14
Examination of witnesses
Professor Jane Clarke and Lord Tarassenko.
Q134 The Chair: Good morning, everybody, and welcome to our two witnesses for this second session today. We are very grateful for you joining us in these difficult weather conditions, and we are looking forward to hearing from you. Could you just start by giving a very short précis of your professional background as it is relevant to this?
Professor Jane Clarke: Thank you so much for inviting me to come to this session. Why I am here is that I am chair of the Royal Society Advisory Committee on Mathematics Education. As you know, the Royal Society is our premium academic scientific body dedicated to promoting excellence in science for the benefit of humanity. The Royal Society is doing a lot of work on educational policy, trying to look at how our education system feeds into delivering this excellence in science for the benefit of humanity, not just to produce great scientists and, of course, the whole ecosystem of technicians and technical supporters, but also to have a scientifically educated public who understand and can use the science that comes out of that.
Interestingly, the Royal Society has divided off mathematical education as a special topic, really because we recognise the foundational position of mathematical and data education. It underpins everything that the Royal Society wants to do. Without this foundational mathematical and data education, we know that we will not get great science or many of the other areas of economic and technological work required for the benefit of our economy. We will not produce active citizens who are able to take advantage of, use and benefit from that. The society produced a new approach to mathematical and data education, a lot of which I will be referring to, because we have addressed that in there.
I am not a mathematician myself, and it might seem really weird that the Royal Society would not put a mathematician in charge of mathematical education. I spent the first 13 years of my professional life as a science teacher teaching in comprehensive schools in Leicestershire, in the London boroughs of Haringey and Enfield, and in Essex. I am one of the few fellows of the Royal Society who has taught mathematics, because an awful lot of mathematics is taught by non-maths specialists.
I started my PhD at 40. I am a chemist, but I worked in the field of biomedicine, so I am also a fellow of the Academy of Medical Sciences. Maths and data underpin the research that we do in so many areas of endeavour, including biomedicine. The fact that the Royal Society chose to put me there represents the attitude that the Royal Society has to mathematical and data education as being so fundamental to everything we do.
Lord Tarassenko: I am not a mathematician either, so we have two expert witnesses who are not mathematicians. I am an electrical engineer, and there is quite a lot of mathematics in electrical engineering. I have been a professor at the University of Oxford for the last 38 years, and chair of electrical engineering for 22 years. My research throughout those 38 years has been in what is now called machine learning. When I started, it was not called machine learning. It was a niche subject. It did require some mathematics. Of course, we all know what happened to that subject in the last three decades.
I am here because I was asked to chair the Maths Horizons project following a national curriculum review led by Professor Becky Francis. She had a huge task of looking at all GCSEs and all A-levels in just over one year. I was asked to chair a review into the teaching of mathematics in our schools, which was called the Maths Horizons project, or maths in the age of AI.
We made seven recommendations, and I believe five of them, in one shape or another, ended up in the Becky Francis report. As a result of that, I got much more interested in mathematics education, where it works and where it may not work so well, and, for those at school, what the implications are for later life if it has not worked so well.
Q135 Baroness Shawcross-Wolfson: We know that, across England, maths skills have got better if you measure them by PISA scores. We know that maths A-level is now, and has been for quite a while, the most popular A-level, and yet we still hear, particularly from employers, about a shortage of mathematical, numerical and analytical skills. I wondered whether you could say a little bit, perhaps starting with Professor Clarke, about where that gap is coming in.
Professor Jane Clarke: In terms of the problem with our current education system, we can define mathematical and data education in three chunks, really. There is foundational mathematics and advanced mathematics. The UK is very good at advanced mathematics, and we have more and more students doing A-level maths, which is fantastic, but we also have something that we call general quantitative literacy, which I think you call numeracy. That is the ability of people to use mathematical and data skills and knowledge to solve everyday problems.
I hope that what we have seen in the curriculum review is addressing the way that foundation mathematics is taught, and they are making recommendations about that moving more slowly. In the whole of that report, there was nothing about data literacy, and yet numeracy involves understanding data. This is how our population meets mathematics in their everyday, and there is virtually nothing in the curriculum review about that.
