Analysis of Selected Cases - Intensive study of selected examples of the phenomenon of interest.
Return to top of page
|
Analysis of Variance - Statistical test employed with interval data to determine if two, or more than two samples came from population with equal means.
Return to top of page
|
Area Sampling - A form of cluster sampling in which areas (e.g., a city block) serve as the primary sampling units.
Return to top of page
|
Banner - A series of cross-tabulations between a criterion, or dependent, variable and several (sometimes many) explanatory variables in a single table.
Return to top of page
|
Brain Wave Research - Research technique that assesses the stimuli that subjects find arousing or interesting by using electrodes fitted to the subject's head that monitor the electrical impulses emitted by the brain.
Return to top of page
|
Branching Questions - A technique used to direct respondents to different places m a questionnaire based on their response to be the question at hand.
Return to top of page
|
Brand-Switching Matrix - Two-way table that indicates which brands a sample of people purchased in one period and which brands they purchased in a subsequent period, thus highlighting the switches occurring among and between brands as well as the number of persons that purchased the same brand in both periods.
Return to top of page
|
Causal Research - Research design in which the major emphasis is on determining a cause-and-effect relationship.
Return to top of page
|
Central-Limit Theorem - Theorem that holds that if simple random samples of size n are drawn from a parent population with mean µ and variance σ², then when n is large, the sample mean x will be approximately normal distributed with mean equal to µ and variance equal to σ² / n. The approximation will become more and more accurate as n becomes larger.
Return to top of page
|
Chi-Square (pronounced "ki-square") Goodness-of-Fit Test - Statistical test to determine whether some observed pattern of frequencies corresponds to an expected pattern.
Return to top of page
|
Cluster Analysis - Body of techniques concerned with developing natural groupings of objects based on the relationships of the p variables describing the objects.
Return to top of page
|
Cluster Sample - A probability sample distinguished by a two-step procedure in which (1) the parent population is divided into mutually exclusive and exhaustive subsets, and (2) a random sample of subsets is selected.
Return to top of page
|
Code Book / Coding Sheet - A document that describes each variable, gives it a code name, and identifies its location in the record.
Return to top of page
|
Coding - A technical procedure by which data are categorized; it involves specifying the alternative categories or classes into which the responses are to be placed and assigning code numbers to the classes.
Return to top of page
|
Coefficient of Multiple Correlation - In multiple regression analysis, the square root of the coefficient of multiple determination.
Return to top of page
|
Cohort - The aggregate of individuals who experience the same event within the same time interval.
Return to top of page
|
Comparative Rating Scale - Scale requiring subjects to make their ratings as a series of relative 'judgments or comparisons rather than as independent assessments.
Return to top of page
|
Computer-Assisted Telephone Interviewing (CA TI) - The conducting of surveys using computers to manage the sequence of questions in which the answers are recorded electronically through the use of a keyboard.
Return to top of page
|
Conjoint Analysis - Technique in which respondents' utilities or valuations of attributes are inferred from the preferences they express for various combinations of these attributes.
Return to top of page
|
Constant Sum Method - A type of comparative rating scale in which an individual is instructed to divide some given sum among two or more attributes on the basis of their importance to him or her.
Return to top of page
|
Convenience Sample - Nonprobability sample sometimes called an accidental sample because those included in the sample enter by accident, in that they just happen to be where the study is being conducted when it is being conducted.
Return to top of page
|
Correlation Analysis - Statistical technique used to measure the closeness of the linear relationship between two or more intervally scaled variables.
Return to top of page
|
Cross Tabulation - Count of the number of cases that fall into each of several categories when the categories are based on two or more variables considered simultaneously.
Return to top of page
|
Depth Interview - Unstructured personal interview in which the interviewer attempts to get subjects to talk freely and to express their true feelings.
Return to top of page
|
Descriptive Research - Research design in which the major emphasis is on determining the frequency with which something occurs or the extent to which two variables covary.
Return to top of page
|
Dichotomous Question - Fixed-alternative question in which respondents are asked to indicate which of two alternative responses most closely corresponds to their position on a subject.
Return to top of page
|
Discriminant Analysis - Statistical technique employed to model the relationship between a dichotomous or multichotomous criterion variable and a set of p predictor variables.
Return to top of page
|
Disproportionate Stratified Sampling - Stratified sample in which the individual strata or subsets are sampled in size and their variability; strata exhibiting more variability are sampled more than proportionately to their relative size, while those that are very homogeneous are sampled less than proportionately.
Return to top of page
|
Double-Barreled Question - A question that calls for two responses and thereby creates confusion for the respondent.
Return to top of page
|
Electronic Test Market - Market or geographic area in which a firm tracks purchases made by specific households that are part of its panel, using identification cards held by panel members and the electronic recording of the products they purchase using scanners.
Return to top of page
|
Experience Survey - Interviews with people knowledgeable about the general subject being investigated.
Return to top of page
|
Exploratory Research - Research design in which the major emphasis is on gaining ideas and insights; it is particularly helpful in breaking broad, vague problem statements into smaller, more precise subproblem statements.
