Content uploaded by Anika Saxena
Author content
All content in this area was uploaded by Anika Saxena on Nov 26, 2021
Content may be subject to copyright.
REGULAR ARTICLE
Designing Unplugged and Plugged Activities to Cultivate
Computational Thinking: An Exploratory Study in Early
Childhood Education
Anika Saxena
1
•Chung Kwan Lo
1
•Khe Foon Hew
1
•Gary Ka Wai Wong
1
Published online: 22 August 2019
De La Salle University 2019
Abstract Educators and policy makers have increasingly
recognized the importance of computational thinking (CT).
Despite the growing body of CT literature, how to cultivate
CT is still underexplored and undertheorized in early
childhood education. Informed by Piaget’s Theory of
Cognitive Development, this exploratory study was con-
ducted with a focus on three CT skills: pattern recognition,
sequencing, and algorithm design. The framework for the
study was developed in two stages. First, we designed two
sets of unplugged activities (relying on tangible materials),
aiming to (1) provide students with more concrete experi-
ences of CT and (2) equip them with the necessary
vocabularies/instructions for the subsequent plugged
activity (with a digital device). The theoretical foundation
for such an unplugged and plugged design comprised
Piaget’s Theory of Cognitive Development and Asher’s
Total Physical Response. In the second stage, we offered
our CT course in a kindergarten in Hong Kong, involving
six teacher participants and a total of 11 students from K1
to K3 (aged 3 to 6). After 10 h of CT training, almost all
students demonstrated their mastery of pattern recognition
and sequencing. However, the K1 students could only
partially complete the tasks of algorithm design while the
others generally reached the target level of achievement.
Strengthening preschoolers’ training on CT language and
differentiated instruction are some possible strategies to
improve the CT instructions.
Keywords Computational thinking Activity design
Early childhood education Preschoolers
Introduction
In the last decade, computational thinking (CT) has
attracted much attention from educators and researchers in
various education contexts (Hsu et al. 2018; Grover and
Pea 2013; Shute et al. 2017). Leveraging the concepts (e.g.,
algorithmic thinking) in computer science, CT is a way to
address real-world situations and solve problems (Buitrago
Flo
´rez et al. 2017). Undoubtedly, CT is essential for pro-
grammers and people in the field of computing and infor-
mation science. With the extensive application of
computing and computers, CT becomes a basic skill for
everyone today (Chen et al. 2017; Grover and Pea 2013).
Wing (2006) even stated that CT is as important as reading,
writing, and arithmetic competencies.
Educators and policy makers have realized the impor-
tance of CT education. Recently, curricular reforms have
been launched to promote CT education in several Asian
regions, such as China, Hong Kong, and Taiwan. In Hong
Kong, for example, the Education Bureau (2016) advo-
cated equipping students with CT skills. As Buitrago
Flo
´rez et al. (2017) asserted, CT skills must be taught at an
earlier age in order to initiate students’ cognitive devel-
opment. In 2017, the Bureau further published a supple-
mentary document of primary school curriculum, kicking
off CT education in Hong Kong primary education (see
Education Bureau 2017 for a review). Indeed, there is
evidence that students can start learning CT at the primary
school level (Hsu et al. 2018; Lye and Koh 2014; Shute
et al. 2017). In children’s earlier years, several early studies
done in some western countries have suggested that
&Chung Kwan Lo
cklohku@gmail.com
1
Faculty of Education, The University of Hong Kong,
Pokfulam, Hong Kong
123
Asia-Pacific Edu Res (2020) 29(1):55–66
https://doi.org/10.1007/s40299-019-00478-w
children as young as 4 to 6 years old can build and program
simple robotics projects (Cejka et al. 2006; Kazakoff et al.
2013). However, the feasibility of cultivating CT in early
childhood education is still underexplored compared to the
primary and secondary school levels in the Asian region
(see Hsu et al. 2018; Shute et al. 2017 for a review). More
recent research work on CT education in children’s early
years is required (Manches and Plowman 2017).
Learning CT well is not necessarily easy for young
learners because it requires a deep understanding of prob-
lem-solving, computer programming, and handling abstract
concepts (Buitrago Flo
´rez et al. 2017; Looi et al. 2018). As
a result, they may not be able to cope with the cognitive-
challenging CT tasks. How can we address this challenge?
Recently, Looi et al. (2018) use ‘‘unplugged’’ or technol-
ogy-free activities that enable some ninth graders to
physically manipulate the object to explore computational
concepts, such as sorting. Their students were asked to sort
several cups from lightest to heaviest. They found that their
unplugged CT activities could serve as scaffolding to
promote students’ CT learning. In early childhood educa-
tion, however, research to explore using both unplugged
and plugged activities together for cultivating CT remains
limited and undertheorized.
This exploratory study aims to overcome the research
gap by (1) designing CT activities for early childhood
education and (2) documenting the findings as the
groundwork for future CT education in preschool settings.
The framework for the study is thus developed in two
stages. First, we draw upon Piaget’s Theory of Cognitive
Development, Asher’s Total Physical Response, and rele-
vant literature to support the design of CT activities for
preschoolers (aged 3 to 6). The first stage of our study
contributes to our knowledge of how we can teach CT—
more specifically pattern recognition, sequencing, and
algorithm design—in early childhood education. Second,
we present our intervention in a kindergarten in Hong
Kong, involving six teachers and their classes. The fol-
lowing research questions guided the second stage of our
study:
1. How do preschoolers perform in the CT activities?
2. How do preschool teachers perceive the CT activities?
Conceptual Framework
We first draw on Piaget’s Theory of Cognitive Develop-
ment as a theoretical foundation for our study. Following
that, we provide an overview of the CT literature. This
section continues to discuss the importance of language
acquisition in CT education. More specifically, a widely
used language learning strategy in early childhood
education, called Total Physical Response (Asher 1977), is
adopted. Based on the theory and relevant literature, sev-
eral CT activities for preschoolers are developed.
