Please help us help you and fill where relevant:
Your LimeSurvey version: [see right hand bottom of your LimeSurvey admin screen]
Version 5.5.1+230103
Own server or LimeSurvey hosting: 
Own Server
Survey theme/template: custom
=========
hello:

We have a survey that starts with a pre-test "group" of 20 questions. These 20 question draw from a hidden group of 64 questions to facilitate randomization.
The subject then views a training video and after proceeds to the post-test.
The Post-test "group" of 20 questions is drawn from a separate hidden group of 64 questions so I can again facilitate randomized of questions.
This all seems to work fine up to this point.

The problem i'm having is the scoring of the pre and post test. I have three assessment rules.
1) Scope: group / Question Group: "Pre-test" / Minimum: Zero Maximum: 200 (correct answers have an assessment value of 10). Message:  Vehicle Pre-Test Score: {TOTAL}!
2) Scope: group / Question Group: "Post-test" / Minimum: Zero Maximum: 180 (correct answers have an assessment value of 10). Message:  Vehicle Post-Test Score: {TOTAL}. Failed!
3) Scope: group / Question Group: "Post-test" / Minimum: 190 Maximum: 200 (correct answers have an assessment value of 10). Message:  Vehicle Post-Test Score: {TOTAL}. Passed!

Accurate scoring does not work well.
So, basically at the end of the survey i'd like the user to see the total score of the pre-test. For the post test, below 189 points the message will show "Failed"; above 190 the message will be "Passed". The user may only get one question incorrect.

Any ideas to adjust/improve the design to make this work?

Thanks

 

eyeballs wrote:

Accurate scoring does not work well.
 

The most common advice about assessment rules is to ditch them and create you own scoring system.

I have read your post twice but still don't get what isn't working
Which shows that I never use assessment rules
If every right answer gets a score of 10, why below 189? The rule uses 180 and 190 as the the thresholds.
Which looks OK.
So what results are wrong or not accurate enough?


I would recommend to check out ExpressionScript which will allow you to create your own scoring rules with a lot more flexibility.
And you have everything saved in the responseset. The assessment parts will never be saved and cannot be checked afterwards.