Hi everyone,

I just wanted to ask how can I do the following with Limesurvey.

I'm looking to create a survey that I will give to every class in a school. Each teacher has 8 classes and there are a total of 40 teachers. I need to give the same survey to every student in each class. I want to be able to look at the feedback from each individual class. Is the best way to create a unique survey for each group or is there something else I could do?

I'd be delighted to hear how I could set this up.

Thanks

Anyone able to help with this?

You can go for both methods. It depends very much on what you want to do with the results and how you want to look at them.

Question is also how the survey is distributed to the students. Via email? One link for all?

If the questions for all classes are the same, then I personally would go for one survey and use tokens and custom attributes to define which teacher and which class the survey is about.

For analyzing the results, I would export the results and look at them in Excel if it is simple or use a more advanced tool to create results like SPSS/PSPP, Sofa Statistics, etc.

If you want the online reporting as the only source for analysis, it is probably best to create one survey per teacher / class, but this is a lot of surveys to manage.

So if you can give us a little bit more information, it would be probably a lot easier to help you.

Hi Holch,

Thanks for your reply.

In the survey there will be 3 main groups of questions.
1. Teacher and class feedback
2. The school
3. Facilities

From the feedback I'd like to be able to view each teacher's student feedback for each class but I'd like last two categories to be viewed as a whole of all surveys.

I was thinking that I could have one survey for each teacher. Within that survey I could have the first question require the students to select what group they are from using a "list dropdown" question. then the rest of the survey will be the same.

This way I can give students one link that the teacher would distribute in class.

Could I export the results and merge the second two categories later with Excel ?
Does this sound like a bad way about doing this? Would you recommend something different? I'm sure I'm not the only person doing something like this so I'd love to hear what others have tried.

Thanks again

OK, if you just want to give one link, I would actually go radical and go for one link for the whole school.

Groups 2 and 3 are the same for all teachers and all classes anyway. I assume that the questions for each teacher and each class are also the same or would there be different questions for each type of class or each teacher?

Because if the questions are the same, I would definitely use one survey for all and do the analysis in Excel later anyway.

First question would be which teacher they students want to evaluate and the second would be which class the evaluate (here you can create a question that contains the class type specifically to the teacher).

If you would be able to invite each student via email, you could actually skip those two questions and include the teacher and the class already in custom attributes of the token management.

Thanks for you input holch. I'm going to give this a try and see how it goes. I'll keep you posted as I'm sure more questions will pop-up at some stage :laugh: . Thanks again. I appreciate your time.

David

Hi!
Found your discussion while searching for a way to merge answers from different surveys.

So, if I have understood everything correctly, I have to choose between having nice diagrams for the whole group and excel-documents for the different "groups in the group" or I can have different surveys for all groups but then I can´t have diagrams from Limesurvey for the whole population?
If I´m right in this, is there anything we can do to change that?

Maybe I should add that I´m going to do environment surveys in various groups. They work in different places but everyone has the same employer.

If you either have in a token (extra attribute in the token table) which group the respondent is a member of or if you just ask which group the respondent is a member of, then you can filter per group and make the usual LimeSurvey graphics from the backend.

So, more or less, it may work from one survey to all respondents.

I would choose to do this another way: use one survey (just as described above) and export the results either to Excel or to PSPP (SPSS export) and make graphics with those tools, but that is because I am not a fan of the standard LimeSurvey graphics.

Thank you for quick answer! I tried to do just that. I made up four persons and put them in to different groups with attribute 1. I also use tokens for them. But I can´t find where I can differ them using the filter I use a swedish version so I don´t know if the picture I attache will help you anything but that´s how it looks when I want to use filters. Nothing like it looks here: www.limesurvey.org/manual/Statistics#Vie..._and_Field_Summaries

My apologies, I forgot 1 thing: you also have to include (at the start of the survey is easiest) a hidden question where you insert the token (using Expression Manager) for the class, so that way you will be able to filter...

Thanks agian but I feel a bit lost I´m not SO good in either english or statistics... Is there any chance that you can explain a little bit more?

I set up a survey (see attachment) with tokens and one extra attribute for the class (numerical, because text will not help you)

The first group of the survey has 1 hidden question, which records the class as it is given in the token. The other question is one you could have asked.

This survey does record both the class as well as the shown question in the database, but I cannot figure out why you cannot filter on the hidden question. Hopefully another LS power user can help you further along the line.

I couldn´t open your file but I think I have found out how to do now except how I do with the hidden question. I have added an extra attribute. I call it Group and you can belong to group 1 or 2. Then I add a hidden question and choose the tokens from the Limesurvey replacement... The question looks like this: {TOKEN:ATTRIBUTE_1} Is that enough? I would have tried it´s easier to ask I hope for luck one more time!

Great minds think alike: what you did what was in the file that I sent you you.

A file that ends with .lss is a file that you can use for importing a survey. So: go to new survey and import this one using the .lss file. Not necessary now, because no new knowledge is in it...