Theory of Cognitive Development
According to Piaget’s Theory of Cognitive Development,
people pass through four primary stages of development:
sensorimotor, preoperational, concrete operational, and
formal operations. His contribution on cognitive develop-
ment provides educators with crucial insights into how
students learn in different ages. Educators can thus design
learning activities according to students’ stage of cognitive
development. Some researchers, however, do not regard
Piaget’s theory as an unproblematic one. They, more
popularly known as the Neo-Piagetians (e.g., Robbie Case
and Kurt Fischer), challenge Piaget’s work and attempt to
create theories that address the criticisms (see Young 2011
for a review). Case (1992), for example, believed that the
age-related nature of Piaget’s theory does not appear to be
correct. In his words, Piaget’s picture of cognitive devel-
opment is ‘‘too monolithic, universal, and endogenous’’ (p.
10). Feldman (2004) compared the theories of Piaget and
Case. He pointed out that there are four large-scale stages
in Case’s theory that superimpose directly onto the afore-
mentioned stages of Piaget but with different labels for
three of the four stages; Case further divided these stages
into substages. Despite this new idea to challenge Piaget’s
work, the Piagetian stages still serve as general guides to
cognitive development (Feldman 2004) and are frequently
used in the research of early childhood CT education (e.g.,
Armoni 2012; Bers et al. 2014; Kazakoff and Bers 2012).
Therefore, in this study we draw on his work as a theo-
retical foundation. Figure 1shows the key characteristics
of Piaget’s four stages of cognitive development (summa-
rized from Sigelman and Rider 2012, p. 49).
In Hong Kong, the age of kindergarten students (i.e.,
preschoolers) ranges from 3 to 6 (K1 = 3 to 4; K2 = 4 to 5;
K3 = 5 to 6). In other words, they are in the preoperational
stage of Piaget. In this stage, children exhibit an increase in
language and symbolic thinking ability. As Sigelman and
Rider (2012) described, they ‘‘can use words as symbols to
talk about a problem and can mentally imagine doing
something before actually doing it’’ (p. 49). Despite their
capacity for symbolic thought, they lack the tools of logical
thought. As a result, they have to rely on their perceptions
which, however, are easily deceived by appearances. To
facilitate student learning in the preoperational stage,
Armoni (2012) and Ojose (2008) suggested tangible
materials, such as blocks, be incorporated with learning
tasks. Ojose (2008) further highlighted the importance of
teacher–student conversation (e.g., questioning) and
observation during lessons. Based on students’ voices and
56 A. Saxena et al.
123
acts on problem-solving, teachers can infer their mecha-
nisms of thinking and offer proper aid or feedback to
facilitate students’ CT learning (Bers et al. 2014; Hsu et al.
2018; Ojose 2008).
Computational Thinking
A highly cited paper by Wing (2006) laid the foundation
for subsequent discussions on CT education. A very first
description of CT, as she outlined, is a process of ‘‘solving
problems, designing systems, and understanding human
behavior, by drawing on the concepts fundamental to
computer science’’ (p. 33). Since then, CT has become
popular and appealing to the academic community (Grover
and Pea 2013; Lye and Koh 2014; Shute et al. 2017).
Compared to the time of Wing (2006), multiple elements
have been added explicitly to the body of CT literature. For
example, Angeli et al. (2016) propose a five-element con-
ceptual framework for CT (i.e., abstraction, generalization,
decomposition, algorithms, and debugging). A recent
review by Hsu et al. (2018) further identifies 19 thinking
steps of CT across studies including pattern recognition,
algorithm design, and simulation, among others.
Through the lens of Piaget’s Theory of Cognitive
Development, pattern recognition and algorithm design are
two possible CT skills that can be introduced to
preschoolers. Pattern recognition involves an observation
of patterns, trends, and regularities in data or other objects
(Hsu et al. 2018). For algorithm design, Buitrago Flo
´rez
et al. (2017) defined it as ‘‘a way of obtaining a solution
through a series of steps’’ (p. 836). Angeli et al. (2016) and
Shute et al. (2017) added that sequencing is an essential CT
concept for algorithm design. Shute et al. (2017) shared an
example of algorithm design activities in which students
have to find the shortest path in a maze that fulfills some
specific criteria. For example, students are required to
insert a sequence of directional arrows to guide ‘‘a lepre-
chaun’’ (a character of a story) to ‘‘a pot of gold’’ (a tar-
geted position to be reached) without hitting obstacles.
Recall that children in the preoperational stage are able to
use symbols (e.g., images and words) to represent objects
and events (Sigelman and Rider 2012). In theory, after
appropriate training, they can (1) recognize patterns that
involve symbols and (2) use symbols or simple words to
present sequences and algorithm designs.
However, children in the preoperational stage rely on
their perceptions to solve problems (Sigelman and Rider
2012). Therefore, tangible materials should be used to
cultivate their CT. One possible strategy is to offer
unplugged (without devices) CT activities prior to their
plugged (with devices) counterpart (Looi et al. 2018). CS
Unplugged (https://csunplugged.org/), for example, pro-
vides various activities that cultivate CT through tangible
materials, such as puzzles and cards, without using digital
devices (Angeli et al. 2016). Having more concrete expe-
riences of CT in unplugged activities can help preschoolers
have a better foundation for cultivating CT in plugged
contexts.
