Ad Hoc :
In basic terms this means ‘as and when’ required. Generally applied to single surveys which are designed as a ‘one-off’ rather than continuous on-going research.
Ad Testing :
Closely, related to promotion testing, ad testing (a.k.a. advertisement testing) refers to various methodologies that focus exclusively on gauging the, perception, effectiveness, or targeting of advertisements in a market be they single adverts or a series. The major difference between ad and promo testing is that promo testing is usually carried out on a larger scale to measure the reaction to a campaign.
Ad testing can deal with adverts of all types across the spectrum and can be employed at any stage throughout the advertisement development process.
The most frequent use for ad testing is to identify the most effective advert(s) at the prototype stage which helps in turn to eliminate the likelihood of expensive and ineffective campaigns. Furthermore, it tests the appropriateness of an advert to its audience.
Some use the method to measure an advert’s competitiveness against a major competitor’s. In this case researchers ask respondents to read magazines or papers or ask them to watch a television programme with an advertisement break and then ask the respondents to recall the adverts that caught their attention or ask them for key information that the adverts contained. A deeper way of using this method is by conducting face-to-face interviews or using focus groups to gain consumer feedback to potential adverts or advertising campaigns.
Researchers aim to analyse the ability of the respondents to recall information, the affect that the advert has on a respondent, its persuasive power, whether respondents identify with the given setting in the advert, understanding of the appeal, and the respondents’ perception of the brand in question.
Used mainly in questionnaire formats, a balanced scale is one on which the number of positive and negative categories are equal, leaving the respondent with a fair decision.
CAPI stands for computer assisted personal interviewing. It is performed with the respondent face-to-face with the interviewer, using laptop computers or other such devices (e.g. IPAQ’s). The interviewer uses the computer to prompt the questions and the interviewee’s responses are immediately inputted back into the computer which enables the program to select the next appropriate question. This enables the interview to pursue many different, yet relevant, routes depending upon the responses given by the interviewee. This form of surveying is particularly effective because due to the fact that the data is already on the computer, the analysis becomes quicker than other methods.
CATI stands for computer assisted telephone interviewing. It is similar to CAPI but conducted over the phone, rather than face to face.
CAWI stands for computer assisted web interviewing. This is based on the same principle as the two above, apart from that it is conducted over the internet.
This is a study which incorporates every available element within a defined population.
Close – Ended Questions :
A question asked which can only be answered specifically. It is used in quantitative rather than qualitative research and is usually in a tick-box style.
Coding is the process of allocating codes to responses collected during fieldwork facilitating analysis of data. Coding normally relates to open ended questions where you have a number of verbatim comments. Coding allows the grouping of these responses so that they can be quantified.
Conclusive Research :
Providing information which helps the manager decide on a correct decision, conclusive research consists of formal research procedures including clearly defined goals and needs. Usually, a questionnaire is designed in conjunction with a sampling plan. There must be a clear link between the alternatives in the evaluation and the information that is to be collected. This line of research can include simulation, surveys, observations and experiments.
Conjoin Analysis :
Conjoint analysis determines the relative importance of product attributes in the consumer choice process. It can determine the make up of the most preferred brand and it also helps to segment the market based on the similarity of preferences for attribute levels.
Consumer Research :
Consumer Research is research where a respondent (for example a member of the public) is asked his/her opinion upon a particular good or service.
Observations or measurements made that apply to an area of the marketing system.
Data Editing :
During the process of data editing, the data collection instruments are tested to ensure that the maximum accuracy in terms of results has been attained. Checks are made on legibility, consistency and completeness and above all the process aims to avoid ambiguity.
Data Processing :
Data processing is the inputting of data into a database format. Once in this format, data can be tabulated and conclusions drawn as to the results of the market research findings.
Data Sources :
There are four main sources of marketing data. These are respondents, analogous situations, experimentation and secondary data. For more information on any of these sources, see glossary definitions.
Depth Research :
A term which is used to describe a multitude of data-collection techniques and procedures although it is usually used in qualitaitive research with individuals.
Desk Research :
This is a research technique which involves the market researcher collating and drawing together secondary sources of information. These secondary sources can serve as complimentary to primary sources or can be collated to form part of a larger unrelated project.
Dairy Methodology :
Diary methodology is a procedure which is used to compile consumer purchase data or media habits. During this procedure, respondents will be required to complete a written report which details their behaviour over a certain period.
