A Review on the Use of Robots in Education and Young Children

Abstract

A systematic review was carried out to examine the use of robots in early childhood and lower level education. The paper synthesizes the findings of research studies carried out in the last ten years and looks at the influence of robots on children and education. Four major factors are examined - the type of studies conducted, the influence of robots on children's behaviour and development, the perception of stakeholders (parents, children and educators) on educational robots, and finally, the reaction of children on robot design or appearance. This review presents the approach taken by researchers in validating their use of robots including non-experimental (mixed-method, anecdotal, cross-sectional, longitudinal, correlational, and case studies) and quasi-experimental (pre- and post-test). The paper also shows that robot's influence on children's skills development could be grouped into four major categories: cognitive, conceptual, language and social (collaborative) skills. Mixed results are shown when it comes to parents' perception of the use of robots in their children's education while design was shown to influence children's perception of the robot's character or capabilities. A total of 27 out of 369 articles were reviewed based on several criteria.

Full text

Toh, L. P. E., Causo, A., Tzuo, P. W., Chen, I. M., & Yeo, S. H. (2016). A Review on the Use of Robots in Education and Young
Children. Educational Technology & Society, 19 (2), 148–163.
148
ISSN 1436-4522 (online) and 1176-3647 (print). This article of the Journal of Educational Technology & Society is available under Creative Commons CC-BY-ND-NC
3.0 license (https://creativecommons.org/licenses/by-nc-nd/3.0/). For further queries, please contact Journal Editors at ets-editors@ifets.info.
A Review on the Use of Robots in Education and Young Children
Lai Poh Emily Toh1, Albert Causo1*, Pei-Wen T zuo2, I-Ming Chen1 and Song Huat Yeo1
1Robotics Research Centre, School of Mechanical and Aerospace Engineering, Nanyang Technological University,
Singapore // 2National Institute of Education, Singapore // emilytohlp@ntu.edu.sg // acauso@ntu.edu.sg //
peiwen.tzuo@nie.edu.sg // michen@ntu.edu.sg // myeosh@ntu.edu.sg
*Corresponding author
(Submitted August 13, 2014; Revised February 24, 2015; Accepted August 27, 2015)
ABSTRACT
A systematic review was carried out to examine the use of robots in early childhood and lower level education.
The paper synthesizes the findings of research studies carried out in the last ten years and looks at the influence
of robots on children and education. Four major factors are examined – the type of studies conducted, the
influence of robots on children’s behaviour and development, the perception of stakeholders (parents, children
and educators) on educational robots, and finally, the reaction of children on robot design or appearance. This
review presents the approach taken by researchers in validating their use of robots including non-experimental
(mixed-method, anecdotal, cross-sectional, longitudinal, correlational, and case studies) and quasi-experimental
(pre- and post-test). The paper also shows that robot’s influence on children’s skills development could be
grouped into four major categories: cognitive, conceptual, language and social (collaborative) skills. Mixed
results are shown when it comes to parents’ perception of the use of robots in their children’s education while
design was shown to influence children’s perception of the robot’s character or capabilities. A total of 27 out of
369 articles were reviewed based on several criteria.
Keywords
Early childhood education, Lower education, Educational robots, Review
Introduction
With the rapid development of technology in the 21st century, the use of multi-media tool in education has become
increasingly popular. Notwithstanding their usual engineering applications, robots are being used more in schools.
According to Beran et al. (2011), children are also playing more with technologically advanced devices during their
playtime. Subsequently, studies were conducted to investigate robot use’s influence on children’s cognition,
language, interaction, social and moral development (Wei et al., 2011; Kozima & Nakagawa, 2007; Shimada, Kanda
& Koizumi, 2012; Kahn et al., 2012). Recent studies (Wei, Hung, Lee & Chen, 2011; Highfield, 2010; Chen, Quadir
& Teng, 2011) reported that robot use encourages interactive learning, making children more engaged in their
learning activities. This increase research on robot application to education needs systematic look at the direction
taken this past decade in order to elucidate a roadmap for future studies.
Recent reviews on the use of robots in education show the challenges faced by researchers in this field. Benitti
(2012) points out that more than 70 papers could have qualified in his review work but only 10 provided quantitative
measurement on the use of robots in education. From these ten papers, only those that discuss the potential of using
robots in all level of education and highlight the non-engineering benefits were selected.
Mubin et al. (2013) analysed research works from through the actual robots used. The major factors identified were
robot’s role, type (physical form), behaviour (capabilities and interaction capacity), learning activity type, and venue
(inside or outside of classroom) where learning takes place. Mubin et al. (2013) and Benitti (2012) find similarity on
the topics where robots were being used in education – learning language, science, and technology. Although Mubin
et al. (2013) differs by pointing out the various roles played by the robot in education – as tutor, tool, or peer.
The reviews provide good starting points for researchers, the criteria (Benitti, 2012) and perspective (Mubin et al.,
2013) taken by these two papers could potentially miss those that could be relevant to researchers in the field.
Moreover, other factors critical in the use of robot in education may have been overlooked, like the effect of design
on interaction or the importance of parent’s perception in the success of implementing a robot-in-education project.
The aim of this paper is to assess the effectiveness of using robots in studies published within the last decade. We
look at effectiveness as having four sub-factors – the study type done by the researcher, the influence of the robots on

149
the behaviour and development of students, the perception of stakeholders (parents, educators and children) about the
robots, and the importance of design or robot appearance. To achieve this aim, we would focus on articles on the
application of robots in early childhood and lower level education and evidence for the factors would be analysed.
The rest of the paper is organized as follows. The review approach, especially the search and selection strategies, is
discussed in details in the next section. The discussions on the four factors above are described in the succeeding
sections. The conclusion provides a summary and presents the remaining challenges in this research field.
Review approach
To limit the papers to be reviewed, we implemented a search and selection strategy using specific keywords in
electronic databases. We started with 369 articles and narrowed it down to 27.
Search strategy
Articles reviewed were limited to those published in English from 2003-2013. To gather as many papers as possible,
five major databases were searched: IEEE Xplore, Academic Search Premier, ERIC (Educational Resources
Information Center, ScienceDirect, and SpringerLink. Only articles published in journals have been included for
review, with some exceptions.
Initially, search terms like “robots” and “education” was keyed in but in order to narrow down the result, we used a
similar approach to what Benitti (2012) employed. Table 1 shows the five databases and the keywords used for each
one.
Table 1. Summary of search protocol
Database
Search protocol
IEEE Explore
(((((“robots”) AND “education”) AND “learning”) AND “ teaching” ) AND “robotic”) under
advanced search options < Journal & Magazines>, < Publication Year : 2003-2013>, < Full
Text and Metadata>
Academic Search
Premier
“Robots” AND “Education” AND “Learning” Search <Full Text>, <Date Published: 2003 to
2013>, <Peer Reviewed Scholarly Journal>
ERIC
“Robots” AND “Education” AND “ Learning” Search < Full Text> <Peer reviewed>
<Journal>, <Date Published: from 2003 to 2013>
Science Direct
Search Terms: ‘Robots’ AND ‘Education’ AND ‘child’ and ‘learning’ AND LIMIT-To
(topics, “child, robot”) AND LIMIT-To (Topics, “child, robot”), <Date Published: Year: 2003
to 2013>
Springer Link
Search Terms: “education” AND “robots”, Search under: <Education and Language>,
<Learning and Instruction>
Selection strategy
This review focuses on articles that reported the use of robot in early childhood education. Selected studies were
relevant from early to secondary education context and focused on robot or robotics influence on learning,
pedagogical and developmental domains. The studies selected should report the use of robots as an educational tool.
Given the broad inclusion criteria, we managed to find 369 articles in all (see Table 2). To further narrow down the
scope of the review, the following exclusion criteria have been implemented:
 Exclusion Critera E1: Article reported the technical use of robots, designs or innovations.
 Exclusion Critera E2: Article reported robotics as a teaching subject.
 Exclusion Critera E3: Article reported studies conducted in higher or university education.
 Exclusion Critera E4: Article reported the use of robots as assistive technologies.
 Exclusion Critera E5: Article did not mention on the use of robots in education.

150
As shown in Table 2, with the above exclusion parameters, only 27 papers were left. A large number of papers were
excluded due to the focus on robots or robotics as the teaching subject (a total of 132 articles based on E2). Most of
the engineering articles excluded mentioned the use of robot in education in passing or as a justfication for its design;
115 articles were removed based on E1. Moreover, around 12% of articles were excluded because robots were
reported as an educational tool for higher education.
Table 2. Summary of selection
Database
Selected
articles
Total
reviewed
Excluded criteria articles
E1
E2
E3
E4
E5
IEEE Explore
11*
59
16
15
10
3
4
Academic Search Premier
5
188
79
70
25
9
0
ERIC
4*
10
0
3
0
0
3
Science Direct
3
46
14
0
0
8
21
Springer Link
4*
66
6
44
11
1
0
TOTAL
27
369
115
132
46
21
28
Note. *One paper in Academic Search Premier and one in ERIC, are repeated in SpringerLink.
From the selected paper, the following details were examined: the purpose of the study, the sample size of the
students involved in the experiments, the description of the setting, data collection and analysis methods, presented
results and the implication of the studies.
Discussion
Four major factors are focused on in this paper: the type of studies conducted, the robot use’s influence on child
behaviour and development, stakeholder perception, and children’s reaction to robot design or appearance.
Types of studies conducted
Majority of the reviewed papers employed non-experimental studies. There were three studies involving the use of
survey, where video was used to record children’s behaviour and interaction with the robots. Four quasi-experimental
studies involved pre-test and post-test, which were conducted with control group. There were ten anecdotal case
studies, five mixed-method studies and one correlational study. There were three experimental studies and one short
review paper. The detail of each study approach is listed in Table 3.
Table 3. Types of study reported in the reviewed papers
Type of study
Papers
Non-experimental (Mixed-method Study)
Williams et al., 2007; Levy & Mioduser, 2008; Liu, 2010; Yo un g
et al., 2010; Sugimoto, 2011
Non-experimental (Anecdotal case studies)
Barker & Ansorge, 2007; Rusk et al., 2008; Highfield, 2010;
Hong et al., 2011; Chang et al., 2010; Chen, Quadir & Teng,
2011; Slangen et al., 2011; Varney et al., 2012
Non-experimental (Cross-sectional survey)
Woods, 2006; Lin et al., 2012
Non-experimental (Longitudinal survey study)
Ruiz-del-Solar & Avi l és, 2004
Non-experimental (Case studies)
Bers, 2010; Bers & Portsmore, 2005
Non-experimental (Correlational study)
Bers, 2010
Quasi experimental (Pre-test & Post-test)
Barker & Ansorge, 2007; Whittier & Robinson, 2007; Chambers
et al., 2008; Kazakoff et al., 2013
Experiment study
Beran et al., 2011; Salter et al., 2004; Michaud et al., 2005
Short review paper
Cangelosi et al., 2010

151
Robot’s influence on children’s behaviour and development
The reviewed articles revealed four major themes where robot was able to aid in child’s behaviour or development.
Theme 1: Problem-solving abilities, team skills and collaboration
Studies by Barak (2009) and Varney et al. (2012) were conducted to investigate how the introduction of robots could
change education, especially to help prepare children with 21st century skills and to increase student interest in
robotics. The study conducted by Barak (2009) showed that high school students were able to come up with
inventive solutions to problems and could benefit from working on project-based programmes. Robotic kits such as
LEGO Mindstorm allowed students to work in teams as they carried out their projects in small groups.
Robotics was further viewed as an effective tool to develop “team skills” in students (Varney et al., 2012). The use of
robots in various activities with young children supports constructivism as a learning method. Students discuss, solve
problems, work with their peers, and combine their knowledge in order to construct their robots. In Chang et al.
(2010), the results from the study further supported that robots could create an interactive and engaging learning
experience.
Robots in elementary school helped promote collaboration and problem-solving skills in children as they became
involved in the process and construction of their artefacts for their robotic projects. This was further highlighted by
Hong et al. (2011) study where robots allowed children to engage in deep reflection as they solve problems and
collaborate with their peers, both of which enhanced their learning experience.
Theme 2: Achievement scores, science concepts and sequencing skills
The study conducted by Baker and Ansorge (2007) examined students’ achievement scores with the use of robots in
their science curriculum. Robots were found to be effective at teaching 9-11 year old students science, engineering
and technical concepts. Results from another experiment study conducted by Kazakoff et al. (2013) supported the use
of the robotic programming such as CHERP, a tangible programme which helped increase sequencing skills in pre-
kindergarten and kindergarten children.
Table 4. Articles that reported on skills development
Papers
Skills
Barker & Ansorge, 2007
Results showed increase mean scores from pre- to post-test, indicating that robotics
was effective at teaching youth about science, engineering, & technology concepts.
Williams et al., 2007
Study shows a significant difference on acquiring physics knowledge but not for
science inquiry skills
Barak, 2009
Study reveals that students often come up with inventive solutions to problem when
learning with robots.
Highfield, 2010
The result significantly showed that children engaged in multiple mathematical
processes; they demonstrated perseverance, motivation & responsiveness.
Whittier & Robinson, 2007
The results showed that all students obtained significant gains in their conceptual
understanding. There is an increase of mean pre-test from 26.9% to post-test
42.3%.
Kazakoff et al., 2013
Results indicated that the sequencing ability of pre-kindergarten and kindergarten
students increases when participating in an intensive robotics and programming
curriculum.
Bers, 2010
The result showed that boys had a higher mean score than girls on more than half of
the tasks. Boys scored significantly higher than girls in properly attaching robotic
components and programming using ‘Ifs’.
Slangen et al., 2011
Robots helped challenge pupils to manipulate, reason, predict, hypothesize, analyze
and test.

152
The use of robot to assist non-English speaking students to improve in their understanding of science concepts was
carried out by the Whittier and Robinson (2007) study. Results showed that all students obtained sufficient gains in
their science conceptual knowledge with an increase from 26.9% in pre-test to 42.3% in post-test. The middle school
students developed problem-solving skills, inquiry and engineering design skills. Robots were also used to develop
and improve learning of science concepts, technology and problem-solving, which was further supported by Barak’s
(2009) qualitative analysis of observations, interviews and reflections of students working on their projects.
Similarly, anecdotal records in the Highfield (2010) study showed that robotic toys could be catalyst for
mathematical problem solving through participation in multi-faceted approach by integrating and inter-relating
concepts and skills through dynamic tasks. The use of robotic to develop of physics content knowledge showed a
significant difference but not for the science inquiry skills, according to the Williams et al. (2007) study. Table 4
shows a summary of the skills where robot has a positive effect.
Theme 3: Language skills development
In the study by Chang et al. (2010), a humanoid robot was used to teach a second language in a primary school.
Results showed that robots could create interactive and engaging learning experiences as the children responded with
high motivation. The use of robots for language development was found to be advantageous as it also allowed for
demonstration of highly mobile behaviour and extensive repetition. Sugimoto (2011) used robot for storytelling,
where the robot was used in students’ learning and provided opportunity for children to learn in a mixed-reality
environment. The children engaged strongly in story expression and acted in a coordinated manner while also being
involved in their story creation with their robots.
Table 5. Articles with focus on language skills development
Papers
Overview of paper on language skills development
Chang et al., 2010
Results indicate that robots could create interactive and engaging learning
experience for students.
Young et al., 2010
Quantitative results showed that 95% have positive attitude towards tangible
learning companions/robots. They become more active in practicing
conversation.
Hong et al., 2011
Students were highly involved and reflective during the construction of their
artefacts.
Varney et al., 2012
Results showed that robot could be used as an effective tool in children to develop
‘team skills’; 75% of students actively raised questions.
Sugimoto, 2011
In the study, the children engage strongly in story expression and acted in a
coordinated manner.
Chambers et al., 2008
Results suggested that providing children with physical experiences were not
sufficient to understand mechanical concepts. Timely & appropriate
intervention is important.
Bers, 2010
TangibleK robotics could be implemented in the early childhood setting in a
developmentally appropriate way by integrating other disciplines.
Rusk et al., 2008
Results suggested multiple paths for engagement of children, teens, families and
educators.
Levy & Mioduser, 2008
The role of adult’s interaction enables children to shift into more complex
technological rules.
Varney et al., 2012
The study presented results on the efficacy of the LEGO robotic programme in
fostering student’s interest.
Ruiz-del-Solar & Avilés, 2004
Social robots were effective in fostering students’ interest in engineering.
Michaud, et al., 2005
Roball, a robot capable of autonomous motion, was used in child-development
studies.
Cangelosi et al., 2010
Studied embodied cognitive agent-humanoid robot. Discussed areas such as
complex sensorimotor, linguistic & social learning skills.
Chen, Quadir & Teng, 2011
The use of robot with computer and book enhanced students’ concentration in
their learning of English, interest and motivation.

153
According to Slangen et al. (2011), students working on projects using LEGO and Mindstorms were found to be
involved in frequent process of comparing their test results with their objectives, expectations, and in refining their
conceptual knowledge and skills. Table 5 summarizes the articles that reported on the use of robots for language
skills development.
Theme 4: Participation
Rusk et al. (2008) introduced Picocricket invention kit program to increase participation from children, teens,
families and educators in robotics-related endeavors via workshops, after-school programs and professional
development programs. The workshops allowed students to work on broad themes based on their own interests. As
these students were given the opportunity to combine art and engineering, encouraged to use storytelling and
exhibition and introduced to new technologies, their interest in robotics increased.
Parents’, educators’, and children’s perception of educational robots
Liu (2012) and Ruiz-del-Solar and Avilés (2004) investigated perception of parents, children and teachers on the use
of educational robots. The results from Lin et al. (2012) revealed that most parents’ would consider educational
robots as beneficial for their children. However, parents felt that they were less confident when playing and teaching
their children on using robots.
Ruiz-del-Solar and Avilés (2004) studied the children’s degree of satisfaction on robot use, their inquired level of
competence and their eventual interest to pursue an engineering career. 700 children and teachers were surveyed in
that study and 86% of the participants would consider studying in an engineering or science university in the future.
In the Bers (2010) study, educators developed computational thinking and learning about the engineering design
process in young children by introducing the TangibleK programme. It integrated other disciplinary learning in a
developmentally appropriate way for young children. Table 6 provides the list of articles and their reports on
stakeholder perception on using robots in education.
Table 6. Perception of different stakeholders
Papers
Perception
Beran et al., 2011
Results from frequency and content analysis suggested that a significant proportion of
children ascribe cognitive, behavioural, and affective characteristics to robots.
Salter et al., 2004
Findings suggested that touch could have an important role to play when developing
natural human-robot interfaces. It further suggested that robot interaction levels could
vary to suit different children.
Woods, 2006
Results showed that although the robots are very human-like, children were still
capable of distinguishing them from humans. However, the robots evoke a feeling of
discomfort or repulsion.
Liu, 2010
Results showed children regard
 Educational robot as a plaything;
 Studying robotics as a source of employment;
 Learning of robotics as a way to high tech.
Male and female perceptions differ.
Lin et al., 2012
Results indicated that parents considered educational robots as beneficial for their
children. But they were less confidence in playing and teaching with educational
robots with their children themselves.
Bers & Portsmore, 2005
Engineering students gained insight into the educational system and issues involved in
incorporating ICT into the classroom. Pre-service teachers saw the potential offered
by technology and what they would need to know to continue using it.

154
Children’s reaction to robot’s design or appearance
Levy and Mioduser (2008) presented rich anecdotal data on children’s descriptions and explanations of robots’
behaviour. Their study involved children in two strands of tasks (description and construction). It also showed that
when adult facilitate and interact with the children, they were capable of shifting into more complex technological
rules. In addition, a study conducted with 184 (Beran et al., 2011) showed that a significant proportion of the
children ascribe cognitive, behavioural, and affective characteristics to robots.
159 children were asked to evaluate 40 images of robot through questionnaires in order to investigate how children
perceive robot’s appearance (Woods, 2006). The study showed that children perceive robots’ intentions and
capabilities based on robot appearance. Children judged human-like robots as aggressive and machine-like ones as
friendly. Sullivan and Bers (2012) showed using the TangibleK programme that the boys scored significantly higher
than girls in properly attaching robot components and programming using “Ifs.” However, as reported for the rest of
the tasks gender differences were statistically insignificant.
Conclusion
The effectiveness of robots in education programme could be analyzed from different aspects: Study design in order
to report meaningful and statistically significant results, robot’s effects on child’s behaviour and development,
relevance of stakeholders’ perception on using robots in and outside of classroom setting, and users’ reaction
(especially the children) to the robot’s design.
Researchers, majority of whom relied on non-experimental methods, implemented various approach to validate their
studies. However this just shows that experimental methods are sorely lacking; quantitative analysis is needed, as
pointed by Benitti (2012).
In education, the use of robots has the potential to help children develop various academic skills like science process
understanding, mathematical concept development and improvement of achievement scores (Barker & Ansorge,
2007; Williams et al., 2007; Highfield, 2010). In addition, the introduction of robotics in curriculum also increases
children’s interest in engineering. As reported in Chang et al., 2010, the use of robots in education allows children to
engage in interactive and engaging learning experiences. Robots seem appropriate to use in language skill
development because it allow for a richer interaction (Sugimoto, 2011; Chambers et al., 2008 ; Bers, 2010; Chang et
al., 2010; Young et al., 2010).
Two new factors have emerged in this review paper: the stakeholder’s perception and the value of robot design.
Aside from the main users (children), parents and educators have to be on-board as well in order to increase the
chances of success of this kind of programmes. Lack of parental support would confine educational robots to
applications only inside the classroom.
Lastly, design is usually the last consideration when incorporating robots into an application. However, as Woods
(2006) and Sullivan & Bers (2013) studies showed, design could make a difference on robot perception and hence,
how the children would interact with it. Unfortunately, not a lot of work has been done yet on this question.
Past studies are like beacons on where research have been and indicates various milestones (e.g., Cangelosi et al.,
2010). This paper shows a possible roadmap and highlights research gaps in this field.
Acknowledgements
The authors would like to thank the Singapore Millennium Foundation for supporting this work.

155
References
Barker, B. S., & Ansorge, J. (2007). Robotics as means to increase achievement scores in an informal learning environment.
Journal Research on Technology in Education, 39(3), 229-243.
Barak, M., & Zadok, Y. (2009). Robotics projects and learning concepts in science, technology and problem solving. International
Journal Technology & Design Education, 19(3), 289-307.
Beran, T. N., Ramirez-Serrano, A., Kuzyk, R., Fior, M., & Nugent, S. (2011). Understanding how children understand robots:
Perceived animism in child–robot interaction. International Journal Human-Computer Studies, 69(7–8), 539-550.
Benitti, F. B. V. (2012). Exploring the educational potential of robotics in schools: A Systematic review. Computers & Education,
58(3), 978–988.
Bers, M. U. (2010). The TangibleK robotics program: Applied computational thinking for young children. Early Childhood
Research & Practice, 12(2), n2.
Bers, M. U., & Portsmore, M. (2005). Teaching partnerships: Early childhood and engineering students teaching math and science
through robotics. Journal Science Education and Technology, 14(1), 59-73.
Cangelosi, A., Metta, G., Sagerer, G., Nolfi, S., Nehaniv, C., Fischer, K., Jun Tani, Belpaeme, T., Sandini, G., Nori, F., Fadiga, L.,
Wrede, B., Rohlfing, K., Tuci, E., Dautenhahn, K., Saunders, J., & Zeschel, A. (2010). Integration of action and language
knowledge: A Roadmap for developmental robotics. IEEE Transactions on Autonomous Mental Development, 2(3), 167-195.
Chambers, J. M., Carbonaro, M., & Murray, H. (2008). Developing conceptual understanding of mechanical advantage through
the use of Lego robotic technology. Australasian Journal Educational Technology, 24(4), 384-401.
Chang, C. W., Lee, J. H., Chao, P. Y., Wang, C. Y., & Chen, G. D. (2010). Exploring the possibility of using humanoid robots as
instructional tools for teaching a second language in primary school. Educational Technology & Society, 13(2), 13–24.
Chen, N. S., Quadir, B., & Teng, D. C. (2011). A Novel approach of learning English with robot for elementary school students. In
M. Chang et al. (Eds.), Edutainment 2011, LNCS 6872 (pp. 309–316). Heidelberg, Germany: Springer-Verlag Berlin Heidelberg.
Highfield, K. (2010). Robotic toys as a catalyst for mathematical problem solving. Australian Primary Mathematics Classroom,
15(2), 22-27.
Hong, J. C., Yu, K. C., & Chen, M. Y. (2011). Collaborative learning in technological project design. International Journal
Technology & Design Education, 21(3), 335-347.
Kahn Jr, P. H., Kanda, T., Ishiguro, H., Freier, N. G., Severson, R. L., Gill, B. T., Ruckert, J. H., & Shen, S. (2012). “Robovie,
you’ll have to go into the closet now”: Children’s social and moral relationships with a humanoid robot. Developmental
Psychology, 48(2), 303-314.
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 Educational Journal, 41, 245–255.
Kozima, H., & Nakagawa, C. (2007). A Robot in a playroom with preschool children: Longitudinal field practice. In Proceedings
of 16th IEEE International Conference Robot & Human Interactive Communication (ROMAN 2007) (pp. 1058-1059).
doi:10.1109/ROMAN.2007.4415238
Levy, S.T., & Mioduser, D. (2008). Does it “Wa n t ” or “Was it programmed to...”? Kindergarten children’s explanations of an
autonomous robot’s adaptive functioning. International Journal Technology and Design Education, 18(4), 337-359.
Lin, C. H., Liu, E. Z. F., & Huang, Y. Y. (2012). Exploring parents’ perceptions towards educational robots: Gender and socio-
economic differences. British Journal of Educational Technology, 43(1), E31-E34.
Liu, E. Z. F. ( 20 10 ) . Early adolescents’ perceptions of educational robots and learning of robotics.” British Journal of Educational
Technology, 41(3), E44-E47. doi:10.1111/j.1467-8535.2009.00944.x
Michaud, F., Laplante, J. F., Larouche, H., Duquette, A., Caron, S., Létourneau, D., & Masson, P. (2005). Autonomous spherical
mobile robot for child-development studies. IEEE Transactions on Systems, Man and Cybernetics, Part A: Systems and Humans,
35(4), 471-480.
Mubin, O., Stevens, C. J., Shahid, S., Al Mahmud, A., & Dong, J. J. (2013). A Review of the applicability of robots in education.
Technology for Education and Learning, 1, 1-7.
Ruiz-del-Solar, J., & Avilés, R. (2004). Robotics courses for children as a motivation tool: The Chilean experience. IEEE
Transactions on Education, 47(4), 474-480.

156
Rusk, N., Resnick, M., Berg, R., & Pezalla-Granlund, M. (2008). New pathways into robotics: Strategies for broadening
participation. Journal Science Education & Technology, 17(1), 59-69.
Salter, T., Te Boekhorst, R., & Dautenhahn, K. (2004, March). Detecting and analysing children’s play styles with autonomous
mobile robots: A Case study comparing observational data with sensor readings. In Proceedings of the 8th Conference on
Intelligent Autonomous Systems (IAS-8) (pp. 10-13). Amsterdam, The Netherlands: IOS Press.
Shimada, M., Kanda, T., & Koizumi, S. (2012). How can a social robot facilitate children’s collaboration? Social Robotics, 98–
107.
Slangen, L., Keulen, H. V., & Gravemeijer, K. (2011). What pupils can learn from working with robotic direct manipulation
environments. International Journal of Technology and Design Education, 21, 449–469.
Sugimoto, M. (2011). A Mobile mixed-reality environment for children’s storytelling using a handheld projector and a robot.
IEEE Trans Learning Technologies, 4(3), 249-260.
Sullivan, A., & Bers, M. U. (2013). Gender differences in kindergarteners’ robotics and programming achievement. International
Journal of Technology and Design Education, 23(3), 691-702.
Varney, M. W., Janoudi, A., Aslam, D. M., & Graham, D. (2012). Building young engineers: TASEM for third graders in
Woodcreek Magnet Elementary School. IEEE Trans Education, 55(1), 78-82.
Wei, C. W., Hung, I. C., Lee, L., & Chen, N. S. (2011). A Joyful classroom learning system with robot learning companion for
children to learn mathematics multiplication. The Turkish Online Journal of Educational Technology, 10(2), 11-23.
Whittier, L. E., & Robinson, M. (2007). Teaching evolution to non-English proficient students by using LEGO robotics. American
Secondary Education, 35(3), 19-28.
Williams, D. C., Ma, Y., Prejean, L., Ford, M. J., & Lai, G. (2007). Acquisition of physics content knowledge and scientific
inquiry skills in a robotics summer camp. Journal Research on Technology in Education, 40(2), 201-216.
Woods, S. (2006). Exploring the design space of robots: Children’s perspectives. Interacting with Computers, 18(6), 1390-1418.
Yo u ng , S . S. C., Wang, Y. H., & Jang, J. S. R. (2010). Exploring perceptions of integrating tangible learning companions in
learning English conversation. British Journal of Educational Technology, 41(5), E78-E83.

157
Appendix
Appendix Table. Details of the selected studies
Paper
Level (age)
Area explored
Robot used
Study detail
Results
Implications
Type of study
Barker &
Ansorge,
2007
32 (9-11 years
old)
Achievement
Scores
LEGO
Mindstorms
28 lessons
conducted using
experiential
learning modes
to teach Science
Engineering
Tec hn ica l
concepts
No significant
results between
Pre-test & Post-
test for control
group.
Robotic group
showed a
significant
increase from
(M = 7.93, SD =
3.71) to ( M =
17.00, SD =
0.88)
Increase of
mean scores
from pre-test
to post-test
indicated that
Robot was
effective at
teaching
youth about
SET concepts.
Quasi
experimental
study
Williams
et al.,
2007
K-12
(21 middle
school
students)
Acquisition
of Physics
Content
Knowledge
and Scientific
Inquiry Skills
LEGO
Mindstorms
and
ROBOLAB
2 weeks robotic
camp as students
work in small
groups to
examine whether
they increase
their Physics
Content
Knowledge &
Science Inquiry
Skills
There is a
significant
difference on
the Physics
Content
Knowledge but
not for the
Science inquiry
skills.
Mixed
methods
Rusk et
al., 2008
Robotic
activities were
arranged for
 museum
workshop for
families
 after-school
program for
girls
 professional-
development
workshop for
educators.
Broadening
of
participation
in robotic
Picocricket
Robotic workshop
for students to
work on themes
to foster their
interest and a
sense of shared
experiences.
Combining art and
engineering
encourage story-
telling,
exhibition &
new
technologies.
The results
suggested
multiple paths
for engagement
for children,
teens, families,
and educators.
Robotic is
introduced in
areas of
students’
interest e.g.,
music, art and
story-telling,
providing new
learning
experiences to
wider
audience.
Non-
experimental
(Anecdotal
Case Studies)
Levy &
Mioduser,
2008
Kindergarten
3 boys, 3 girls,
randomly
selected, (5yrs -
6yrs old)
Children’s
perspectives
LEGO mobile
robots
Two sets of
instruments
have been
developed for
the study: a
computerized
control
environment
and a sequence
of tasks
To investigate
children’s
perspectives.
Children took part
in a sequence
braided of two
strands of tasks:
Description and
Construction.
Five 30-45 minute
session.
Data collected on
children’s
description and
explanations of
robots’
behaviour.
The role of adult
during
facilitation:
with adult’s
interaction,
children shift
into more
complex
technological
rules.
Learning is
viewed as
enculturation
and
knowledge is
socially
constructed.
Differentiate
between
technological
and
psychological
points of
view.
Mixed-
method
Barak,
2009
Junior High
School, 80
students
To improve
learning
concepts in
Science,
Tec h nol ogy
and problem-
LEGO
Mindstorms
Data are collected
through
qualitative
analysis of
observations,
interviews &
Students often
come up with
inventive
solutions to
problems. They
are likely to
Non-
experimental
(Anecdotal
Records,
Case studies)

158
solving
reflections as
students work on
the projects.
benefit from
implementing
informal
instructions in
project based
programme
Liu, 2010
Elementary:
Grades 4-6
Survey: 318
Students.
Interview: 48
(24 boys, 24
girls)
Early
adolescents’
perspectives
of
educational
robots and
learning of
robotics.
To d ev el op a
scale to
collect
students’
perception.
Experiences in
using LEGO
Mindstorms &
in using robots
The study was
conducted with
the use of
questionnaire.
The tool was
developed with
high validity &
reliability.
Results showed
(1) children
regards
educational
robot as a
plaything,
(2) learning
about robot as
source of
employment,
(3) Learning of
robotics as a
way to high
technology;
Differences
between male
and female
perception.
Mixed-
method
Highfield,
2010
33 (3-4years
old)
22 (Year 1)
Robotic toys
as a catalyst
for
mathematical
problem-
solving
Bee-bots &
Pro-bots
2 hrs/week over 12
weeks of study.
Children were
required to
complete 3 tasks
(1) Structural tasks
(2) Exploratory
tasks
(3) Extended tasks
Study to examine
tasks as
sequenced as
possible;
learning
framework to
support the
development of
mathematical
processes.
The result
showed
significant
children
engagement in
multiple
mathematical
processes; they
demonstrated
perseverance,
motivation &
responsiveness.
A multi-faceted
approach,
integrating
and inter-
relating
concepts,
processes and
skills through
dynamic tasks
could provide
rich
mathematical
thinking and
sustained
engagement.
Anecdotal,
Case Studies
Chen,
Quadir &
Ten g
(2011)
Elementary
School, 5
students
Using robot
to teach
English
Robot, Zigbee,
computer and
book
Observations and
Interviews
Use of computer
with robot and
book provided
interactive
experiences to
students
Anecdotal
Case studies
Yo un g et
al., 2010
Elementary
School
68 (Grade 3-4);
6 students (2
boys & 4 girls)
were selected
as a focus
group
To investigate
children’s
perception of
robot as
learning
companion
Rocky robot
Questionnaire
survey
conducted in an
elementary
school in Taiwan
Quantitative
results: 95%
have positive
attitude
towards using
tangible
learning
companions.
The students
became more
active in
practicing
conversation
with Rocky.
The children
were active in
practising
conversation
with the robot
Mixed-
method
Hong et
al., 2011
Elementary
School
Collaboration
of learning in
technological
project design
POWERTECH
robot
Students took part
in a
POWERTECH
contest in
Tai w a n .
Each pupil was
highly
involved
during the
process and
Reflection
essential for
problem-
solving were
often raised
Non-
experimental
(Anecdotal
records)

159
Cooperation in
learning basic
technical
processes.
Collaborative
problem-solving
to improve
design.
construction of
the artefact
with deep
reflection.
among the
team
members
during design.
Collaboration
enhances
learning.
Lin et a l.,
2012
Junior High
School’s
parents: 39;
17 male, 22
female
Parents’
perceptions
towards
educational
robots
Questionnaire
survey about the
parent’s attitude
was conducted.
Gender and socio-
economic
differences were
also examined.
Results indicated
that parents
considered
educational
robots as
beneficial for
their children.
But they were
less confident
in playing and
teaching with
educational
robots with
their children
themselves.
Parents were
willing to
provide
chance and
encourage
children to
learn with
educational
robots.
More training
for parents in
this area are
required to
boost their
confidence.
Non-
experimental
(Cross-
sectional-
survey
study)
Ruiz-del-
Solar &
Avi l és,
2004
K-12
700 children
and teachers in
Chile
Children and
teachers’
perception of
educational
robots.
BEAM robot
Parallox robot
and
LEGO
Reviews on use of
robots since
2000 through
surveys with
children and
teachers.
 Tested the degree
of child
satisfaction with
the workshop
 Inquired the
level of
competence.
 Determined
children’s
interest in
eventually
pursing an
engineering
career.
92% satisfied
with the
workshop,
88% finished all
the basic tasks
during the
workshop,
86% indicated
they would
follow an
engineering or
science career
in the future.
Non-
experimental
(Longitudinal
Study-using
survey)
Va rn e y e t
al., 2012
Elementary
school
TASEM
summer camp
to raise
interest in
STEM
LEGO
Working in small
groups, 1
hr/week session
Robots are
effective tool
for children to
develop “team
skills” (75%
actively raised
questions)
The robotic
programme
allowed
students of
different
socio-
economic and
cultural
backgrounds
to participate.
Anecdotal
Sugimoto,
2011
Elementary
school
25 (11-12 years
old),
13 boys and 12
girls; randomly
allocated into 5
groups.
A mobile
mixed-reality
environment.
Study
conducted
over 2
weekends.
GENTORO
robot
Children took part
in a story
creation by
manipulating a
robot and a
handheld
projector. The
study involved 2
previous pilot
studies.
Study-1 :
COGAME
Study-2: Software
modules,
involving scene
drawing tasks to
The children
engage
strongly in
story
expression
processes and
acted in a
coordinated
manner.
Mixed-
method

160
support story-
telling.
Quantitative
results were
collected with
the use of
Creative Product
Semantic Scale
on their story
creation.
Chambers
et al.,
2008
Elementary
(9-10 yrs old)
10 girls and 12
boys
Developing
conceptual
understanding
through
robotic
LEGO
Mindstorms
Hands on
experiences of
robot
construction and
gear
configuration
manipulation;
6 weeks – 3
sessions about
120 minutes
using LEGO
robotic
materials.
Pre and post
interviews
conducted.
Intervention
consisted of
semi-structured
guided scientific
inquiry
approach.
Results suggest
that providing
children with
physical
experiences
were not
sufficient to
develop
mechanical
conceptual
understanding,
of the
importance of
timely and
appropriate
intervention.
Results
confirm that
there is
variability
among children
in how they
reason about
gears &
conceptual
development.
A g ui de d
inquiry
instructional
approach is
proposed for
the conceptual
understanding
development
Quasi-
experimental
Chang et
al., 2010
Three classes
of 5th graders
Instructional
tool for 2nd
language
Humanoid
robot
5 scenarios were
tested, one per
week:
 Story telling
 Oral reading
mode
 Cheerleader
mode
 Action command
mode
 Question-and-
answer mode
(1) The
humanoid
robot performs
rich gestures.
Non-verbal
signals are
important part
of
communication
(2) The robot can
change
intonation or
speech rate
(3) The human
appearance of a
robot attracted
attention, even
from weaker
students. This
may motivate
them to
participate
more in the
language class.
(4) Robots’
ability to
interact and
recognize
students’
commands
offer a more
natural way to
perform
The children’s
reactions and
the teachers’
opinions
indicated that
robots could
create an
interactive
and engaging
learning
Non-
experimental
(Case
studies-
observational
records)

161
language drills.
Bers, 2010
Prekindergarten
to 2nd grade
To d ev el op
computational
thinking &
learning
about the
engineering
design
process in
young
children
Tan g ib le K -
programme
Assessment:
Student’s
portfolio,
Video journals,
SSS rubic levels of
understanding.
After 6 TangibleK
sessions,
students create a
final project by
working
individually or in
pairs
Tan g ib le K
robotics was
implemented in
the early
childhood
setting by
integrating it
with other
disciplinary
learning in a
developmentall
y appropriate
way for young
children.
Development
of evidenced-
based
systemic
account of
children’s
learning
according to
positive
technology
development
framework
Non-
experimental
(Case
studies-
observational
records)
Whittier &
Robinson,
2007
Middle school
(Grade 7-8),
29 students (16
Grade 7, 13
Grade 8)
Using
robotics to
teach non-
English
proficient
students in
developing
their
understanding
of science
concepts
LEGO,
Evobots
12 sessions of 60-
minute lessons.
Teachers use
LEGO robotics
to address state
science
standards.
The results
showed that
students having
significant
gains in their
conceptual
understanding.
An increase of
mean pretest
26.9% to
posttest 42.3%
Students
developed
many science
processes,
problem-
solving,
inquiry, and
engineering
design skills.
Quasi
Experiment
study
Beran et
al., 2011
184 children, 5-
16 years old,
98 female,
86 male
Children’s
perception of
animism
A 5 d eg re e
freedom robot
arm,
performing
block stacking
task
Semi-structured
interviews
conducted with
the children.
9 questions were
asked whether
the robots
referenced
humanistic
qualities.
Results from
frequency and
content
analysis
suggest that a
significant
proportion of
children
ascribe
cognitive,
behavioural,
especially
affective
characteristics
to robots.
Experiment
Study
Salter et
al., 2004
6 Children (5-7
years old)
AuRoRA
project
develop for
use with
children with
autism in
therapeutic
and
educational
context.
Pekee robot
Children grouped
into clusters
according to
their
psychological
classification.
Sensor captures
children’s
interaction level.
Children playing
style with the
robot were
examined by
using sensor
data.
Findings suggest
that touch
could have an
important role
to play in
developing
natural human-
robot
interfaces.
Also, robot
interaction
levels could
vary to suit
different
children.
Results
indicated that
robot’s
behaviour can
be adapted to
a different
children. It is
suggested for
future to use
robot to
quantify and
assess
children’s
behaviour.
Experiment
study using
sensor and
observational
techniques
Woods,
2006
159 children
To examine
children’s
perception of
robots’
appearance
Evaluate 40
robot images
by completing
a
questionnaire
on appearance,
personality
and emotions
Results showed
that depending
on appearance,
children clearly
distinguished
robots in terms
of their
intentions,
Some robots are
human-like but
still
distinguishable
from humans
and evoke a
feeling of
discomfort or
Study implies
the value of
robot design
and reaction
of users to it.
Non-
experimental
(Cross-
sectional
survey)

162
understanding
capabilities and
emotional
expression.
Children judged
human-like
robots as
aggressive, but
human–machine
robots as
friendly.
repulsion.
Kazakoff
et al.,
2013
Early
Childhood,
29 total
participant,13
pre-
kindergarten,
16 kindergarten
Sequencing
skills test
after robotic
intervention
using a
picture-story
sequencing
task
New York
STEM School,
CHERP
tangible
programme, 1
week intensive
programme
A p ai re d t-test was
conducted on the
children
sequencing
abilities using
sequencing
cards. A pre-test
and post-test
conducted. There
was a control
group.
Results indicated
that it was
possible to see
increases in the
sequencing
ability of pre-
kindergarten
and
kindergarten
students
participating in
a robotics and
programming
curriculum in
as little as 1
week.
Robotics offer
children and
teachers a
new
way to tangibly
interact with
traditional
early
childhood
curricular
themes
Quasi-
Experiment
Study
Bers &
Portsmore,
2005
Pre-service
early childhood
teachers and
engineering
students
To e ng ag e
early
childhood
teachers to
have hands on
experiences
in robotics
Pre-service
teachers working
in partnership
with engineering
students during
their training.
The goal is to
develop a model
and approach for
this teaching
methodology.
Three models
were
evaluated:
 Developer’s
Model
 External
Consultant’s
Mode
 Collaborator’s
Model
From
engineering’s
perspective,
students
gained insight
into the
educational
system and
issues
involved in
incorporating
ICT into the
classroom.
Pre-service
teachers saw
the potential
of the
technology
and resources
needed to
continue
using it.
Non-
experimental
(Case
studies)
Bers, 2013
Early
childhood, 53
children,
3 different
kindergartens
A study on
gender
differences in
robotics and
programming
achievement
Tan g ib le K
programme
The study
examined
whether
kindergarten
boys and girls
were equally
successful in a
series of building
and
programming
tasks. The
Tan g ib le K
Program
consisted of a six
robotics lessons.
Pearson product-
moment
correlation
Results showed
that boys had a
higher mean
score than girls
on more than
half of the
tasks but very
few differences
in the results
were
statistically
significant.
Boys scored
significantly
higher than
girls in only 2
areas: properly
attaching
Correlational
study

163
6pt Likert-scale
assessment tool
components
and
programming
using “Ifs.”
Slangen et
al., 2011
10-12 year olds
Developing
of
technological
literacy in
working with
robot
LEGO
Mindstorms
NXT
Study on
conceptual and
cognitive
analysis to
develop a
reference frame
to determine
students’
understanding of
robotics
Study concluded
that robotic
DMEs
challenge
pupils to
manipulate,
reason, predict,
hypothesize,
analyze and
test. Students
frequently
compare test
results with
their objectives
and
expectations to
refine their
conceptual
knowledge and
skills.
Non-
experimental
(anecdotal
study)
Michaud,
et al.,
2005
12-24 months
children
(12-18 months:
3 girls, 1 boy)
(18-24 months:
3 girls, 1 boy)
To study
children’s
interaction
with robot.
Roball
To examine the
potential of
using robot to
help children in
areas of their
language,
affects, motor,
intellectual &
social skills
development
Trials were
conducted with
the children
while
interacting
with Roball
Roball could
capture
children’s
attention,
enabling
interaction
studies
Experimental
Cangelosi
et al.,
2010
Humanoid
robot
Study of embodied
cognitive agents
to understand
cognitive
development,
complex
sensorimotor,
linguistic and
social learning
skills
Review of
specific issues
and progress,
with a series of
milestones is
translated into
a practical
roadmap for
future research
The milestones
on the
roadmap
directs future
work of
cognitive
developmenta
l robotics
Short Review
paper