Can I use ACA data with CBC/HB to generate Utilities

Question

I have generated utilities for my ACA study from utility option of Ordinary least squares.
Can I use this data in SMRT to get counts? If yes how to do that ?

Answer 1

CBC/HB cannot be used to analyze ACA data, as ACA is ratings-type data vs the choice-type data required by CBC/HB.  We do offer an ACA/HB module which can effectively analyze the data.  You may also choose to use the OLS routine in SMRT, which still does a good job of estimating at the individual level.

I am not aware of a way to do counting analysis using OLS utilities.

Comments

Well, with some advanced data processing work, a person could indeed analyze ACA results using CBC/HB.  But, it would be so much additional work, that the user would do better probably just to buy the ACA/HB software.

Here's how I would do it in CBC/HB.  The self-explicated information (ratings within attribute) could be coded using an approach described by Lattery in a Turbo CBC session, where additional choice tasks are developed for each attribute, indicating which levels are preferred over other levels, in paired comparisons.  

Then, the ACA Pairs questions could be created to look like CBC pairs, but using chip-allocation between left and right concepts.  The sum of chips for any 9-point ACA Pairs question is 10.  Choosing a "9" means allocating 9 points to the right, and 1 point to the left concept.  Choosing a "5" in the ACA Pairs means allocating 5 points to both concepts.

So, with a lot of extra data processing work to create a new .CHO file (or .csv file) suitable for CBC/HB software, one could analyze ACA results in CBC/HB.  But, it's so much additional work, that one should just buy the reasonably affordable ACA/HB software.