Return to top of page
|
Eye Camera - Camera used to study eye movements while the subject reads advertising copy.
Return to top of page
|
Factor Analysis - Body of techniques concerned with the study of interrelationships among a set of variables, none of which is given the special status of a criterion variable.
Return to top of page
|
Fixed-Alternative Questions - Questions in which the responses are limited to stated alternatives.
Return to top of page
|
Focus Group - Personal interview conducted among a small number of individuals simultaneously; the interview relies more on group discussions than on directed questions to generate data.
Return to top of page
|
Funnel Approach - An approach to question sequencing that gets its name from its shape, starting with broad questions and progressively narrowing the scope.
Return to top of page
|
Halo Effect - Problem that arises in data collection when there is carry-over from one judgment to another.
Return to top of page
|
Histogram - Form of bar chart on which values of the variable are placed along the X-axis, or abscissa, and the absolute frequency or relative frequency of occurrence of the values is indicated along the Y-axis, or ordinate.
Return to top of page
|
Interval Scale - Measurement in which the assigned numbers legitimately allow the comparison of the size of the differences among an between members.
Return to top of page
|
Judgment Sample - Nonprobability sample that is often called a purposive sample; the sample elements are handpicked because they are expected to serve the research purpose.
Return to top of page
|
Latin-Square Design - Experimental design in which (1) the number of categories for each extraneous variable we wish to control is equal to the number of treatments, and (2) each treatment is randomly assigned to categories according to a specific pattern. The Latin-Square design is appropriate when there are two extraneous factors to be explicitly controlled.
Return to top of page
|
Leading Question - A question framed to give the respondent a clue about how he or she should answer.
Return to top of page
|
Multichotomous Question - Fixed-alternative question in which respondents are asked to choose the alternative that most closely corresponds to their position on the subject.
Return to top of page
|
Multicollinearity - Condition said to be present in a multiple-regression analysis when the predictor variables are not independent as required but are correlated among themselves.
Return to top of page
|
Multidimensional Scaling - Approach to measurement in which people's perceptions of the similarity of objects and their preferences among the objects are measured, and these relationships are plotted in a multidimensional space.
Return to top of page
|
Nominal Scale - Measurement in which numbers are simply assigned to objects or classes of objects solely for the purpose of identification.
Return to top of page
|
Nonparametric Tests - Class of statistical tests, also known as distribution-free tests, that are applicable when the data reflect nominal or ordinal measurement or when but the assumptions required for the appropriate parametric test are not satisfied.
Return to top of page
|
Nonprobability Sample - Sample that relies on personal judgment somewhere in the element selection process and therefore prohibits that any population element will be included in the sample.
Return to top of page
|
Nonresponse Error - Nonsampling error that represents a failure to obtain information from some elements of the population that were selected and designated for the sample.
Return to top of page
|
Nonsampling Errors - Errors that arise in research that are not due to sampling; nonsampling errors can occur because of errors in conception, logic, misinterpretation of replies, statistics, and arithmetic; errors in tabulating or coding: or errors in reporting the results.
Return to top of page
|
Omnibus Panel - Panel in which the information collected from the participating panel members varies from study to study.
Return to top of page
|
Open-Ended Question - Question characterized by the condition that respondents are free to reply in their own words rather than being limited to choosing from among a set of alternatives
Return to top of page
|
Ordinal Scale - Measurement in which numbers are assigned to data on the basis of some order (for example, more than, greater than) of the objects.
Return to top of page
|
Outlier - Observation so different in magnitude from the rest of the observations that the analyst chooses to treat it as a special case.
Return to top of page
|
Panel (omnibus) - Fixed sample of respondents who are measured repeatedly over time but on variables that change from measurement to measurement.
Return to top of page
|
Panel (true) - Fixed sample of respondents who are measured repeatedly over time with respect to the same variables.
Return to top of page
|
People Meter - A device used to measure when a TV is on, to what channel it is tuned, and who in the household is watching it. Each member in a household is assigned a viewing number, which the individual is supposed to enter into the people meter whenever the set is turned on, the channel is switched, or the person enters or leaves the room.
Return to top of page
|
Personal Interview - Direct, face-to-face conversation between a representative of the research organization (the interviewer) and a respondent, or interviewee.
Return to top of page
|
Physiological Reaction Technique - Method of assessing attitudes in which the researcher, by electrical or mechanical means, monitors the subject's response to controlled introduction of some stimuli.
Return to top of page
|
Plus-One Sampling - Technique used in studies employing telephone interviews in which a single randomly determined digit is added to numbers selected from the telephone directory.
Return to top of page
|
Primary Data - Information collected specifically for the purpose of the investigation at hand.
Return to top of page
|
Probability Sample - Sample in which each population element has a known, nonzero chance of being included in the sample.
Return to top of page
|
Protective Technique - A method of questioning respondents using vague stimulus that respondents are asked to describe, expand on, or build a structure around; the basic assumption is that an individual's organization of the relatively unstructured stimulus is indicative of the person's basic perceptions of the phenomenon and reactions to it.
