Marketing Research Glossary - L

Labor Force: All U.S. civilians 16 or older who are working or actively looking for work (in the past four weeks). Institutional populations (e.g., prisoners) and military personnel are not counted.

Labor Force Participation Rates: The percentage of the population working or looking for work.

Laboratory Experiments: Scientific experiments conducted in a carefully controlled environment (i.e., laboratory) where all (or most) variables can be controlled.

Laboratory Test Market (LTM): A method of estimating first-year sales volume for new products, without going to a test market. Concept tests and product tests are used to estimate trial and repeat purchase rates, and the results are fed into a mathematical model to forecast year-one sales. A Laboratory Test Market (or LTM) is the same thing as an STM or Purchase Lab. Most LTM forecasts are related to CPG brands.

Laddering: A qualitative probing technique. The interviewer or moderator pursues each answer, seeking the reason for the answer (i.e., the reason for the reason) until a basic human need or motivation (such as ego or status, etc.) is uncovered . Laddering may be thought of as peeling back the layers of an onion until the inner core is reached.

Latent Class: A class of respondents determined by similarity with regard to some latent variable or combination of latent variables. Latent variables are unobservable variables that underlie groups of explicit or observable variables. Learn More

Latent Class Cluster Analysis: A type of statistical modeling that assigns units or individuals to unobserved (latent) clusters or segments based on their probability of cluster membership. Learn More

Latent Variable Analysis: Analysis of a data set on the basis of one or more latent (hidden, inferred, unobservable) variables, such that the values of the latent variable(s) may be inferred from observed variables. A common form of latent variable analysis is factor analysis, in which the factors represent hypothetical concepts that are represented by the aggregate of the variables that load on the factors.

Lead Selection: In database marketing, lead selection is the process of selecting a subset of records from the database for solicitation. Leads are normally selected on the basis of criteria, which may be dictated by experience or embodied in a segmentation or formal mathematical model.

Layout: Also referred to as Data Layout. A source document that lists the data storage locations for all answers to a survey.

Leading Question: A question that suggests or implies a particular answer by the way the question is sequenced, structured, or worded.

Leg: A segment of a project, or portion of a project. It can be a sample quota cell, or it can be a phase of a project in which a series of respondents will participate.

Lelly Triads: Also known as Repertory Grids. Used especially by advertising agencies to elicit consumer language for the products in question. Products (or services) are described or pictured on cards that are dealt three at a time. The respondent is invited to pick the "odd one out" and explain why it is different from the other two cards. The consumer's language and key discriminators are noted.

Length Of Interview: The number of minutes it takes for a respondent to complete a survey or interview.

Level Of Significance: Likelihood that a given statistical result is due solely to chance variation.

Lifestyle Research: Research that attempts to analyze target consumers' attitudes, hobbies, activities, perceptions, and opinions. Psychographic Research is a similar term. Learn More

Lifestyle Segmentation: A segmentation technique where respondents are segmentated based upon multivariate analyses of consumer attitudes, values, behaviors, emotions, perceptions, beliefs, and interests. Also known as Psychographic Segmentation. Learn More

Lifestyle Selectivity: Segmenting a population based on hobbies, interests, product usage or ownership, magazine readership, etc.

Likert Scale: A type of agree-disagree scale used to measure a respondent's agreement-disagreement with a series of statements about a product or service.

Linear Programming Models: A mathematical model aimed at identifying the maximum or minimum of linear functions in many variables subject to constraints. Linear programming models provide both the values of the decision variables for the optimal solution and the shadow prices for the constraints (i.e., the change in value of the objective function per unit change in the value of the constraint).

Linear Regression Analysis: A type of regression analysis using variables that are thought to have a linear relationship. Learn More

Listed Sample: A sample that contains only telephone numbers listed in a telephone directory, or any list of published phone numbers for a target sample.

Listed Telephone Households: Households that are listed in published telephone directories.

Location Analysis: Also known as Real Estate Site Selection. Refers to a study that evaluates the business opportunity offered by a specific real estate site. Learn More

Log Report (Data Entry): The function that is built into the MPA data entry package that can be printed out to ensure all questionnaires have been entered and checked (double entered). This report also tells us the number of keystrokes, the keystroke-per-hour average of each entry person, and the number of errors that were caught and changed by the person who verified the data.

Logician®: One of the world's most advanced online survey programming and reporting system. Three Decision Analyst computer programmers work continuously to improve and upgrade the Logician® system. Logician® is a registered trademark of Decision Analyst. Learn More

Logician® Simulated Shopping: Decision Analyst's Logician® Simulated Shopping with 3D animation creates a virtual in-store environment that emulates the retail shopping experience in an online survey. This online simulated shopping experience provides a setting where all possible variables are controlled (price, packaging design, package size, shelf location, and competitive set offered). Different product offerings, prices, etc. can be tested, and multiple shopping scenarios can be explored, based on a choice-modeling experimental design. Learn More

Logistic Distribution: The logistic distribution closely describes the probability distribution of binary responses (e.g., yes/no) and is commonly used in logistic regression or logit models.

Logistic Regression Analysis: A version of regression used to analyze the relationship between predictor variables and a dichotomous (binary) outcome variable. This type of regression fits a logistic function to the data.

Logit Model: A version of regression analysis using a specific S-shaped curve (the logistic curve) instead of a straight line. Used to model probability of an outcome when outcome variables are binary; e.g., yes/no rather than continuous numbers.

Log-Normal Distribution: The log-normal distribution describes variables with normally distributed logarithms, such as personal income or age at first marriage.

Long Census Form: During each decennial U.S. census, some households receive a longer questionnaire (the long census form). Data from the long census form are available only for large geographic areas.

Longitudinal Study: The same respondents are interviewed repeatedly over time, or matched samples of consumers are surveyed over time. The purpose of longitudinal studies is to monitor changes in awareness, trial, usage, attitudes, and other variables over time. Tracking studies are a type of longitudinal study.

Loyalty Research: Survey research conducted to measure the loyalty (typically repeat purchases) with a product or service and related variables. Typically, these are long-term tracking studies, so that changes can be monitored over time. Related to Customer Satisfaction Research. Learn More

LTM: See Laboratory Test Market.

Contact Decision Analyst

If you would like more information on Marketing Research, please contact Jerry W. Thomas by emailing or calling 1-817-640-6166.