Language in Computational Thinking
In addition to CT skills, Bers et al. (2014) pointed out that
‘‘children must understand in general that people use
symbolic language to communicate with computers, and
they must select specific instructions to accurately repre-
sent their intended outcome’’ (p. 150). Although children in
Fig. 1 Piaget’s four stages of
cognitive development
Designing Unplugged and Plugged Activities to Cultivate Computational Thinking: An Exploratory…57
123
the preoperational stage can use symbols and language
(Sigelman and Rider 2012), CT activities (e.g., program a
robot) may involve some specific computer-related lan-
guage that they are not familiar with. Therefore, students
should first be introduced with the necessary vocabularies
and instructions prior to CT activities.
From the Piagetian perspective of language and educa-
tion, educators should provide younger children with
opportunities for interactions with the physical environ-
ment (Sigelman and Rider 2012). In a similar manner,
Asher (1977) emphasized the essence of learning language
through physical actions. His Total Physical Response is a
widely used language learning strategy in early childhood
education, suggesting that ‘‘understanding should be
developed through movements of the student’s body’’ (p.
4). The premise of Asher’s Total Physical Response theo-
retical perspective is consistent with the embodied cogni-
tion view (Choi and Kim 2015). According to the
embodied cognition theory, motor movements or gestures
can activate images in working memory and help facilitate
encoding (Richardson et al. 2003). In other words, mean-
ingful gestures with speech have a positive influence on
verbal information memory and thus support young chil-
dren’s cognitive development (Macedonia et al. 2011).
Bui (2018) continues to add that pre-class preparations
and classroom teaching are two important stages of
implementing Asher’s Total Physical Response. Before the
lesson, teachers should set achievable objectives with ref-
erence to students’ language ability. Relevant visual and
audio materials can be prepared accordingly. During the
lesson, teachers can demonstrate an action and give its
corresponding commands. To strengthen students’ associ-
ation between the action and commands, vocabulary drills
are the major activities in which meaning can be clarified
via physical movements (Bui 2018). In early childhood
education, Er (2013) states that Total Physical Response is
most effective when the learning activities are reinforced
with games, songs, and stories. This kind of activities can
also create an enjoyable, fun, and interesting environment
to engage preschoolers in the learning process (Er 2013).
Designing CT Activities for Preschoolers
In this study, the plugged CT activity that we chose was
Bee-Bot (https://www.bee-bot.us/). Bee-Bot is a pro-
grammable robot with a mat (Fig. 2) which is suitable for
the children in the preoperational stage (Angeli et al. 2016).
The device (i.e., Bee-Bot) can be controlled by several
buttons, such as backward/forward and rotation to the left/
right buttons. Users can enter their programmed sequence
for executing a series of commands. The goal of the Bee-
Bot activity is to guide using commands of the Bee-Bot to a
targeted position, such as a treasure as printed on the mat,
without hitting obstacles (e.g., waterfall and forest).
Referring to Shute et al. (2017), the minimal CT skills
involved in the Bee-Bot activity are sequencing and algo-
rithm design. To establish a foundation for learning CT, an
introduction to pattern recognition can serve as a starter to
cultivate students’ sense of order (e.g., an array of sym-
bols). As discussed in the previous section, children in the
preoperational stage are able to learn these three CT skills.
They, however, largely rely on their perceptions (Sigelman
and Rider 2012). Therefore, teaching in an unplugged
environment using tangible materials can provide them
with more concrete experiences to cultivate their CT (Looi
et al. 2018).
The first set of our unplugged activities includes LEGO
pattern and sequencing stories (Fig. 3). LEGO pattern is a
hands-on pattern building activity for students to learn
pattern recognition. Starting with some simple color pat-
terns of LEGO bricks (e.g., orange–blue–orange–blue), the
sense of patterns can be cultivated. Teachers can further
ask their students to continue the pattern using suit-
able LEGO bricks. When students are able to recognize
those simple patterns, teachers can stretch their ability
using more complicated patterns (i.e., with both colors and
shapes varied). For sequencing stories, the main CT focus
is sequencing. Students are required to arrange several
pictures of story scenes in a correct sequence. Taking
‘‘Everyday events’’ as an example, there are six pictures of
daily routine. Students should arrange the pictures in the
order of (1) wash and brush teeth, (2) breakfast, (3) school,
(4) lunch, (5) playtime, and (6) go home. With the ability
of pattern recognition and sequencing, students are better
prepared to design a path with its corresponding sequence
of commands in the Bee-Bot activity.
The second set of our unplugged activities includes
Vocabulary building songs, Direction game through cards,
and Tic-Tac-Toe (Fig. 4). This set of activities leverages
the language learning strategy of Total Physical Response
(Asher 1977; Bui 2018;Er2013), aiming to visually and
verbally introduce students with necessary language to
Fig. 2 A Bee-Bot (lower right corner) with a Bee-Bot mat
58 A. Saxena et al.
123
express and apply their CT in the Bee-Bot activity.
Therefore, physical movements are incorporated in this set
of activities. Taking ‘‘Tic-Tac-Toe’’ as an example, it is a
game in which a student acts as a robot and a teacher (or
another student) gives verbal commands. These commands
are some positional and directional language, such as ‘‘turn
around’’ and ‘‘six steps forward.’’ Following these com-
mands, the student moves from one position to another.
According to Asher (1977), students can better acquire the
language through this kind of physical actions.