Return to top of page
|
Proportionate Stratified Sampling - Stratified sample in which the number of observations in the total sample is allocated among the strata in proportion to the relative numbers of elements in each stratum in the population.
Return to top of page
|
Psychographic Analysis - Technique that investigates how people live, what interests them and what they like; it is also called life-style or AIO analysis, because it relies on a number of statements about a person's Activities, Interests, and Opinions.
Return to top of page
|
Q-Sort Technique - General methodology for gathering data and processing the collected information. The subjects are assigned the task of sorting various statements by placing a specific number of statements in each sorting category; the emphases are on determining the relative ranking of stimuli by individuals and deriving clusters of individuals who display similar preference orderings of stimuli.
Return to top of page
|
Quota Sample - Nonprobability sample chosen in such a way that the proportion of sample elements possessing a certain characteristic is approximately the same as the proportion of the elements with the characteristic in the population; each field worker is assigned a quota that specifies the characteristics of the people he or she is to contact.
Return to top of page
|
Random-Digit Dialing - Technique used in studies employing telephone interviews in which the number to be called are randomly generated.
Return to top of page
|
Ratio Scale - Measurement that has a natural or absolute zero and that therefore allows the comparison of absolute magnitudes of the numbers.
Return to top of page
|
Recall Loss - A type of error caused by a respondent forgetting that an event happened at all.
Return to top of page
|
Regression Analysis - Statistical technique used to derive an equation that relates a single criterion variable to one or more predictor variables.
Return to top of page
|
Reliability - Similarity of results provided by independent but comparable measures of the same object, trait, or construct.
Return to top of page
|
Response Rate (R) - Measure used to evaluate and compare interviewers in terms of their ability to induce contacted respondents to participate in the study; R = number of interviews/ number of contacts.
Return to top of page
|
Sample - Selection of a subset of elements from a larger group of objects.
Return to top of page
|
Sample Survey - Cross-sectional study in which the sample is selected to be representative of the target population and in which the emphasis is on the generation of summary statistics such as averages and percentages. Also called a field survey.
Return to top of page
|
Sampling Distribution - Distribution of values of some statistic calculated for each possible distinguishable sample that could be drawn from a parent population under a specific sampling plan.
Return to top of page
|
Secondary Data - Statistics not gathered for the immediate study at hand but for some other purpose.
Return to top of page
|
Semantic Differential - Self-report technique for attitude measurement in which subjects are asked to check with cell between a set of bipolar adjectives or phrases best describes their feelings toward the object.
Return to top of page
|
Sentence Completion - Questionnaire containing a number of sentences that subjects are directed to complete with the first words that come to mind.
Return to top of page
|
Sequence Bias - Distortion in the answers to some questions on a questionnaire because the replies are not independently arrived at but are conditioned by responses to other questions; the problem is particularly acute in mail questionnaires because the respondent can see the whole questionnaire.
Return to top of page
|
Sequential Sample - Sample formed on the basis of a series of successive decisions. If the evidence is not conclusive after a small sample is taken, more observations are taken; if still inconclusive after these additional observations, still more observations are taken. At each stage, a decision is made about whether the information should be collected or whether the evidence is sufficient to draw a conclusion.
Return to top of page
|
Simple Random Sample - Probability sample in which each population element has a known and equal chance of being included in the sample and in which every combination of n population elements is a sample possibility and is just as likely to occur as any other combination of n units.
Return to top of page
|
Simulated Test Marketing - Test marketing done by firms in shopping malls or consumers' homes as a prelude to a full-scale marketing test for the product.
Return to top of page
|
Snake Diagram - Diagram (so called because of its shape) that connects with straight lines the average responses to a series of semantic differential statements, thereby depicting the profile of the object or objects being evaluated.
Return to top of page
|
Snowball Sample - Judgment sample that relies on the researcher's ability to locate an initial set of respondents with the desired characteristics; these individuals are then used as informants to identify still others with the desired characteristics.
Return to top of page
|
Spurious Correlation - Condition that arises when there is no relationship between two variables but the analyst concludes that a relationship exists.
Return to top of page
|
Storytelling - Questionnaire method of data collection relying on a picture stimulus such as a cartoon, photograph, or drawing, about which the subject is asked to tell a story.
Return to top of page
|
Stratified Sample - Probability sample that is distinguished by the two-step procedure in which (I) the parent population is divided into mutually exclusive and exhaust subsets, and (2) a simple random sample of elements are chosen independently from each group or subset.
Return to top of page
|
Syndicated Search - Information collected on a regular basis that is then sold to interested clients (for example Nielsen Retail Index).
Return to top of page
|
Voice Pitch Analysis - Type of analysis that examines changes in the relative frequency of the human voice that accompany emotional arousal.
Return to top of page
|
Word Association - Questionnaire containing a list of words to which respondents are instructed to reply with the first word that comes to mind.
Return to top of page
|