Switch course ▾

5.1 Understanding How Students Learn 5.2 Lesson Planning Fundamentals 5.3 Classroom Management 5.4 Student Engagement 5.5 Assessment Fundamentals 5.6 Differentiated Instruction 5.7 Communication with Students & Parents 5.8 Digital Teaching Fundamentals 5.9 AI Literacy for Teachers 5.10 Reflective Teaching Practice 6.1 Multilingual Classrooms: Uzbek, Russian & English Instruction 6.2 Teaching Large Classes Effectively 6.3 Early Literacy & Numeracy Foundations 6.4 Classroom Technology Basics 6.5 Working with Parents in a Multicultural Community 6.6 Wellbeing & Stress Management for New Teachers Programme (64) →

Session 4. Fair and Reliable Marking

Objective. Name the main sources of bias and inconsistency in marking and use simple routines to bring marks closer together.

Why this matters

A mark is a promise: the pupil, the family and the next teacher all believe it says something true about the work. If the same work would have received a very different mark from a colleague, or from you on a tired evening, the promise is weak. No teacher marks perfectly, and that is not the point. This session helps you measure how close your marks really are and shows low-cost ways to bring them closer.

Key ideas

Valid, reliable, fair

Valid: the assessment measures what you meant it to measure. A geography answer should not lose marks for handwriting.
Reliable: the same work gets a similar mark from different markers and on different days.
Fair: every pupil has an equal chance to show what they know, in the language, context and conditions of the task.

Where bias creeps in

The halo effect: a neat, long answer or a strong pupil's name makes us read generously. Order and tiredness matter too: the last scripts in a pile often get a different standard from the first.
Teachers may also be influenced by a pupil's gender, family background or how they behave. A study in which university teachers in Iran reflected on their own marking found they reported such biases and suggested training and self- and peer-assessment.
Naming these risks does not mean you are a bad teacher. It means you can design a routine against them.

Routines that make marks more reliable

Write or fix the mark scheme before marking, and decide in advance what counts for each mark.
Cover names, mark one question across all scripts, and shuffle the pile now and then.
Re-mark five scripts after a break. If two marks differ by more than one point, your scheme needs clearer words.

Moderation and fair conditions

With a colleague, mark three shared scripts alone, then compare. Agree on one anchor script for each level and keep it for next time.
Check conditions too: the same time and help for everyone, instructions in the pupils' language of instruction, and contexts all pupils know.
Plan reasonable adjustments in advance, such as extra time or large print for a pupil who needs them, so the test still measures the same outcome.

Steps toward consistent marking

Private judgementEach marker decides alone

Written schemePoints fixed before marking

Careful routinesCover names, one question at a time

Shared standardAnchor scripts, colleague check

Each level adds a little time and a lot of consistency.

Consistency is checked by comparison

You cannot see your own reliability by looking at your marks alone. You see it only when you compare: with a colleague on the same scripts, or with yourself a week later. A regulator's review of teacher marking in England noted that internal standardisation in schools is often weak, and that methods such as exemplar work, training and agreement trials are used although their effectiveness has been little tested.

That honesty is useful. The goal is not perfect agreement but a smaller spread. A realistic target for a 6-mark question is that two careful markers differ by no more than one mark on most scripts. When they differ by three, the scheme words, not the markers, are usually the problem.

Fairness starts before marking

A fair mark needs a fair task. Check whether the instructions are in the language the pupils learn in and use words they have met in lessons. Check whether every pupil has had the chance to learn the content: a test on a topic you skipped for half the class is not a fair measure of that pupil.

Check the conditions as well. If one class hears a listening passage read quickly and another slowly, or one class has 40 minutes and another 30, the marks differ because of conditions. Write a short procedure for each test: time, help allowed, how instructions are read, and what adjustments are available, and use the same one every time.

Examples from the classroom

Three teachers, one essay

Scene. Three Grade 6 teachers in Samarkand each mark the same short story on a 12-point scheme. Their marks are 5, 8 and 10. They talk through the three marks and discover that 'uses interesting vocabulary' means very different things to each. They replace it with 'uses at least five words from the unit list' and mark three more essays. The spread falls to within two points.

Insight. A shared look at the same work finds the unclear word in the scheme.

Names under a sticky note

Scene. Aziz covers the names on 30 Grade 8 composition scripts in Fergana, marks them, then marks five again the next morning. Two marks differ by one point because he was tired at the end of the pile. He now marks the longest answers in the morning, and starts the pile from the opposite end every other time.

Insight. Re-marking a few scripts after a break shows the drift in your own standard.

The same passage, read twice

Scene. Dilnoza reads a listening passage to her Grade 3 class and her colleague reads the same passage to hers, faster and only once. Dilnoza's class scores higher. They write a short procedure: the passage is read twice, at the same speed, with a 10-second pause between questions. The next test has similar averages in both classes.

Insight. Equal conditions are part of fair marking: otherwise the scores measure the reader, not the pupils.

Common pitfalls

Believing that a long experience means your marks are automatically consistent.
Marking all the scripts of one pupil together, so a strong first answer lifts the others.
Letting some pupils redo or get hints while others do not, without a clear rule.
Leaving the scheme words as 'good' or 'enough' instead of describing what the page must contain.

Hands-on activity · ~50 min

Mark the Same Scripts: How Far Apart Are We?

Alone (10 min): mark the three pupil scripts in your Activity Kit out of 6 with the mark scheme. Do not talk. Write your three marks on a slip.
Plot the marks (10 min): the facilitator draws a line for each script on the board and each participant adds a dot. Look at the spread.
In groups of four (15 min): explain your marks, find the word in the scheme that caused the difference, and rewrite that word. Agree on one anchor mark per script.
Plan (15 min): choose three routines from the list in the Kit and write when you will use each in the next two weeks.

Your facilitator will guide you through this activity in the session.

Questions to discuss

How big was the spread on each script, and which one surprised you?
Which words in the scheme caused disagreement?
Which routine can you use without extra time?

Reflect

When was I last surprised by how a pupil's mark changed when I re-read the work?
Which of my tests might be unfair to some pupils because of language or context?
Which colleague could I compare marks with this month?

Try it tomorrow

Pick five scripts, mark them, and mark them again tomorrow without looking at the first marks. Count how many changed.
Ask a colleague to mark one question from your next test with you and compare.
Write a five-line procedure for your next test: time, help, reading of instructions, adjustments.