With the CT skills and necessary language acquired in
the above unplugged activities, teachers can introduce
algorithm design in the Bee-Bot activity (i.e., Direction
game with Bee-Bot). As a transition to this plugged
activity, teachers can first conduct the unplugged Direction
game using the Bee-Bot mat (Fig. 5). Arrow cards are used
when designing algorithms to guide the Bee-Bot. In the
words of Armoni (2012), ‘‘The goal is that this concrete
knowledge will in due time evolve or transfer to more
general and abstract contexts’’ (p. 19). Meanwhile, students
can get used to the setting and rules of the Bee-Bot activity.
Upon the completion of the unplugged Bee-Bot activity,
students can start their algorithm design and input their
commands into the Bee-Bot. To stretch their ability of
algorithm design, teachers can alter the difficulty levels of
the direction game, such as defining additional treasures/
obstacles and bringing the Bee-Bot back to the starting
position.
Method
Research Design and the CT Course
Figure 6summarizes the procedures for implementing our
study. In the first stage, we designed several unplugged and
plugged activities as mentioned in the ‘‘Designing CT
Activities for Preschoolers’’ section. With reference to
prior research in early childhood education (e.g., Hsu et al.
2018; Israel et al. 2015), we expected that not all kinder-
garten teachers were familiar with CT education. They
might get frustrated by new instructional practice. In fact,
the lack of computer skills and pedagogical knowledge are
also some major teacher perceptions about teaching CT in
other contexts, such as primary and secondary schools
(e.g., Ling et al. 2017; Wu et al. 2018).
To address the possible challenges to new instructional
practice, we drew on a highly cited framework for pro-
fessional development and teacher learning by Borko
(2004). His framework is based on the situated learning
theoretical perspective. In the words of Adler (2000), sit-
uated theorists view teacher learning as ‘‘a process of
Fig. 3 Major unplugged
activities to cultivate CT
Fig. 4 Major unplugged
activities to acquire language
for the Bee-Bot activity
Designing Unplugged and Plugged Activities to Cultivate Computational Thinking: An Exploratory…59
123
increasing participation in the practice of teaching, and
through this participation, a process of becoming knowl-
edgeable in and about teaching’’ (p. 37). From the situative
perspective, what people learn is grounded in the contexts
and activities in which they learn (Greeno et al. 1996).
Borko (2004) thus summarized that teacher learning occurs
in multiple aspects, involving facilitators, teachers, pro-
fessional development programs, and contexts (Fig. 7).
In Borko’s (2004) framework, teachers must first
understand the central facts about the subject they teach. It
is therefore important that professional development pro-
grams, led by experienced facilitators, focus explicitly on
the specific subject matter. The facilitators must understand
the goal of the program well. Borko (2004) further argued
that the use of the teachers’ authentic classrooms provides
essential contexts for facilitating teacher learning because
they can fully relate and apply what they have learned into
actual practice.
In this study, the first author (the facilitator) offered a
2-h training workshop as a professional development pro-
gram for five kindergarten teachers, during which the
concepts and instructional strategies of CT were intro-
duced. The facilitator is an experienced CT teacher who
had received doctoral level training in conducting CT
activities. In addition to the workshop, the teachers learned
how to conduct the CT activities by observing the CT
lessons of the facilitator (Teacher A).
In the next stage, we administered our CT course in a
kindergarten in Hong Kong. Three individual classes were
run, involving three groups of preschoolers (an average of
3 to 4 students each). The CT course was 1 week in
duration, consisting of five 2-h lessons (i.e., a total of 10 h).
We delivered five CT activities in each lesson. Each
activity lasted for 20 to 30 min. Although some activities
were rerun throughout the course in difference lessons, the
difficulty level increased progressively. For example, the
first activity of each lesson was LEGO pattern. During the
first few lessons, students were required to describe some
simple LEGO patterns (i.e., with different colors only) and
continue the patterns using LEGO bricks. Toward the end
of the course, more complicated patterns (i.e., with both
colors and shapes varied) were presented. A detailed course
rundown is provided in Appendix 1.
Research Context and Participants
As mentioned above, this study was conducted in a
kindergarten. In Hong Kong, pair teaching is a common
practice in preschool settings. Such a practice not only
facilitates student learning but also enhances classroom
management. Therefore, each class was taught by two
teachers; of the three classes, a total of six teachers par-
ticipated in the study (Table 1). Their teaching experience
ranged from 3 to more than 15 years. All the teacher par-
ticipants were novice CT educators, except for Teacher A
who had received doctoral level training in CT instructions.
Our CT course was offered as an enrichment course
during summer. Student participation was entirely volun-
tary. All K1 (aged 3 to 4), K2 (aged 4 to 5), and K3 (aged 5
to 6) students in the kindergarten could enroll in the course.
We successfully obtained a parental consent for study from
Fig. 5 Direction game through arrow cards with a Bee-Bot mat
Fig. 6 Procedures for implementing the study
Fig. 7 Framework for the professional development (PD) system
60 A. Saxena et al.
123
11 student participants (N
K1
=3; N
K2
=6; N
K3
= 2).
Despite the small number of student participants, this study
lays some important groundwork for us to examine and test
our unplugged and plugged CT activities in early childhood
education. It can provide insights for other researchers to
scale-up our study in other preschool contexts.
Data Collection and Analysis
Our research questions were addressed using three major
sources of data, including performance assessments, lesson
observations, and teacher interviews. Figure 8provides an
overview of the data sources and their corresponding
purposes.
To assess students’ CT learning, we adopted the 6-point
Likert scale (ranged from 0 to 5) rubric of performance
assessments by Bers et al. (2014). As they defined, a score
of 4 or above is ‘‘the target level of achievement’’ (p. 149).
