Poster #28 - Robot Coding Toys as a Context to Cultivate Spatial Thinking and Computational Thinking Skills

• In Event: Poster Session 15

Abstract

Spatial thinking (ST) entails understandings of space and objects’ positions in space, reasoning with objects or representations in space, and operations on spatial relationships (NRC, 2006). The Common Core State Standards of Mathematics (CCSSI) emphasize children’s use of language to describe shapes and positions of shapes in space, but the critical underlying concepts are about developing children’s spatial orientation. Spatial orientation is the understanding of different positions in space, and children first develop spatial orientation concepts in relation to their own position in space and later develop external based reference systems using landmarks outside themselves (Sarama & Clements, 2009). Yet the CCSSI standards mainly focus on applying spatial knowledge as relational from their own perspective with reference to another object. Children are not asked to impose their own perspective on that of an agent or individual.

Children learn ST through play with blocks, puzzles, and, as we will discuss, coding with robot toys. Coding with robot toys involves navigating the robot’s movement through physical space, which draws upon spatial orientation knowledge and allows children to explore orientation from their own perspective as well as from the robot’s. For example, coding a robot to traverse a path requires children to understand movement based on the robot’s orientation and perspective- not their own. However, coding is primarily researched as a context for teaching computational thinking (CT). CT involves the thought processes involved in formulating a problem and expressing its solution(s) in such a way that a computer—human or machine—can effectively carry out (Wing, 2014). Very few studies have explored the relationship between ST and CT in early childhood and cognitive models of CT in early childhood do not include ST. In our research, we have found that ST is foundational to CT.

We will present research on the relationship between ST and CT in early childhood, specifically, kindergarten. We developed curricular tasks that involve coding with robot toys as a context to cultivate CT and ST, developed a cognitive model that integrates CT and ST, and designed a performance-based assessment to measure these skills. We validated our assessment with a sample of 272 children aged 5-8. We found that our items fit well to two-parameter unidimensional IRT model, confirming that both model- and item-level model fit indices were acceptable. Item analyses showed average item discriminability was high (M = 2.261, SD = 1.121), ranging from moderate (a = .909) to perfect (a = 4.657) level, and average item difficulty was moderate (M = -.213; SD = .864), ranging from very easy (b = -2.404) to hard (b = .930). Reliability analysis showed high internal reliability across 19 items (α = .91, ω = .92) and inter-rater reliability in scoring process (Cohen’s κ = .91). We will present our cognitive model of CT and ST, our assessment, and share examples of how coding with toy robots help children understand abstract concepts like spatial orientation in tangible ways. Our findings support the importance of integrating ST and CT in early childhood.

Authors

• Jody Clarke-Midura, Utah State University
  Presenting Author

• Deborah Silvis, Utah State University

• Jessica Shumway, Utah State University

• Chungsoo Na, Utah State University

• Victor R. Lee