What we want to see as part of our recommendations here is general quantitative literacy and data literacy embedded across the curriculum. It is in geography lessons, science lessons, and PE and technical lessons that we learn to apply mathematics in a problem-solving way. It is through that that we will find that our children develop a confidence. If we explicitly say, “You are using maths here”, we can develop a confidence and a competence in using mathematics and applying it to the everyday.
For example, one of the problems that we have—and I am a scientist—is that the amount of practical work happening in schools is dropping like a stone. From 44% of GCSE students doing practicals every fortnight, we are now at 25%. Students are now more likely to watch a video of a practical, or see the teacher do a practical, than to do it themselves.
It is in doing this practical work that students learn how to collect data, what good data looks like, how to keep it there, and how to analyse it. These are key skills that we are not addressing because we are putting maths in a little pot and saying, “How many students do GCSE maths and how well do they do?” as this entry qualification, rather than looking at what maths is good for. That is a starter for 10.
Lord Tarassenko: I would not disagree with any of that, but I will not repeat it either. There is no doubt that we produce good mathematicians in this country, but we do seem to suffer from a problem. Of course, there is going to be a distribution of abilities in any school. We seem to do reasonably well at the top end and okay in the middle, but we have a big tail.
I tried to understand that as much as possible once I got involved in the Maths Horizons project and, indeed, with Baroness Bull when discussing the issues of dyscalculia and so on. I do not think it is just dyscalculia. It is almost a cultural thing that does not exist in other European countries, in that you can be labelled very early on in this country as not being good at maths. If you have that label, it is very difficult to overcome it. You probably need a very dedicated teacher who takes an interest in you and enables you to get to maybe the median level of the class.
For too many people—and it could be the family they come from—there is this culture that we have. Some people are quite proud of it. You get people quite high up in the Civil Service telling you, “I have had a brilliant career, but I was never any good at maths”, and they are almost proud of it. A cultural shift needs to be done. Everybody can be good at maths if taught properly.
It is very difficult, when you teach, to keep the very bright ones interested, as well as thinking about what you do with those who are, say, in the lower quartiles, so most teachers will teach to the median level. I do not advocate AI as the ultimate panacea in this, but at least my understanding of AI tutors is that you would have some tools that allow somebody who is struggling with particular aspects to be brought to the median level so that they could be fully integrated with the class teaching. That is where I see the help of AI in lifting people who might need dedicated coaching, as it were, and to bring them to the median level of the class, so that they can be fully integrated in the teaching of that class.
We need to act early, because, if you have this label of not being good at maths at age five or six, the correlation between being given that label and failing GCSE maths, or not getting a grade 4, is extremely high. We need to stop it very early, identify those who, for whatever reasons, may be struggling, and give them the help. You should be able to teach the whole class at the median level so that the people who are very bright can thrive at the same time without the teacher having to spend so much time on trying to bring along the students who are in the lower quartiles.
The Chair: What age is early? Is that primary or key stage 3?
Lord Tarassenko: It is as early as possible, because it is very hard to lose labels. I have known all my life, for example, the myth that Oxford was a place where you went to read humanities; if you wanted to read sciences, you went to Cambridge. That is a myth. If you look at the performance in engineering or computer science, it has probably been better in Oxford than Cambridge. The whole country believes that, if you are really good at humanities, you should go to Oxford and, if you are really good at sciences, you go to Cambridge. Myths persist for a very long time, and it is the same when a child is labelled as not being good at maths. We need to intervene as early as possible in primary school.
Professor Jane Clarke: The evidence that we have suggests that the early years are critical as well. We are talking about preschool mathematics and the work that is done in nurseries. The evidence is clear that it is that early on that the beginnings of mathematical numeracy absolutely begin. We have evidence to show that it starts from very early on.
Viscount Stansgate: Is there evidence that you have been able to provide to us on the committee about the very early preschool years?
Professor Jane Clarke: I will certainly make sure that we send it through to you as soon as possible.
Q136 Baroness Alexander of Cleveden: My question was about the purpose of numeracy today. Professor Clarke, you have touched already on whether our approach to numeracy education is fit for purpose in a world shaped by AI, data and digital technologies. Can I ask you just to expand a little bit on where in the curriculum we try to make sure that that focus on data, AI and digital technologies features? Is it in the maths curriculum or is it that we have to find space in other subject domain knowledge?
My second question is about how we might adapt the maths GCSE so that it fulfils that function of stretching the most able children, but also can act as a gateway qualification for those who perhaps, at the moment, are struggling.
