AI-Powered Assessment

Scoring the papers is not where the work ends: item difficulty, discrimination and distractor analysis measure the exam itself, not only the students. In most institutions the process closes when the sheets are scanned and the report cards go home, leaving the question of whether an item measured anything, or just confused everyone, to staffroom opinion. This is written for the people who set the paper and then have to explain the result to a parent: the assessment office, department heads and academic coordinators.

It fits inside the exam calendar you already run — your department still writes the paper, the sheets are still scanned, and the difference begins once reading is done. A department meeting then opens with a list of items to retire and outcomes to teach again, instead of an argument about which questions felt hard. The same data tracks progress by outcome through the term, puts a predicted LGS or YKS score and percentile band beside a student's mock results, and gives the guidance team a table to open in a parent meeting. Next term's exam plan is then built from those tables rather than from memory.