Sign-up for our Newsletter

Glossary

This glossary contains terminology specific to market research. We offer it as a tool to better understand market research, whether you are a user, buyer or just an interested party. This glossary, like our entire resource section, is a continual work in progress. We will update our definitions and add new ones as new terminology evolves and new methodologies are developed.

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

A

A Priori Segmentation- Segmenting markets into smaller units, typically based on standard practices, assumptions or hunches.
AAU (Attitude, Awareness & Usage) Study- A type of longitudinal study that monitors changes in consumer attitudes, awareness and usage levels. Typically used for a product category as a whole or a specific product or brand.
Accuracy- The degree of conformity of a sample statistic to the population.
Ad Concept Testing- Testing used to gauge reaction by the targeted audiences to different advertising approaches or preliminary ad concepts.
Ad Tracking Research- Measuring the impact (awareness, attitudes, sales, etc) of advertisements over a period of time.
Aided Recall- Providing possible response options to a respondent in order to encourage recollection.
ANOVA (Analysis of Variance)- An analysis method for determining the level of statistical significance of differences among the means of two or more populations.
Anthropomorphic- A research technique in which participants describe a product, service or brand in terms of human characteristics so that the participants' feelings about the test object can be determined.
Applied Research- Any research which is used to answer a specific question or solve a particular problem. In contrast, basic research is driven by curiosity and gathers knowledge for knowledge's sake.
Area Samples- Samples which include geographic areas as part of the sample design.
At-Home Testing- A product is provided for participants to use at home under the same circumstances that they would actually use the product. Feedback is often obtained afterwards by survey or interview.
Attempt- Every effort to reach a potential respondent, regardless if contact is made or a survey completed.
Attitudinal Scaling- A moderation technique in which participants are instructed to conceptualize the product or service on a two-dimensional scale, such as price and quality. The goal is to better understand the participants' feelings about the product or service.
Attribute- The quality or characteristic of the item being evaluated.
Awareness- Knowledge of the subject or object of question. Often refers to knowledge or familiarity with a product, brand, or advertisements.

B

Balanced Question- A question with both sides of the issue equally represented. A balanced question would have a balanced scale by definition.
Balanced Scale- A response scale which presents the same number of positive and negative response options.
Benchmark- A control source or standard which the object of study can be judged against. For example, you may compare the test results of one school to the results of all schools in the state.
Bernoulli Response Variables- Each element has one of two possibilities (e.g. yes/no or on/off).
Bias- A systematic tendency to misrepresent or distort the the population. Biases can be caused by several sources, including improper sampling, interviewing techniques, and wording of questions, to name a few.
Bimodal- A distribution in which the frequency curve has two peaks. A single peak is called a mode.
Blind Test- Assessment of a product, service, etc, without revealing the company or brand.
Brand Associations- Opinions and knowledge that consumers have regarding a brand.
Brand Equity- The level of awareness and consumer commitment generated by the qualities of a company's brand.

