Marketing Research Glossary - B
B2B: Business-to-business marketing or marketing research
B2C: Business-to-consumer marketing or marketing research
Baby Boomers: The large generation of Americans and Europeans born after World War II. Usually defined as those born between 1946 and 1964.
Baby Boomlet: The children of the U.S. Baby Boomers that resulted in a peak in births between 1977 and 1990.
Baby Bust: A time period (such as between 1965 and 1976) when birth rates drop rapidly and remain low.
Back Office: A reference to operation functions in marketing research companies, such as sampling, field services, mail processing, and tabulation.
Back Room: A room with a one-way mirror from which clients observe and listen to focus group respondents. Also called the Observation Room or Viewing Room. Term is sometimes used to mean “back office” functions.
Back Translation: Typically used for international surveys. Questionnaire is first translated into another language and then translated back into the original language by a different translator. The objective is to ensure that the original translation is accurate. Decision Analyst believes that simply having two independent translators work on a questionnaire yields the most accurate questionnaire.
Backup Sample: See Emergency Sample.
Balanced Incomplete Block Design (BIBD): An experimental design in which respondents see only subsets of the experimental variables. The design is considered balanced when the number of times each variable appears and the number of times each pair of variables appears are constant.
Balanced Scales: Scales with the same number of positive and negative categories.
Bandwidth: The volume of data transmitted over an Internet connection, or digital device, in a given time. Typically measured by bits per second (bps) or bytes per second. At Decision Analyst’s data-collection web servers, for example, the bandwidth is so great that up to 200,000 surveys per day can be conducted.
Banner: The heading for a page of cross-tabulations.
Banner Ad: An ad placed on a website, usually at the top of the computer screen.
Bannerpoint: The heading for one column of data on a page of cross-tabulations.
Base: The sample size or number of respondents on which the percentages in a table are calculated (i.e., it’s the divisor).
Baseline Market-Segmentation Study: The first market-segmentation study conducted; the benchmark for the future.
Basic Research: Research aimed at expanding knowledge rather than solving a specific problem. This term is most often used to reference scientific research, and it is often contrasted with Applied Research.
Batch (Data Entry): A group of paper questionnaires to be entered, usually between 10 and 50 (depending on the number of pages and/or complexity). Questionnaires are signed out to data-entry personnel in batches, and they are signed back in when completed (as tracking and quality-assurance procedures).
Batch Number: A unique number assigned to each batch of paper questionnaires for data entry.
Batch Ticket (Data Entry): The cover sheet for each batch of paper questionnaires (contains the study number, batch number, and identification numbers of the questionnaires in the batch, as well as the initials of the data-entry person and the data-check person).
Bayesian Statistics: Statistics that incorporate prior knowledge and accumulated experience into probability calculations. Decision Analyst uses Bayesian statistics extensively in modeling and simulation work.
Before-And-After Control Group: True experimental design that includes random assignment of people to an experimental (or test) group and to a control group, with identical premeasurements and postmeasurements of both groups.
BehaviorScan: A research service that reports market share and other variables for consumer packaged goods based on a large household panel for data collection.
Benchmark: A “control” to compare study results against. For example, you might compare the results of a study in one country (the benchmark) to the results of a study in another country, or you might compare the first wave of a tracking study (the benchmark) to the second wave of that study, or you might test your advertising against your competitor’s advertising (the benchmark). Decision Analyst always recommends that some type of control or benchmark be incorporated into every research project.
Bernoulli Response Variables: Also known as dichotomous or binary variables. The value of each element is one of two possibilities such as yes/no or on/off.
Beta Distribution: A family of continuous probability distributions defined on the interval [0, 1]. Because the beta distribution is bounded on both sides, it is often used to model the distribution of order statistics or ranks, or for representing processes with natural lower and upper limits.
Bias: The difference between “truth” and the estimates of truth based on a survey.
Biased Sample: A sample that does not as accurately represent the target population. Decision Analyst begins every study by screening a nationally representative probability sample of the U.S. population (or other country), to identify the users of a product or service category to be surveyed. This procedure always guarantees a truly representative sample of the survey population.
Bibliographic Database: An index of published studies and reports with citations of author, publisher, dates, etc.
Bid: Estimated price to conduct a custom marketing research project, given a set of requirements and/or specifications.
Big Data: While there are many definitions on the Internet, big data is data that is large in volume, contains a large variety of information that can be either numeric or text, and can be retrieved, assembled, and accessed very rapidly. As these data sets are constantly growing, so too is the definition of how large big data is.
Big Data Analytics: Refers to the use of advanced analytics and predictive modeling to cull and retrieve information out of multiple big data sets. Ideally, this information would be useful and aid in decision-making.
Bimodal: A distribution in which the frequency curve has two peaks. A single peak is called a Mode.
Binomial Experiment: An experiment that consists of repeatedly drawing independently from the Bernoulli population; the sequence of Bernoulli trials.
Bipolar Scale: A scale with opposite end points, such as “sour” versus “sweet” or “good” versus "bad."
Birth Rate: Number of births a year per 1,000 members of a population.