Professor Jane Clarke: To your first question, the work that we are doing suggests that fundamental mathematics should be taught in a mathematics class, but there is then a responsibility to go across the curriculum to show that you can use that mathematics and gain mathematical competencies by using the mathematics elsewhere. It makes mathematics seem real. It makes mathematics seem of intrinsic value.
We are doing some projects at the moment to get schools and maths specialists working with other curriculum specialists, because there is a problem with teacher support here. If you are expecting somebody in a science class to embed the mathematics that they have learned in maths, they need to know exactly what the mathematics learned is. They need to use the same language, because, often, one term is used in maths and a different one in geography, so how would students know that they are talking about the same thing—for example, “mean” and “average”? There is a really big issue here about teacher support, teacher training, and teachers’ continuing professional development.
We are very strongly of the opinion that we need to look at the mathematical journey from 14 to 18 as a continuous one. The problem with GCSE maths is that it is a gatekeeper. If you have grade 4, you have passed, and all these doors are open to you, even though, on the higher paper, you can get a grade 4 with 25%, which means that you really do not have great maths, and we do not exactly know which competencies you have in that 25%. If you do not have it, you have to repeat it again and again. We have done some work—and we can talk about this—on the gateway and the progress from there.
What we would like to see is building what we call domain-specific competencies, or general mathematical competencies, into the vocational courses and what goes on again. There is the foundational maths on which all of this is built, and which I am arguing should be in a maths lesson. Then there is this general quantitative literacy, which I am arguing should be built up through a child’s experiences in school across the board. When we start to move on to vocational courses, we need to make sure that these have the mathematical competencies built in them that are specific for whatever this vocational course is.
We have produced a list of general mathematical competencies that were used when T-levels were being developed and have been incorporated into T-levels. A T-level in health and social care will have domain-specific mathematical competencies that you need to go in there, which will be different for those people doing an apprenticeship in engineering. We are working with some people in the Department for Education to bring these general mathematical competencies to the design of the new V-levels and that sort of thing. These things are built in.
As to GCSE, we see it as a gateway, if you like, and having a transactional value but not necessarily being of a useful value. We would like to see some kind of assessment done, where a child going through school can build a portfolio of mathematical competencies. Imagine how much more confidence that would bring to somebody if they go, “I have shown in my science lessons that I can collect and analyse data. I have a competency in that. I have shown that I understand ratio”, and these sorts of things.
That will give them confidence in what they can do, instead of spitting out a third of our students and saying, “You have failed maths”. If you are on free school meals, 50% of you will end up failing GCSEs. It is a terrible thing to do to our young people. Maths is so foundational, and it would be possible to build in assessments based on these practical applications of mathematics, so that a child leaves education with a portfolio of, “These are things I can do in mathematics. I can use a spreadsheet to analyse data”.
This is what we would like to see, because, if we want a numerate population, we have to have a population who are confident in using mathematics to analyse the world. The fact that we have measles back is telling you that our people do not understand risk.
Lord Tarassenko: One has to try to think of what the levers are in order to effect change. As you may know, I was involved, after the Maths Horizons project, with Professor Simon Peyton Jones of the University of Cambridge, in proposing a level 3 qualification in data science and AI. At the moment, we are working that up in a project with the Raspberry Pi Foundation, funded by DfE.
Following the curriculum review, there are some various working groups that are going to report in 2028. Changes will be made at the earliest in 2029, so there is a very long time lag. It is very clear that perhaps the key high-level recommendation from the Becky Francis review is that it should be evolution rather than revolution, so what is going to be possible?
From my limited experience, it seems to me that where we really have the greatest levers is between 16 and 18 or 19. Therefore, we should think about what we do for students between those ages. Some of them will be doing V-levels and T-levels. Some will be doing A-levels. Given what happened to them from the ages of five to 16, studying mathematics, and then sitting GCSE at age 16, how can we optimise the course of what they study between five and 16 in order to allow them to thrive in these different categories and groups?
I do not think that we can redesign the GCSE syllabus from scratch in order to try to really have a big impact. The conclusion that we came to in Maths Horizons is that, while there is AI and so on, you need to put it somewhere else, because what is far more important is for people to develop what you called the mathematical skills that enable them to thrive in everyday life. If it could be integrated with physics, biology or geography, that would be brilliant.