C

Callback- A follow up attempt to reach a possible respondent who could not be reached on a previous try.
CATI (Computer Aided Telephone Interviewing)- Survey administered by an interviewer via the telephone using a computer-based questionnaire. The computer-based survey allows for questions to be asked based on questions already answered or information already known about the respondent.
Causation- The inference that a change in one variable is responsible for an observed change in another variable.
Census- A count of the entire population.
Census Blocks- Census areas roughly equivalent to city blocks.
Census Tract- Small areas within a county with relatively homogeneous households with respect to household characteristics. Tracts, on average, have approximately 4,000 residents.
Chi-square- A test of statistical significance which tests one measure of how well your model of expected distribution fits the observed distribution.
Clarifying- A follow up technique for getting more complete and in-depth responses to questions (often open-ended questions) by asking respondents to explain their answers. Also known as probing.
Classification Questions- Questions used to categorize respondents based on demographics or other defining variables.
Closed Ended Question- Questions that provide the respondent a limited number of pre-listed answers from which to choose.
Cluster- A group that is relatively homogeneous based on key characteristics (often demographic, social, and economic characteristics).
Cluster Analysis- A statistical classification technique for grouping individuals of a population by making quantitative comparisons of multiple characteristics. The differences within any group are less than the differences with other groups. Commonly used for consumer segmentation and brand or product positioning.
Coding- The process of translating responses questions into another form to facilitate data analysis. Often done with open-ended questions to group similar responses.
Cohort- A group of individuals as defined by a common factor (most often age).
Completion- A fully completed survey instrument that is used in the analysis.
Completion Rate- The percent of qualified respondents who complete an interview or study.
Concept Board- Visually rendered products or services for the purpose of testing the concept with potential customers without having the physical product present.
Conceptual Mapping- An interview or focus group moderation technique in which participants are asked to place the products or services being tested on a grid to help visualize relationships.
Conclusions- The final outcomes, findings or results of the research project.
Conditional Probability- The probability of a given event when additional information about that event is known. For example, the probability of rolling a one on the toss of a die when we are given the additional information that the roll landed on an odd number.
Confidence Interval- The interval estimate around a result for which there is a high statistical probability that it contains the true population parameter.
Confidence Level- The probability that a confidence interval will include the population's true value.
Confidentiality- The restriction of information that is released. In research, this can include both keeping the client and/or research participant's information private.
Conjoint Analysis- A statistical technique used to quantify how people value different attributes, or different levels of attributes, that make up a specific product or service. By trading product attributes against each other, a product preference and the relative importance of attributes can be determined. Better for functional than fashionable brands.
Constant Sum Scales- Scales that ask the respondent to divide a given number of points, typically 100, among two or more attributes based on their importance to the individual
Consumer Drawings- A response technique that has respondents draw their responses in order to convey their feelings or perceptions of an object.
Consumer Expenditure- The amount spent on goods and services by consumers.
Consumer Expenditure Survey (CEX)- Data gathered in an ongoing survey for the Bureau of Labor Statistics on the buying habits of consumers. Consists of an quarterly interview survey and a diary survey.
Consumer Price Index (CPI)- Compares the current cost of purchasing a fixed basket of goods and services with the cost of the same set at a specific base year.
Contamination- The inclusion of individual(s) in a sample who do not represent the population.
Content Analysis- A technique used to study communication materials (often advertising copy) by carefully dividing it into meaningful units.
Continuous Variable- A quantitative variable that can assume an infinite number of values within the variable range. A noncontinuous variable is called discrete.
Convenience Sample- A sampling method that leaves the selection of respondents totally to the interviewers, with no quotas or qualifications imposed. Typically members of the population that are most accessible are those that are selected.
Cooperation Rate- The proportion of respondents who meet the study criteria that actually participate.
Copy Testing- Technique to measure the target audience's perception, understanding, opinions, etc of a proposed advertisement. Can include all forms of advertisements, such as print, audio, and commercials.
Correlation- Analysis of the degree to which changes in one variable are associated with changes in another.
CPI (Cost per Interview)- The overall cost for conducting one interview. Calculated by dividing the total number of interviews by the total cost.
Cross-Tab (X-Tab)- Analysis of the responses to one question relative to responses to one or more other questions (often demographics).

D

Data Collection- The compilation of information to be used for analysis. Most often used in conjunction with the execution of surveys.
Demographics- Description of the vital statistics or objective and quantifiable characteristics of an audience or population. Demographic designators include age, marital status, income, family size, occupation, and personal or household characteristics such as age, sex, income, or educational level.
Dependent Variable- A variable whose observed change is caused by the independent variable.
Diary- A style of survey in which respondents keep a log, or diary, of what they watch, listen to, or purchase, etc., over a set period of time.
Dichotomous Questions- Question format that ask a respondent to choose between two provided answers.
Digital Fingerprint- The compilation of computer settings (both hardware and software) that together can create an unique identifier for that particular computer device or machine.
Discrete Variable- A quantitative variable that can only assume a finite number of values, opposed to a continuous variable that can take on any value.
Discriminate Analysis- A multivariate statistical technique for analyzing which variable(s) are the best predictors of an independent variables.
Disguised Observation- The process of watching or monitoring research subjects (e.g. people, objects, or occurrences) that are unaware that they are being watched.
DMA (Designated Marketing Area)- A region in which the population receives approximately the same media offerings (e.g. television, radio, print, etc).
Door-to-Door Survey- Consumers or consumers are interviewed face to face at their homes.
Double-Barrel Question- A single question that poses two separate questions or concepts that should be asked individually.

