Panel discussion around the role of data scientist, particularly in the field of education, as well as the presumed anticipated shortage. Already touted as the sexiest career in the 21st century, there’s a lot to live up to…
Ryan Baker: we need more. Recruitment is happening right now in industry and academia. A new phenomenon. Educational data is not the same as ‘other’ data.
John Behrens: educational data scientists are… nice people. What is data and how (easily) do we interpret the same data differently? Be a philosopher. Be practical. Make mistakes. Revisit what has already been learned.
Martin Hawksey: data is the new big commodity. Be creative. Be curious. Make mistakes.
Naomi Jeffery: visualisation is important. Need for robustness. An ability to communicate the meaning is key.
Taylor Martin: Finding new ways to answer questions. It’s an interdisciplinary critter. It’s about asking good research questions about data. We’re all going to have to play nice (even with industry).
Questions posted earlier: interdisciplinary vs multidisciplinary and where are educational data scientists coming from. Should focus be on building teams from existing cross functional experts? Or on teaching people to be interdisciplinary thinkers? Ans: both! JB: is anyone really only a ‘pure’ single role? NJ: need to be a scientist. It’s not just about pretty pictures. SB-S: does it have to be one person?
Question: Haven’t we been doing this for years? What is new or different about an educational data scientist and an educational researcher. Ans: nothing. JB: but techniques have moved on. NJ: because it’s cool, there can be an earlier involvement and a greater impact. RB: there are valuable similarities to other fields which we can learn from.
Question: who in the real world needs to be equipped to be able to read and interpret data for ordinary students? Ans: TM: school administrators need to be able to support teaching staff who have enough to do. JB: different stakeholders will have different needs. MH: if the dashboard isn’t easily understandable then the dashboard is wrong.
Question: Do you agree that, in education, the outcomes are less well specified than in business? Seems to have an impact on the related analytics. Ans: JB: we measure things that people can agree on not necessarily what is most important.
Question: We think of students as accountable for their learning. With the advent of learning analytics, who is now accountable for the quality and outcomes of education? Ans: more choice now in how students and teachers interact with learning materials. But teachers are still the ones in control. TM: there is a need for a common framework, a set of KPIs that can be used across a range of materials.
So, educational data science, sexy or no?