Hello Community,

I'm trying to create a questionnaire where there are two questions:

1) one where you have to rate a product (scale from 1 to 5)

2) second one where you have to rate from 1 to 3 the three most important benefits of the previous products.

In the second question I would like to reduce the number of products based on the selection done in the first question: I would like to have just only the products that have been rated with 3 or more.

Is it possibile to do that?

I attached a file in order to better understand my textual explanation.

Best regards,

Andrea

See relevance equation on sub questions : www.limesurvey.org/manual/Adding_answers...fic_for_subquestions

Hi,
1. first you should swap benefits and products in the second question.
Not only because of consistency but... there is no subquestion relevance on columns.

2. You should change your alphanumerical codes "A1", "A2" to numerical 0,1,2,3...
In the second question you are going to compare

3. Having done this you can set the subquestion relevance in s1001
s1001_SQ001: s1000_SQ001>2
s1001_SQ002: s1000_SQ002>2

With your codes "A1",... you had to enter: s1000_SQ001=="A4" OR s1000_SQ001=="A5" OR s1000_SQ001=="A6"
Furthermore: I am sure you want to calculate some statistics, like mean, t-test, or so.
What do you think is the mean of "A1", A2" and "A4". It is not "A2,33".

Best regards
Joffm

Hi Joffm,

I swapped products and benefits in the second question and I also changed the codes (in the first question - edit answer option).

Then I set the relevance equations in the second question but i have probably done some errors because it didn't worked.

I share a picture about the second question's setting.


Thanks in advance,

Andrea

Hi, Andrea,

s1001_SQ001: s1000_SQ001>2 should mean:

In subquestion SQ001 of question s1001 you enter into the relevance equation s1000_SQ001>2

Joffm

Yes Joffm,


I had also tried that solution but it didn't work so I thought it wasn't correct.


I attach the file. I launched the survey preview but the second question's rows still also appear independently by the first question's answers.

If you call your question "s1000", please use "s1000" in the relevance equation, not "S1000"



Some more words:
Remeber to check in each row that only three benefits are selected with different values.
This is done by subquestion validation.

Joffm

Thank you Joffm,

Ops, I made a terrible mistake with that "s".

I will try to set the subquestion validation as you suggested to me.

Thank you again for your help.

Andrea

Keep calm, Andrea,
that's not a terrible mistake. That's a typo, but unfortunately you will never find yourself.

Similar to my "remeber" in the last post.

Joffm

Regarding to the subquestion validation i was trying to use the following logic:

1) Set the sum of the subquestions' answers must be equal to 6 (because there are only three possible value: 1,2,3)

2) There must be only 3 answers for each row, the others must be empty (because they have to choose the three most important benefits among all the possible one).


Is it a good way to do it or there are other better solutions?

Best Regards,

Andrea

Yes,
these are the two conditions:
Sum of answers ==6
Count of answers =

Joffm

You have also to check uniqueness to avoid "2+2+2"

Joffm