What makes an inventor?

With support from AlphaPlus, an educational consultancy, we carried out a literature review on factors that influence pathways to invention. Educational ability is clearly relevant: maths test scores in early childhood are strongly predictive of inventive activity in adulthood.[11] Meanwhile, STEM knowledge is important for technological invention: individuals with technical, medical and natural science qualifications are more likely to file a patent.[12]

There is also a range of skills and attributes associated with inventiveness. The US-based Lemelson Foundation, whose mission is “to support the next generation of inventors”, argues that young inventors need “to think critically, and identify real-world problems and possible solutions” as well as being able “to turn ideas into solutions”.[13] In a similar vein, earlier work commissioned by Nesta identified five attributes that underpin innovative behaviour in young people: creativity, self-efficacy, energy, risk-propensity and leadership.[14]

However, it’s not only personal characteristics that make a difference, but also factors in children’s environments. Inventors are likely to have had a better-than-average start in life: the ‘lost Einsteins’ study shows that children from families in the the top 20% of the income distribution are much more likely to become inventors, while a study of Finnish inventors finds that parents’ education levels are a significant driver of children’s rates of invention.[15] Family occupation plays a key role: children of inventors are nine times more likely to become inventors themselves as children whose parents are not inventors.[16]

It also matters where you grow up. The ‘lost Einsteins’ study, for example, found that children from areas where there are more inventors are more likely to become inventors themselves. In fact, they find that ‘exposure to innovation’ in childhood influences not just whether individuals invent, but the type of inventions they come up with. Children who grow up in Silicon Valley are more likely to apply for patents in software as adults, even if they later move to another part of the country. Meanwhile, evidence from Sweden shows there are clusters of future inventors born in places that are close to prestigious academic institutions and have good support for entrepreneurs. The researchers suggest that ‘inventive cultures’ during childhood help form the career paths of Swedish inventors.[17]

Schools and other learning environments also play a role in fostering future innovators, by helping children gain what could be loosely termed ‘practical knowledge’. This could be gained for example through exposure to scientific discoveries and technological breakthroughs, information on how to develop and execute innovation strategies and learning from other innovators.[18]

Why are boys from privileged backgrounds more likely to become inventors?

So how do we explain differences in likelihood to be an inventor between demographic groups? Disparities in educational attainment play a role. We know for examples that science attainment in the UK is significantly lower among disadvantaged socio-economic groups.[19]

But attainment is only part of the story. Researchers at University College London have used the sociological concept of ‘capital’ to explore differences in attitudes to science and science careers among young people aged 10-19. 'Social and cultural capital' are resources that individuals can use to gain social advantage, such as social networks or qualifications that are seen as being ‘high status’.

By tracking students’ attitudes and aspirations over time, the researchers identified eight key factors that affect likelihood of wanting to be a scientist. The researchers propose that these factors are specific forms of cultural and social capital, which they call ‘science capital’. Students with high levels of science capital are much more likely than others to consider becoming scientists.[20]

Moreover, the researchers found systematic differences between groups in levels of science capital. Boys were much more likely to have high science capital than girls. There were significant differences between ethnic groups, with South Asian students having the highest levels. Pupils with high science capital were most likely to come from better-off households. Meanwhile, students with low science capital were more likely to be girls from lower-income backgrounds.

What is science capital?

Summarised as ‘what you know, what you think, what you do and who you know’, researchers at University College London have identified eight dimensions of science capital that help predict young people’s likelihood to consider science as a career. They are:

• Scientific literacy
• Science-related attitudes, values and dispositions
• Knowledge about the transferability of science
• Science media consumption
• Participation in out-of-school science learning contexts
• Family science skills, knowledge and qualifications
• Knowing people in science-related roles
• Talking about science in everyday life

The more science capital a student has, the more likely they are to aspire to post-16 science and have a ‘science identity’. This can help explain why, for example, while girls in the UK are just as likely as boys to study chemistry and biology at A-level, there are far fewer girls studying physics and computer science - despite girls outperforming boys in nearly all subjects at GCSE.[21]

These disparities in science capital are not necessarily down to different levels of interest in, or experience of, science. Those with low science capital might actually have considerable scientific knowledge - but of types that are rarely recognised or valued as such. The team at UCL give the example of pupils who like cooking, or whose parents are mechanics. These activities involve a lot of STEM expertise, but are not usually seen as being ‘scientific’.

In fact, the way that science is often taught at school prioritizes academic language, which benefits children who are comfortable using it, and excludes or marginalizes others. It is those who can ‘do’ science in specific ways - those that are recognised by dominant scientific elites - who are more likely to be able to use their science capital to their advantage.

One way, therefore, of thinking about how to create more diversity in innovation is that this involves helping those with low levels of relevant ‘capital’ to acquire more, as well as shifting societal attitudes and norms that means some people’s forms of capital are valued more than others’.

We know there are many other things that need to be done to improve diversity in science and innovation. For example, the concept of the ‘leaky pipeline’ is well-established. This refers to the pattern - observed worldwide - which sees more women drop out of the STEM workforce at every step up the ladder of seniority. Barriers to progression include discrimination, a competitive culture, long working hours and difficulties in managing career advancement around having children.[22]

Research on why women’s rates of patenting are so low shows how important professional relationships and networks are in patenting - most patents are filed as group applications - and how women are often missing or excluded from these networks.[23] However, there are compelling reasons to think that if we want to tackle diversity in innovation we also need to focus on what happens in childhood.