Understanding Your Assessment Options: AI-Enhanced Comparative Judgement vs. Double Marker

At No More Marking, we leverage cutting-edge technology to reduce teacher workload while maintaining high assessment standards. If you are exploring our solutions, you might be wondering about the difference between our foundational AI-Enhanced Comparative Judgement (CJ) and our new Double Marker system.

While both use artificial intelligence to streamline marking, they are designed for different assessment scenarios and produce different types of data. Here is a guide to help you understand how each system works and how they differ.

1. AI-Enhanced Comparative Judgement (CJ)

The Foundation of Relative Assessment

What it is:

AI-Enhanced CJ is a specific assessment methodology that replaces traditional rubrics with a simple, comparative approach. Instead of looking at a piece of writing and assigning it a grade out of 10, human teachers and AI are shown two pieces of student work side-by-side and asked to decide which is "better."

How it works:

• The Hybrid Model: An AI is trained to mimic expert human decisions. The marking load is split (for example, 10% human and 90% AI), keeping teachers in the loop for quality control.
• The Maths: Millions of these relative, pairwise comparisons are fed into a sophisticated statistical model to produce a highly reliable, scaled score for each piece of writing.
• The Benefit: Humans (and AI) are naturally much better at comparing things than making absolute judgments. This method prevents AI "hallucinations," avoids the pitfalls of superficial grading, and anchors scores to established national standards.

2. Double Marker

The Complete Mock Exam Solution

What it is:

Double Marker is a comprehensive, hybrid assessment system that takes the underlying logic of AI-Enhanced CJ and applies it to traditional marking against a standard mark scheme. It is specifically designed to help departments run full-scale mock exams (currently optimised for GCSE English Language and Literature) while reducing the marking workload by about 90%.

How it works:

• Unified Booklets: Students take exams in physical booklets featuring unique barcodes, which are then scanned into the system.
• AI as First Marker: The AI evaluates the scanned exams against official exam board mark schemes, providing initial scores and detailed feedback.
• Targeted QA: Teachers only need to mark a small sample of scripts (typically 5–10 per cohort) to standardise the results. The system automatically flags any problematic scripts (like unusually short answers or tricky handwriting) for human review.

Key Differences at a Glance

While Double Marker utilises CJ logic under the hood, the user experience and the outputs are quite different. Here is how they compare across five key areas:

1. The Assessment Approach

• AI-Enhanced CJ relies purely on relative judgments ("Is Essay A better than Essay B?"). It entirely bypasses traditional rubrics.
• Double Marker yields absolute judgments ("This response gets 6/8 marks on Question 2"). It is designed to emulate traditional exam board marking.

2. The Final Output

• AI-Enhanced CJ provides a psychometrically valid, scaled score that shows where a student ranks and links their performance to a massive dataset of national standards.
• Double Marker provides granular, question-by-question marks, overall mock exam grades, and specific written feedback linked directly to the mark scheme criteria.

3. Workflow and Logistics

• AI-Enhanced CJ is highly flexible and typically used for assessing digital text or shorter pieces of writing throughout the school year.
• Double Marker is a specialized physical-to-digital workflow. It requires printing specific barcoded exam booklets, guillotining, and scanning them back into the system for processing.

4. Teacher Involvement

• AI-Enhanced CJ requires teachers to actively participate in the comparative judging process (completing their 10% share of the comparisons).
• Double Marker focuses teacher time purely on Quality Assurance (QA). Teachers mark a tiny representative sample to ensure consistency and review specific scripts that the AI flags for human attention.

5. Feedback and Insights

• AI-Enhanced CJ allows for deep statistical analysis of where human and AI judgments disagree, ensuring fairness and transparency.
• Double Marker focuses heavily on student and cohort feedback. It can even take audio notes recorded by a teacher and synthesize them into a "whole-class" feedback summary. Furthermore, schools retain complete control to set their own custom grade boundaries.

Which one is right for you?

• Choose AI-Enhanced CJ if you want to run highly reliable, standardized writing assessments across your school or trust to track progress over time without the constraints of a rigid rubric.
• Choose Double Marker if you are preparing for formal mock exams (like GCSE English), need traditional marks and student feedback, and want to drastically reduce the administrative and marking burden on your department.

Updated on: 22/05/2026