Hi,

I'm totally new to LimeSurvey and wondering if my survey and myself are LimeSurvey-compatible.

My survey has 3 sections: (A) demographic questions; ( 10 scenarios with the same 3 questions after each scenario; and, (C) final questions. I need to be able to: send invitations and reminders, track who has (not) responded, and export the data and/or analysis. Participants cannot respond more than once.

Part B is the complex area:
- There are 9 versions of Part B. This must be randomised for each respondent.
- If possible, I would also like to randomise the order of the 10 scenarios in each version of Part B. However, if this cannot be done, so be it.

What is possible / not possible in LimeSurvey? How and where do I start? Unfortunately, time is not on my side so I have limited time to indulge in learning LimeSurvey properly (hence, the paying options (competitors of LimeSurvey) look attractive).

Thank you very much.

I think it's possible to do it in LimeSurvey. However, you may need to learn how to use Limesurvey first by reading the manual if this is first time using it. If you don't have time to learn it, then I guess go for paying options as you mentioned. Good luck

How and where do I start? Unfortunately, time is not on my side so I have limited time to indulge in learning LimeSurvey properly (hence, the paying options (competitors of LimeSurvey) look attractive).

Any survey tool will need some learning for what you are trying to do. It is not very complicated to do, but you need to get the basics done and then need to understand how to randomize (not too complicated, depending on what you need to do).

What I did not understand is Part B: First you say that there are 10 scenarios with the 3 same questions for each scenario. OK, no problem.

However, later you say that there are 9 versions of part B. So actually you have 9x10x3 questions in Part B? Do I understand this correctly? Who will fill in 180 questions?

All sounds possible to do, but we would need to understand the part B better to make sure that all your randomization works. Because you will have work with group randomization.

Will you really create 9 times 10 scenarios and every respondent needs to see 9 times 10 scenarios?

Thank you, david2013 and holch for your prompt feedback .

Just to clarify, each respondent would see only 1 version of Part B (10 scenarios, 30 questions). No one could be expected to answer 180 questions plus the questions of Parts A and C.

That said, I'll have a prize drawing to motivate and reward participants.

The problem is that you would actually need two levels of "question groups".

You would need to greate 9 groups for each version of B and then within each version of B you would have another 10 groups with the different scenarios.

As you only have one group level available in Limesurvey you will need to think about some kind of workaround. Here some options:
- 9 different questionnaires (1 for each version of
- Hide all other versions of B via conditions or better relevance equations. This means quite a bit of work. But your design will always cause quite a bit of work.

Wowww!!! Thanks for thinking this through, holch. It sounds like really high level stuff (double Dutch in fact :blink: )... I'll give it a shot next week when I'm back at my desk.

Just out of interest, how long do you think you would take to do this?

Difficult to say. But if the structure of the questions / groups is quite repetitive the basic structure should be quite quick if you are working on it straight and find a good rythm.

From what I understand your whole questionnaire in the B section is very similar with minor changes to the scenarios within a version of B. I don't know how the different versions of B are differing from each other. Would be good to know what makes the versions of B different. Within the version from what I understand only the text for the scenary changes, right? So this should be quick. You create a text display question for the scenary and then the 3 questions which are the same for the scenarios. Once that is done you just copy the groups (I think you need to export and import them, there is no copy question group function, as far as I remember).

Now the question is, what makes the version of B different from each other?

Hi again, holch! I appreciate your time and interest, but I must be frank that I'm very likely not to use LimeSurvey this time around. And I don't want to waste your time...

To answer your last question, the scenario consists of a short text (5-6 lines). There are 5 changes to be made each time, either a word or a short phrase. Currently, this is all in SAS code and I was going to transform it into a Word document.

Would you (or another LimeSurveyer who knows what they are doing) be interested to create the survey for me? (Is such a question allowed on the forum?)

Thanks again.

dkwok83 wrote: Hi again, holch! I appreciate your time and interest, but I must be frank that I'm very likely not to use LimeSurvey this time around. And I don't want to waste your time...

It is totally up to you to use Limesurvey or not. We can only explain how things work in LS. If this is not working for you, so be it.

dkwok83 wrote: Would you (or another LimeSurveyer who knows what they are doing) be interested to create the survey for me? (Is such a question allowed on the forum?)

Not me. I am only helping out in the forum in my spare time, but I can't dedicate too much time on Limesurvey. But here are many Limesurvey users that provide professional support in creating surveys (e.g. Denis from "Sondages Pro" or Mazi from "Limesurvey Consulting", which both are also involved in the development of Limesurvey). Have a look at the professional partners link on the website. "Survey creation" should be what you are looking for: www.limesurvey.com/survey-creation

Thanks holch, I'll ask for quotes.

Have a great day!