Tables with scripted subquestions don;t work properly

More
1 year 1 month ago #141864 by Malgorzata_Zub
Malgorzata_Zub created the topic: Tables with scripted subquestions don;t work properly
Hallo everyone
Beware of the ew function which is supposed to work in LimeSurvey 2.51.3+160907. The version offers an opportunity to add relevance equations to subquestions, but this doesn't always work for survey participants. I've done this in a table question with relevance equations on rows (subquestions), and while the survey worked for some participants, it crashed on this question for a number of others.
Best wishes,
Małgorzata

Please Log in to join the conversation.

More
1 year 1 month ago #141868 by holch
holch replied the topic: Tables with scripted subquestions don;t work properly
OK, this is a little vague, but should it be true of course needs further investigation.

Do you have a more concrete example?
What doesn't work?
What does "crashed" mean?

I'm not a LimeSurvey GmbH member. I answer at the LimeSurvey forum in my spare time. No support via private message.
Some helpful links: Manual (EN) | Question Types | Workarounds

Please Log in to join the conversation.

More
1 year 1 month ago #141875 by Malgorzata_Zub
Malgorzata_Zub replied the topic: Tables with scripted subquestions don;t work properly
Well, to be more detailed:

One question was a table with responses ranging from 0 to 10 and "no answer".
The second question was a similar table which showed those rows, which were related to rows from the 1st table. Eg if in the 1st table in the 3rd row people answered between 8 and 10, they saw the 2nd table with the 3rd row visible.
The scripts were like this:
Relevance for the entire group (which included the 2nd table)

((P1.NAOK == "A1") and (((P3_SQ001.NAOK >= "8") and (P3_SQ001.NAOK <= "10")) or ((P3_SQ002.NAOK >= "8") and (P3_SQ002.NAOK <= "10")) or ((P3_SQ003.NAOK >= "8") and (P3_SQ003.NAOK <= "10")) or ((P3_SQ004.NAOK >= "8") and (P3_SQ004.NAOK <= "10")) or ((P3_SQ005.NAOK >= "8") and (P3_SQ005.NAOK <= "10")) or ((P3_SQ006.NAOK >= "8") and (P3_SQ006.NAOK <= "10")) or ((P3_SQ007.NAOK >= "8") and (P3_SQ007.NAOK <= "10")) or ((P3_SQ008.NAOK >= "8") and (P3_SQ008.NAOK <= "10")) or ((P3_SQ009.NAOK >= "8") and (P3_SQ009.NAOK <= "10")) or ((P3_SQ010.NAOK >= "8") and (P3_SQ010.NAOK <= "10")) or ((P3_SQ011.NAOK >= "8") and (P3_SQ011.NAOK <= "10")) or ((P3_SQ012.NAOK >= "8") and (P3_SQ012.NAOK <= "10")) or ((P3_SQ013.NAOK >= "8") and (P3_SQ013.NAOK <= "10")) or ((P3_SQ014.NAOK >= "8") and (P3_SQ014.NAOK <= "10")) or ((P3_SQ015.NAOK >= "8") and (P3_SQ015.NAOK <= "10")) or ((P3_SQ016.NAOK >= "8") and (P3_SQ016.NAOK <= "10")) or ((P3_SQ017.NAOK >= "8") and (P3_SQ017.NAOK <= "10")) or ((P3_SQ018.NAOK >= "8") and (P3_SQ018.NAOK <= "10")) or ((P3_SQ019.NAOK >= "8") and (P3_SQ019.NAOK <= "10")) or ((P3_SQ020.NAOK >= "8") and (P3_SQ020.NAOK <= "10")) or ((P3_SQ021.NAOK >= "8") and (P3_SQ021.NAOK <= "10")) or ((P3_SQ022.NAOK >= "8") and (P3_SQ022.NAOK <= "10")) or ((P3_SQ023.NAOK >= "8") and (P3_SQ023.NAOK <= "10")) or ((P3_SQ024.NAOK >= "8") and (P3_SQ024.NAOK <= "10")) or ((P3_SQ025.NAOK >= "8") and (P3_SQ025.NAOK <= "10")) or ((P3_SQ026.NAOK >= "8") and (P3_SQ026.NAOK <= "10")) or ((P3_SQ027.NAOK >= "8") and (P3_SQ027.NAOK <= "10")) or ((P3_SQ028.NAOK >= "8") and (P3_SQ028.NAOK <= "10")) or ((P3_SQ029.NAOK >= "8") and (P3_SQ029.NAOK <= "10")) or ((P3_SQ030.NAOK >= "8") and (P3_SQ030.NAOK <= "10")) or ((P3_SQ031.NAOK >= "8") and (P3_SQ031.NAOK <= "10")) or ((P3_SQ032.NAOK >= "8") and (P3_SQ032.NAOK <= "10"))))

Relevance for subquestions in the 2nd table:
((P3_SQ001.NAOK >= "8") and (P3_SQ001.NAOK <= "10"))
((P3_SQ002.NAOK >= "8") and (P3_SQ002.NAOK <= "10"))
etc etc.

It worked when I tested it on 2 different computers, one with Windows 8 and one with Windows XP, both in Google Chrome.

But I've got information from some respondents, that when they went through the 1st table and got to the 2nd one, two things happened:
1) the 2nd table - the one with scripted rows - worked very slowly. It took respondents ages for an answer whcih they wanted to indicate to get indicated. When I'd tested it before sending, I;d also noticed that getting answers into this 2nd table was very slow. But eventually I could complete it and move on to the next screen, whereas...
2) my respondents found, that when they clicked "go further", they didn't get to the next screen, but to the previous one and their answers to the 2nd table were lost.

I hope this helps.

Małgorzata

Please Log in to join the conversation.

More
1 year 1 month ago #141876 by Malgorzata_Zub
Malgorzata_Zub replied the topic: Tables with scripted subquestions don;t work properly
And one more thing:
When I removed the table with scripted rows and turned each subquestion (row) into a separate question, with the same relevance equations (this time for questions, not for subquestions), the survey worked like it should (at least for one respondent, for whom the other version crashed).
MZ

Please Log in to join the conversation.

Start now!

Just create your account and start using Limesurvey today.

Register now