Check out the LimeSurvey source code on GitHub!

Psychological Assessment Difficulty with Likert scales

5 years 7 months ago - 5 years 7 months ago #62151 by RichTatum
I'm trying out LimeSurvey and am trying to set up a survey with 46 Likert-scale questions that relate to four different measurements. Here are my problems:

  • I would like to present the 46 questions randomly. However, I think I have to set them up in one of four "groups" to correspond to the psychological aspects being aggregated/measured. And the easiest and most usable way to set up these questions is to create them as "subquestions." Unfortunately, even though I've assigned each question a random group string, it appears that while subquestions can be randomized, and questions can be randomized, and groups can be randomized, I cannot treat all 46 questions as a single "group" subject to random presentation. Suggestions?

  • When showing the user public answers at the end of the survey, I am not interested in showing detailed summaries of answers to each of the 46 questions. Rather, I would like to aggregate the scores for each of the four psychological "groups" the questions belong to (like a personality survey). As far as I understand the documentation, this does not seem possible. Is it?

  • I would like some of my questions to use a reverse-score for the Likert scale (e.g., for regular scoring, option 1 is assessed a value of 1, but for reverse scoring option 1 is assessed as a value of 7.) Unfortunately, this requires a separate question/subquestion. It seems the only way to get around this is to require each of the 46 scales to reside as independent questions. Is this correct?

  • Do I have to enable tokens/registration in order to send the user a confirmation email at the end? This is not clear to me. I disabled tokens/public registration and anyone can take the survey now, but there seems to be no ability to send email confirmation.

Regards, and thanks in advance for any help you can provide. I'm hoping LimeSurvey will fit the bill for my client - it seems to be the best thing going right now. Unfortunately, I've hit some serious snags here.


Last Edit: 5 years 7 months ago by RichTatum. Reason: Added bulleted list

Please Log in to join the conversation.

3 years 3 months ago #100627 by deprem
sorry , no way
lime survey is not suit for psychology research :whistle:

Please Log in to join the conversation.

3 years 1 month ago #102380 by FuzzyGman
could you explain what you base this on and why...

Please Log in to join the conversation.

3 years 1 month ago #102385 by fransmarcelissen
What do you mean, Deprem, is this a joke, or so? Limesurvey is very suited for psychological research.

Please Log in to join the conversation.

3 years 1 month ago #102386 by fransmarcelissen
Hi Rich,
I think everything you mention is possible.
1. You say that there are four groups of items. But that does not mean you have to create 4 limesurvey groups. It is more simple to make 1 group with all items, and later on make calculations on item1-7, item 8- (etc). In this way you can do complete randomization.
2. If you want to show the aggregated results on-line, use expression manager. With that you can create new, aggregated variables and show these.
3. Within one array all the subquestions have the same scores. But later on, when you use expression manager to calculate score scores, you can reverse these (trait=v1+(6-v2).....)
4. In general you need tokens to send email confirmations. In the confirmation you can also show the (aggregated) scores. I suppose it is possible (using javascript?) to do it without tokens, but I did not try this (and will not try this).


Please Log in to join the conversation.

Imprint                   Privacy policy         General Terms & Conditions         Revocation information and revocation form