Sawtooth Software: The Survey Software of Choice

After more than a year of effort, we have finally finished and shipped SSI Web version 8.3! Most of the improvements for v8.3 focus on ACBC (Adaptive CBC). We’re especially excited that ACBC can now handle alternative-specific designs. This means that certain attributes can only apply to (say) certain brands. This gives you much more flexibility in ACBC, similar to the flexibility offered through the advanced design module for CBC.

With v8.3, you can now customize the interview flow of the sections.  The standard approach is to ask three sections: BYO + Screener Tasks + Choice Tournament Tasks.  Now, you can show any combination of these sections (for example, if you just want to show Choice Tournament Tasks).

Here are the highlights regarding new ACBC capabilities:

  • Alternative-specific designs
  • Improved experimental design algorithm (typically 10-30% more efficient than before), including the option to avoid dominated concepts
  • Ability to customize section flow (drop BYO and/or drop Screeners)
  • New Test Design routine for generating robotic respondents and assessing the quality of your design (including the impact of attribute prohibitions)
  • Interaction search tool (works for CBC too!)
  • Ability to Customize the "At Most" and "At Least" Text by Attribute
  • Option to Take BYO Concept Forward to Choice Tournament
  • Ability to Exclude Certain Attributes from "Must Have/Unacceptable" Questions
  • Covariates in ACBC/HB Utility Estimation
  • More Flexible Utility Constraints

We have also thrown in a few goodies for standard CiW, CBC and MaxDiff:

  • General SSI Web Improvements: Able to export data for massive surveys (well past 60K fields) much better than before.
  • CBC: We’ve added an automated interaction search tool that examines all potential interaction effects and prioritizes which ones are most significant. This routine relies on the 2LL test and aggregate logit. We’ve also added an option to the Latent Class routine for constraining the solutions to those where the groups have equal scale. (See http://www.sawtoothsoftware.com/1378 for more info)
  • MaxDiff: We’ve added the same option as described above to the Latent Class routine for constraining solutions to those where the groups have equal scale.

For more details about the v8.3 release, please visit: http://www.sawtoothsoftware.com/657#version-history

For current v8 and annual subscription users, v8.3 is a free upgrade. If you are using v7 or prior, please write This email address is being protected from spambots. You need JavaScript enabled to view it. and ask for a quote to upgrade.