This tool is for use on machines where CBC/HB v5.5.4 or later is installed. It interacts with CBC/HB’s command interpreter to automatically run multiple instances of CBC/HB to search for optimal priors and first-order interaction effects that can improve hit rate. Hit rate is computed by jack-knife resampling of a subset of existing choice tasks for holdout validation. This tool should be unzipped into a convenient folder on your hard drive (there is no installation program). Brief documentation is provided in the included PDF file. We also recommend you read the white paper entitled, What Are the Optimal HB Priors Settings for CBC and MaxDiff Studies? (2016).
This program is for taking allocation-based CBC data (where the sum of "chips" is not forced to be constant) and reformatting a new .CHS for developing a "volumetric CBC" model.
The data could represent how many boxes of each type of cereal the respondent would "purchase" within a CBC task. The program scans the respondent's choice tasks to see the maximum volume ever "purchased" in any one task. A new "None" concept is formatted for each task with sufficient chips allocated to the None to "make up the slack" between the maximum volume ever allocated and the volume allocated for the current task. A .CSV file is also written that contains the caseid followed by the maximum volume allocated for each respondent (to use as a weight in weighted simulations).
This program reports the distribution of Summed Prices for products in the ACBC design. It is computed based on respondents in the specified dataset (both Completes and Incompletes). It includes only the near-neighbor product design (not the BYO-selected products). Studying the distribution of prices is helpful when selecting breakpoints for fitting piecewise price functions under HB analysis.
NOTE: This utility is for SSI Web v7 and earlier. SSI Web v8 and later has this functionality built in.
This utility is designed to automate testing surveys created in SSI Web 7. It can simulate respondents both randomly and programatically answering the survey questions. It can simulate multiple respondents taking the survey simultaneously, and allows survey programmers more sophisticated testing of surveys.
This utility automatically takes SSI Web surveys. It generates random data enabling you to test the survey. Recommended for SSI Web Version 6 and earlier.
Calibrates CBC utilities for use in purchase likelihood simulations. Supports up to 150 parameters. Officially incorporated into CBC/HB v5.
We have a 21 megabyte library of images that SSI Web users may use within their SSI Web surveys. The images currently include five different models, and are saved in different formats:
These images are owned by Sawtooth Software Inc. and we claim copyright. Your SSI Web license permits you to use these only within SSI Web surveys.
You should only download this library if you are licensed for SSI Web and agree to the above terms.
Download the legacy styles used in older versions of Lighthouse Studio.