I am not quite sure how we get schools to do that, given how stretched teachers are, but it would be wonderful if teachers were able to have that kind of integration, so that you can apply a topic that you have learned in mathematics. It may be biology first, because there is a lot of mathematics underpinning modern biology, or it may be in geography, given that geography relies these days on very modern techniques such as earth observation from satellites. This integration would be wonderful. How you implement it is a different question.
Given those challenges of implementation, how can we optimise the teaching of maths between five and 16, so that people can then go into these different subgroups and really have their mathematical numeracy education optimised, depending on which group they are? My view is that everybody should be doing something on mathematics, data science and AI between the ages of 16 and 18.
If you are somebody who has, sadly, not been able to get a grade 4 at GCSE, is it really the right thing to keep asking you to re-sit every six months until the age of 19? It probably is not. Therefore, maybe allow them one re-sit, because there may have been circumstances in June, but think about those vocational qualifications. There is a focus on numeracy, but I want numeracy to be slightly wider than it is in the GCSE maths syllabus. It is one of six topics, but there are at least two others that are very important for vocational qualifications.
One is called ratio, proportion and rates of change, which are very important for students to understand, even if they do not want to be mathematicians—for example, the differences between percentages and ratios. I know that it was an issue during Covid when our Prime Minister did not really understand the difference between ratios and percentages, so that should not happen. That particular element of the GCSE maths syllabus should be added to the numeracy topic.
Even more important than when Thomas Bayes was around a couple of centuries ago is probability. Probability underpins so much of everyday life these days. Therefore, within vocational qualifications, there should be probability. So that is a qualification for those for whom it may not be useful to keep re-sitting maths GCSE.
Then there is the qualification for those who are going to go on to study medicine, who are going to be lawyers, who are going to be doing economics, and so on. That is where we want to pitch data science and AI. It is not for those who are going to read computer science or engineering at university, but for those who are going to be professionals in their career. What should they understand about this new technology of generative AI in order to thrive at university and then in their careers?
For those who are really at the top end of the distribution and are doing maths and physics, there needs to be more provision of further maths in schools that do not do it. That is how I would tackle the various groups across the ability spectrum.
The Chair: We are going to run out of time. Can I ask you to be a bit more succinct? We have only about four minutes for each question from now on.
Q137 Baroness Garden of Frognal: What you have to say is really interesting. We had a committee on 11-to-16 education here a few years ago, and our strong recommendation was that maths should not focus on algebra and geometry, which I used to love but have never used in later life. Instead, it should cover financial literacy and things that kids could see had a relevance to later life. It was a Conservative Government; they threw out all our proposals, but let us not focus on that. What would it take now to change numeracy teaching from the focus on speed and procedure to things that have a real relevance to everyday life? How would we do that? What barriers would there be to changing the syllabus in that way?
Professor Jane Clarke: We would argue that incorporating this across the curriculum is what we need to do. The foundational maths needs to be there in the maths lessons, and we then need to incorporate that understanding of mathematics across. As a really good example, there has been a proposal for a natural history GCSE. Whatever you think about that, what is at the end of it—and it is out for consultation—is a whole list of the mathematical content of that GCSE. We could easily do that. The GCSE programmes are under consideration. We could recommend looking at the mathematical and data content of the GCSEs explicitly.
Baroness Hamwee: How do we get from here to there?
Professor Jane Clarke: The GCSE syllabuses are going to be rewritten really soon. We could make strong recommendations.
Baroness Hamwee: What about the training of teachers?
Professor Jane Clarke: We are doing some projects at the moment through ACME to work with some of the big MATs, which are very enthusiastic about this. They do not have enough maths teachers, but they can see that they could work together to get the maths specialists and leads working with other curriculum leads to support the development of maths in other areas. We desperately need more maths teachers. We are never going to get enough mathematicians, but we can work, through continual professional development of people like me, a biology and chemistry teacher, to deliver really good maths lessons, given the right professional development in the right areas.
Lord Tarassenko: Given the fact that we want to cover as many questions as possible, I am happy to leave it there.
The Chair: It is all fascinating stuff. I am sorry to railroad it through, but I then get told off if I let the sessions overrun, so I am between a rock and a hard place.
