Assessment statistics

In the “Evaluation assessments” > Assessments Bank (AMI)” menu, the “Statistics” tab allows you to analyse the data in detail once the test has been completed.

The window shows the global statistics as well as those for the various test folders.

1. Global statistics

Statistical data for the examination, including all folders, is displayed.

This information is provided at the end of the test.

Various graphs complete this page:

  • Number of participants by average (weighted by questions and by folders)
  • Number of students with averages below 8, between 8 and 10 and above 10 (weighted by questions and by folders)

The global statistics also include diagrams showing the average marks for each skill involved in the test (by folder and by question). Details of the skills involved and the average for each can be accessed by clicking on “Legend”.

2. Statistics by folder

2.1 Overall analysis

When you select a folder, you will see the same elements as before, presenting the data for the folder in question.

2.2 Question analysis

At the bottom of the page is a table analysing each of the questions in the folder. There are several elements:

  • Success rate: The percentage of participants who answered the question correctly.
  • Sup: The success rate of the group of participants who obtained the best results over the whole test. This sample represents 30% of the total number of participants.
  • Inf: The success rate of the group of participants who obtained the worst results on the test as a whole. This sample represents 30% of the total number of participants.
  • Disc: This is a value calculated from the “Sup” and “Inf” values. The aim is to compare the success of the question between two groups.
  • Discrimination: Comment on the “Disc” column. If the gap is positive and large, discrimination is considered strong. If the difference is zero or negative, the discrimination is deemed insufficient.


The “Display numbers” button displays the number of users who correctly answered the question in relation to the group. In the example of MCQ #1, 15 of the 19 participants in the group with the best overall results in the test answered the question correctly (i.e. 80%). By the same token, 4 out of the 19 participants in the group with the lowest overall results answered the question correctly (i.e. 20%).
By clicking on the “Action” icon, it is possible to visualise more precisely the behaviour of users with regard to the proposals.

Did you find this article useful ?