Their assessment rubric was developed in the context of
early childhood CT education, and thus suitable for our
study (see Appendix 2). Three assessments were conducted
toward the end of the CT course, including LEGO pattern
(pattern recognition), Sequencing stories (sequencing), and
Direction game with Bee-Bot (algorithm design). To
enhance the reliability, student performance was rated by
two researchers. Inter-rater reliability was high (91%). In
the event of disagreements, the two researchers would
review the lesson recordings together to come to a
consensus.
Lesson observations had been done throughout the
course as children’s verbalization and actions on CT
materials reflect their thought processes (Ojose 2008).
Similar to Israel et al. (2015), we took detailed field notes
of student performance and interactions as well as teachers’
instructional practices during the CT activities. Lessons
were video-recorded and transcribed in order to detect
excerpts that could provide information to address the
research questions (Fessakis et al. 2013). To protect stu-
dents’ privacy, we had ensured that their faces were outside
the camera view or having their images blurred in any
forms of public disseminations. All field notes were typed
and shared among researchers for analysis (Israel et al.
2015).
Finally, all the six teacher participants were interviewed
to understand, from their perspectives, students’ CT
learning and their implementations of the CT activities. We
adopted a semi-structured interview approach, and the
interview protocol was developed based on Israel et al.
(2015). For example, ‘‘Have you faced any challenges in
implementing [the CT activities]? Probe for additional
information and examples’’ (p. 278). We first transcribed
the interview data, and then performed a series of quali-
tative data analysis procedures proposed by Creswell
(2012). The interview data were thematically analyzed and
organized into categories by the second and third authors.
To enhance the consistency of coding, exemplary quotes
were identified to illustrate each constructed category/sub-
category. Any disagreements between the two coders were
resolved through discussion to come to a consensus.
Fig. 8 Overview of the data
sources
Table 1 Information on the teacher participants
Class Teacher Teaching experience Experience of CT instructions
Class 1 A (the first author) Over 15 years Received doctoral level training in CT instructions
B 7 years Novice
Class 2 C 6 years Novice
D 3 years Novice
Class 3 E 13 years Strong interest in teaching CT and coding
F 6 years Novice
Designing Unplugged and Plugged Activities to Cultivate Computational Thinking: An Exploratory…61
123
Findings
The findings are presented in two subsections according to
the sequence of the research questions.
RQ1: How do Preschoolers Perform in the CT
Activities?
Figure 9shows that almost all students were able to reach
the level of complete achievement or mostly complete
achievement in the assessments of pattern recognition
(LEGO pattern) and sequencing (Sequencing stories). For
algorithm design (Direction game with Bee-Bot), however,
only 7 out of 11 students could be rated as reaching the
target level of achievement. Those students who had got
partially complete achievement were from K1 (n= 3) and
K2 (n= 1).
From the teacher interviews, further qualitative evidence
could be identified to support the above quantitative results
of student performance. Teachers’ comments included
•Pattern recognition: After the LEGO activity, they (the
students) learned patterns and shapes. (Teacher F)
•Sequencing: In the unplugged sequencing activities,
children were able to give the correct sequence of story
cards. (Teacher A)
•Algorithm design: Children could follow the instruction
and design a path to move the robot on the mat.
(Teacher B)
Regarding the K1 students’ failure to reach the target level
of achievement, some teacher participants offered the
following explanations:
•I would say directional understanding of K1 students
was challenging. (Teacher F)
•K1 children faced difficulty in saying or linking
directional vocabularies to the directions. (Teacher A)
RQ2: How do Preschool Teachers Perceive the CT
Activities?
Overall, the teacher participants reported positive senti-
ments about the use of our CT activities with their students.
First, some teachers (n= 3) found teaching CT to be fun
and interesting. As they expressed during the interviews,
Teaching them [the students] sequencing and directional
games was fun, and they learned a lot through those
unplugged activities. (Teacher B)
Second, all teachers (n= 6) confirmed that the unplugged
activities could provide their students with concrete
experiences to cultivate CT:
•It was good to start from concrete to abstract learning.
(Teacher F)
•Visual orientation and visual cards activities were
done. Tangible cards helped them to plan the path.
(Teacher E)
Third, several teachers (n= 4) explicitly mentioned that
they found the use of the unplugged activities to be useful.
More specifically, these activities helped the students apply
the CT skills to the plugged activity (i.e., the Bee-Bot
activity). Their views are extracted as follows:
•In my class, children learned more about patterns and
applied it in coding through arrow cards and eventu-
ally the Bee-Bot. (Teacher F)
•Taking some concepts through unplugged activities that
they already have some experiences with and using
these to apply with the technology helped them to
respond quickly and understand better. (Teacher E)
Despite the positive sentiments, three major practical
challenges of our CT intervention were identified. First,
most of our teacher participants (n= 5) lacked the
knowledge of CT instructions. As they expressed during
the interviews,
•I went to the bookfair and I got exposed to this term
[computational thinking]. I don’t know much but it
looks like high-tech things. (Teacher B)
Regarding the training (i.e., workshop and class observa-
tion) that we provided, the teacher participants (n=5)
found that class observation was an effective way for them
to equip the instructional strategies of CT education. As
one teacher mentioned,
•I have got learning and teaching strategies from
[Teacher A]. Observations help when I don’t have
experience of how to teach CT. (Teacher C)
The second challenge was about learner diversity in class.