Q138 Lord Massey of Hampstead: AI has come up. Lord Tarassenko, you mentioned it in terms of potentially improving, in a rather bespoke way, the people who are not getting to the basic or median level that one needs to get to. Your premise was that everyone can get there if they are taught properly. That is a very important premise, because, if we take that, there could be real potential here for AI to bespoke teach individuals who are not getting to that level. What are your feelings, both of you, about AI and how we can integrate these tools to also improve people’s mental numeracy abilities? As well as teaching them, will it help them in later life to intellectually improve their mental maths?
Lord Tarassenko: I mentioned AI tutors. It is an interesting philosophical question, because, if you improve everybody to the median, the median shifts, but we will not go into that. The whole distribution will shift, which is fine. You will always have a distribution. The median will now be higher, and the bottom part of the distribution will be at a level that will ensure that everybody passes GCSE. I am very much in favour of that, but that is an AI tutor to help develop mathematical reasoning and understanding. It is not about understanding AI. That is quite different. That comes with data science and AI. That comes post 16.
The reason why I have put such an emphasis on probability is that, unless you understand probability, you will not understand why a large language model occasionally gives you the wrong answer. As I am sure every member of the committee knows, with a large language model, you get the wrong answer about 10% of the time from Gemini, which runs Google’s AI overviews, for three reasons.
First, the prediction of the next token or word is probabilistic. It is a menu of choices of probability. Secondly, there is conflicting data in the training dataset. Thirdly, because of the way these models are trained, they tend to be sycophantic and try to agree with you. Everybody should understand that, but, in order to understand that, you need to understand probability. That is why I place such importance on it. I do not know whether you have it within your numeracy definition, but I would definitely have probability.
I want to distinguish between AI to help you develop the relevant mathematical skills, and then, differently, post 16, understanding how AI itself works. That is not necessarily for everybody, although I do think there should be a basic level of understanding, whether it is the vocational courses, or whether it is the level 3 qualification in data science and AI, which, by the way, we are developing as an A-level for 2029.
My concern about that is that it would almost universally be taken as a fourth A-level. Many schools find it very difficult to put on four A-levels, so it may never feature in certain schools, which would be a pity. Therefore, we are trying to adapt it to making it also an extended project qualification—EPQ—which is half an A-level. Even if you are at a school that offers only three A-levels, you could do the EPQ in data science and AI, and we are piloting that this September. At the moment, I am working with the Raspberry Pi Foundation on designing the syllabus for that. That is about understanding AI.
I do not want to go on for too long, but there is one important factor that we need to start thinking about, which is that the overuse of AI diminishes cognitive skills. Therefore, even though we are going to be introducing AI in tutors and so on—without a shadow of doubt, it will feature in our schools in five years’ time—we also have to be concerned by the fact that the striving to understand how to solve a maths problem, and the getting it wrong four or five times, means that, by the time you get it right the sixth time, it is really embedded in your cognitive understanding of mathematics.
One of the dangers of AI is that you remove that. You go straight to giving the answer. Again, I am not an expert. I work with a few people in this domain, but I can only talk to the people I work with. I work with Isaac Maths in Cambridge, and the Raspberry Pi Foundation. Their whole teaching is based on the use of hints. It is technology. It may have AI underpinning it. It is some form of tutoring delivered via either a smartphone or a laptop. It gives you a hint around the solution, and that is going to be very important. We have to start to think about how we use AI to help people become better mathematicians. It is not about giving them the answer, but giving them hints.
What is very interesting about Isaac Maths—I was talking to the professor in Cambridge who is behind it, Alastair Beresford—is that, if you design it properly, the number of times the students use hints decreases as they go through the course. As they build confidence, they find that they need to use the hints fewer and fewer times. That is the kind of thing that we really need to start thinking about as we make these tools available, so that they do not become just the source of the answer.
Q139 Baroness Bull: You have led on really well to what I wanted to ask about, which is about the potential on civic engagement and effective participation in democratic processes in this age of data, misinformation, statistical claims and, indeed, as you have already said, Lord Tarassenko, people not having the inbuilt number sense, as I might call it, or probability to see when they have snake oil in front of them rather than something that is legitimate.
Perhaps I will come first to you as you led on to it. Is this something that you are looking at and thinking about? What might we do to offset the risk of a lack of numeracy impacting people’s ability to effectively engage in democratic processes and civic life?