E

Eligibility- The criteria used to determine whether a possible respondent qualifies to participate in the study.
Error- Any inaccuracy between study findings and the true state.
Ethnography- A qualitative research methodology in which subjects are studied in their own environments. The researcher attempts to be a part of their environment to understand the conditions around their actions.
Executive Summary- The portion of a research report that explains why the research was done, how it was done, key findings (and what those findings mean), and resulting recommendations. See also conclusion.
Exhibit- Anything shown to respondents during the execution of the research, such as an interview or focus group. Examples may include a print advertisement or slogans.
Expected Value- The sum pf the probability of each potential outcome (of an experiment) multiplied by the outcome value. It is the value "expected" of repeated random trials.
Experiment- The process of investigating certain research questions. Typically done by altering one variable (i.e. the independent variable) and observing the change in another variable (i.e. the dependent variable).
Experimental Design- The research design used to collect data through experiments.
Experimental Unit- The basic element, or unit, on which the experiment is conducted to observe change. Often referred to as the respondent, participant, or subject.
Exploratory Research- Research that is conducted for the sole purpose of clearly defining the problem. Exploratory research often completed through secondary research sources.

F

Face Validity- An intuitive test of whether a measurement appears to measure what it is suppose to measure.
Factor Analysis- Statistical method for data simplification through reducing the many variables used by the researcher to a smaller set of factors or composite variables by identifying interdependencies between the data.
Fielding Period (see response period)
Focus Group- A qualitative research technique in which a group of participants (commonly 8-10 and share some common trait(s)) are guided through a discussion of a particular topic(s) by a trained moderator.
Frequency- A measure of how often an event occurs; a count of the number of subjects falling in the different categories.
Fresh participants- Focus group participants who have never participated in a session previously (or at least not for several years), or have never participated in a session of similar purpose or subject.
Frugging- "Fundraising under the guise of research." An unethical practice in which an organization tries to use research as a cover for fundraising.
F-Test- A statistical test to determine the chance of a particular calculated value could have been due to chance.

G

Geocoding- The process whereby data are segmented by geographic identifiers (e.g. county, MSA, etc.), in order to map or analyze the data.
Geodemographic Segmentation System- A multivariate statistical classification technique for discovering whether the individuals of a population fall into different groups by making quantitative comparisons of multiple characteristics with the assumption that the differences within any group should be less than the differences between groups. Often used for consumer segmentation and brand positioning. Often known as cluster analysis.
GIS (Geographic Information System)- A system for interpreting spatially referenced data.
Group Dynamics- The interaction among individuals in a group. In focus groups, a moderator works to both enable group dynamics to promote helpful discussion and to minimize the potentially negative effects of group dynamics.

H

Hedonic Scale- A scale for measuring a respondent's general likeness of a product. Often used in sensory research, such as food tests.
History- External variables that change during the course of an experiment.
Homogeneous Group- A groups in which the individuals have characteristics that are the same.
Honorarium- The payment provided to focus group - or other type of research - participants. The amount varies significantly, based on the difficulty of recruiting the participants. Also called the co-op payment or incentive.
Household- All persons who occupy a housing unit. Persons may or mat not be related.

I

Incentive (see Honorarium)
Incidence- The percentage of the population in a specified category.
Independent Variable- A variable that is controlled or manipulated by the researcher to cause change in the dependent variable.
Intercept Survey- A survey method in which an interviewer stops possible respondents in a public setting to administer a survey.
Internet Survey- A survey conducted online, typically via a panel of respondents or a pre known list of respondents (e.g. employees of a company or customers in database).
In-Depth Interview- One-on-one interviews that probe and elicit answers to questions with greater detail. Often used to get at the "why" behind a respondent's answers.
Interval Scale- Ordinal scale with the additional property that the equal distances between measurements are equivalent, and therefore, can be meaningfully compared. One such example would be temperature.
Interviewer- The person administering a questionnaire.
Interviewer Bias- Error resulting from either the conscious or unconscious bias in the interviewer's interaction with the respondent.
IVR (Interactive Voice Response)- Technology used for surveying in which respondents complete the survey via the telephone by interacting with an automated system. The key pad is typically used to record responses, though voice recognition could also be possible.