Editing is the procedure that takes place in order to ensure that a questionnaire has been correctly completed.
Elasticity is used to measure the exact amount of volume shift responding to the variable under scrutiny.
Element Sampling :
During this procedure, every unit in a specific population has an equal probability of being chosen for research.
This is a technique that is used within a qualitative interview that allows the respondent to articulate their views.
An estimate is an assumption based upon a smaller sample that is then applied as a hypothesis for a whole population.Science projects for kids Science projects and science experiments for kid’s grades 1 to 8 that is safe and fun to do.
Various e-books range containing full step by step instructions. These science projects can be used to teach children the basics of robust research.
This is a test whereby the researcher performing, it has control over one, or a few independent variables and manipulates them to retrieve the most relevant data.
Used in the early stages of the decision-making process, exploratory research is used to assess the situation at hand with the minimum cost and time possible. The process must be flexible due to the unknown quantities that may be encountered. Versatility and a wide-ranged approach to the preliminary investigation are key. The exploratory research can draw on interviews, observations, group interviews, secondary data sources and case histories.
A F Test explores the probability that a specific result may have been down to chance and therefore inconclusive.
This is a simple test which ensures that a measurement or research actually targets the areas that it should.
Factor analysis of data identifies the underlying factors that explain the correlation among a set of variables. It is used to identify a new, smaller use of uncorrelated variables to replace the original set of correlated variables. It also identifies a smaller set of salient variables for use in subsequent research. It is used in various types of research, including product research, customer satisfaction, market segmentation and consumer profiling.
False accuracy occurs when a collection of data gives the impression of being accurate when in reality, only a low degree of accuracy exists. This problem can arise when results are being rounded to fit a pattern. E.g. Rounded to decimal places etc.
Fieldwork can be conducted by observation, surveys (such as face to face interviews, telephone interviews and web interviews) or experiments. It is the basic term for the live collection of primary data from external sources. Fieldwork is either co-ordinated by an in-house fieldwork department within a market research agency or an external fieldwork company. Once fieldwork has been conducted data processing is usually the next step.
Focus Groups (aka Group Discussions)
In a focus group, respondents (normally between 8-10 people) are gathered together in order to gauge their responses to specific stimuli. Groups are guided by a research moderator who often uses a topic guide to control the discussion to ensure it meets the initial research objectives. The data generated is probably most applicable to exploratory work. The technique falls under the broad category of qualitative research.
Freelance Market Researcher
These self employed market researchers usually work alone. They would typically have many years experience in the field of work and are contracted by market research agencies and consultancies in busy periods, effectively as an extra resource. In many cases, they usually help out with qualitative research techniques (for example, conducting depth interviews or group discussions) but they can also be utilised for quantitative research projects.
Frequency measures the number of times that an incident or event occurs. Used in data-analysis, it details the number of results that fall in the various categories.
During this procedure, a number of respondents are invited to attend a fixed location (traditionally a hall from which the term originates) and then are asked to respond to stimuli, usually individually.
This scale is used to measure the generalized overall views and opinions about a product or service.
Heteroscendasticity is the term used to express a non-consistent variance during regression analysis.
Groups in which the people or objects are extremely similar, if not identical in their characteristics or behavior.
The opposite of heteroscendasticity. It is used to describe a state of constant variance in regression analysis.
The generic term used to describe every person living in one unit of housing.
The in-depth interview is a method used primarily in qualitative research and is carried out upon individuals (one-on-one), rather than in a group discussion scenario. The interview takes part with a researcher and a respondent (the interviewee). The researcher normally uses a topic guide to guide the discussion and it usually lasts between 30 minutes to an hour. The interview helps the researcher to gain detailed and in depth information.
This is the payment made to a respondent in return for their participation in a research project and depending on the complexity and duration of the research, the incentive will vary.
The use of internet research has grown massively over the last few years due to the increasing popularity of the internet for both business and leisure purposes. The research itself can be qualitative and quantitative. Quantitative internet research can be performed as a web-survey where the respondents reply to questionnaire-based emails. This means that with the click of a mouse, the results are back in the inbox of the researcher ready for analysis. Qualitative research can be carried out for example by setting up group forums on the internet, or by setting up group discussions using webcams, thus mimicking the conventional group-discussion format.
