This report compares the highest and lowest performance on each question. It selects the upper 27% and lower 27% of students who have selected each option in a question. (The percentage can be overridden in the cohort dropdown on the previous screen). A question's difficulty p is calculated as (average score)/(maximum achievable score). It ranges from 0.0 (nobody got it right) to 1.0 (everybody got it right). Values of p<0.25 or p>0.75 indicated questions with limited differentiating power.
The report calculates the discrimination d as the Pearson product-moment correlation coefficient between the student's overall score and their performance on the question itself, over the cohort of students, and takes the value -1.0 < d < 1.0. High correlations mean that the question is measuring the same thing as the test. Low correlations, d< 0.25, mean that the question is not related to a good test score.
Each report has a header section that lists basic statistics:
It then lists each question as they appear in the report headed by theme:
In this case, d is low for this question and so is not testing in the same way that the exam is. The examiner can choose to exclude the question by clicking the red pencil icon.
An orange rectangle indicates that standard setting is being applied to the question.
allows examiners to toggle the specific question's inclusion in the paper. Questions are included by default
Lists the overall F&D statistics for the paper.