L

Laboratory Experiment- Research conducted in a controlled setting.
Laddering- A probing technique, used in qualitative research such as interviews and focus groups, designed to arrive at the real reason(s) for participants' attitudes and behavior toward the topic. The interviewer seeks the reason behind each answer until he or she arrives at a basic human need such as ego or status.
Likert Scale- A response scale in which the respondent specifies a level of agreement or disagreement with the statement provided.
Listed Sample- A sample developed from a source that has aggregated individuals with a common trait. Examples include telephone books, membership directories, and voter lists. Typically used for harder to reach populations where pure random sampling would be inefficient. Not ideal for true public opinion research.
Listed Telephone Households- Households which are listed in published telephone directories (i.e. telephone book).
Long Census Form- The decennial census questionnaire that was answered by a sample of the population (approximately one in sex households). The form contained the same basic questions as the short form, plus additional population and housing questions. Replaced by the American Community Survey, which is an ongoing survey, rather than once every ten years.
Longitudinal Study- A study conducted multiple times to measure changes in the same item at different points in time.

M

Market- The total of all individuals, organizations, or other entities that represent potential buyers for a product or service.
Market Research- The planning, gathering, and analysis of data relevant to marketing related decisions.
Market Segmentation- The process of dividing a market into sub-groups, or segments, of consumers who exhibit similar qualities or preferences as they relate to product or brand in question.
Margin of Error- The measure of deviation between research findings and actual state of what is being measured.
Mean- The sum of all values from observations of a variable divided by the total number of observations. Also known as the average.
Mean Square Error- A method to measure the amount of error from the sample estimate to the true value.
Measurement- Process of assigning a unit of measure to things in accordance with defined rules to represent an attribute.
Measurement Error- Error that results from a variation between the actual value and the information actually obtained by the measurement process.
Media Testing (see Copy Testing)
Median- The middle observation that divides the distribution of observations in half.
Methodology- The specific procedures used to conduct research. Includes all steps of the research process, from design, recruiting participants, analysis, and so on.
MIS (Marketing Information Systems)- Systems used to collect, analyze, and report marketing data.
Mix-Mode Research- Research that involves multiple types of data collection to maximize response from the target population.
Mode- The most frequently occurring value.
Modeling- The formulation of mathematically-expressed variables to represent a process or operating environment. For example, a model could be formulated using a company's customer database and their financial data, combined with census data, to select new markets that have similar attributes of their current, profitable markets.
Moderator- The person hired by the client to lead the focus group.
Moderator's Guide- The moderator's outline used to lead the discussion in the focus group session. It is developed by the research team on the basis of the research questions to determine specific topics to be covered during the focus group session.
Monitor- Any type of quality control designed to ensure quality data. May include check such as auditing results or observing interviews.
MSA (Metropolitan Statistical Area)- Counties containing a highly urbanized area at its core and adjacent counties if they have strong social and economical ties to the urban area.
Multiple Choice Questions- Questions that ask a respondent to choose from a response list with more than two options.
Multiple Regression Analysis- Statistical technique that studies multiple (as opposed to one) independent variables simultaneously to identify a pattern or patterns. See also regression analysis.
Multivariate Analysis- Any statistical procedure that analyzes several variables at the same time.
Mutually Exclusive- Each observation needs to be mutually exclusive in order to be organized into categories.
Mystery Shoppers- People employed to pose as consumers, often with specific tasks, and shop at specific stores to compare prices, measure service, note displays, etc.

