Sagitta Market Research Ltd

Consumer Quantitative Research

Tel: +44 (0)1303 262259 - Email:

Decoding Market Research Jargon A-Z (Part I, A-C)

By Helen Lester, Sagitta Market Research Ltd.

Market research, like most industries, has its own jargon. However, whilst as research professionals we use the terms readily, we appreciate clients outside the industry may not be so familiar with some terms. Over the coming months, we will therefore be providing a helpful guide with the aim of demystifying some of the terminology often used in the industry. This month, we are covering terms beginning with A, B or C. If you have any terms you would like explained or added, do contact us at

Analysis. Evaluating respondents’ answers (research results) in aggregate form or by various characteristics, in order to provide an understanding and recommendations in a response to the research objectives.

Brand mapping. A research technique where respondents are asked to position different brands based on the relationship between perceived key characteristics (e.g. product quality, innovation, environmental friendliness or price). This helps clients to understand the relative strengths and weaknesses of different brands and how they are perceived in relation to one another.

CAPI (Computer-Assisted Personal Interviewing). When face-to-face interviewers record answers to questions using laptops or tablets (e.g. iPads) in place of the more traditional pen and paper method.

CATI (Computer-Assisted Telephone Interviewing). CATI involves telephone interviewers typing respondents’ answers directly into a computer-based questionnaire, rather than recording them on a paper-based questionnaire.

CAWI (Computer Assisted Web Interviewing). Similar to CATI, but rather than the results being stored on the computer or iPad, they are entered onto a website browser and stored directly on a server for instant, real time results.

Closed question. A closed question is one where the respondent is not given the opportunity to elaborate on their answer (as opposed to an open-ended question), the answer being recorded against a predefined list of answers. An example would be: Could you tell me which, if any, foundation brands you have heard of? A closed question may be unprompted, as per our example (where the respondent answers without the assistance of a list of answers) or prompted (where the respondent is given a list of answers to choose from).

CLT (Central Location Testing or a Central Location Test). A type of research where consumers test a product in a central venue (see hall test).  Participants are recruited (e.g. in the street) according to set criteria based on the characteristics of the target consumer profile.

Cluster analysis. A method used to classify people or items into mutually exclusive groups based on two or more attributes (e.g. social classification).

Coding. Part of data processing process, whereby responses from open-ended questions are grouped with comparable answers and these are then categorised using numerical codes.

Concept boards. Used in qualitative research (e.g. focus groups, hall tests or in-depth interviews) concept boards depict designs of products, packaging, adverts or brand names.

Confidence interval (or margin of error). The range within which the percentage (value) giving a specific answer would fall if the whole population were interviewed. Usually calculated to the 95% confidence interval.

Conjoint analysis. A statistical technique used to demonstrate the relative importance of one rated characteristic over another. Often used in product development research to ascertain which product features are critical to purchase intention (versus others which could be dropped without significant consequence).

Continuous research (or longitudinal research). Research conducted in several phases (or waves) to establish trends and evaluate how opinions may be shifting over time. Continuous research typically involves asking the same questions to the same individuals each wave or asking the same questions to people who share similar characteristics (e.g. customer profile).

Correlation. The relationship between two attributes.

Cross-break (see cross-tabulations). Analysis of data tabulations by two or more attributes. Data tabulations typically include cross breaks so researchers can look at how people differ in their opinions depending on their profile (e.g. gender, age, product usage etc.).

Cross-sectional research. Cross-sectional research is the opposite of continuous research (or longitudinal research). A cross-sectional research study collects data in a single phase, as opposed to over several periods of time (phases).

Cross-tabulations (or contingency table). When one question is crossed with another question in the data tables. It shows researchers how the answers to one question are relative to the answers of another question. For example, one question may record the respondent’s gender (male/female) and another may ask purchase intention of a product. This cross-tabulation would show the purchase intention by gender.

CUT (Consumer Use Testing or a Consumer Usage Test). CUT can be carried out in-home, at a central location (e.g. hall test) or at another venue suitable for testing the product.