Browse Space    
Version 2 by vegard
on Jul 17, 2008 05:07.
<< Changes from 1 to 2
compared with
Current by vegard
on Jul 17, 2008 05:16.
Key
This line was removed.
This word was removed. This word was added.
This line was added.

Changes (7)

View page history
...
..this as there can be reasons for not filling up the database with this information at every test.
h3. Improve test statistics
h3. Improve how you can use statistics
 
For some test types, the statistics needs to be improved. This is the statistics for a graphical matching question, where you are asked to match "dog" with "woof" and "cat" with "miau"
In some situations, specifically for surveys, it's interesting to get cross reports on answers to a test. For instance you might want to know how many "people between 20 and 25" (alternative on one age question) has "daily access to Internet" (alternative on a different question). what has "doctors" (alternative) answered on questions X, Y and Z (three different questions) etc.
 
It's not at all obvious what is the best way to implement this, but one should probably have a look at other test & survey tools to get some ideas.

h3. Improve test statistics for certain question types.

For some test types, the statistics needs to be improved, specifically matching and ordering questions. This is the statistics for a graphical matching question, where you are asked to match "dog" with "woof" and "cat" with "miau". The problem with the stats are that you do not see the alternatives, and it's not obvious how many matched what rightly or wrong.
!screenshot1.png!
And this is the statistics for a sorting question asking the user to sort 4 countries from North to South:
And below is the statistics for a sorting question asking the user to sort 4 countries from North to South. Again, how many users got it right and how many got it wrong?

!screenshot2.png!