N

Neutral/Neutrality- The absence of holding a position or supporting a side in research.
Nominal Scale- A scale that provides categories of variables, rather than a quantitative descriptor. Examples include male/female or user/nonuser. Data cannot be ordered in any meaningful way (i.e. male > female).
Nonbalanced Scale- A response scale weighted toward one end or the other.
Nonprobability Sample- A sample that was not created through non-random sampling methods. Results using a nonprobability sample cannot be inferred to the larger population.
Non-Random- Outcomes which do not have an equal probability of occurring. In non-random sampling, bias is almost always introduced. Examples of non-random sampling techniques include convenience sampling and snowball sampling.
Nonresponse Bias- Error that results from systematic differences between respondents and non-respondents to a survey instrument.
Nonsampling Error- All the sources of errors that arise during a study besides sampling error.
Normal Distribution- A continuous distribution that is both bell shaped and symmetrical about the mean.

O

Objective- The information to be learned from a research study to meet the project's goal.
Observation- An individual recording of a measurement for one variable of a singly unit of the sample.
Observational Research- Qualitative research that monitors a subject's actions and behaviors in their natural setting.
Omnibus Survey- A survey in which costs and other resources of executing the survey are shared among multiple clients who each get to ask a few of their own specific questions.
One-Way Mirror- A mirror that allows observers to watch respondents, most often in focus groups, without the respondents being able to see the observers.
Open-Ended Question- A question which allows the respondent to answer in their own words, rather than to a predefined set of answers.
Optimization- Mathematical modeling that seeks to minimize or maximize a function without violating predefined constraints.
Ordinal Scale- A scale with categories of the variable which can be ordered, or ranked, but the differences between values is insignificant.
Over Recruit- Recruiting extra participants for a study (usually a focus group) to compensate for participants who will be no-shows.

P

Paired Comparison- A test in which a range of options are evaluated. Each option is compared to each of the other options to determine the preferred option overall.
Panel- A pre assembled group of possible respondents for research studies. Panels can be used to recruit for surveys, focus groups, and other research methods.
Participant- An individual participating in the research. Also known as a respondent, unit, or subject.
Pearson's Correlation Coefficient- The most common measure of association, or correlation, between two variables.
Penetration Analysis- Study of market share held by a product or company within a specific population.
Percentile- A value (on a scale of 100) that indicates the percent of observations that are equal to or below it.
Perceptual Mapping- A visual technique used to display consumers' perceptions and preferences regarding a product or brand. Can be useful to determine a brand or product positioning among other brands or products, or to determine consumers' ideal attribute combinations.
Photo Sort- A research technique that has respondents organize different types photos to illustrate their perceptions. An example may be a respondent who organizes photos to show different perceptions of vacation spots.
Poll- A collection of opinions on a subject.
Population- The entire collection of all objects or subjects that are of interest to the statistician.
Precision- The degree to which repeated measurements show similar or the same results.
Predictor Variable- The variable (that is manipulated) that explains or predicts the differences in dependent variables. See also independent variable.
Pre-Test- A trial run of the research methodology (commonly a survey) to check for problems with the research mode or execution and to gauge response rate.
Primary Data- New data gathered to help answer a research question. An example would be data collected through a survey.
Primary Research- Research conducted for the purpose of obtaining new data to answer a research need.
Probability Distribution- Describes all possible values and probabilities with a discrete random variable
Probability Sample- A sample in which every unit has an equal and known probability of being selected. Also known as a random sample.
Product Concept Testing- The testing of new product ideas in front of possible consumers to test the viability of the product before introduction to the market.
Product Positioning Research- Research used to determine how a target market perceives a brand relative to competitor brands.
Professional Respondent- A research respondent who participates in many research studies, such as responding frequently to surveys or participating in many focus groups. Often characterized by the participant's effort to be included in research by signing up on many survey panels or focus group recruitment lists.
Projectability- The ability to take research results and extrapolate them to a larger population. In order to do this, the sample must be representative of the total population.
Profile- An outline or summary of respondents, most often by demographic characteristics.
Psychographics- Attributes that relate to values, attitudes, and lifestyles of an individual, opposed to demographics that are comprised of quantifiable characteristics.
PUMS (Public Use Microdata Sample files)- US Census micordata files that show the full responses made on questionnaires of a statistically valid sample of all data collected.

