Check out the LimeSurvey source code on GitHub!

maxdiff scaling

More
5 years 11 months ago #57714 by abfab
is it possible to carry out best-worst question with lime survey?
thx :)

Please Log in to join the conversation.

More
5 years 11 months ago - 5 years 11 months ago #57717 by paulfiner
Hi

It is possible to do this type of question but it's not completely straightforward.
Here's the link to the workarounds section: docs.limesurvey.org/tiki-index.php?page=...axDiff_question_type

This workaround details how to make the layout work but when doing a MaxDiff question you need to handle different versions. I have successfully managed it in a large survey.

Let me know if you have any further questions.

Cheers

Paul
Last Edit: 5 years 11 months ago by paulfiner.

Please Log in to join the conversation.

More
5 years 11 months ago #57718 by abfab
thanks a lot! i will try to make it work...and thank you for your offer to answer further question! what do you mean with different versions? we are also conducting a large survey using a balanced incomplete block design...you helped me a lot in stage...i think the next thing to deal with is how to handle the results in spss...:)

Please Log in to join the conversation.

More
5 years 11 months ago #57719 by paulfiner
How did you generate the block design...was it with the SawTooth software?

When creating a MaxDiff question you typically need about 5-6 different versions of the block designs to show randomly to respondents.

I'll see if it's possible to export this question from my survey so you can see all the Javascript that's required.

Please Log in to join the conversation.

More
5 years 11 months ago #57721 by abfab
wow...thank you! no...i designed it by myself and i think or hope that it it is valid....:;) i have a (5 attributes ,5 blocks,4 items ,4 times occuring,3 times pair occuring) and therefore symmetric or unreduced design...

First Block 1,2,3,4
Second Block 2,3,4,5
Third Block 1,2,4,5
Fourth Block 1,3,4,5
Fifth Block 1,2,3,5

it would be very helpful if you could send me your javascript! :)

Please Log in to join the conversation.

More
5 years 11 months ago #57722 by abfab
occurring not occuring

Please Log in to join the conversation.

More
5 years 5 months ago #65752 by ricardo01
I'm trying this with LS 1.91 build 10951, but can't get it to work. Have you tried with the new version of LS?

Please Log in to join the conversation.

More
5 years 5 months ago #65753 by ricardo01
Never mind, It works great! I seems my dyslexia is kicking in...I had the question id wrong.

Cheers

Please Log in to join the conversation.

More
5 years 5 months ago #65841 by Mazi

ricardo wrote: Never mind, It works great! I seems my dyslexia is kicking in...I had the question id wrong.

Cheers

You're welcome!

If our hints have been helpful and you enjoy limesurvey please consider a donation to the team .
We do all this in our free time and you don't have to pay a penny for this software.

Without your help we can't keep this project alive.


Best regards/Beste Grüße,
Dr. Marcel Minke
(Limesurvey Head of Support)
Need Help? We offer professional Limesurvey support
Contact: marcel.minke(at)survey-consulting.com'"

Please Log in to join the conversation.

More
5 years 1 month ago #72307 by bort123
how did you analyze the data?

does anyone know a freeware solution for that?
seems a little bit counter intuitive to use a free survey tool and then invest several thousand dollars for the sawtooth tools (or one of the other commercial tools, all in the same price range) ;)

counts (worst case, i do it manually :whistle: )
hierarchical bayes
latent classes

Please Log in to join the conversation.

More
5 years 1 month ago #72308 by paulfiner
In my case, my colleague already had the Sawtooth software and he analysed the data.
I agree that having to use very expensive software goes against the grain but I couldn't find any free tools out there.

I did look into the possibility of analysing the data without Sawtooth but it's fairly complicated (hierarchical bayes) and I don't think you can just use counts.

I might try an investigate further when I get some free time.

Please Log in to join the conversation.

More
5 years 1 month ago #72309 by bort123
happy to learn about your findings!

as a student the commercial tools are sadly a little bit out of my price range for one time use ;)

Please Log in to join the conversation.

More
5 years 1 month ago #72358 by Mazi
Wel, Limesurvey is a survey tool, no analysis tool. That's why we offer several export formats such as Excel, SPSS or "R" which can be used to analyse data.


Best regards/Beste Grüße,
Dr. Marcel Minke
(Limesurvey Head of Support)
Need Help? We offer professional Limesurvey support
Contact: marcel.minke(at)survey-consulting.com'"

Please Log in to join the conversation.

More
4 years 9 months ago #79527 by ricardo01
Actually you create the BIBD in R using the crossdes package

Check this link for more info www.r-bloggers.com/generating-balanced-i...-block-designs-bibd/

Please Log in to join the conversation.

More
4 years 9 months ago #79528 by ricardo01
You can generate the block design using R with the crossdes package. Check www.r-bloggers.com/generating-balanced-i...-block-designs-bibd/

Please Log in to join the conversation.

Imprint                   Privacy policy         General Terms & Conditions         Revocation information and revocation form