The teachers (n= 4) from two classes specifically pointed
Fig. 9 Student performance of each CT skill (Note: a score of 4 or
above is the target level of achievement)
62 A. Saxena et al.
123
out that the K1 students got confused with the direction
while the others did not:
•Mixed ability kids—K2 knows direction but K1 students
were not able to follow direction. So, it was challeng-
ing. (Teacher C)
•It was hard as it was mixed age group. …Most of the
K2 and K3 students were good but I think the K1
students were struggling a bit with the activities.
(Teacher F)
•When they need to send the Bee-Bot to a specific
location on the map, some children got confused. They
had difficulty in visualizing multiple steps of the Bee-
Bot. (Teacher E)
Several teachers (n= 3) therefore suggested (1) strength-
ening the teaching of directional language and (2) dividing
students into different classes by age:
•Direction from Nursery rhymes was a great idea if it is
done for longer. (Teacher F)
•Next time, please use separate age group to learn as
their abilities are different. (Teacher C)
Besides the above challenges and their suggestions, a few
teachers (n= 3) foresaw that the current resources were
only enough for a week and suggested more resources be
developed:
•The CT course materials for this lesson planning were
good enough. But for longer coding class, we should
have more resources. (Teacher F)
However, the design and production of both plugged and
their corresponding unplugged activities were time con-
suming. In the words of one teacher participant, ‘‘At first I
was a bit overwhelmed’’ (Teacher F). Especially, designing
appropriate unplugged activities and their corresponding
materials to cultivate preschoolers’ CT required significant
intelligent input (Teacher A).
Discussion
The findings pertaining to each research question are dis-
cussed in two subsections: (1) student attainment and les-
sons learned about the activity design, and (2) practical
challenges and possible solutions. After that, we
acknowledge several limitations of this study and provide
recommendations for future research.
Student Attainment and Lessons Learned
about the Activity Design
CT is a basic skill that everyone, even preschoolers, should
equip (Chen et al. 2017; Grover and Pea 2013). In order to
cultivate their CT, we first provided them with some hands-
on learning experiences using the unplugged activities
(e.g., LEGO pattern). The theoretical foundation of such a
practice is based on Piaget’s Theory of Cognitive Devel-
opment. In early childhood education, the theory suggests
the use of tangible materials which can provide
preschoolers with more concrete experiences to cultivate
CT (Armoni 2012; Ojose 2008). We found that almost all
of our students could accomplish the tasks of pattern
recognition and sequencing. Extending previous research
on early childhood CT education (e.g., Bers et al. 2014;
Fessakis et al. 2013; Kazakoff and Bers 2012), this study
provides more concrete evidence that children in the pre-
operational stage could acquire these two CT skills in
unplugged environments.
Besides the unplugged activities for cultivating CT, we
designed several unplugged activities to equip our students
with the necessary language for the subsequent plugged
activity (i.e., the Bee-Bot activity). The importance of these
activities can be reflected in their possible confusion about
directional language. More specifically, even though we
had gone through this set of activities with our K1 students
(aged 3 to 4), they did not fully comprehend those
vocabularies/instructions (e.g., turn left/right) used in the
Bee-Bot activity. By contrast, Bers et al. (2014) did not
report such a problem when relevant language and
instructions were introduced. However, it is important to
notice that their student participants were 5 to 6 years old
while our study involved some younger students in K1
(aged 3 to 4) and K2 (aged 4 to 5). The findings of our
study indicate that more training on directional language is
needed especially for K1 students (aged 3 to 4) to establish
a solid foundation for subsequent CT activities.
In the plugged activity, most K2 and K3 students could
apply the concepts learned in the pattern recognition and
sequencing activities. The students could design a correct
path to guide the Bee-Bot even in some complicated
problems (e.g., having multiple treasures/obstacles defined
and bringing back the Bee-Bot). This finding echoed with
the existing literature that children at age 5 could start to
program (Bers et al. 2014; Fessakis et al. 2013; Kazakoff
and Bers 2012). Most importantly, our study provides
preliminary evidence that the use of the unplugged activ-
ities could foster most students’ accomplishment in the
plugged CT activity.
However, not all students could demonstrate their
mastery of algorithm design. We found that the K1 and a
few K2 students had difficulty in using directional lan-
guage during the Bee-Bot activity. Also, our teacher par-
ticipants pointed out that they ‘‘had difficulty in visualizing
multiple steps of the Bee-Bot’’ (Teacher E). Besides
strengthening their training on CT language, we can allow
these students to use arrows (as a temporary substitute of
Designing Unplugged and Plugged Activities to Cultivate Computational Thinking: An Exploratory…63
123
verbal commands) to represent their algorithm design.
Furthermore, differentiated instruction is another possible
strategy to cater to learner diversity in CT education. While
more advanced problems (e.g., up to 5 to 7 steps) can be
given to the more capable students, teachers can first offer
some problems that consist of 2 to 4 steps for the younger
or less capable ones. In other words, achievable goals
should be set according to students’ CT and CT language
ability, even though they are in the same age group or stage
of cognitive development.
Practical Challenges and Possible Solutions
Apart from learner diversity in the CT course, two practical
challenges were identified. First, our teacher participants
doubted their competence of CT instructions. Consistent
with Hsu et al. (2018) and Israel et al. (2015), there is a
need to provide professional training especially for those
teachers with no or limited CT knowledge. In addition to
CT workshops, another possible strategy that we used was
lesson demonstration. As our teacher participants expres-
sed, they could learn a lot (e.g., executing lesson plans and
CT activities) from observing authentic CT lessons. This
finding confirms Borko’s (2004) framework for profes-
sional development and teacher learning. More specifi-
cally, the facilitator had guided the teacher participants to
construct new knowledge and practices of CT instructions.
