Gongshang New Optical Marking for Exams
Print the official 45-question optical answer sheet for the assessment.
Load the authorised school roster from FlexiList so scanned Class + Index resolves to names, form classes and Teaching Groups.
GNOME can also read an existing FlexiList .xlsx locally. Teaching Groups are available only if those columns exist in that workbook.
After the school namelist loads, select one or more form classes and/or Teaching Groups. Overlaps are de-duplicated.
Roster files stay in this browser. The GNOME Namelist bookmark runs only inside your authenticated SCMobile tab and uses read-only same-origin FlexiList GETs.
Scan the response sheets on the Canon feeder, classify every generated PDF with Microsoft Purview, then load all PDF chunks together.
With the scan-bed down, place your stack of OMR responses into the feeder.
Log into the Canon printer.
Click on Main > Scan and Send > Send to Myself >
Select all scanner PDFs in Windows Explorer → Show more options → Apply sensitivity label with Microsoft Purview → OFFICIAL (CLOSED) → NON-SENSITIVE.
Select multiple PDFs in one action. If the Canon created another chunk, use Add more PDFs before processing.
Confirm what GNOME read from each script. The teacher is authoritative; any overwrite immediately becomes the value used downstream.
Set the question count, then enter or paste the answers and marks. The 45-question capacity is shown as three compact 15-question panels.
Paste directly from Excel if useful. Questions beyond the declared count are dimmed and ignored.
| Q# | Ans | Marks | Q# | Ans | Marks | Q# | Ans | Marks |
|---|
Run the marking pass, inspect the preliminary cohort and item-level results, then download the complete results pack.
Any material change to cohort, identity, responses, answer key or marks makes the run dirty; simply re-mark.
Includes Scripts PDF, Annotated PDF, Result Slips, Analysis.xlsx, Answer Key.xlsx and Session JSON in one ZIP.
Scripts [timestamp].pdfAnnotated [timestamp].pdfResult Slips [timestamp].pdfAnalysis [timestamp].xlsxAnswer Key [timestamp].xlsxSession [timestamp].jsonA quick-and-dirty guide to reading the Functional Index (Fi) and Discriminatory Index (Di) in the analysis workbook.
This is simply the proportion of students who gave the correct response.
In this example, about half the student made a mistake for Q3. Most of the students gave the correct answer for Q2, so Q2 is considered to be 'easy' for this group of students.
A well-balanced test should have a good mix of easy, moderate and hard questions.
This is a measure of how well the question can distinguish strong vs weak students.
DI is calculated by (proportion of high overall scorers getting this question correct) - (proportion of low overall scorers getting this question correct).
For e.g., if ALL the high-scorers got Q1 correct, and ALL the low-scorers got the question wrong, then DI for Q1 = 1 - 0 = 1.
A problematic extreme case can be when ALL the high-scorers got Q2 wrong, and ALL the low-scorers got the question correct, then DI for Q2 = 0 - 1 = negative 1.
The setters for this test should study if items Q6, Q7, Q9, Q12, Q14 are problematic.
This is meant as a quick-and-dirty guide. If you wish to go more in-depth, you should consider SEAB Certificate in Assessment (CEA) courses.