Content uploaded by Alexander Repenning
Author content
All content in this area was uploaded by Alexander Repenning on Jun 30, 2021
Content may be subject to copyright.
Copyright © 2020, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
DOI: 10.4018/978-1-7998-1479-5.ch011
A difficult challenge to computer science education is the systemic professional development of teachers.
K-12 computer science education models limited to voluntary in-service teacher professional develop-
ment may not reach a critical majority of teachers who are skeptical towards information technology,
computer science, programming and computational thinking. The inclusion of computer science in a
national K-12 education standard in Switzerland has made it possible to move beyond voluntary K-12
computer science education for in-service teachers to mandatory pre-service teacher education for all
elementary teachers. This chapter describes the vision of the Digital Polymath as a digitally enabled
person empowered by computational thinking to connect computer science with other disciplines. The
course design, combining game design activities, computational thinking tools and the 7 big ideas from
the computer science principles framework is outlined and experiences are reported.
Alexander Repenning
School of Education Northwestern Switzerland (PH FHNW), Switzerland
Anna Lamprou
School of Education Northwestern Switzerland (PH FHNW), Switzerland
Patrick Wigger
School of Education Northwestern Switzerland (PH FHNW), Switzerland
Contemporary educational methods, strongly flavored by notions of efficiency popularized by the rise of
public education during the industrial revolution, still tend to value discipline specialization over disci-
pline integration. But the 21st century workforce is likely to benefit from an educational transformation
shifting from discipline specialization to discipline integration. The notion of a Digital Polymath is a
vision revisiting the ideals of disciplinary integration found in the renaissance polymath and projecting it
into the digital future. This notion of a Polymath has nothing to do with math but with the competency to
think and act interdisciplinary. Leonardo da Vinci, or Hildegard of Bingen were two exemplary polymaths
capable to solve hard problems by drawing on complex bodies of knowledge from different disciplines.
Similarly, our vision of the Digital Polymath is to employ digitalization as tools to enable humans with
interdisciplinary thinking and acting. The shift from the Renaissance Polymath to the Digital Polymath,
outlined in Table 1, is not trying to turn people into Leonardo da Vinci, or Hildegard of Bingen, but to
create a digital learning enabled 21st century society willing and capable to deepen as well as to connect
knowledge from different disciplines.
The ability to connect knowledge from different disciplines, i.e., the transformation of education
models from narrowly specialized disciplines, popularized in the industrial revolution (Johns, Laubscher,
and Malone 2011), to deeply integrated disciplines, is key in building a knowledgeable 21st century
digital society. There are two main challenges that prevent interdisciplinarity within the K-12 education
level. The first one is the lack of self-guided learning skills. The second is the lack of ability to connect
disciplines. Computer Science Education (CSed) will play a pivotal role in the making or breaking of
the digital polymath. Much depends upon how successfully the two main challenges that prevent CSed
integration to other disciplines in the K-12 educational level will be addressed. If CSed is set up to pro-
mote the development of a mindset that can learn without input from a teacher and is focused in teaching
Computational Thinking (CT), the Digital Polymath could become a reality.
However, before CSed can offer a way to the digital polymath, it has to find a place into the K-12
curriculum. The main challenge preventing K–12 CSed to advance from teachers who are technology
enthusiasts to pragmatists is perhaps best characterized by Crossing the Chasm, a notion anchored in
Table 1. The shift from the Renaissance Polymath to the Digital Polymath
the diffusion of innovation literature (Rogers 1962). This chasm appears to exist for Csed (Repenning,
2018). It suggests it is difficult to move beyond early adopters (Figure 1, red and orange stages) of a
new idea, such as K–12 CSed, to the early majority (Figure 1, green stage). The three fundamental CSed
stages (Figure 1) are described in detail below (Repenning, 2018).
Stage I focused mostly on the “right” tools. In the 1990s the overall negative children’s perception of
programming, best described as “hard and boring,” suggested cognitive as well as affective challenges
(Repenning, 2016, 2014). Programming in schools was typically marginalized into after school contexts
such as Friday afternoon clubs. These clubs, in turn, attracted only the usual suspects, that is, self-selected
kids (mostly male) instructed by self-selected teachers (also mostly male). To make computing more
relevant to schools it was crucial to focus less on tools supporting programming per se but to create tools
to forge explicative ideas through computing, or as Papert started calling it, Computational Thinking
(CT) (Papert 1996). We call these tools Computational Thinking Tools (Repenning, 2017).
Once CT Tools sufficiently addressed the cognitive and affective challenges dimensions of CSed, it was
time to shift the research focus from tools to curricula and teacher professional development. Research-
ers started not only to design and conduct teacher professional development but also to systematically
Figure 1. Crossing the computer science education chasm with mandatory pre-service teacher education
evaluate its efficacy. The teachers attending were mostly self-selected, but in many cases the classes they
would teach were aimed at all students and no longer limited to after-school contexts. Stage II started to
reach a much broader audience than Stage I. Teacher professional development gradually scaled from
local face-to-face summer institutes, to online and blended professional development (Basawapatna et
al. 2013).
Compelling curricula can spread through networks of excited Computer Science (CS) teachers surpris-
ingly quickly (Bradshaw and Woollard, 2012), but how will they cross the CSed chasm (Figure 1) and
persuade more conservative audiences? Switzerland, a highly affluent, but in terms of K–12 CSed some-
what conservative country, is radically shifting its strategy to cross this chasm by introducing mandatory
pre-service teacher CSed, starting at the elementary school level. In 2013, in response to the new Swiss
national curriculum Lehrplan 21 (LP 21), that mandates through the module “Medien und Informatik”
K–12 CSed, the school of education of northwestern Switzerland (PH FHNW), Switzerland’s largest
School of Education, started exploring the systemic impact on K–12 CSed through mandatory pre-service
teacher education. This project engages all PH FHNW pre-service elementary school teachers in CSed,
through two mandatory CS courses starting in September of 2017.
The current situation in Switzerland provides ideal circumstances not only for crossing the CSed chasm
but also for making the digital polymath vision a reality by bringing interdisciplinarity, through CT, into
K-12 education. Because this situation involves not just the self-selected elementary school teachers it
provides an ideal testing ground to explore a systemic introduction of computer science education. At
the very same time this offers high potential for impact but also significant risks. The high potential for
impact consists of the fact that elementary school teachers already have to teach many different disci-
plines. Consequentially, an interdisciplinary computer science education approach connecting computer
science with other disciplines is, at the elementary school level, a more natural fit to teaching practice
than at the middle and high school level. On the other hand, there is the risk that elementary school
teachers find computer science education too difficult as subject. Our data suggest that the vast majority
of pre-service teachers have no experience with programming at all and that they did not expect having
to teach any kind of programming to kids.
Developing an understanding of interdisciplinarity and the ability to apply this understanding to
teaching is difficult. An interdisciplinary culture of collaborations does not just happen (Schanzer et al.,
2015). Simply because elementary school level pre-service teachers, i.e., students at a school of education,
get exposed to many disciplines (e.g., language, science, math, music, art, sport) with the goal to teach
these disciplines does not imply that they think or can teach interdisciplinary. We cannot blame these
future teachers to not develop a deep interdisciplinary understanding given that current teaching praxis
at many schools of education. Training of pre-service teachers typically consists of taking courses from
different professors, e.g., the language professor, the music professor, etc. This teaching practice does
not stress interdisciplinary perspectives and, consequently, is no likely to advance much towards the vi-
sion of a digital polymath. Short of revolutionizing our school of education, the only realistic short-term
option was to devise a course design bringing interdisciplinary thinking into the foreground through an
approach described in this paper.
The paper outlines some of the related work, and focuses mostly on the design, and evaluation of two
courses mandatory for pre-service elementary school teachers. During the first course, the pre-service
teachers develop of sense of computational thinking including practical skills such as being able to pro-
gram simple games and simulations During the second course, they connect computer science to other
disciplines through CT and begin to appreciate the added value resulting from establishing this connec-
tion. The students start by applying predefined interdisciplinarity, continue by designing interdisciplinary
scaffolding, and conclude by making interdisciplinarity explicative. In this paper we describe in detail
the two courses and we present findings that showcase the success of our approach.
The introduction of CS into the school classroom came with the development of personal computers.
Personal computers could change education and provide new possibilities for learning. Starting in 1970,
Papert spent a decade developing LOGO, a programming language that gives children the ability to com-
municate with computers. Papert viewed CSed not just as teaching programming but as an opportunity
to revolutionize education, making learning a fun activity, not defined by curriculums and instruction.
He viewed computers as tools that can be used by kids to resolve math problems by learning to speak
the language of math (Papert, 1980). In the 80s an emerging hype about learning programming was not
focused on programming per se but an anticipated collateral benefit towards learning in other disciplines
including math and science or even discipline independent problem-solving skills. Early studies explor-
ing these collateral benefits found little evidence supporting these hopes. Pea even compared the idea
of learning programming in schools with the “overzealous prescriptions for studying Latin in Victorian
times” (Pea and Kurland, 1984). Even though promising, with analyses offering reasons for this hope,
there is still very little evidence for automatic transfer from programming to other disciplines. Actually,
research comparing the types of decisions made during planning processes showed that no differences
were found between students who were exposed to programming and those who had not (Pea, 1983).
More recently, however, some studies have found early evidence of transfer of the connections between
disciplines are explicitly scaffolded. For instance, the Bootstrap research (Seaburn et al., 1996) has found
some evidence of scaffolded skill transfer from computing to algebra.
In the 90s, the term Computational Thinking (CT) was used for the first time by Papert who argued
that CT is about forging ideas through explicative programs: “the goal is to use computational thinking
to forge ideas that are at least as “explicative” [as the Euclid-like constructions (and hopefully more
so) but more accessible and more powerful]” p. 110 (1996). The term computational thinking got later
popularized by Wing in her frequently cited paper from the mid-2000s (2006).
In terms of K-12 CSed, CT has defined much of the discussion surrounding the field. CT, a skill
acquired through CSed, is described by many (Grover and Pea 2013, Wing 2006) as a key skill for the
workforce of the 21st century, but has been the subject of intense and long discussions within the CSed
community (Grover and Pea 2018, Denning 2017). The debate revolves around questions about CT’s
definition, importance but more intensively around how this skill can be effectively taught. With regards
to the definition debate, Jeannette Wing defines CT as the thought process involved in formulating a
problem and expressing its solution(s) in such a way that a computer—human or machine—can effec-
tively carry out (2014). Based on Wing’s definition, the CT process can be segmented into three stages:
1. Abstraction: problem formulation
2. Automation: representation of a solution
3. Analysis: execution and evaluation of the solution representation (Figure 2).
Despite the ongoing debates surrounding CT’s meaning, importance and way of teaching, commonly
used definitions are gradually emerging and there is broad acknowledgement that our global economy
is based and shaped by computing. There is enough literature about teaching and learning programming
and CS but mostly for the undergraduate college level. Despite many countries introducing CSed in their
K-12 curricula, such as: UK’s Computing at School movement (Crick and sentence, 2011), the partnership
of the University of Adelaide with Google to successfully implement Australia’s Digital Technologies
Curriculum (Adelaide), and the CS for All mandate in the US (Smith, 2016), and other initiatives aim-
ing at introducing CS into schools, such as the CS4HS and Computing in the Core, CSed has not been
systematically introduced to K-12 education (Grover and Pea, 2013; Franklin, 2015).
In Switzerland, up until the LP 21, CSed in primary and secondary (K-12) schools mainly stood for
learning to use applications like Microsoft Office. There are a number of organizations that have been
aiming at improving CSEd K-12 education for many years such as the Hasler foundation. However, most
of these programs and initiatives are focused on secondary schools, and some even target only talented
students. Switzerland has an early CSEd history including programming at the high school level. Between
1985 and 1995, CS was included into the high school curriculum with a focus on programming and was
obligatory for all high school students. In the 1990s, however, more and more the focus shifted towards
application programs that became “CS lessons”. However, because the computer was also increasingly
used as a tool in other subjects (e.g. word processing, spreadsheets), the scientific discipline “CS” in-
creasingly faded into the background (Burkhart et al., 2013). Since 2008, CS is part of the Swiss high
school’s curriculum as an elective subject (Roos et al., 2014).
Figure 2. Three stage process (the AAA CT process) describing computational thinking
With regards to the teaching of CT, many believe that CT cannot be taught in a traditional manner
and has to overcome both a pedagogical and a systemic challenge, e.g., (Duncan and Bell, 2015). With
regards to the pedagogical challenge, the question about what to teach in what school level and how, is
to this day open. Even though CSed has been introduced especially at the high school level mainly as
programing, previous research has shown that even though CT has close connections to programming
the first cannot be automatically learned through teaching the latter. Duncan and Bell summarized a
large pilot study with primary school students in New Zealand with “We had hoped that CT skills would
be taught indirectly by teaching programming and other topics in computing, but from our initial ob-
servations this may not be the case” (2015). The relationship between programming and computational
thinking remains unclear but traditional approaches to teach programming may not necessarily be effec-
tive means to yield computational thinkers. Putting it a bit more crassly: Expecting students to turn into
computational thinkers from teaching them to program is like expecting people to turn into architects
from teaching them to assemble IKEA furniture. Evidence, e.g., (Duncan and Bell, 2015), supporting
the programming ≠ computational thinking conjecture helps to support the notion that if programming
tools are for programmers, perhaps there should be computational thinking tools for computational
thinkers (Repenning, Basawapatna, and Escherle, 2016). Computational thinking tools assist the CT
process (Figure 2), without the introduction of complicated and difficult programming and can make
programming exciting and accessible.
Interdisciplinarity can play an important role in introducing and teaching CS in terms of CT within
K-12 education. Interdisciplinarity itself however is hard to achieve, mainly because of lack of know-how
and also because of being perceived as overhead intensive. Previous research has shown that despite
evidence that when students in higher education are exposed to interdisciplinary experiences, begin to
act more wisely (Weinberg and Harding 2004), and especially pre-service teachers show an increased
appreciation for the subjects taught (An, Ma, and Capraro, 2011, An et al., 2016), interdisciplinary initia-
tives are difficult to maintain, even when institutions have interdisciplinarity as an explicit aim (Chet-
tiparamb, 2015, Franks et al., 2007, Pharo et al., 2012). Barriers to interdisciplinarity identified within
the literature expand from structural or formal differences between disciplines to differences in cultures
and frames of reference (Franks et al. 2007). Moreover, especially in the case of CS, integration with
other subjects has been viewed as problematic and as a factor weakening the content of CS, especially
because the teachers of other subjects are responsible for teaching CS and they are often insufficiently
trained (Roos et al., 2014; Kohlas, 2013).
With regards to the systemic challenge, the question is about how, with CS being up until recently
mainly an extracurricular activity, taken up by interested in the subject (self-selected) teachers and
students, we can cross the chasm (Moore, 1999) of CSEd. That is, how can we be shifting from self-
selected teachers and students to ALL teachers and students (Figure 1)? Within Switzerland, research
has shown that the teachers who teach CS in the high school level do not have degrees in CS but rather
in other STEM areas (Parriaux and Pellet 2016). With most of K-12 teachers not having a background
in CS, teaching CS and CT has proven a challenge. Currently most efforts are focused in providing
professional development of in-service teachers (Lamprou, Repenning, and Escherle, 2017). There are
worldwide various initiatives and organizations providing professional development courses and support.
For example, the Beauty and Joy of Computing (BJC), an introductory CS curriculum developed at the
University of California, Berkeley and intended for non-majors, offers weeklong summer courses for
teachers (Berkeley). The Scalable Game Design (SGD) Summer Institutes in Colorado, brings together
teachers with CS researchers who develop highly effective and easy to use CT curricula for classrooms
(SGD). Finally, the Computer Science Teachers Association (CSTA), a membership organization, sup-
ports and promotes the teaching of CS globally. It provides opportunities for K–12 teachers and their
students to better understand CS and to more successfully prepare themselves to teach and learn, provid-
ing resources, courses and community support to interested teachers (Association 2005).
Even though there are many excellent examples of professional development initiatives, the crossing
of the CSEd Chasm requires not only finding a place for CSEd into K-12 curricula and supporting in-
service teachers through professional development, but also finding a place for CS in pre-service teachers’
educational curricula. Literature on CSed courses for pre-service primary level teachers is sparse. With
the exception of pilot projects involving selected pre-service teachers’ CS training (Huett and Varga,
2016, Yadav et al. 2014), there is not much literature about pre-service teacher CSEd (Döbeli Honegger
and Hielscher, 2017). Even though the need to educate pre-service teachers has been previously ad-
dressed in literature (Darling-Hammond and Bransford, 2007; Yadav, Stephenson, and Hong, 2017), most
research involving pre-service teachers and CS is focusing on questions about attitudes towards the use
of computers (Teo, 2008). The introduction of LP 21 in Switzerland is changing this. In anticipation of
its implementation, the School of Education in Schwyz Switzerland started offering the first obligatory
CSed course for its pre-service teachers (Döbeli Honegger and Hielscher, 2017), followed by the PH
FHNW which has been offering two mandatory courses since 2017 in 4 states with over 600 students.
In the following chapters we describe in detail the two courses taught at the PH FHNW and we present
first findings from data collected during the first year of the courses’ lives that highlight their success.
The Scalable Game Design (SGD) courses are mandatory for all the students that study to become
primary level teachers at the PH FHNW and are taught in two consequent semesters. The first course is
focused on establishing computer science as a subject. The second, is focused on didactics. The goal of
the SGD courses is dual: the focus of the first course is, that the pre-service teachers (students) become
Computational Thinkers. The pre-service teachers (students) do not need any previous knowledge of
CS for the course. The focus of the second course is for the pre-service teachers, to learn to support,
through interdisciplinary practices, their future primary level students to become Computational Think-
ers themselves. The second course is built to strengthen interdisciplinarity in connecting CS through
CT to other subjects/ disciplines.
Assuming that students had essentially no computer science / programming background the first course
had to establish computer science as a subject area and provide students with an understanding of com-
putational thinking. Additionally, the course materials and competencies developed had to be connected
to the Lerhrpan 21. We used a three-pillar approach: 1. Motivational and learning strategy: Scalable
Game Design (SGD); 2. Tools specifically designed to teach and support CT in schools: CT Tools; 3.
The 7 big ideas of CS: CS Principles. The first two pillars are also used to guide big part of the second
part of the course on didactics. The following sections describe in detail the three pillars.
The motivational and learning strategy of Scalable Game Design (SGD) has been developed and tested
over many years. Experiences with SGD in the canton of Solothurn show the great motivation of students
and teachers (Lamprou, Repenning, and Escherle, 2017). These concepts were originally developed and
tested in the USA in various countries and cultures including Mexico, Brazil and Japan. Not only are the
students enthusiastic but also the teachers are often surprised to discover a new energy in their students
(Repenning et al., 2015). The motivational model of Scalable Game Design is based on the Zones of
Proximal Flow framework (ZPF) (Figure. 3) (Basawapatna et al., 2013). ZPF combines Csíkszentmi-
hályi’s (1997) concept of flow and Vygotsky’s (1978) zones of proximal development concept. Flow is
the feeling of a happy mental state in complete concentration and complete dissolution in an activity.