Second, although our teacher participants suggested
creating more CT resources for future practice, the design
and production of instructional materials required a con-
siderable investment of teacher effort. Especially, the
connection between unplugged and plugged CT activities
should be carefully established, making the preparation
work overwhelming for some teachers. However, Glass
et al. (1981) point out that the practical importance of an
intervention relies on its costs and benefits. Despite the
significant amount of start-up effort, the CT instructional
materials can be reused when rerunning the courses. Res-
onated with Looi et al. (2018), such effort could facilitate
students’ CT learning. The production of learning resour-
ces is therefore cost-effective in the long run. In future
practice, teachers can produce and accumulate CT resour-
ces progressively so that the preparation workload is
manageable.
Limitations and Recommendations for Future
Research
Although this study provides evidence that the unplugged
and plugged activities could cultivate preschoolers’ CT,
our findings should not be over-generalized due to two
limitations. First, our CT activities mainly focused on
pattern recognition, sequencing, and algorithm design.
Further research is required to examine whether our CT
activity design is applicable to cultivate other CT skills
(e.g., decomposition and debugging) in early childhood
education.
Second, although we presented both quantitative and
qualitative evidence of student performance in different
grades (i.e., K1, K2, and K3), the generalizability of our
findings was limited by the small number of student par-
ticipants. We therefore suggest future studies be scaled up
by involving more student participants with different ages
in the preoperational stage. Researchers may also consider
testing the feasibility of cultivating CT in pre-kindergarten
contexts.
Conclusion
This study tested the feasibility of cultivating CT in early
childhood education and has laid a useful preliminary
groundwork for the implementation of CT education in
preschool settings in Hong Kong. Based on Piaget’s The-
ory of Cognitive Development, we focused on cultivating
three CT skills (i.e., pattern recognition, sequencing, and
algorithm design) and developed several unplugged and
plugged CT activities for kindergarten students. Using
tangible materials, our unplugged activities aimed to pro-
vide students with more concrete experience of CT. Lev-
eraging Asher’s Total Physical Response, another set of
unplugged activities was designed to acquire students with
the necessary language for subsequent CT learning. Stu-
dents could thus have a better foundation for the plugged
CT activity that involved a digital device (i.e., the Bee-
Bot). We found that the K2 (aged 4 to 5) and K3 (aged 5 to
6) students could generally demonstrate their ability of
pattern recognition, sequencing, and algorithm design. By
contrast, the K1 students failed to design a correct algo-
rithm in some complicated problems. However, one should
exercise caution when viewing our findings because of the
small sample size. In future research, we suggest scaling up
this study by introducing more CT skills and involving
more student participants with different ages in preschool
settings.
Funding This research did not receive any specific grant from
funding agencies in the public, commercial, or not-for-profit sectors.
Appendix 1
See Table 2.
64 A. Saxena et al.
123
Appendix 2
See Table 3.
References
Adler, J. (2000). Social practice theory and mathematics teacher
education: A conversation between theory and practice. Nordic
Mathematics Education Journal, 8(3), 31–53.
Angeli, C., Voogt, J., Fluck, A., Webb, M., Cox, M., Malyn-Smith, J.,
et al. (2016). A K-6 computational thinking curriculum frame-
work: Implications for teacher knowledge. Educational Tech-
nology & Society, 19(3), 47–57.
Armoni, M. (2012). Teaching CS in kindergarten: How early can the
pipeline begin? ACM Inroads, 3(4), 18–19.
Asher, J. J. (1977). Learning another language through actions: The
complete teacher’s guidebook. Los Gatos, CA: Sky Oaks
Productions.
Bers, M. U., Flannery, L., Kazakoff, E. R., & Sullivan, A. (2014).
Computational thinking and tinkering: Exploration of an early
childhood robotics curriculum. Computers & Education, 72,
145–157.
Borko, H. (2004). Professional development and teacher learning:
Mapping the terrain. Educational Researcher, 33(8), 3–15.
Bui, G. (2018). Total Physical Response. In J. I. Liontas (Ed.), The
TESOL encyclopedia of English language teaching (pp.
927–932). Hoboken, NJ: John Wiley & Sons Inc.
Buitrago Flo
´rez, F., Casallas, R., Herna
´ndez, M., Reyes, A., Restrepo,
S., & Danies, G. (2017). Changing a generation’s way of
thinking: Teaching computational thinking through program-
ming. Review of Educational Research, 87(4), 834–860.
Case, R. (1992). The mind’s staircase: Exploring the conceptual
underpinnings of children’s thought and knowledge. Hillsdale,
NJ: Lawrence Erlbaum Associates.
Cejka, E., Rogers, C., & Portsmore, M. (2006). Kindergarten robotics:
Using robotics to motivate math, science, and engineering
literacy in elementary school. International Journal of Engi-
neering Education, 22(4), 711–722.
Chen, G., Shen, J., Barth-Cohen, L., Jiang, S., Huang, X., & Eltoukhy,
M. (2017). Assessing elementary students’ computational think-
ing in everyday reasoning and robotics programming. Computers
& Education, 109, 162–175.
Choi, D., & Kim, M. (2015). The effects of visual stimulation and
body gesture on language learning achievement and course
interest. Educational Technology International, 16(2), 141–166.
Creswell, J. W. (2012). Educational research: Planning, conducting,
and evaluating quantitative and qualitative research (4th ed.).
Boston: Pearson.
Education Bureau. (2016). Report on promotion of STEM education:
Unleashing potential in innovation. Hong Kong: Education
Bureau.
Education Bureau. (2017). Computational thinking – Programming
education: Primary school curriculum supplementary document
[In Chinese]. Retrieved November, 30, 2018, from
https://www.edb.gov.hk/attachment/tc/curriculum-development/
renewal/ct/supplement_ct_chi_draft.pdf.
