Audience Dialogue

Glossary of research methods

This glossary lists the most common methods and approaches, particularly for quantitative research. See also the glossary of qualitative methods. Overlapping methods are listed in both glossaries.

Action research
A multi-stage type of research, in which a problem is researched, changes are made, the problem is researched again, more changes are made, and so on until the problem is solved.

Ad hoc
A one-off survey done for a specific client - cf.omnibus and syndicated surveys.

Audience research
Shares its techniques with market research and social research but focuses on audiences and communications. Mainly used by broadcast media, publishers, network computing, and community organizations. To many people these days, media research has an identical meaning.

Audience response cultivation
A group of methods for use mainly by broadcasters: not quite formal market research, not quite marketing, but something between the two. See also dialogue techniques.

Auditorium testing
Getting a large group of people (usually 100 or more) together in an auditorium, showing them TV or radio programs, and getting them to rate these - either with questionnaires or electronic gadgets. Sometimes called theatre testing. Similar to hall testing.

Business to business
Research whose respondents are businesses rather than consumers. The same as industrial research.

Survey of a whole population. Most countries have a Population Census (with a capital C) every 5 or 10 years, but a researched population can be much smaller. Thus a census (with a small c) of all staff of an organization would be a survey where everybody was sampled.

Central location
A type of research method where respondents are all interviewed at one venue - as opposed to having interviewers go out and interview respondents in their own places. See intercept, hall testing.

A workshop, often lasting several days, which involves a community in its urban planning process. Similar to a clinic or a search conference.

A type of research, usually done at a central location, where a product is displayed and respondents asked to use it and react to it. See also charrette.

Communication research
An umbrella term for market research, social research, media research and audience research combined.

Co-discovery conference
A method of qualitative research (developed by Dennis List, founder of Audience Dialogue) in which audience and producers discover each other's needs and use this knowledge to plan new programs. Based on the search conference and consensus group.

Consensus group
A type of group discussion, similar to a focus group, in which participants try to form a consensus on an issue. In contrast to surveys, which seek differences between people, this technique (originated and developed by our founder Dennis List) seeks similarities.

Content analysis
Analysing the content of media - e.g. publications and broadcast programs to determine the main themes being represented. This is a quantitative technique, which usually involves counting the number of times a word or theme appears.See also Media research.

Customer satisfaction measurement
A rapidly growing branch of market research: assessing the satisfaction level of an organization's customers. See also mystery shopping.

Delphi method
A way of estimating future measures by asking a group of experts to make estimates, re-circulating the estimates back to the group, and repeating the process till the numbers converge.

Depth interview
A type of qualitative research, which involves long, probing interviews without the use of a formal questionnaire. Sometimes called simply a depth: e.g. "As well as the survey we'll do 20 depths." (Doesn't that sound shallow?) Also known as in-depth interviews.

Desk research
Research done by summarizing published sources - a form of secondary research.

Dialogue techniques
Similar to audience response cultivation, but with a more qualitative emphasis. A systematic way of improving and making use of customer feedback. often involving a panel.

A type of qualitative research which treats a group of people as an anthropologist would an unknown tribe, with detailed descriptions of how they live.

Event survey
A type of audience research, where people attending an event (concert, sports match, etc) are surveyed. More commonly known as a visitor survey.

Focus group
A common type of group discussion, in which a moderator encourages a small group of people (usually 8 to 10) to gradually focus on a topic.

Group discussion
A generic type of qualitative research in which a small group of people provide information by discussing a topic. Some variants are the consensus group, focus group, nominal group, co-discovery conference.

Hall test
Getting a group of people (typically about 50 of them) together in a public hall, usually to see a product demonstration and to fill in questionnaires on the spot. A type of central location study. Similar to auditorium testing, except that with hall testing, people don't always need to be there at the same time.

Industrial research
Market research in which the respondents are organizations, not consumers. Much the same as business to business research.

Key informant interviews
A method of collecting information (usually factual) about a community or group of people, by finding and interviewing key informants. These are people who are likely to be well informed about an issue, and willing to answer without bias. For example, if you wanted to research TV reception in remote towns, you could telephone servicing shops. A sample of 3 per community is often enough (if they all agree).

Market research, marketing research - what's the difference?
Most people don't distinguish between these two terms, but "marketing research" (used more by academics) tends to have a broader meaning than "market research" (used by commercial researchers). The latter term often applies only to primary research,while the former sometimes adds secondary research and desk research. See also audience research, media research, social research, communication research.