Lord Tarassenko: I am going to join two dots, if I may, at the start of my answer. Sitting on the Communications and Digital Select Committee, we have been thinking about the BBC charter renewal. There is this wonderful aspect to the BBC, which is really important in this age, which is BBC Verify. These fact-checking organisations are becoming more and more important.
As you say, Baroness Bull, some of the facts have a mathematical basis. You do not really understand the facts unless you have the right understanding of numbers and especially probability, and so it is very important that these fact-checking organisations do not shy away from giving the numerical data that underpins the evidence that they are using.
I have heard them talk about triangulation, for example. Do the viewers of the BBC understand what they mean when they are triangulating in order to verify what a location might be? That is fairly simple mathematics. If you are a sailor, you would understand what that is, but not necessarily everybody watching BBC Verify is a sailor. Those are the kinds of concepts that are explained when they are trying to say, “We believe that this particular footage comes from this location because we are able to triangulate from the following points”. These are the concepts that underpin understanding and fact checking, and especially probability.
One of my heroes in this space is Professor David Spiegelhalter, who was the Winton professor of public understanding of risk, and very often on our TV sets during Covid. He talks about how to explain probability. It is very important that you should do that by defining the baseline explicitly.
This is just a very simple example, but it really brings it across. “This medical treatment increases your stroke risk by 50%”. It is very hard for people to understand what that means unless you say, “Out of 100 people who are not taking this drug, two will have a stroke. Out of 100 people who take this drug, three will have a stroke”.
Even this communication of what the probability means is very important, so we need experts such as Professor David Spiegelhalter to make sure that the organisations that do fact checking and try to minimise the level of misinformation that is available do so in a way that is understandable by as many people in the general public as possible.
Baroness Bull: Do you think that we will always have to rely on a fact checker or that we should increase our own skills for at least sniffing that something is wrong, perhaps?
Lord Tarassenko: We need both.
Professor Jane Clarke: I will leave with you what the Royal Society wrote about needing data-literate citizens, which may be of value. I want to make a bit of a point that it is through getting children in school to do this data science and the use of data that they learn these skills. If you are asked to design an experiment to do something or to collect data, what is the relevant data? How confident are you in the data?
It is through encouraging children, in schools, to collect, use, collate and work out the validity of data—this data circle—that we are teaching children to do it. Fundamentally, if we present it as just a maths problem, we cut people off, whereas, if we enable people to do that, we give them the confidence that they have those skills to understand and answer those questions. I used the word “risk”. You used the word “probability”. This is different language meaning the same, but I do think it is important.
Coming back to the last question, it is very important to navigate the difference between data literacy, which is the ability to read, understand, use and communicate with data—that is the question you are asking, essentially—versus digital skills, which is the ability to consume, create and communicate with digital stuff. They are very different.
The curriculum review talked a lot about gaining digital skills, and nothing about data literacy, which, to us, fundamentally underpins the importance of numeracy.
Baroness Bull: We had that in our briefing notes. That was raised, and I am glad you put it on the record today. Thank you.
Q140 Lord Hannett of Everton: I heard the discussion earlier, where you mentioned culture as an important factor, and you have just covered the word “confidence”. How can numeracy teaching across all age groups better address the issues of confidence, belonging, and a perceived difficulty that prevent many people, especially those who are disadvantaged—and I emphasise disadvantaged—or difficult-to-reach groups from engaging and, ultimately, succeeding?
Professor Jane Clarke: It is such an important question. We know that, across children with the same level of deprivation in the north-east and in London, those in the north-east do significantly worse at GCSEs than those in London. That is partly to do with the educational landscape in London that has been worked on significantly.
It is also, I would guess, due to the culture in a post-industrial northern town compared to, maybe, a population in London that will have a higher proportion of immigrants from different communities and cultures there. I do not know, but it is really important. This is why we want to aim our education at giving people confidence. That is our proposal to have this gain of qualification portfolio. We think that that would help towards it.
It also speaks to the teachers. You are far less likely to be taught maths by a mathematician if you are in the north-east than if you are in London. It also speaks to the quality of teachers, and the same for physicists, biophysicists, chemists and biochemists. It also speaks to these areas of deprivation. You need great teachers who are supported and given the confidence and the tools to do that. Some of these differences are there.
