Good day everyone!

I am having a little bit of trouble to finish a survey.
Concretely, we are evaluating some classes and would like to do the following:
1) participants select the class(es) they have participated in (multiple choice answer)
2) ask the participants to evaluate the quality of class XYZ they participated in

The trouble is the following: I would like the question about the quality of the class depending on which class(es) they have selected in the first question, so that the graphs will also automatically show us the evaluation of a particular class.

I could use the condition "Subquestion 007 (e.g. Monday yoga class) = Yes" but that has two major inconveniences:
1) There are over 50 classes, it will take forever
2) The participant will not know which "quality evaluation" question corresponds to which class, if they select several classes since it will not be shown in the question itself (= high error margin)

Is there an easier, more sensible way to do this in your opinion?
I would appreciate any help!

Have a good day!

Hi,

There are over 50 classes, it will take forever

Why?
As you have to create the group for the several classes, it willl take not more than five seconds to enter the relevance equation.

You didn't really talk about the outdated "condition builder"? Nowadays we use relevance equations.

And of course you can show the selected class in the question text of the validation group.
Read the manual here.
manual.limesurvey.org/ExpressionScript_-...#Access_to_variables

To get a more detailled answer, send a prototype of your survey (only some relevant questions) as lss export.

Joffm

Hello Joffm,

thank you for your swift reply!
About the conditions -it could be I am simply not understanding it- from my page, I can only choose the input "Which class(es) have you participated in" and then the class in question.
Or the class in question Y/N.
I am attaching what I am seeing and the .lss file as well, thank you.

I will have a look at the question text input, thank you! That helps a lot!

Best

Hi,

1. Why do you double identical questions? (for staff and students)
You have a variable to split these groups in your analysis.

2. How do you think to be able to calculate some statistical values of your "quality" question (a scale), if you use strings as codes?
Use numerical values to be able to calculate a mean, a std.deviation, statistical tests like t-Tests, or ANOVAs.
I did this in my sample survey.

3. As I said. Forget the condition builder.
Just enter the relevance equation here


In my example I removed the double questions, changed your alphanumerical codes in the scale, improved the layout of the answer options,
and showed how you have to proceed for the rest of your classes.
There are only about 5 clicks.
Copy the question,
change the question code,
change the subquestion code in the relevance equation
and - if you want - change the subquestion code in the equation that shows the class name in the question text.

If there are more questions to each class you should create a group for each class.


And if you really have only this one question and want to show all selected classes in one array, use subquestion relevance as I showed in G5.


Here the lss


Joffm

Hello Joffm,

my apologies but when I uploaded your .lss file it did not contain any questions?

In any case, I will take the time to read through all of your comments and suggestions. I am by no means an expert and I thank you for your valuable input.

Well, of course there are questions.

And you use version 4.x., don't you?
You have the rights to import lss files?
The download of my file has a size of 104kB?

After the import you should see this.

Joffm

No unfortunately we have a Version 3.19.3+191023. I cannot change this since it is managed directly by my employer.
However, with the directions you've given me, I've been able to come much further already (a few last things to figure out and I should be fine!)

As I said, I am by no means trained in informatics or with this kind of software so it has all been learning by doing: it takes me a bit to figure things out so again, thank you for your patience with the explanations in this case, they have been very helpful indeed!

Hi,
Therefore the most important information is the version.
Anyway, I will send the sample in 3.x
As you know, you can't import a 4.x lss into 3.x.
Joffm

Here the survey for version 3.x.



Joffm