Greetings,

I have a survey which consists of 10 questions (10 points for each question). If the participant score all questions correctly then the assessment result will be saved, else a message will prompt out to ask the participant to retest but without saving the assessment result. Is it possible to do so?

Thanks & regards,
 

What is your definition of "retest"?
1. Start a new survey?
2. Go back to the first question?
LS has no feature to skip back except the index.
Or you have all questions on one page.

In my opinion the best is to screenout participants who didn't answer correctly and let them start a new survey.

And about the saving of the assessment score.
If it is only saved if all questions were answered correctly, the will be a constant score.
So why save it? Either everything is correct -> the survey can be completed or not all questions are correct -> the survey can not be completed.

Joffm

Hi Joffm,

Thanks for the reply. Yes, "retest" to go back to the first question. Would happy to see if LS or any conditions able to make the survey unable to complete if not all questions are answered correctly.

Thanks again for your suggestions.

Hmm, just an idea, not 100% thought through, but you could think about creating something like a screen out via a zero quota. You could calculate the score in a hidden equation question and then create a zero quota that closes when the score is not 100%. You could then put the survey url in the quota url.

Hi Holch,

Zero quota? Never heard about this before. Will look into on how to implement this.

Thanks for your suggestions.

Hi,
LimeSurvey has no feature to distinguish between "Quota Full" and "Screenout".
You see, usually you set a quota if you want to receive 100 responses, 50 male and 50 female. Here the quota limit of both is "50".

Now to screenout participants which you do not want to participate (as "too young", "driving the wrong car brand",...) you also have to set a quota, this time with limit "0". You want to get 0 participants of this group.
This is what @holch called "zero quota".
And what I explained as "the best is to screenout participants who didn't answer correctly".

So, what do you have to do?
You calculate the score in an equation (let's call it "eqScore")
Afterwards you create a second equation (let's call it "eqQuota") with the content
{if(eqScore==100,1,0)}  So variable eqQuota will have the value of "1" if all answers were correct (a score of 100), or a value of "0", if not.
Now you set your quota accordingly.
1. Set the quota
 
with the link to your survey. Don't forget the parameter "&newtest=Y"

2. Set the criteria.
 
Here the "answers" - "0" means, the quota fires if the value of eqQuota is "0" = "not all answers correct".

Remember, quota only work in an activated survey. You cannot test without activating the survey.

And read about quotas in the manual:
[url] www.limesurvey.org/manual/Survey_quotas [/url]

Joffm

@Joffm: I am curious why you implement a second equation. Couldn't the zero quota be set on the first equation? Everything that is not 100? Any specific reason or just because it is "cleaner"?

Hi, holch,
I tried, but it doesn't work
 
Here you CAN enter something like "<100", "!=100", but you are successful only with discrete values.

Therefore the additional equation.

Joffm

 

Hi Joffm,

Thanks for the guide. Really appreciate it.

Regards,
Jourdan

Another addition.
Of course you can combiine the two equations like
{if(sum(if(Q1==1,10,0),if(Q2=,10,0),...)==100,1,0)} where the red part is the calculation of the score.
Advantage. Only one equation
Disadvantage: The score is not stored.

Joffm

Here you CAN enter something like "<100", "!=100", but you are successful only with discrete values.

Understood. Makes sense for quota. Didn't think of that.

Joffm wrote: Hi, holch,
I tried, but it doesn't work
 

Here you CAN enter something like "<100", "!=100", but you are successful only with discrete values.

Therefore the additional equation.

Joffm


 

It's a new fearture in 4.X ? Use Equation for quota ? In 3.X : only single choice can be used for quota.

No, as far as I know this works in LS 3.x already to use equations for quota (I have learned that from Joffm).

I just see that !

Great !