CBC Latent Class

Overview

The latent class standalone program is a tool for discovering segments of respondents who tend to have similar preferences in CBC (choice-based conjoint) data.  It uses the same approach as the Latent Class procedure that is integrated within Lighthouse Studio, but lets you analyze data from a .CSV file that weren’t necessarily collected within Sawtooth Software’s tools.  Latent class MNL classifies respondents into different segments and estimates the utilities for each segment. Latent class is included with your subscription to  the CBC component within for Lighthouse Studio. 

CBC/Latent class

• Up to 1000 attributes or parameters per individual
• Up to 1000 alternatives per choice task
• Maximum of 1000 levels for any one attribute
• Up to 1000 tasks per respondent
• Effects-coding, linear coding, or user-specified coding
• Can impose utility (monotonicity) constraints
• Can include first-order interactions
• Supports chip-allocation, MaxDiff, dual-response None
• Latent Class is designed to run on Microsoft Windows XP or later.

Methodology

If there are distinct segments of respondents with different preferences, a model that recognizes them produces more accurate results than an aggregate (single group) solution. 

The Latent Class Standalone System uses choice data for the simultaneous development of segments and estimation of utilities. For example, one segment might be composed of price-sensitive shoppers, and another might be composed of those who usually select premium brands. Each respondent has a probability of belonging to each segment, but can be classified into the most likely segment for subsequent tabulation.  Latent Class is often used to find needs-based segments for target marketing purposes. 

The Latent Class Standalone permits weighting of respondents, and you can specify attributes for which utilities should be monotonic, such as for levels of price or quality. Choosing the number of segments is facilitated by specifying a range to investigate, such as from 2 through 30 segments. Statistics are provided for assessing goodness of fit for each solution.