Letting Computational Thinking Drive

August 11, 2016

Last week, we described why we are simultaneously taking three approaches to develop learning trajectories. Today, we’ll tell you about the approach that puts computational thinking (CT) in the driver’s seat.

To develop this approach, we started with two fundamental questions:

1. What do we know about computational thinking and how it might be developed in elementary school students?
2. How do we best leverage what we know to construct a learning trajectory?

We have many sources of information for answering the first question. Research on what children are able to do in friendly programming environments, how work in those environments affects the way children think, and the kinds of non-computer-based activities that can foster computational thinking has been ongoing since the 1980s (e.g. ^{1 2 3 4}). Organizations such as the Computer Science Teachers Association (CSTA) and the International Society for Technology in Education (ISTE) have produced standards and reference materials to guide teachers as they address computational thinking in their classrooms. Many sets of instructional materials have also been created and piloted to facilitate computational thinking instruction (e.g., KELP CS, code.org, ScratchJr, ScratchED, CS Unplugged, etc.).

Having rich sources of information, though, is only half the battle. To create a learning trajectory, we must have a strategy for organizing the information into a coherent picture. A number of factors make this a challenging task. Interpretation of research on computational thinking is hindered by a lack of consensus on a definition of computational thinking and how it is different from computer science. Lack of consistent definitions and vocabulary make it difficult to draw connections across studies. And not only is computer science a relatively new field from a professional standpoint, it is a brand new subject to elementary schools, making it difficult to study CT in authentic school contexts or understand how laboratory research might guide implementation of CT instruction in school settings.

While examining what we know about CT in elementary school, we came to the conclusion that to answer question (2) – to leverage this information to create trajectories – we needed a powerful and flexible organizational tool. To that end, we developed a database to store, organize, and connect seemingly disparate pieces of information into coherent collections. Because we are viewing learning trajectories as an ordered set of learning goals, we chose learning goals as our basic unit of analysis.

Our approach to populating the database can be roughly described in four steps: Collect, Sort, Order, and Illustrate. We’re collecting learning goals by entering each into the database as a separate record, whether they are explicitly stated in a standards document, implicitly communicated via the description of the purpose of an instructional activity, or emphasized via the measures used in an empirical study. We’re preparing to sort the goals by tagging them according to CT concepts and domains and the type of support the literature provides for them (theoretical, empirical, and so on). We’re preparing to order them by noting the grade levels for which researchers or educators believe they are appropriate and any information provided about their level of complexity. In the end, we hope to synthesize groups of goals into “super-goals” and illustrate them with activities that could be used to help students reach the goals.

The LTEC database, when fully populated, will be a well-organized collection of CT learning goals, along with descriptions of where the goals came from and the evidence that exists to support their appropriateness for elementary school. We hope that the database will become a widely used, dynamically updated resource for the CS education community. Stay tuned for updates on when aspects of the database will be available for public view and comment!

References