Written by mfreund the 6 Jul 10 at 22:08.
Related project: 1.x all.
New
It would be great if questions containing subquestions (e.g. array questions with multiple questions) could display in statistics and graphs together on a single chart.
Also, pie charts are a poor way to display data that is not categorical in nature. Ordinal (e.g. education level) or scale (e.g. age) variables are hard to make sense of in pie charts.
For example, I often ask respondents to agree or disagree with a number of related statements. It would be good to view the responses together and displayed in a way that conveys that the responses are on a continuum.
Written by atavei the 7 Apr 10 at 19:06.
Related project: 1.x all.
New
Nowadays, all research with human subjects require informed consent and authorization. The use of a program such as LimeSurvey in academic, government, or business environments require a mechanism of informed consent. In the specific case of academic environments, Institutional Review Boards WILL NOT APPROVE any type of research with human subjects without a mechanisms to verify the informed consent. Restrictions such as this are already found (or likely to be found in the near future) in government, medical and some business settings. Usage of LimeSurvey in academic--or other-- settings will be hindered by the absence of this kind of functionality.
Written by pmonstad the 10 Oct 09 at 19:00.
Global category: Logic & Conditions.
Implemented
Making conditions in LS are very time consuming.
The existing condition rules are only useful when asking a very few additional questions, as the condition has to be entered as a rule to each question. There is no way for example to easily make whole question groups conditional.
Written by JoyLaneResearch the 29 Nov 10 at 22:17.
Related project: 1.x all.
In development
In concept testing (marketing research) it is sometimes necessary to randomize or rotate the order in which groups of questions are presented, such as when exposure to questions about one concept change feelings about another.
While not a feature of LimeSurvey, there was a workaround available that met this need in LS 1.87+. The workaround was described in a document called "randomize_groups.pdf."
Unfortunately, the workaround described in the document no longer works in LS 1.90+.
I have two currently active clients that rely on me to have this capability. Without it, I may loose their business.
Written by newleo the 6 Nov 10 at 14:13.
Related project: 2.x all.
New
I wanted to use Limesurvey for Assessment, but I would have to download the data to do it in excel, because there aren't radar charts. Is there any chance these are in your roadmap?
Written by peterbrown05 the 12 Oct 09 at 09:57.
Related project: 1.x all.
Implemented
When a survey has been completed, we have chosen to receive the full notification including questions + answers.
It would be nice if the email could be formatted via html/css so that it is easier to read in email client. presently it is essentially just plain text.
Written by tiprichard the 25 Sep 09 at 17:17.
Related project: 2.x all.
New
For assessments we need statistics about howmany people passed the assessment. (like an exam)
For auditing purposes we need reports how many points are scored per group as percentage of the total possible points. If possible shown in a spiderweb graph, where each group equals one line of the web.
Any question that totals to 100% gets a graphic representation in the form of a pie chart.
I tried to change the code to make it possible to choose a bar chart instead.
I did this because the pie chart is terrible to intepret when it depicts more than 3 variables. It sure looks nice, but it does not convey information in a nice manner: it flabbergasts.
(This link tells it much better than I ever could:http://www.perceptualedge.com/articles/08-21-07.pdf)
Written by ppuls the 15 Oct 10 at 10:14.
Related project: 1.x all.
New
Dear LS-Team!
I am using the "Send invitations" with additional parameters and i am never sue, if they are incorporated in the email correctly. this can be quite a nerve wrecking situation, if doing this for a large number of invitations.
Whenever i do this with my testrecords, and have to reimport the token table, i am never quite sure, if i checked all the flds, if i installed all the correct additional attributes...
Written by mfreund the 5 Oct 10 at 03:01.
Related project: 1.x all.
New
Currently, LS defaults to display 50 tokens, and it would be good to be able to easily change that setting without going into the PHP file.
I often have surveys with over 50 respondents, and it is inconvenient to always have to change the number of tokens displayed in order to see them all.
Written by mfreund the 27 Sep 10 at 12:19.
Related project: 1.x all.
Implemented
Currently the assessments only seem to allow reporting based on a total value for each group (or perhaps the percentage of total possible score?).
