3.2 Standardised tests and Dyslexic Thinking

Standardised tests, S-A-T's, or SATs, measure standardised thinking with their emphasis on memorising information, assimilating knowledge, and relaying it in timed tests. And this sits firmly in our dyslexic challenge wheelhouse.

Our research found that 94% believe standardised tests and exams disadvantage dyslexic thinkers. We know AI is brilliant at assimilating knowledge, but it turns out it's great at relaying it in timed tests as well, doing it faster and better than most humans. AI was asked to take US SATs and scored in the top percent.

ChatGPT scored 700 to 800 in the SAT math, so in the top 10%. It scored 85% in the LSAT, the exam you take to get into law school. It came in the top 10% in the bar exam, the exam taken at the end of law school. And it scored 4 and 5 on the AP Calculus and AP Biology exams, 5 being the highest score possible. So machines can already ace the high-stakes tests that we currently use to measure and treasure the skills in young people.

In a world where humans and AI work side by side, it's the human ability to think differently that employers are placing value on. They are looking for the soft skills like analytical, creative, and innovative thinking that can take this assimilated knowledge to the next level. And that's exactly what dyslexic thinking can do.

Over time, educators will need to rethink how we assess learners to prepare them for the workplaces they will enter. And these changes should benefit dyslexic thinkers. But here are five ways we can support dyslexic thinkers in tests and assessments today.

One, offer a range of ways for them to show what they know in a strengths-led way. Don't rely solely on written information. Lean into their communication strengths.

Two, focus on more coursework and project-led assignments, and include these in your assessments wherever possible.

Three, acknowledge that standardised tests are difficult for dyslexics, but their dyslexic thinking skills are the skills the workplace needs, so they will lead to success in life.

Four, give every dyslexic thinker the adjustments they need to help level the playing field, like extra time and using technology and devices.

And, five, place more emphasis on preparing learners for life, with a 360 view of that child and their strengths, rather than the narrow filter of standardised assessments. Our education experts, Gavin and Josh, agree.

Through the work of Made by Dyslexia, it's becoming obvious what we need, and many workplaces are recognising the change that is needed. However, we still have an assessment system here in the UK which is typical of those around the world. There are many different countries, many different languages, but sadly, the system of standardised assessment for children before they leave formal education is more or less the same. And it hasn't changed much, and certainly not enough, over the last 100 years.

Those education standardised assessments typically require massive amounts of knowledge recall in a very set period of time. They place an unnecessary emphasis on grammar and spelling. And also, the way in which information is provided to those students can be confusing and often uses language which is obfuscatory or makes it difficult for children to access what the question actually is.

Now, what can we do to make that different? It's really straightforward. We can generate a system of assessment which assesses the skills that are important to us. We can acknowledge that through the work of Microsoft and others, there are plenty of tools out there that level the playing field for dyslexic learners. So why on Earth do we have examination systems which ignore the existence of those tools? It feels to me counterintuitive.

And also, we've spoken already about that amazing spiky profile that dyslexic learners will have. They will often be in the top percentile in some areas of intelligence and perhaps in the bottom percentiles in other areas of intelligence. And our education system typically rewards them based on the lowest aspect of that profile. Couldn't we have a system which rewarded them for the highest aspect of that profile instead and acknowledged that later in life, AI would probably take care of the lower one, because that education system therefore brands typically dyslexic learners as failures at that point of completion, whether that be at 16, 17, or 18.

Here in the UK, dyslexic learners achieve 50 percentile points' poorer examination results on average in English and mathematics than those without dyslexia. That isn't because they're 50% worse. That's because the examination system is failing them and is prejudiced against them.

And at Millfield, we've been able to show over years that we can level that playing field. And in actual fact, dyslexic learners at Millfield exceed national outcomes because we've been able to level the playing field. And that is possible with small changes in all schools.

Standardised tests have never been fair to dyslexic learners. They've never showcased what they know or what they can do. And now with AI, we have definitive proof that computers can do these things better than any human being. And so it's an opportunity for dyslexics to finally showcase their creativity, their big-picture thinking, their out-of-the-box notions, because that's exactly what AI cannot do.

And as educators, we need to think about, what do we actually want to instil in our students, what do we want them to be good at, and now, how do we assess that? And AI has proven that a standardised assessment is not the tool that we need.