Hello

I am developing a survey that has simulated store visits where participants select from various product features as illustrated in the attached image.


I have reviewed question types but cannot find anything close.

Do you have suggestions? Is a custom Question Type required?

Regards,
Wayne

I am sorry but I don't even understand the displayed question - looks way too complicated to me.

Remember: Don't put the analysis on the back of the participant by fitting the question to the analysis.

Whoever came up with this most likely needs to go back to the drawing board.

It seems to me that you can use a normal select-radio question. Your table of "features" can be created in the question text.

I have worked on similar solutions before. One can use some JavaScript to connect a tabular being placed at the question text with some radio buttons to answer the questions:


In a more advanced setup one can even grab the features from a table made for conjoint analysis.

I have worked on similar solutions before. One can use some JavaScript to connect a tabular being placed at the question text with some radio buttons to answer the questions:

Yep, or use JS to physically move the radio/checkbox/text inputs up into the table in the question text.

.

Mazi, you example looks great.

@Carsten: This is something quite common in market research: conjoint analysis. Some online research tools have their own sections for this, as in some cases you do not just need the layout (would be solved with a simple question type), but you need to calculate the combinations of attributes and the analysis can also be quite tricky.

But it would be great if Limesurvey would have at least the possibility to create the question layout.

@Mazi: Bei einer ordentlichen Conjoint-Analyse sollte aber dein Beispiel gar nicht auftauchen. Eine Nikon mit Vollformat für Euro 500,00 und eine Canon mit APS-C-chip für Euro 2.500. Da dürfte die Wahl ohnehin klar sein und die Kombination ist auch nicht wirklich realistisch. Aber das ist mehr ein Problem des Umfragedesigns denn ein Limesurvey-Problem.

@Tony: Moving the answer option into the above table is exactly what we did for this.

@holch: That data was provided by the customer, don't blame me
I think the whole DB contains some hundred thousand data sets.

We kind of hacked the conjoint feature into Limesurvey by coding new question attributes to adjust the layout (meaning calling the JavaScript to adjust the layout if the attribute was set) as well as connecting to an external DB to dynamically load the HTML of the comparison table into Limesurvey.
You can even tell which data set the system should load, how many features it should list and the like.

Additionally, some special export was coded to provide a custom format to later analyze all that stuff with statistic software "R".

Ahh, sorry, I don't know why I switched to German.

@Mazi: I am not blaming you. I assumed that this was data delivered by the client. If you just compare all possible options, you'll ask the obvious in some cases. But it might be necessary in some methodology, I guess.