We need to work harder to support from early years. The evidence was that Sure Start raised GCSE marks. There is absolute evidence that children who went through Sure Start got higher GCSEs. We now have the data, so let us go in early. Let us get in early. Let us support early.
Primary schools are really important because it is the same teacher, and many primary school teachers are not confident in their own ability to do maths. How do we support them to become great maths teachers? It is to help them see how many ways they use mathematics themselves.
We need to think about going in early and making sure that whoever is delivering this has the confidence and the support needed to support the mathematical learning. Maybe it involves parents too. If you have an adult who is not confident in doing mathematics, they are not going to be confident in supporting their children to learn mathematics and mathematical skills. I think you are seeing Sam Sims next week, and I am sure there is a role here for supporting adult numeracy, particularly for parents of children.
Lord Tarassenko: I would add role models to that. I have talked about sport on a number of occasions. I remember, for example, going to give a talk at Old Trafford about engineering. Of course, mathematics underpins engineering. You get children who come to that sort of talk who would not go to a talk about maths or engineering at school.
You could take other examples. Lewis Hamilton is a fellow at the Royal Academy of Engineering now. He comes from a background that you would not normally associate with mathematics, but, in a Formula 1 operation, with the cars, the algorithms that they use, and the way that they map out every individual circuit, there are huge amounts of mathematics.
You need to expose kids to those role models early and see what things they are interested in. It may be Formula 1. It may be football. Footballers attract computation of what is called “expected goals” now. It is all mathematical. Go in at a very early age and talk to them about things that excite them.
I am sorry that this is slightly biased from my own knowledge. I was director of a football club for a short time, so I am keen on that. Being a fellow of the Royal Academy of Engineering, I do know Lewis Hamilton, so it is biased knowledge. I realise, of course, that I am covering only a percentage of the population, but there are others.
I do not know how many of you have read the biography of Demis Hassabis. He started from a disadvantaged background. If you can understand that people such as that are able to do wonderful things, those labels can be removed. I agree entirely with what Professor Clarke has said, but we should do more about having role models who children at an early age can identify with, and getting them to understand that there is a lot of mathematics, engineering and science underpinning this.
Of course, nowadays, because of videos and TikTok, we do not have to be physically in Old Trafford or the Emirates—other stadia are available. They can see it on short little videos on TikTok. That is really important because there is still this image that, if you are going to be a mathematician, you are going to be somebody who does not play football and is not interested in the kinds of activities that normal kids are, and that is not true. We need to correct that and get them interested in the subject by giving them these wonderful role models who they will be able to identify with.
Q141 Viscount Stansgate: I quite agree with you about maths and football. For people watching Scotland play Brazil last night, every goal that Brazil scored reduced the mathematical probability that Scotland could proceed to the next stage, and we will have to see. I wish them well, but it is a tough job. My question is really simple. We are a committee. We have to make a report. In 20 seconds, what is the one major recommendation you would like us to make?
Lord Tarassenko: Very simply, think about 16 to 19, and the mathematical or numeracy type of education for the various groups I described in the earlier answer at 16 to 19. That does not mean you have to re-implement a new curriculum for maths GCSE and so on, but for the different types of outcomes at 16, from those who do really well at the top end who are going to be the Demis Hassabises of this world, to those who did not get a grade 4, what you do with them from 16 to 19, whether they stay at school or go on to do V-levels and T-levels? Let us really think about that in the context of mathematics, numeracy and everyday skills to understand the modern digital world.
Professor Jane Clarke: Mine would be to invest in teaching. We need more maths teachers. We need to enable the non-maths specialists to teach the maths that is in their syllabus. We need to give them the resources that they need to do experiments. Without them, we will go nowhere. It is about investing in maths teaching, and not just specialist maths teachers but the primary maths teachers. Invest in the primary teachers who are not necessarily mathematicians. Invest, invest, train, train, support, support, and give them the resources that they need to deliver this.
The Chair: That brings us to the close of our session. If you have other thoughts on recommendations, do feel free to write to us.
Professor Jane Clarke: Thank you for doing this. It is such an important topic.
The Chair: You must thank Lady Bull and Lord Tarassenko, because, between the two of them, they put the bid in for this inquiry.
Professor Jane Clarke: Really, thank you, because it is just so fundamental to everything and the whole well-being of this country. We are failing our country unless we address this, so thank you for doing it. We do appreciate it.
The Chair: Thank you.