It would be great if it were possible to construct assessments using a mean or other statistic (e.g. median or mode)
Even better would be to be able to create more complex calculations. Some applications might be to allow for differential weighting of questions (beyond simply assigning greater or lesser assessment values), transformation of values (e.g. combining a question with a 4-point scale and one with a 5-point scale), or calculation of new variables using question values (e.g. ratio of two question scores).
Written by partic the 14 Aug 10 at 16:09.
Related project: 2.x all.
New
The assessment options in Limesurvey are currently reasonably limited, and a personality type assessment would be useful.
The type of survey I'm trying to build at the moment is a personality survey where a user goes through a number of questions, and based on the totals of their answers to options A, B or C, their personality trait can be identified.
eg Based on the percentage of A, B and C a particular trait might be highlighted. This trait would be then displayed to the survey participant, as well as being recorded as the result of the survey along with the answers and assessment score ideally.
> 50% A = Type X
> 50% B = Type Y
> 50% C = Type Z
> 40% A & > 40% B = Type XY
> 40% B & > 40% C = Type YZ
> 40% A & > 40% C = Type XZ
> 30% A & > 30% B & > 30% C = Type XYZ
Written by lovepade the 3 Aug 10 at 08:59.
Related project: 1.x all.
New
Currently the system can send out reminder emails, which is great. But the emails which people receive are identical regardless of whether they have already begun the survey (but not finished) or wether they have not begun the survey.
This is a problem because some people actually think that they have finished the surveye - or maybe forgot that the got interrupted. Some of theese people get annoyed when the recieve a confirmation mail.
Written by linxin the 25 May 10 at 11:10.
Related project: 2.x all.
New
As we know, there are may applications and games are become more and more popular with the integration with Facebook.
If we can make LimeSurvey become a widget, then it would make LimeSurvey become more easy to integrate with SNS or open source applications, such as Joomla, WordPress, etc.
Written by Mazi the 23 Oct 09 at 16:24.
Related project: 1.x all.
New
When creating / testing / editing a survey mandatory question are really annoying. I can think of these situations where making questions of a survey (temporary) non mandatory will be a great improvement:
1. (Developer) Testing possible bugs
When importing a survey for testing mandatory question can drive you nuts.
2. (User) Previewing a survey which is in development
You just want to take a look at the questions and survey design, having to answer all mandatory questions is a mess.
Written by lemeur the 3 Jan 10 at 14:33.
Related project: 1.x all.
Implemented
Limesurvey tokens already have a field named token status that takes the "OK" default value. When this field is updated withj any other value, the token email can be automatically bypassed by LimeSurvey invitation/reminder process.
The goal of this idea is to design a new process that would make it possible to an external probe script, to send a failed-email-address notification to LimeSurvey so that the token status is automatically updated (right now this is only a manual update).
The probe script is external to limesurvey since its operation depends on the mail-server system used by the Administrator Bounce address.
Special care must be take to prevent anyone on the internet from sending a specially crafted fake bounced email that would disable valid tokens.
Written by nobrain the 21 Jan 10 at 02:11.
Related project: 1.x all.
New
I need to collect more contact information on registration such as phone number, address, etc.
I can add an attribute for it (tokens) but then when the user visits, it does not ask for this info.
So, basically my idea is to have a setting "Public" and "Required" for attributes in tokens. Then, I can add an attribute such as "Phone Number" and set it to public and required and the user will be required to provide it to take the survey.
Written by mkhines the 20 Jan 10 at 16:19.
Related project: 1.x all.
Implemented
There are surveys we take that require some degree of geographic information to be submitted.
Rather than ask novice/general public users to use a GPS and submit coordinates (which invariably end up incorrect), we would like to have them use a map-style interface like Google maps, where they can drag and drop the pin where they were when they saw what they are reporting, and then have the question type save the latitude and longitude of their dropped pin.
Written by dynia the 23 Oct 09 at 21:02.
Related project: 1.x all.
Implemented
In dual-scale array questions there is no clear visual separation between scales, what may lead participants to make errors when answering such questions (especially when there are many answers).