Creating tools to fine-tune learning

Image
john stachurski

Professor John Stachurski from the ANU Research School of Economics (RSE) has been awarded A$1.3 million from the prestigious Alfred P. Sloan Foundation. The grant will fund the development of tools for creating next-generation scientific textbooks and lecture series.

John’s work is based around the rapidly growing Jupyter platform, which combines state of the art web publishing with an emphasis on open access and open science – a practice that encompasses unhindered access to scientific articles.

Traditionally we have used hardcopy textbooks to convey scientific ideas to students but this mode of communication is becoming less suitable each year. 

Our team aims to develop a set of tools for creating documents that replace traditional textbooks in the scientific domain. 

These tools will be powerful, elegant and easy to use. They will fully exploit modern technologies such as web publishing and cloud-based computing,” explained John, whose team comprises of Chris Holdgraf from the Berkeley Institute for Data Science, Greg Caporaso at the Northern Arizona University and Matthew McKay from RSE.

Traditionally we have used hardcopy textbooks to convey scientific ideas to students but this mode of communication is becoming less suitable each year.

The documents produced from John’s new generation tools will be called “executable books.” He hopes they will become “an attractive alternative” for students to learn from rather than “outdated” textbooks.

“If we look at current science textbooks, many of them contain executable computer code, which is used to communicate scientific phenomena, or the way we have come to think about scientific ideas. The problem with this scenario, even for very recent publications, is that computer code embedded in printed textbooks is inaccessible to students, hard to execute and often out of date. Traditional tools for creating textbooks are unable to fix this,” John asserted.

John identifies three primary advantages of the executable books: students can test scientific concepts while learning; they can freely access the study material; and the tools will contain integrated techniques that will ensure a better learning experience for them.

“Through executable books, students will have the option to run the code embedded in the lectures live in their browser. They can also edit that code and simultaneously observe the effects. This makes it very easy for them to experiment and test out their own ideas, which is a great way to learn about scientific concepts,” he said.

John observes that academics are searching for better ways to convey scientific ideas to students, such as publishing lectures on the web. He hopes that these tools become “a de facto standard” across all sciences, replacing traditional textbooks with open source executable books.

Given the capability and capacity of the tools he is creating, John has also started conversations with organisations outside of academia who are already keen on using his research.

“One of the early adopters of our tools is a company of consultants who teach computer programming and machine learning to private sector companies in the US. They want to use our tools to create training material. Within the field of economics, there are also many policy practitioners who are trying to learn more about computational methods and how these techniques can help them better understand economic phenomena,” he shared.

John is an internationally recognised economist and has spent his career contributing to quantitative economic modelling. He is a co-founder of QuantEcon, a non-profit organisation that is focused on improving economic modelling by enhancing computational tools for economists.

The College is always keen to explore research collaborations with the public and private sector and to reconnect with Alumni. Please get in touch if you would like to know more.