Glossary of Research Terms

Market researchers, like all specialists, have their own peculiar language. From time-to-time, you may come across one of the following research terms. Here's what they mean:

Access Panel: Households recruited by a research company which agree to take part in research several times a year.

Ad hoc studies: Customised research conducted for a specific client. Ad hoc studies differ from recurring research such as indexes.

Advertising research: Assessing the effectiveness of advertising. This discipline includes evaluating the quality of advertising through pre-and-post tests, assessments and campaign tracking as well as brand assessment and brand equity.

Advertising tracking: Analysis of a brand's communication dynamic and its impact on consumer attitudes and behaviours. In general, information is collected weekly and uses rotating samplings (unlike panels). Brand tracking based on the same model recently appeared in France (analysis of the changes in image and recognition parameters, etc.).

CAPI (Computer-assisted personal interviewing system): A face-to-face computer-assisted interview. The interviewer uses a laptop computer, reading the questions from the screen and keys the replies directly into the memory.

CATI (Computer-assisted telephone interviewing system): The interviewer conducts the questionnaire by telephone and records the replies directly into a computer.

Copy testing: Testing of an advertisement either before or after it runs, or both (e.g.. Ipsos-ASI Next*TV, Ipsos' global film pre-test product).

Data processing: Taking the raw data gathered by interviewers, finding and correcting any errors, tabulating it and creating spreadsheets similar types of documents so that response patterns can be read and interpreted.

Face-to-face: Research conducted in person whether at home, office or in the street).

Fieldwork: All of the operations involved in data collection and in overseeing the work of the interviewers (interviewing, instructing interviewers on respondent selection, keeping track of quotas, verification or monitoring of interviews, etc.).

Field & Tab:  A basic level of research service that includes fieldwork and data processing.

Multiclient studies: As the name suggests, a multiclient study is research that is sold to several clients. There are two kinds: (i) syndicated research conducted for groups of clients (e.g. media audience measurement studies carried out for industry associations in each medium) to which the findings of these studies belong and (ii) subscription research conducted by Ipsos in its own name, whose findings are then sold to several clients (e.g. the La France des Cadres Actifs).

Omnibus study: A type of quantitative research in which only the sample of persons to be surveyed is defined beforehand and in which each participating client may ask the questions of its choice. This type of study also belongs to the multiclient study category.

Panel: A representative sample of individuals or professionals regularly surveyed on identical variables. The information may be collected automatically, since it does not require the direct participation of the participants (e.g. television viewer panels in passive audience rating). Information may be gathered directly from professional sources using sales statements (e.g. distributor panels).

Pre-test: A group of techniques designed to check the value of the communication of an advertising message before it is aired or published.

Qualitative research: An exploratory study (to explore an unknown sector, identify the main dimensions of a problem, draw assumptions, understand motivations) or operational study based on in-depth analysis of interviewee responses (in a group or individually), typically in what's known as "focus groups." It most often deals with a restricted sample of individuals that does not necessarily need to be representative. It may be the preliminary phase of a quantitative study or stand alone research.

Quantitative research: A study that aims to quantify attitudes or behaviours, measure variables on which they hinge, compare, and point out correlations. It is most often conducted via a survey on a sampling that must be representative so that the results can be extrapolated to the entire population studied. It requires the development of standardised and codifiable measurement instruments (structured questionnaires).

Questionnaire: A group of questions submitted to the persons surveyed in a quantitative survey.

Survey: A survey on a representative sample of the target population.

Survey/Index: Study designed to follow indicators on markets, companies, brands, products, services, etc. over time. It is conducted using the same methods (same sampling structure, same questions, etc.). It is used to make comparisons from one period to another.