes I get the impression that the authors are a little ashamed of the methods they used for qualitative research, so they use long words and appeals to authority to hide the fact that what they've done is so obvious that it hardly qualifies as a Method. That applies to academic researchers: commercial ones are too busy researching to write about it. The commercial researchers, I suspect, are secretive because they don't want potential clients to think it's easy to do this kind of work.

My own view is that qualitative research is a craft, perhaps like making wooden furniture. Anybody can buy the tools, anybody can make furniture, but to make high-quality furniture takes a lot of skill and practice. But buying a service in which you're not an expert - anything from medical or legal advice, to qualitative research - is like buying a of furniture that you can see only from a distance. You can't tell how well it's been made, so you have to choose a supplier based on trust.

As it's incredibly time-consuming to compile a reading list, and even more so to keep it up to date, I decided to cheat, and simply make links to some other web pages with annotated bibliographies on qualitative research.

A bibliography by Jan Nestor, divided into general books, observation, interviewing, life history, case studies, analysis, writing up research, personal accounts of fieldwork, access and ethics, uses of photographs, statistical methods, evaluation, and a few other bits and pieces.

General references on qualitative research in information systems. This site at www.qual.auckland.ac.nz has a wealth of useful information. Fortunately, qualitative research into information systems uses the same methods as qualitative research into anything else, so the usefulness of this site is broader than it might seem.

Sociosite - another useful resource in the Netherlands, with lots of links to pages on research methods and data - not only qualitative. [#link to be added]

Bob Dick's Action Research site at Southern Cross University in Australia [#link to be added]

Software for qualitative research

The section about software outgrew this page> Now there is a separate Qualitative Software page, which discusses the different types of software that can be used in qualitative research.