An interview is the general term for the method used to draw information from a respondent. It can take several guises e.g. Face-to-face, in depth, telephone interview, group discussion etc.
The term used for a a sample which consists of respondents who have been selected with the understanding that their opinions, behavior, characteristics and so on, will be representative of the population as a whole.
This is the process by which two people input the same data to avoid the distortion of data at the data input stage.
The Likert Scale allows the respondent to specify an opinion on statements which relate to the subject being researched.
Longitudinal research is used in most instances where continuous performance monitoring is taking place which consists of repeatedly analyzing a fixed sample of population elements.
Market Research Brief
Also known as a research brief, this is a basic plan which guides the data collection and analysis phases of the research project. It acts as a framework which details the type of information to be collected, the data sources and the data collection procedure.
Modelling / Simulation
The application of specific assumptions to a number of variable factors and the relationships that exist between those factors. Usually used in hypothetical situations, the models can be verbal, mathematical or graphical.
This is where a respondent is asked to rate an item on some form of scale without comparison to another product or service.
A question which the respondent may only answer within a predetermined selection of responses.
This is a term for a whole range of analysis techniques that are used to examine quantitative data. These techniques such as cross-analysis of data (such as age, gender and socioeconomic grouping) allow data to be analyzed in far more depth than usually is the case. The data analysis is performed in such a way as to allow trends and patterns to emerge from within the data itself.
Mystery shopping is an objective system for measuring and monitoring what happens in the customer interface through a pre-determined scenario. It is a process of observation and recall and can be done on any scale. It is a process used by many groups such as clearing banks, building societies, restaurants, motor manufacturers, government departments and many more. For example, somebody posing as a customer in a supermarket (a mystery shopper) could analyze the performance of the staff, environment, length of queues etc. It is a form of objective, factual, research – the information that is recorded by the mystery shopper either happens or does not happen.
A scale of measurement in which numbers don’t represent value, instead their use is to label or categorize an object or an event.
A non-balanced scale is one which is unfairly weighted towards one end of the response spectrum and can often cause a response bias.
This is a sample which has been chosen to disregard in terms of representing the wider population.
A non-sampling error refers to any cause of bias or inaccuracy with the results that has been caused by anything other than the sampling.
Observation is the collection of data through non-verbal means which can stand alone as research on its own or compliment other relevant research. It is the process of recognizing and recording relevant objects and happenings.
These allow the recording of behavior when it takes place, therefore allowing for the elimination of such errors associated with the recall of behavior. It allows for the data to be more precise with less margin for error.
An omnibus survey covers a variety of topics, usually for different clients. The majority of the samples used are generally nationally representative and are made up of people who are in general demand. The charges for this element of research are based upon the size of the questionnaire produced in either space or questions required. It is a form of cost effective quantitative research – clients can share much of the cost of the survey. Questions normally cost around £1,000 each.
A question to which the answer has no definite response.
A panel is a long-standing sample that is retained by a market research agency from which data can be attained. It is most useful for continuous research whereby the same set of respondents is used on a continuous basis over time.
Once a marketing plan has been implemented, the performance monitoring, research simply monitors the progress of the plan. It ensures that the progress of the program is as planned as a deviation from the initial projected route can lead to the plan being carried out improperly or the result pattern inexplicably changing.
A circle which is divided into segments which vary in size relating to the value which they represent.
Postal research is carried out by sending questionnaires or journals, which are to be completed and returned, through the post. It is perhaps the most traditional form of data collection. The main drawback with postal research is response rates – it is difficult to achieve a high response rate.
These are used by moderators or research interviewers. They are used in interview situations when the interviewer will use the projective technique to enable the respondent to distance themselves from their personal views. It enables the researcher to obtain ‘the real views’ that perhaps wouldn’t be revealed without using the interview technique.
Qualitative research is in-depth research performed on a small scale to provide details in depth results and data. Qualitative research can be performed over the phone, via group discussions or through one-to-one interviews. Discussions are normally aided by using a ‘topic guide’ – this outlines the basic structure of the interview/group and indicates the general direction in which the interview/group should be led. Questions area of open nature as opposed to closed questions which are used in quantitative research. (See FAQ insert)
Quantitative research is performed on a far larger scale compared with qualitative research (in terms of the sample size) and helps to provide accurate statistical data from which conclusions can be drawn. Questions tend to be closed as opposed to open. (See FAQ insert)
This contains a group of questions and is used as an investigative market research tool in order to gain information from a respondent. It is used mainly in quantitative research. Questionnaires normally contain closed questions, but open questions can also be included.
