I attended a lecture by Professor Sonia Livingstone this evening which was fascinating and thought provoking. The lecture focused on the topic of data privacy for children in an online context. There was lots to unpack and I’m sure I’ll be having “a-ha!” moments for weeks to come.
However, my initial “a-ha” was the parallel that Professor Livingstone drew between the education and awareness children are (at least in theory) taught about media and how it works and the education they need to have into how the data economy and digital society works and the broader issues of what lies behind the tools and applications they interact with.
I tweeted at one point that the future of the digital age of consent probably has parallels in the evolution of sex education – practical and informed education to allow young people to make smart and aware choices in line with their level of personal maturity.
Professor Livingstone discussed the need to educate children about the business models of online businesses and potential uses of their data, and the potential for misuse and abuse. This “data literacy” is an important aspect of developing digital awareness in children so that they can make what m’kiddo calls “smart choices”.
However, this aspect of data literacy is only part of the equation in my view. There is a range of data illiteracy that needs to be addressed as part of developing the skills needed to safely navigate our increasingly data-driven world. These include:
- Statistical awareness: Developing an understanding of what statistics are, and what a statistic they might hear quoted actually means so that they can put the “fact” in context. For example, understanding that the UK Brexit referendum was carried by 52% of those who voted, but only represented approximately 25% of the total population of the United Kingdom.
- Algorithmic awareness:Â Developing an understanding of how the data that is given by them can, in turn, lead to data that is inferred about them, and how that data is in turn processed against patterns of historic behaviour to infer future actions, traits, or behaviours. Also, understanding how the algorithmic curation of news sources or entertainment can lead to “filter bubbles” or worse. (At a minimum we need to educate people to understand how Amazon’s algorithms will keep trying to sell them toasters even after they have bought a toaster).
- Data Quality Awareness:Â Even after over 20 years in this field I am taken aback at how people in organisations struggle with basic concepts of data quality. This is increasingly important in an environment where algorithmic decision making is being applied and the quality of training data leads to bias, but it also affects basic issues in data management. We need to educate children (and adults) that the outcomes we get from information processes are often the result of a chain of data quality issues ranging from poor data capture processes, to botched data cleanups, to data being used for purposes it wasn’t originally intended for. In m’kiddo’s case, it’s understanding that the accents in her name are important and the implications of that in how her information is processed.
This is a non-exhaustive list, and it is before the issues of actually using software or writing code. But there are some key building blocks that are needed to help give children (and adults) tools to comprehend the data-driven world that we are living in and to assess the issues and risks arising from new technologies like semi-sentient robot dogs to artificially intelligent digital assistants around the house.
To focus on a “stay safe online” approach to data privacy for children is to repeat the mistakes of grown up data privacy by conflating privacy with security. However, it is also akin to simply teaching teenagers about contraception without engaging in education about other aspects of sex and sexuality and then expecting them to cope with complex questions of choice, consent, and power dynamics in relationships.
Data Literacy education is key to both keeping children safer online as well as enabling them to make smarter and more informed choices about how they allow their data to be used and the consequences of the choices the make or actions they take. It might also contribute to a generation of people working with data and information who are better able to understand and engage with the fundamental disciplines of data governance, data ethics and data quality as part of the adoption of new technologies.