Computational Thinking (CT) Meets Young Children: Critical Review of Research on CT in Early Childhood

• In Event: Computational Thinking in Early Childhood: The State of Our Burgeoning Field

Abstract

Viewed as the foundational processes of computer science with broader application for learning, computational thinking (CT) is increasingly promoted and researched in early childhood (EC; ages 2-8) (Bers, 2018). In contrast to secondary education (the site of most current CT pedagogy and research), EC education prioritizes development of the whole child through integrated interdisciplinary pedagogy. There is a need for an informed debate about benefits and tensions of introducing CT in EC (Manches & Plowman, 2017). To address the need, we conducted a systematic review guided by the following questions: (1) What are the general characteristics of the existing studies on CT in EC? (2) How is CT defined and operationalized in EC? (3) How is CT promoted and supported in EC?

Following Alexander’s (2020) guidance for systematic reviews, we employed a three-step process to select studies. First, we developed a keyword search using 35 articles identified by domain experts as an initial validation set. The final query was: (“computational thinking” OR “Robotics” or “Unplugged” or “Computing Education” or “Computer Science Education”) AND ("children" or “kindergarten” or “early childhood” or “early years”), published between 2006 and 2021. Second, we ran a keyword search across five databases (ERIC, ScienceDirect, SpringerLink, IEEE, and ACM) and yielded 8398 unique results. Third, we filtered these with a Bayesian classifier trained on article abstracts and hand-coded the results, producing a corpus of 125 studies. To analyze the selected studies, (1) we used content analysis to summarize characteristics of the selected studies’ research questions, research design, theoretical framework, and participants’ demographics; (2) we categorized how CT was defined, operationalized, or measured; and (3) we coded CT pedagogies including activity types, duration, teaching methods, technical tools, and effectiveness.

Our preliminary results reveal: (1) there is a lack of diversity in study populations, in research agendas (with a dominant focus on feasibility and effectiveness of CT activities), and in theoretical frameworks (constructionism or sociocultural theories dominate); (2) these studies narrowly define or operationalize CT as a set of cognitive concepts and skills; and (3) developmentally appropriate practice (DAP) in CT learning context is employed superficially. Programming is the dominant context for CT learning.

Situating the review results in the fields of early childhood education as well as K12 computing education (Kafai et al., 2019; Tissenbaum et al., 2021), we propose ways to diversify the research agenda and study populations, to broaden definitions of CT to better align with the idea of educating a whole child in EC while preparing children for future CT learning, and to expand the pedagogical repertoire with practices consistent with evidence from EC as well as K12 computing education.

Going beyond synthesizing research on CT in EC, this paper offers theoretical suggestions on how to better-align conceptualizations of CT for early childhood, as well as practical guidance for teachers and researchers on teaching developmentally appropriate CT. Finally, we suggest ways to diversify research agendas and study populations to address equity and inclusion.

Authors

• X. Christine Wang, University at Buffalo - SUNY

• Christopher Proctor, University at Buffalo - SUNY