I’m an “accidental” database admin/data analyst (learning on the job rather than via formal education) with over 20 years of experience. Because of this informal training, I have spent a lot of time mentoring new data analysts and database administrators (DBAs), with the hope that they can learn from my experiences and see what a “real-world” role of this type involves. I am intrigued by how today’s students learn about various technologies, software, and hardware--especially considering the pace with which these things change. Although I have been in various roles/industries and have mentored dozens of young people, I haven’t read any articles covering the educational aspects of this area, so Moon et al.’s paper was eye-opening and enlightening.

As Moon et al. state, having computer- and data-literate young adults will allow for a dramatically improved society: “Data literacy can help youth interpret the information they consume, make evidence-based decisions, and become more socially and civically engaged.” The paper also highlights that, as algorithms continue to become more and more ingrained in our society, having data-literate students in kindergarten through 12th grade (K-12) is critical. This will help students become more aware of how data--their data--is being used and how those on the margins of society are being pushed further out.

Moon et al. divide data collection and analysis tools into four main categories--spreadsheets, visual interfaces, scripting languages, and other interfaces--and talk about how even as early as kindergarten, children can start becoming familiar with how data is collected and used. This will lead to the standardization of data science teaching in universities/colleges--something that Moon et al. highlight in a discussion about a survey done on data science courses. Standardization would ensure that all students taking these courses are exposed to the same types of software and methodologies, allowing for easier transitions into the workforce; this is something I’ve struggled with when mentoring, as students learn a very specific piece of software that may not be commonly used in the real world.

Moon et al. then go into an exploration of the types of datasets that students tend to engage with when doing coursework, and highlight the importance of this “especially for youth from populations historically excluded from computing and data-intensive fields.” One of the main findings highlighted in this section is that, although students often use “authentic tools” (RStudio, Python, and so on), they are often navigated toward (or only given access to) smaller datasets, thus missing an opportunity to learn how to handle big data. As an aside, creating simulated data is something that I cover with the people I mentor, and I find this useful for many reasons: it is a practical skill to have, it creates a bespoke dataset for testing or other purposes, and it allows for creativity and flexibility in the kind of data being used.

As technology becomes more invisible in our day-to-day activities (for example, AI-generated news reports) and yet drives more of our subconscious decision-making (for example, ads on social media), having young people both aware of and involved in data-based roles becomes more critical. Having the next generation of data scientists prepared at an early age, especially those from marginalized groups, will ensure that the technology is being used in a responsible, ethical, and equitable manner.