A quota sample is one which has been selected to represent the population, sometimes incorporating certain control characteristics.
Also known as a market research briefing, this is a basic plan which guides the data collection and analysis phases of the research project. It acts as a framework which details the type of information to be collected, the data sources and the data collection procedure.
The research report is the compilation of findings from a piece of research. These findings are normally presented in the form of a report or a PowerPoint document.
A respondent is the person whose views and opinions are required by the researcher.
Response bias is the term for inaccurate response data which could have been caused by a number of factors such as monotony, tiredness, aiming to please its
In research terms, a sample refers to a group of interviewees or respondents who are chosen to represent the population as a whole. The sample provides the data within the market research project.
Secondary research utilizes data sources that already exist. The technique used to collect the information is called desk research (for a definition please see desk research). Using the web or visiting libraries would be two forms of desk research.
Used in research, semiotics is a type of social description and analysis which places specific emphasis upon an understanding and exploration of the cultural context within which the particular work is taking place.
Social Grade/Socioeconomic Grouping
Socioeconomic groupings are a way of separating various sections of society. There are six social grades in the UK socioeconomic grouping system. These are:
- A: Upper middle class. Higher level managers, administrators and professionals.
- B: Middle class. Intermediate manager’s administrators and professionals.
- C1: Lower middle class. Supervisory or clerical workers and junior manager’s administrators and professionals.
- C2: Skilled working class. Skilled manual workers.
- D: Working class. Semi and unskilled manual workers.
- E: Those at the lowest level of subsistence. State pensioners or widows (no other earnings), casual or lowest grade workers.
A sampling procedure which is carried out in two stages. The population is divided into two separate groups, mutually exclusive and collectively exhaustive, and a random sample is taken from each of the two.
The systematic gathering, analysis and elucidation of information about any aspect of study. In reference to market research it specifically applies to the gathering of information through processes such as sampling or interviews with selected candidates, respondents or interviewees.
The tabulation is simply placing the results and data obtained from research into data tables. (See cross tabulation)
The basic term for the population that is being studied.
A test market is used as a trial market for a new product or service.
Time series analysis
During this form of analysis, data is recorded at set intervals in time.
A topic guide is a brief guide which is used by a researcher in qualitative research interviews or group discussions. It enables the interview to retain relevance and ensures that the project meets the initial research objectives.
Tracking refers to studies that monitor consumer behavior towards a product, brand or service over a period of time and is carried out over a longer timescale than ad-hoc research. (See FAQ insert)
The trimmed mean is obtained by taking a percentage of the high and the low levels of data and then finding the mean.
Two Way Focus Groups
This is where one focus group observes another and then discusses what they have taken from the observation.
An unbiased estimator is when the mean of the sample distribution and the estimated distribution is equal.
These are samples that derive from an unbiased source, thus ensuring that any sampling error is a result of the randomness and nothing else.
These are procedures that are designed purely to measure only one single attribute of an object or respondent.
This is an ordinal scale which has a positive end and a negative end.
Unstructured observation is where a study takes place with an observer or moderator simply watching and taking notes on the behavior on display.
This is the process of market segmentation which uses data and analysis when no previous ideas or views are held regarding the segmentation of that particular market in any way.
A post-research essential; Validation in the process by which a respondent verifies that the interview was conducted correctly.
Validity in reference to market research is concerned with whether the purpose of the research was fulfilled accurately.
A verbatim is a record of actual spoken comments by a respondent. These are used to back up any research findings in the final report.
Video Focus Groups
These are focus groups that instead of meeting face-to-face, carry out the group via a video conferencing link.
Viewing facilities are the premises used for conducting group discussions or depth interviews in qualitative research. These premises can contain rooms where, when inside, the respondent can be observed by the client using one-way mirrors or via a video link. It is thought that because the client is not physically present in the discussion, the respondents may be less inhibited and more honest and open about their opinions and feelings.
A technique in which the respondent is confronted with a word and is asked to respond with the word that immediately comes to mind.
A test which draws comparisons between a sample test and a hypothesized test. It is used when working with interval data and any data on a large scale.