Media research
This term seems to have developed two slightly different meanings. In the 1980s, media research referred mainly to researching the media, not their audiences. Thus media research was almost synonymous with content analysis. These days, many people use "media research" to include audience research as well.

A mechanical or electronic device that can count people's behaviour. A well known example is the TV peoplemeter.

Continuous research which regularly measures the change in some indicators. Similar to tracking.

Mystery shopping
A systematic way of assessing customer satisfaction, by having research staff pretend to be potential customers, and noting how frontline staff respond to their demands. Sometimes called shadow shopping. After the encounter, the interviewer fills in a questionnaire, so mystery shopping is a type of survey.

Nominal group
A type of group discussion in which participants work independently (on paper) at first, then present an idea at a time to each other. Often abbreviated to NGT.

A research technique in which no direct questions are asked, but people in a public place (e.g. shoppers and drivers) are watched and their behaviour recorded.

A type of survey (done regularly by most large market research companies) on which organizations can place a few specific questions. It's like a bus, on which a lot of people can travel at once.

Opinion poll
A type of survey in which people's opinions are asked, specially on topics related to politics.

A group of respondents who are surveyed a number of times, in order to detect changes in their behaviour or opinions.

PRA = Participatory Rural Appraisal
A qualitative method for involving communities (specially rural ones in developing countries) in their own futures. Not a single technique, but an approach to research, usually involving a number of simple stages. Now (2006) more often known as Participatory Learning and Action or PLA.

Usually the same as opinion poll, but sometimes loosely used to mean any type of informal survey.

Push polling
A type of pseudo-research whose intention is to change opinions (usually on voting) rather than measure them, often by asking leading questions. For example "When did you first become aware that Candidate A is the son of a criminal?"

Research in which questions are open-ended, and results are expressed in non-numerical terms. Contrasts with quantitative research. Often shortened to qual.

Quantitative research
Methods of research can be broadly divided into qualitative and quantitative. the basic difference is that quantitative research reports findings as numbers, while qualitative research reports them as words. The main quantitative research technique is the survey, with all its variants. There's a much wider variety of qualitative techniques - see the separate glossary for qualitative research.

RRA = Rapid Rural Appraisal
A simpler version of Participatory Rural Appraisal, with less participation by the population involved, with the appraisal done more (but not only) by experts.

Reception analysis
Also known as Reception theory. A type of audience research that focuses on what audiences perceive in the media - as opposed to what broadcasters think they produce. Similar to Uses and gratifications.

A questionnaire designed to be filled in by respondents - also called self-administered. Thus a self-completion survey is a survey using this type of questionnaire. See also diary.

Shadow shopping
Same as mystery shopping.

Combining different kinds of question in one survey, e.g. TV audience and product use. The opposite of fusion.

Social research
Uses the same techniques as market research, but focuses less on business and more on public issues. See also audience research.

A form of pseudo-research: Selling Under the Guise of research. This happens when somebody rings you up pretending to do a survey, but in fact trying to sell you something. Market research companies hate sugging, and will have nothing to do with it.

A whole exercise of measuring public opinion. Don't confuse a survey with a questionnaire: some people say "The interviewer did 50 surveys" when they mean 50 interviews, for one survey. As a verb, "to survey" is used much more loosely, and often means the same as "to interview."

Syndicated research
Research originated by a research company, with data sold to anybody who is interested - unlike an ad hoc survey, which is a one-off survey for an individual client.

A series of repeated surveys in which the same questions are asked, so that a measure can be tracked over time. Often used in measuring the reach of advertising. A form of monitoring.

Taking a variety of different research approaches to an issue, as if you're seeing it from different angles. Though different methods come up with different results, the results should be similar enough that they might be plotted on a graph as a small triangle. Somewhere inside that triangle is the real truth.

Usability testing
Originally, this referred to methods that measured the usability of electrical equipment. These days, most usability testing is of web sites, but it's also possible to test written instructions using the same methods.

Uses and Gratifications
An offshoot of audience research that developed in the 1980s. Instead of studying the content, or what media to do audiences, the Uses and Gratifications people study how people use programs. For example, when children watch sitcoms without laughing, maybe they're learning how (they think) adults behave. Five main types of uses and gratifications have been defined: for information, aesthetic feelings, personal needs, social needs, and escapism.See also Reception theory.

Visitor survey
A survey of visitors at a venue; also known as an event survey.