Birthday/Anniversary Emails: Decision Analyst sends birthday cards and anniversary cards to its panel members via email, to strengthen its relationship with the panel members and to help create a sense of community.
Bivariate Data Set: Data set that consists of two measurements (variables) on each experimental unit or respondent.
Bivariate Regression Analysis: Analysis of correlation between two variables where one is the independent variable and the other is the dependent variable (the variable you are attempting to explain or predict).
Bivariate Techniques: Statistical methods of analyzing the relationship between two variables.
Blacklisting: The process of ISPs blocking research companies from sending emails or contacting their panel members. Decision Analyst subscribes to Habeas, a service that guarantees that Decision Analyst's emails reach its panelists. This prevents distortion of the sample related to ISP-blocking activities.
Blind testing: The testing of products or ads with all brand identity removed. Brand identity can introduce bias into a research project. (In fact, that is the whole purpose of brands-to create a positive bias toward a brand.)
Block Numbering Area (BNA): Prior to Census 2000, statistical subdivisions within nonmetropolitan counties for grouping and numbering. BNAs were discontinued for Census 2000, when they were replaced by census tracts.
Blocked Calls: Telephone-survey calls that receive busy signals.
Blocks: Census areas usually equivalent in size to a typical city block.
Blog: (Also called weblog) An online journal or diary.
Boost/Booster: Same as a sample Augment or Sample Supplement.
Bootstrapping Modeling Techniques: Bootstrapping is a general approach to statistical inference based on building a sampling distribution for a statistic by resampling from the data at hand. Bootstrapping modeling creates a large number of data sets, then computes the statistics for each data set. These statistics from each data set are analyzed for standard errors. Bootstrapping modeling techniques are used in Predictive Analytics. Learn More
Bounce-Code Processing: Decision Analyst processes and codes email “bounces” continuously during surveys, so that potential problems in email delivery (and, hence, potential problems in sampling) can be identified and corrected.
Bounce Codes: Codes assigned to undelivered emails returned from other Internet servers. Decision Analyst tracks every email “bounced,” in order to identify potential survey problems during the execution of a survey.
Boundary: The border around a market area that is being studied, or the border of a sampling area.
Boundary Files: Geographic features, such as streets, railways, or blocks, described or coded in a way understandable to a computer.
Box Plot: A graphical tool used to picture the distribution of the data.
Brainstorming: A creative method of coming up with new ideas or solutions to a problem by generating a large number of ideas, without subjecting them (or the person who suggested the ideas) to critical evaluation. A similar and related term is “ideation.” Decision Analyst owns and operates Imaginators®, the largest creativity panel in the world. This panel is used for new product ideation, advertising concept development, new positioning concepts, etc.
Brand: The name of a product or service that identifies that product or service and, hopefully, distinguishes that product or service from competitive products/services. Over time, a brand can become a reservoir (or symbol) of values and psychological benefits above and beyond the product/service itself.
Brand Associations: Images, emotions, colors, values, and other meanings that consumers attach to, or associate with, a brand.
Brand Awareness: The degree to which consumers are aware of (i.e., have seen or heard of) a brand. It is typically expressed as the percentage of the target population that is aware of the brand.
Brand Awareness, Aided: The percentage of respondents aware of a specific brand when asked, “Which of the following brands have you ever seen or heard of?”
Brand Awareness, Unaided: The percentage of respondents aware of a specific brand when asked, “When you think of peanut butter (i.e., category), what brands come to mind?” PROBE: “What other brands of peanut butter can you think of?”
Brand Equity: The level of awareness and consumer “goodwill” generated by a company's brands and/or products.
Brand Equity Monitor™: Decision Analyst’s proprietary model that measures relative brand preference based on all aspects of the brand, including both rational and emotional perceptions of the products/services. Learn More
Brand Footprint: A European term for Brand Image.
Brand Image: The total impression created in the mind of a potential consumer by a brand and all its functional, perceptual, and psychological associations. Brand personality, brand character, and brand expression are terms with similar meaning.
Brand Loyalty: The preference by a consumer of one brand over another, often resulting in repeat purchases for the preferred brand.
Break Off: The act of a respondent aborting a survey during its execution.
Briefing: A training session, prior to starting work on a study/survey, in which all of the survey specifications, questionnaire parts, and details of the interview are reviewed, explained, and clarified for all interviewers assigned to the project. This is generally followed by practice interviews being administered by one interviewer to another.
Broadband: See Bandwidth.
Browser: A software program that allows the user to access the web. Firefox, Chrome, and Internet Explorer are some examples of browsers.
Bulletin Board: A software system that permits people to have discussions online, read postings, add comments and feedback, upload and download files, etc. Decision Analyst conducts Time-Extended™ Online Discussion Forums using bulletin board software. Learn More
Buying Intent: Also called Purchase Intent. A five-point scale used to measure the likelihood that a respondent will purchase a product or service: “definitely buy,” “probably buy,” “might or might not buy,” “probably not buy,” and “definitely not buy.”
Contact Decision Analyst
If you would like more information on Marketing Research, please contact Jerry W. Thomas by emailing email@example.com or by calling 1-817-640-6166.