Er, S. (2013). Using total physical response method in early childhood
foreign language teaching environments. Procedia – Social and
Behavioral Sciences, 93, 1766–1768.
Feldman, D. H. (2004). Piaget’s stages: The unfinished symphony of
cognitive development. New Ideas in Psychology, 22(3),
175–231.
Fessakis, G., Gouli, E., & Mavroudi, E. (2013). Problem solving by
5–6 years old kindergarten children in a computer programming
environment: A case study. Computers & Education, 63, 87–97.
Glass, G. V., McGaw, B., & Smith, M. L. (1981). Meta-analysis in
social research. Beverly Hills: Sage Publications.
Greeno, J. G., Collins, A. M., & Resnick, L. B. (1996). Cognition and
learning. In D. C. Berliner & R. C. Calfee (Eds.), Handbook of
educational psychology (pp. 15–46). New York, NY: Macmillan.
Grover, S., & Pea, R. (2013). Computational thinking in K-12: A
review of the state of the field. Educational Researcher, 42(1),
38–43.
Hsu, T. C., Chang, S. C., & Hung, Y. T. (2018). How to learn and
how to teach computational thinking: Suggestions based on a
review of the literature. Computers & Education, 126, 296–310.
Table 2 Activity plan of the CT course
Time Day 1 Day 2 Day 3 Day 4 Day 5
20 min LEGO pattern LEGO pattern LEGO pattern LEGO pattern LEGO pattern
20 min Story telling Sequencing stories Direction game with
cards
Direction game with
cards
Direction game with
cards
30 min Sequencing stories Direction game using a Bee-
Bot mat
Direction game with
Bee-Bot
Direction game with
Bee-Bot
Direction game with
Bee-Bot
30 min Vocabulary building
songs
Vocabulary building songs Vocabulary building
songs
Vocabulary building
songs
Vocabulary building
songs
20 min Direction game with
cards
Direction game with cards Tic-Tac-Toe Tic-Tac-Toe Tic-Tac-Toe
Table 3 The rubric of performance assessments for CT activities
(Bers et al. 2014, p. 149)
Score Description
5 Complete achievement of the goal, task, or understanding
4 Mostly complete achievement of the goal, task, or
understanding
3 Partially complete achievement of the goal, task, or
understanding
2 Very incomplete achievement of the goal, task, or
understanding
1 Did not complete the goal, task, or understanding
0 Did not attempt/Other
Designing Unplugged and Plugged Activities to Cultivate Computational Thinking: An Exploratory…65
123
Israel, M., Pearson, J. N., Tapia, T., Wherfel, Q. M., & Reese, G.
(2015). Supporting all learners in school-wide computational
thinking: A cross-case qualitative analysis. Computers & Edu-
cation, 82, 263–279.
Kazakoff, E., & Bers, M. (2012). Programming in a robotics context
in the kindergarten classroom: The impact on sequencing skills.
Journal of Educational Multimedia and Hypermedia, 21(4),
371–391.
Kazakoff, E. R., Sullivan, A., & Bers, M. U. (2013). The effect of a
classroom-based intensive robotics and programming workshop
on sequencing ability in early childhood. Early Childhood
Education Journal, 41(4), 245–255.
Ling, U. L., Saibin, T. C., Labadin, J., & Aziz, N. A. (2017).
Preliminary investigation: Teachers’ perception on computa-
tional thinking concepts. Journal of Telecommunication, Elec-
tronic and Computer Engineering, 9(2–9), 23–29.
Looi, C. K., How, M. L., Longkai, W., Seow, P., & Liu, L. (2018).
Analysis of linkages between an unplugged activity and the
development of computational thinking. Computer Science
Education, 28(3), 255–279.
Lye, S. Y., & Koh, J. H. L. (2014). Review on teaching and learning
of computational thinking through programming: What is next
for K-12? Computers in Human Behavior, 41, 51–61.
Macedonia, M., Mu
¨ller, K., & Friederici, A. D. (2011). The impact of
iconic gestures on foreign language word learning and its neural
substrate. Human Brain Mapping, 32(6), 982–998.
Manches, A., & Plowman, L. (2017). Computing education in
children’s early years: A call for debate. British Journal of
Educational Technology, 48(1), 191–201.
Ojose, B. (2008). Applying Piaget’s theory of cognitive development
to mathematics instruction. The Mathematics Educator, 18(1),
26–30.
Richardson, D. C., Spivey, M. J., Barsalou, L. W., & McRae, K.
(2003). Spatial representations activated during real-time com-
prehension of verbs. Cognitive Science, 27(5), 767–780.
Shute, V. J., Sun, C., & Asbell-Clarke, J. (2017). Demystifying
computational thinking. Educational Research Review, 22,
142–158.
Sigelman, C. K., & Rider, E. A. (2012). Life-span human development
(7th ed.). Belmont, CA: Wadsworth, Cengage Learning.
Wing, J. (2006). Computational thinking. Communications of the
ACM, 49(3), 33–35.
Wu, L., Looi, C. K., Liu, L., & How, M. L. (2018). Understanding
and developing in-service teachers’ perceptions towards teaching
in computational thinking: Two studies. In J. C. Yang et al.
(Eds.), Proceedings of the 26th international conference on
computers in education (pp. 735–742). Philippines: Asia-Pacific
Society for Computers in Education.
Young, G. (2011). Development and causality: Neo-Piagetian
perspectives. New York, NY: Springer.
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
66 A. Saxena et al.
123
