Fostering Children's Computational Thinking in Early Childhood: A Systematic Review

• In Event: Understanding and Supporting Young Children's Computational Thinking

Abstract

Computational thinking (CT) has drawn considerable attention in early childhood education in recent years. With burgeoning efforts to promote CT in early childhood (ages 0-8), there is an urgent need to systematically review empirical studies in the field. To address this need, our systematic review synthesizes the existing research and charts future directions for CT in ECE.

Through a four-step process, we selected 35 studies of CT in early childhood that have been published during the last decade (2010-2019). Following Alexander’s (2020) guidance for quality systematic reviews, we first categorized the selected studies’ main characteristics (research questions, theory, research designs, CT concepts, tools or platforms used, and findings), and then analyzed these articles in depth to answer these four main review questions: (1) How is CT defined? (2) What CT components do children develop or struggle to develop? (3) What kind of technological tools are used to promote CT? and (4) How effective are the programs in promoting CT?

We identified these three trends in the characteristics of the reviewed studies: (1) increasing number of publications in recent years but dominated by super contributors (e.g., Dr. Marina Umaschi Bers authored and co-authored 19 of the 35 studies); (2) lack of theoretical grounding or diversity in adopted theoretical frameworks. Almost a half of the studies (16) did not cite any theories and Papert’s Constructionism (1980) was featured in 11 out of 35 studies; and (3) relatively even distribution of research methodologies among qualitative, quantity, or mixed methods.

To answer the four review questions, we found (1) the reviewed studied generally agreed on the CT definition but disagreed on specific CT skills/concepts or practices for young children. (2) CT concepts (e.g., sequence and loops) and practice (e.g., debugging) seemed the most accessible to young children. Some inconclusive results regarding CT outcomes indicate potential developmental differences and the role of teachers’ instruction and scaffolding (Wang, et al., 2020). (3) Tools with three different interfaces—tangible, graphical and hybrid—were all appropriate for young children and each interface afforded different experiences and learning. (4) Three features defined the effective CT programs/curricula: (a) introducing CT concepts in developmentally appropriate ways and progressing gradually, (b) integrating explicit CT instructions with children’s free exploration, and (c) providing opportunities for children to apply and practice their CT concepts/skills.

Based on our review, we advocate the following ways to advance this field: (1) diversify researchers in the field to bring in fresh perspectives, (2) broaden theoretical frameworks, (3) continue to employ a range of research designs and methodologies. In addition, we should consider (1) adopting a unified CT framework to guide studying of different components of CT, (2) exploring those understudies CT components for young children as well as examine how they are related, (3) continuing to explore children’s CT learning with different programming tools/interfaces, and (4) making effort to standardize developmentally appropriate CT assessment for young children and providing more consistent reporting on CT program dosage and intensity to allow rigorous assessment of their effectiveness.

Authors

• X. Christine Wang, University at Buffalo - SUNY

• Youngae (Young) Choi, Nevada State College