Pattern Recognition: Following Data Trails and Discerning the Signals from the Noise

Topics: Future of Learning

For better or worse, the data we generate when using digital technology has become almost indispensable for navigating the world around us. This data creates trails and patterns that give us personal recommendations, steer viewing and listening recommendations, give us driving directions and so much more. Because of their ubiquity and utility, these same data trails and data patterns are also extremely valuable, with the tech companies that gather and assemble them having become the new titans of industry.

Such an expansion of data creation and use was a central theme of the Pattern Recognition driver of change from KnowledgeWorks’ 2020 Forecast: Creating the Future of Learning. This post, part of an ongoing look at back at that forecast, reflects on the possibilities it described for data and its possible applications in learning.

The 2020 Forecast: Creating the Future of Learning, published ten years ago, revealed how many of our fundamental relationships — with ourselves; within our organizations; and with systems, societies and economies — were being reimagined and re-created in ways that could disrupt the status quo and challenge our usual assumptions.
Learn More

The Pattern Recognition driver of change described the increasing output and flow of data being generated from sources such as social media activity, the GPS in cars and smart phone usage. It forecast that these data would combine to create personal data trails as we went about our daily lives and engaged in educational activities. This driver of change also detailed the need to make sense of all this data. It forecast the emergence of new tools and skills to help discern patterns out of all the noise being created as the volume and types of data increased over time.

There are connections between the drivers of change identified in the 2020 Forecast to those identified 10 years later in Navigating the Future of Learning. For instance, in the newer forecast, we explore how forces of change such as Pattern Recognition are related to new forces of change like Automating Choices.

Data proliferation continues

Looking back at the Pattern Recognition driver from our perspective in the year 2020, many of the key themes that it described still seem highly relevant. Data and information proliferation continues, with TechJury estimating that in 2020 individuals will generate 1.7 megabytes of data a second and with the World Economic Forum putting the entire digital universe at around 44 zettabytes in size, over 40 times more bytes than there are stars in the observable universe.

As our data trails have grown longer and longer, we have also seen the emergence of jobs and roles associated with making sense of them. Postings for data scientists, data analytics professionals and other jobs associated with big data are now common. A job report published in January 2018 by Indeed highlighted the growth of such jobs, noting that postings had skyrocketed 256% since December 2013, more than tripling.

The Pattern Recognition driver of change also asked two questions that are helpful to revisit in order to get a better sense of whether its key themes are still relevant today. I explore these questions, slightly revised for the sake of looking back at the 2020 Forecast, below.

How have ubiquitous, visible data impacted teaching, learning and the assessment of learning experiences?

Many current examples point to how data trails have impacted teaching, learning and assessment. They include adaptive learning platforms such as Dreambox and Cognitive Tutor. Such platforms are typically online, computer-based educational systems that modify curricular material in response to how a student performs. These platforms capture and use student data to customize content for individuals. In addition, the Mood Meter from Yale’s Center for Emotional Intelligence offers a different perspective on the use of data trails in learning. The Mood Meter is used in classrooms to assess the students’ emotional states, enabling teachers to customize content and delivery based on how members of the class are feeling.

While these examples all highlight how data trails are being leveraged in education, the current policy landscape in relation to student data, specifically in the K-12 sector, may be a limiting factor to fully realizing the possibilities imagined in the Pattern Recognition driver of change. Current policy typically limits a student’s data trail to the types of data generated within the context of their formal learning journey. Fully utilizing a student’s data trail would mean broadening the sources of data that the K-12 system can use.

How have we used data to enhance human decisions rather than automate them?

This question is a bit harder to answer than the first. While we are creating more and more data, the decision points presented to us when it comes to that data are often narrowed down by machine learning. On one hand, we still have options and choices to make. On the other hand, many of those choices are never made available to us because we are often presented with options for which an algorithm figures we are best suited.

Data will be critical for the coming decade

The changes explored in the Pattern Recognition driver of change are continuing to unfold. We live in a world that is rich with data, and we rely on other people and algorithms to help make sense of it all. Education has also benefited from the increased flow of data, having used it to personalize and customize learning for students. However, a policy shift may be needed to fully realize the potential for tapping students’ data trails.

As we look ahead to the next ten years, we can expect to see more data trails, more noise and a growing need to separate the signals from the noise. Questions about the role of human agency in the decision-making process may also become pressing as we increasingly rely on algorithms to make sense of the data that we generate.

Ultimately, how we think about data, who owns it and the policies that we put in place around what types of data can be collected and used in learning will have profound implications for shaping the future of learning.

Dive deeper into how your learning community can improve the performance of the mind in a changing environment with customized talks and workshops.

2020 Forecast Retrospectives