AI Marking for Teachers Achieves 0.88 Correlation for Edexcel's Post 1914: 40 Mark Question

"Can we really trust AI to mark English Literature essays?" It's the question that comes up time and again in our conversations with schools and teachers.

As such, we've performed comprehensive evaluations to demonstrate the accuracy of the Top Marks' English Literature AI marking tools really are. The results speak for themselves!

We're examining the performance on Edexcel English Literature -- specifically, the Post-1914: 40 Mark Question.

Edexcel makes available numerous exemplar essays for their exam papers and we've put our tool to the test using 78 of those very same exam board approved standardisation materials. These exemplars showcase a broad spectrum of answer quality. These exemplars are used for standardisation, showing teachers what responses at various levels look like.

We took 78 of these essays and ran them through our dedicated marking tool. Then we measured the correlation between the official marks the board awarded each essay, and the marks Top Marks AI assigned to those same essays.

We employed the Pearson correlation coefficient. In short:

• • A value of 1 would mean perfect correlation -- when one marker assigns a high score, the other always does too, and when one assigns a low score, the other always does too.
• • A value of 0 means no correlation whatsoever -- knowing one marker's score tells you nothing about what the other marker awarded.
• • Negative values would mean the markers systematically disagree - when one assigns high scores, the other assigns low scores.

For context, how do humans perform?

What sort of correlation do experienced human markers achieve when marking essays already marked by a lead examiner?

Cambridge Assessment conducted a rigorous study to measure precisely this. 200 GCSE English scripts - which had already been marked by a chief examiner - were sent to a team of experienced human markers. These experienced markers were not told what the chief examiner had given these scripts. Nor were they shown any annotations.

The Pearson correlation coefficient between the scores these experienced examiners gave and the chief examiner was just below 0.7. This indicated a positive correlation, though far from perfect. If you are interested, you can find the study here.

How did Top Marks AI perform?

The results showed Top Marks achieving a correlation of 0.88 -- an incredibly strong positive correlation that far outperforms the experienced human markers in the Cambridge study. (Top Marks AI was also not privy to the "correct marks" or any annotations).

Moreover, 71.79% of the marks we gave were within 4 marks of the grade given by the chief examiner.

Another interesting metric is the Mean Absolute Error, for which our system scored 3.13. On average, the AI differed from the board by 3.13 marks, which is comfortably within 4 marks. As a percentage, that's an average of 7.8% difference.

In contrast, in that same Cambridge study, experienced examiners marking a 40-mark question showed a Mean Absolute Error of 5.64 marks, that's a difference of 14.1%. These results highlight the exceptional accuracy of Top Marks AI compared to traditional marking practices.

We don't claim that Top Marks is infallible, but when it does get things wrong, just how bad is it? Well, let's turn to the Root Mean Square Error to find out. Root Mean Square Error (RMSE) is a measure of the severity of large errors. When you square the number 1, you still get 1, and when you square 2, you still only make a small jump to 4. But square 5, and you're suddenly all the way up at 25. That's how RMSE works - it (essentially!) highlights large errors by squaring them.

Top Marks AI's Root Mean Square Error was 4.38, meaning even when larger errors occur, they remain remarkably small relative to the 40-mark scale.

You can see the full side-by-side human and AI scores below.

Can I see a graph to help me visualise this?

Absolutely.

First, here's a scatter graph to show you what a theoretical perfect correlation of 1 would look like:

Now, let's look at the real-life graph, drawn from the data above:

On the horizontal axis, we have the grade given by the exam board. On the vertical, the grade given by Top Marks AI. The individual dots are the essays -- their position tells us both the mark given by the exam board and by Top Marks AI. You can see how closely it resembles the theoretical graph depicting perfect correlation.

Discover how Top Marks AI can revolutionise assessment in education. Contact us at info@topmarks.ai.