Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

how to calculate group differences with alpha draws and covariates


The values after the intercept level show me the change in relation to the intercept level, right?

Lets assume I have a covariate with 3 levels and I want to find out if there are differences between the groups. Can I just use the values of change and e.g. count how often level 2 shows values above/below zero and how often level 3 shows values above/below zero and compare them to each other? Or do I need to calculate the final values first (as described in the original post)?

Also I am wondering if this works, as e.g both level 2 and level 3 could potentially be greater/lower than the reference level, but still are different to each other. Ho do I capture that? May I compare level 2 and level 3 directly (after the utilities are calculates)?

So, if attribute level A in Covariate level 3 shows in 9500 rows, that the values are higher than attribute level A in Covariate level 2, means that level 3 is significantly different from level 2 and we are 95% confident that the level 3 has higher values than level 2. Is that right?

Thanks a lot!
related to an answer for: CBC/Hb - Covariates in the alpha file
asked Apr 16, 2019 by bs77 Bronze (1,035 points)
edited Apr 16, 2019 by bs77

Your solution to the original question

Please only use this to answer the original question. Otherwise please use comments.
Your name to display (optional):
Privacy: Your email address will only be used for sending these notifications.
Anti-spam verification:

To avoid this verification in future, please log in or register.