Q

Qualitative Research- Research data that is not gathered for the purposes of quantification or quantitative analysis, rather it is gathered to gain knowledge on the "why" and "how." Qualitative methodologies include focus groups, one-on-one interviews, and open ended-questions.
Quantitative Research- Research carried out with the purpose of obtaining empirical observations of attitudes, behavior or performance through measurement. Designed to generate numerical data about a topic that can be modeled, relationships analyzed, and projections made. Survey research is prime example of a quantitative methodology.
Questionnaire- A compiled set of questions designed to generate the data needed to answer the research questions of the project.
Quota Sample- A sampling method that dictates that a specific number of respondents have a certain set of characteristics (often demographics). Once the quota for a group is filled, participants are no longer recruited for that particular quota group.

R

Random Error- Error that affects measurement in an unpredictable and inconsistent manner. Random error can be caused by fluctuations in measurement or the researcher's interpretation of reading the measurement.
Randomization- The random assignment of subjects or treatment conditions to ensure equal representation of subject characteristics in all groups.
Random Sampling- (see probability sample)
Range- An indication of statistical dispersion of the results, calculated by taking the highest value for a variable and subtracting the lowest value for that variable.
Rank-Order Scales- Scales in which the respondent compares one item with another or a group of items against each other and ranks them.
RDD (Random Digit Dialing)- A sampling procedure for telephone surveys that generates phone numbers at random. By not using a listed source of telephone numbers, such as the phone book, better coverage can be obtained because all numbers (from land lines), including unlisted, have a chance of being selected.
Recruitment- The process of securing participants for research, most often referring to focus groups or interviews.
Regression Analysis- A multivariate statistical technique that models numerical data consisting of dependent and independent variables in order to look at the relationships between those variables.
Reliability- The consistency of measures or the measurement instrument from one test to the next.
Representative Sample- A sample in which each unit in the population has an equal chance of being selected. See also probability sample.
Research Design- The framework of all research activities to be used in order to successfully answer the research objectives.
Research Ethics- Rules of conduct that guide research in an honest, professional, and just manner, that encourages confidence in respondents.
Research Objective/Statement- The purpose for which the research is being conducted. Typically the first step in the research process that will guide research decisions.
Residual Error- The variability that still cannot be explained, after all the main effects and interactions have been segregated.
Respondent- The research participant from whom data is collected. Also known as the subject or experimental unit.
Response Bias- Error that results from respondents answering a question falsely, either because the respondent is answering what they believe is the right way to answer or what they believe the questionnaire or interviewer wants to hear.
Response Period- The timeframe in which participants can respond to a survey.
Response Rate- The proportion of respondents who participated to all potential respondents in the sample.

