Sagitta Market Research Ltd

Consumer Quantitative Research

Tel: +44 (0)1303 262259 - Email: office@sagittamr.com

Decoding Market Research Jargon A-Z (Part 4, N-Z)

By Helen Lester, Sagitta Market Research Ltd.

Following on from our previous introduction to research jargon, we continue our guide to demystifying research terminology, with terms beginning with N, O, P, Q and R. If there is anything you would like explained or added, please contact us at office@sagittamr.com

Normal distribution A term used to describe a bell-shaped probability distribution, often referred to as a ‘bell curve’. A normal distribution is symmetrical and curved around the mean point. More specifically, around 68% of the values in a normal distribution lie within one standard deviation of the mean. Moreover, 95% fall within two standard deviations of the mean. In other words, in a study, 95% of the values are within +/- two standard deviation points of the mean.

Omnibus A cost-effective research methodology. Typically involving interviews with the general public at regular intervals (e.g. weekly or monthly), multiple clients purchase survey questions, thus sharing the research costs (e.g. sampling, interviewing and data processing).

One-way mirror Used in research studios. One-way mirrors enable researchers and clients to view the reactions and behaviour of respondents/participants during qualitative research (e.g. focus group, product testing). The respondent/participants can only see a mirror, whilst the researchers/clients observe them from the room behind the mirror.

Online survey (or web survey) A questionnaire completed via the internet, typically self-administered by the respondent, but sometimes administered by an interviewer.

Open-ended question Sometimes known as unstructured questions, the aim of an open-ended question is to ascertain a respondent’s detailed opinion. The respondent’s answer is recorded verbatim (either by an interviewer or the respondent directly) as opposed to selecting the ‘best fit’ answer against a list of predetermined answers (see closed question). An example of an open-ended question is: “What do you like about the fabric conditioner you have been testing over the past week?”.

Paired interview An in-depth interview involving two ‘paired’ respondents; typically two spouses or a parent and child.

Panel A pre-recruited group of respondents who have agreed to take part in research surveys. Typically, respondents complete a screener questionnaire before joining the panel, so the agency can select respondents based on a future survey’s demographic/household requirements.

Pilot study A small-scale study with a limited number of respondents to test how the questionnaire works before commencing fieldwork for the main study.

Product placement test (or in-home test) Respondents are given a product to test at home (or sometimes place of work) in order to gain their opinions about it in the environment in which it would be used. Typically, the respondent completes a self-completion diary during the placement and/or an interviewer gains their feedback during a follow-up interview.

Postal survey Self-administered surveys mailed to respondents with a prepaid return envelope. Sometimes an incentive is included (e.g. pen or voucher).

Professional Codes of Conduct Self-regulatory codes developed by professional bodies which outline industry best practice. Examples include those by the MRS (Market Research Society), ICC/ESOMAR (International Chamber of Commerce/ World Association of Opinion and Marketing Research Professionals). There are also industry specific guidelines developed by associations such as EphMRA (European Pharmaceutical Market Research Association).

Prompted question A question which provides the respondent with a list of possible answers.

Proposal A detailed outline of a research project, as designed by a research agency. Used by clients to i) select the agency they wish to appoint to undertake the research and ii) review recommended research options (concerning the most appropriate methodology, sampling method, sample size and associated analysis/reporting options and costs). Apart from outlining the proposed research project, a research proposal enables agencies to demonstrate their understanding/interpretation of the research brief, as well as showcase any relevant experience.

Qualitative research An unstructured and embryonic approach to research, where the researcher (interviewer or moderator) uses a topic guide or semi-structured questionnaire to explore people’s perceptions and motivations. Focus groups and depth interviews being two examples of qualitative research methodologies. The main purpose of qualitative research is to offer detailed understanding through open and exploratory discussion.

Quantitative research A structured and quantifiable approach to research. Whereas qualitative research provides insights into behaviour and attitudes, quantitative research seeks to provide statistically robust conclusions from structured questions.

Questionnaire design The process of developing a questionnaire for the fieldwork stage. This design phases often involves several draft documents before a questionnaire is finalised. It is important to ensure the questions fulfil the research brief and that the questionnaire will work ‘in field’ (e.g. that it flows logically, questions are not repetitive or ambiguous, etc).

Quota sampling A non-probability sampling method used to gather representative data from a group of individuals. Quota sampling involves setting target quotas for specific subsets within the population group. For example, if interviewing people after they have seen a show at the theatre, you might set quotas on gender, age and social class – perhaps even where they sat in the auditorium – to ensure you interview a representative sample.

Random sampling A sampling method where each individual in the population has an equal chance of being selected. For example, if interviewing people after they have seen a show at the theatre, random sampling could involve selecting people randomly as they leave the building.

Regression analysis A multivariate analysis procedure which examines the relationship between a dependent variable and one or more others in order to make quantitative associations and derive prognoses.

Research brief A document outlining a client’s research objectives. From this document, the researcher will prepare a research proposal.