Examination Sets

Examination Sets

A set is one question paper within one examination. If your candidates all sat the same paper, you have one set and can mostly ignore this page. If they sat different papers — by post, subject or category — the division runs through attendance, answer keys and reporting, and getting it right at the start saves a great deal later.

Where a set shows up

Stage	What is per-set
Project	The sets are defined here: a code and a description each
Template	The sheet carries a Set region, so recognition reads which paper a candidate sat
Scan	Each sheet's recognised set code appears in the table
Attendance	One attendance workbook per set
Answer Key	One verified key per set
Results	Marks are computed against the candidate's own set's key
Reports	One report per set, built on that set's own attendance workbook

Defining them

Application menu → File → Project Configuration…

Each set has:

• a code — the machine-readable key, which must match what is printed on the answer sheets, because that is what recognition reads;
• a description — for people.

Sets are persistent and uniquely identified, so renaming a description does not break anything already keyed to the set. A set that something references cannot be deleted; OMRFlow says what is blocking it.

Per-set attendance

Each set gets its own attendance workbook, on the Attendance stage. That is not merely organisational: candidates who sat different papers are different populations, with their own rosters, their own keys and their own merit orders. A single combined roster would make "rank" meaningless.

The workbook you import for a set also becomes that set's result template — see Results & Reports.

Set-aware reporting

Generate All Sets on the Reports stage produces one workbook per set, each built from that set's own template and roster. A set missing either is refused by name, rather than generating something incomplete.

Status

This is a cross-phase enhancement delivered in two parts, at different maturities:

Part	Status
1 — Project configuration (examination name, sets)	✅ Implemented and tested
2 — Per-set attendance and set-aware reporting	🟡 Implemented; synthetic testing only

Part 2 awaits real-data validation: no real attendance workbook and no real scanned cohort has been processed end to end. It is the least proven part of OMRFlow. See Known Limitations and the Development Roadmap.

Related

• Projects
• Attendance
• Results & Reports