S

Sample- A subset of the population selected for a research study.
Sampling Design- The method of selecting a specified portion, called a sample, from a population, from which information concerning the whole can be inferred.
Sampling Error- The error in the results of a study that results from observing a sample of the population, rather than the entire population.
Sampling Frame- A listing of all identified elements in the sample.
Sampling Interval- Sampling a given number of units selected equally over the full population of study. The interval is calculated by dividing the number of sampling unit by the desired sample size
Sampling Unit- The element or set of elements available for selection during the sampling process.
Sample Size- The number of people (or other unit) chosen from the population.
Scaled-Response Questions- A question format in which the available choices are designed to capture the intensity of the respondent's answer.
Screener- Questions used to determine qualified respondents.
Secondary Data- Data that has been previously gathered and/or analyzed. May include sources such as the Census and organizational data.
Secondary Research- Collecting and analyzing existing, available information from previously conducted research.
Segmentation- Dividing a population (often markets) into different segments, or divisions, based on similar characteristics. Characteristics may include demographics, geography, behavioral or other criteria.
Selection- The process of choosing individual units from a population.
Selection Bias- Systematic differences between the sample and the population because of bias in the selection process.
Semantic Differential- A rating scale designed to compare two items by having respondents rank it between dichotomous pairs of words or phrases used to describe the items.
SES (Socioeconomic Status)- A person's or family's status as derived from income, education, occupation and community standing.
Short Census Form- The U.S. Census Bureau's questionnaire that all Americans answer every 10 years.
Skewed- A distribution whose frequency curve is not symmetrical about its mean. If the left tail is longer than the right, it is called left or negative skew. If the right tail is longer, it is called right or positive skew.
Skip Pattern- The planned logic of having a respondent pass over certain questions based on the respondent's answer to previous question(s).
Snowball Sample- A sample in which the selection of additional respondents is based on referrals from earlier respondents.
Standard Deviation- A measure of dispersion of the sample values
Standard Error- The standard deviation of the error for a method of measurement.
Statistical Inference- Using statistics to draw conclusions about a populations based on a random sample.
Statistical Significance- The likelihood that a result could have occurred by chance.
Statistics- The mathematical science of collecting, organizing, analyzing, and/or drawing conclusions from data.
Stratified Random Sample- A statistical sampling procedure used when subpopulations vary considerably. The population is divided into relatively homogeneous groups, called strata, then each strata is randomly sampled.
Subject- The element that is being observed in the experiment or study. Also known as a participant, experimental unit, or respondent.
Sugging- "Selling under the guise of research." An unethical practice in which a company tries to use research as a cover for a sales pitch.
Survey Instrument (see questionnaire)
Survey Research- Research utilizing questionnaires that are either self-administered or conducted by an interviewer to quantify data regarding facts, opinions and attitudes.
Systematic Sample- A statistical sampling procedure that selects every Nth unit, where N is the skip interval, (starting from a random starting point), of a sampling frame until the desired sample size is reached.
Systemic Error- Error in measurement that can lead to results being consistently inaccurate. Can result from poor research design or execution.

T

Target Population- The population of interest in a study.
Temporary Population- Non permanent residents of an area. May include groups such as tourists and seasonal residents.
Test Market- A trial market used to assess a new product or service offering.
Test Statistic- A calculated value from the data which is used to test the hypothesis.
Third-Person Techniques- A method of learning respondents' true opinions and feelings by asking them to project their answers onto another person (i.e. the third person). This method is used to minimize social pressures to give "correct" responses.
Tracking- Studies repeated over time to observe any changes, typically in behavior, awareness or attitudes.
True Experimental Design- Very accurate form of research using a test group and a control group, where assignment of test units to both groups is randomized.
T-Test- Hypothesis test about a single mean when the sample is assumed to be normally distributed and the sample size is small.
Type I Error (A Error)- Rejecting the null hypothesis when it is true. Also known as a false positive.
Type II Error (B Error)- Failing to reject the null hypothesis when it is false. Also known as a false negative.

U

Unaided Recall- Not providing possible response options or other cues to a respondent when asking a question. No response options are provided or read.
Unbiased Samples- Samples in which every unit in the population has an equal chance of being selected and any sampling error comes from random sources.
Unit of Analysis- The primary entity that is being studied, most often individual people, also known as the respondent, participant, or subject.
Universe- (see population)
Usability Testing- Assessing a product by having actual users try out the product. Products, instructions, services, software and web sites can all be tested in such a manner.

V

Validation- The process of verifying that data collected is in fact accurate. May involve re-contacting respondents to confirm responses.
Validity- The degree to whether what the study aimed to measure was in fact measured.
Variability- The differences or spread in the measurement of variables.
Variable- Any characteristic, element, or attribute that can be measured for a unit of the population.
Variance- A statistical measure of how similar or spread out a population is in the characteristic being measured.
Verbatim- The exact reproduction of a participant's response. Typically from focus groups, interviews, or open-ended survey questions.

W

Weighting- Assigning a numerical coefficient to a unit to emphasize some aspect relative to other units.
Word Association Tests- Research in which the interviewer or moderator says a word and the participant replies with the first thing that comes to their mind.

Z

Z-Test- A statistical test used to determine if the difference between the sample mean and population mean is significantly different, rather than having been caused by chance.

[Disclaimer] The definitions and terminology on this page are meant for general reference only. These definitions are compiled by Corona as they relate specifically to Corona's line of work and may or may not be relevant or accurate for your specific use.