Tracking the Trailblazers

Our workshop looked at the hidden history of 'innovation indicators' -- tracking knowledge flows and big data on everything from biotech to nanotech...

"The most important step is to learn and combine. Breakthrough innovations often emerge from a combination of different types of knowledge from different areas... The next step is to track these innovations via relevant data sources and comparative indicators."
Dr. Lili Wang, UNU-MERIT Researcher

The second UNU-MERIT workshop on ‘Data Science and STI Indicators’ gathered experts from 14 different innovation organisations from 8-9 May 2019. They joined us to discuss new data sources and research methods; to explore the construction of science, technology and innovation indicators; and to link scientometric research with innovation studies. We caught up with lead organiser Dr. Lili Wang to find out more.


Why did we hold this workshop and what exactly did it cover?

UNU-MERIT is one of the world’s leading research institutes on innovation and technological change, which means we have world-class researchers exploring the social, political and economic factors driving technological innovation. A major part of this ‘exploration’ is to build powerful indicators to help shed light on the complicated system of innovation and knowledge production.

Three years ago, back in January 2016, we organised the first STI indicator workshop at UNU-MERIT, with great contributions from colleagues such as Fred Gault and Hugo Hollanders, as well as external experts like co-organiser Ismael Rafols [second speaker image] and Robert Tijssen. Because of the success of last time, we always wanted to organise it again. Thanks to the support from our institute, we were happy to host this workshop again.

First we looked at innovation concepts, then at applications like innovation systems and knowledge production in both developed and developing countries. On the second day, we had a session talking about data mining and technology prediction, and then in the end we discussed data matching and research relevance. The workshop covered different countries and different industries, including both public and private sectors.


There’s a lot of talk about ‘new’ sources of big data – but what exactly is new here?

The fun side of doing research in this area is that, along with the availability of new data sources, we always see lots of challenges and possibilities for new research. In recent years, big data and associated software have enabled us to evaluate science, technology and innovation from very different and new perspectives. For instance, publication and patent data, as well as social media data, involve huge databases – all of which can help re-evaluate innovation from different angles. The intention of the workshop was to gather experts to sit together and discuss the development and use of new indicators, based for the most part on new data sources.

"... if knowledge is published in an open source format then it's available to anyone, which means developing countries can also learn from the knowledge."
Dr. Lili Wang, UNU-MERIT Researcher

How exactly should we define data mining, among other aspects of innovation?

There's already a lot of data available, but the important thing with mining is to use the data wisely and really construct appropriate indicators. That's actually a very important step – to do high quality research. Here I mean work by among others Fred Gault [first speaker image], who has focused on innovation for decades. To put this into perspective, he was involved with the Oslo Manual of innovation definitions from its inception in 1992.

In terms of definitions, lots of people talk about innovation but not everyone fully understands what it means. Just one example: Fred showed a table during the workshop giving details of the numerous innovations based on ‘Zero R&D’ – i.e. innovations created without any R&D input. In the past, people would use R&D as a proxy measure of innovation – but it’s really not that simple. R&D is a capital investment, but there are many Zero R&D innovations. In other words, if you do not invest in R&D, you can still innovate. This showed up time and again during the workshop, particularly in relation to knowledge flows.

The importance of Zero R&D innovations is also recognised in the European Innovation Scoreboard (the European Commission's main tool for tracking innovation performance across the continent and beyond). One of the Scoreboard indicators specifically covers 'Non-R&D innovation expenditures'.


What exactly do you mean by knowledge flows – simply the transfer of knowledge from universities to industry?

From anywhere to anywhere, actually. What’s more, if knowledge is published in an open source format then it's available to anyone, which means developing countries can also learn from the knowledge. But to create new knowledge you generally have to learn from – and build on – old knowledge. I really think this is very important for developing countries because they don’t have much money to invest in basic research. Knowledge flows from other countries are so important, particularly for African nations.


Some developing countries are particularly innovative, according to recent research. What are these countries doing differently in terms of knowledge flows?

For many developing countries, including Kenya, Rwanda and Tanzania, a very important step is to learn from partners in advanced countries. International collaboration includes technical collaborations, in-country training, like our DEIP innovation workshops, as well as bright young fellows going abroad to do PhDs, such as those who join our doctoral programme – but then also going back after their studies in order to benefit their home economies.


What are you working on personally in terms of knowledge flows?

I’m focusing on multi-dimensional knowledge flows, which can happen in many different ways. For instance, Chinese inventors have historically learned a lot from the USA, but more recently they have tended to build up their own capacity. From another angle, more advanced regions help less developed ones, as per the case of nanotechnology development in China. The most important step is to learn and combine. Breakthrough innovations often emerge from a combination of different types of knowledge from different areas, as well as knowledge from public science – which is something I’ve written a lot about recently. The next step is to track these innovations via relevant data sources and comparative indicators.

It's clear, then, that combining different types of knowledge is often the most fruitful approach. In China, in the past, they only learned from the USA, but more recently China has used a lot of combined knowledge and technologies from different areas and different countries – not just the USA but also European countries, as well as Korea and Japan. See, for example, the upgrading of biotech in China.

Ultimately, innovation is a story of (re)combination from many angles!


Tracking the trailblazers? A heatmap from a popular fitness app, showing user density across Asia -- a rich source of data.

Tracking the trailblazers? A heatmap from a popular fitness app, showing user density across Asia -- a rich source of data.

See the workshop programme below: