cbc_overlap help description possibly incorrect

[TCEagleInSV]

I tested a design I built using the SAS macros such that no level of any attribute appeared more than once in a choice task (choice question). When I examined the cbc_overlap 100% of every attribute showed up under the labeled 1 of the output from the function. The help says that the count under the 1 is the count of tasks where every level of an attribute is the same across all alternatives. I think you really meant a 3 9assuming the # of alts was 3) is the count of tasks where the attribute level is the same across all 3 alternatives and a 1 is the count of times there was NO OVERLAP.

I would be glad to share my code and the design for you to double check my work. Please let me know. My email is teagle@tceagle.com

Tom Eagle

[jhelvy]

Thanks for brining this up Tom. Could you send over the design? You should be able to upload it in here and I can take a look.

[TCEagleInSV]

Here is my R code where I generate a random design and the design file where I used Warren Kuhfeld's SAS macros to produce a design

TEcbcTools.zip

Best

Tom

[jhelvy]

Okay great, thanks for sending this. It's finals week coming up, so this will probably have to wait a week or so until I can find time to look into it.

[TCEagleInSV]

No worries! Whenever you can get to it. I’m pretty sure I’m right. There is code at the end of the r file where I count the overlaps in the design I produced using cbcTools at the beginning of the r file. It matches the cbc_overlap function. Cool code by the way. Is there any way to handle restrictions across alternatives in a cbcTools design – for example preventing dominance or never allowing the same brand to appear more than once in a choice task? Best! Tom From: John Helveston ***@***.***> Sent: Friday, December 8, 2023 10:30 AM To: jhelvy/cbcTools ***@***.***> Cc: teagle ***@***.***>; Author ***@***.***> Subject: Re: [jhelvy/cbcTools] cbc_overlap help description possibly incorrect (Issue #32) Okay great, thanks for sending this. It's finals week coming up, so this will probably have to wait a week or so until I can find time to look into it. — Reply to this email directly, view it on GitHub<#32 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ABTQPWOF54YZ7NR7PWJJR6TYINMC3AVCNFSM6AAAAABAL5C5QGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNBXGY2DENRWGY>. You are receiving this because you authored the thread.Message ID: ***@***.***>

[jhelvy]

These are definitely things I want to add to improve the package. Right now it mostly relies on other packages to come up with designs, like D optimal designs. But I think a lot of those designs don't work out too well, and a lot of users want more control over specific things like avoiding large overlap, removing "dominant" alternatives, etc. All things I want to add, but it'll be a while til I have time to do some overhaul work.