This state is localized in the context where challenges are proportional to existing competencies (green
area in Figure 3). If the challenges are too small in comparison to competences, boredom appears as a
state (blue-grey area in Figure 3). If, on the other hand, challenges are extremely higher compared to
competencies, this results in a state of anxiety (red area in Figure 3). In contrast to Csikszentmihalyi,
Vygotsky was mainly interested in learning processes. He sees another condition or zone, the “zone of
proximal development” (orange area in Figure 3), as an ideal condition for learning success. In this zone,
students are consciously brought to the limits of their own competences in order to be provided with
missing knowledge or relevant concepts at the right moment, thanks to the targeted support of another
person. This type of social learning can be supported by traditional teacher/student models, but also
by student/student peer learning. Zones of Proximal Flow show how students can ideally navigate the
challenges/competencies space (Basawapatna et al., 2013).
Figure 3. The Zones of Proximal Flow (ZPF)
The motivation model and the learning strategy of SGD are closely intertwined. Competencies are
viewed as the result, and to a certain extent as a by-product, of motivation. Research has shown that game
design is extremely motivating for children regardless of gender. The motivational aspect is strongly based
on the possibility of creating something of their own for example objects and worlds in 2D (Repenning
and Ambach, 1997) or 3D (Repenning et al., 2012) that are interesting and relevant for them. The great
motivation of game design, for example, leads to children building skills that seem to be out of their
reach according to traditional learning approaches. The learning strategy of SGD includes the use of
gradually increasing in their complexity and cognitive demands (hence the term “scaling”) Computational
Thinking Patterns (CTPs). CTPs are game design patterns that students first learn in game design but
then apply to the creation of STEM simulations (Koh et al., 2010). These CTPs describe fundamental
models that define the interaction between objects or agents or the interaction of objects or agents with
users. Due to their high degree of abstraction, these constellations of coordinated rules of conduct can
be used independently of specific projects and can therefore easily be used to create agent interactions
in simulations. The SGD curriculum builds students’ CT skills by introducing them from simpler games
(such as Frogger) to more challenging games that require additional, more complex CTPs.
CT is not the same as programming. It rather refers to the conceptual understanding of object interac-
tions across programming languages. In contrast to CT, each programming language is characterized
by its own syntax, which users must master in order to be able to write functioning programs. However,
Figure 4. The cognitive/affective challenges space
syntactic details of a programming language are not as important as the conceptual understanding of
general programming concepts. Previous research (Duncan and Bell, 2015), shows that specific pro-
gramming languages and environments can be inefficient for teaching CT. In contrast to commonly
used programming languages and environments, CT tools are not aimed at future programmers, but at
so-called computational thinkers. That means, for example, games and simulation can be created by
students with very little effort and maximum support by the tools.
Many students, especially women, find programming difficult and boring (2014). “Difficult” is a
cognitive dimension and “boring” an affective dimension of this fundamental challenge. These persis-
tent concerns can be interpreted as a two-dimensional research space called the Cognitive / Affective
Challenges Space (Figure 4) (Repenning et al., 2012). CT Tools address these two fundamental chal-
lenges. The “hard” part, the cognitive challenge, requires programming to become more accessible. The
“boring” part, the affective challenge, requires programming to become more exciting. In other words,
the big question is how does one transform “hard and boring” into “accessible and exciting?” To make
programming more accessible and exciting, it is necessary to understand complex interactions between
affective and cognitive challenges. Children may be quite excited to build a game, simulation, or robot,
but if the tools are too complex then there is a good chance they will give up because the return on in-
vestment is not clear. Good CT tools combine creativity support tools that make programming exciting
and programming support tools that make programming accessible (Repenning, 2017).
In terms of content, the SGD course is oriented towards competences in data structures, algorithms and
CS systems in accordance with the requirements of LP 21. With these skills, digital media and infor-
mation technology are critically examined with regard to their social significance. This is one of the
main reasons why computer literacy is nowadays part of basic education. To address the content and
learning objectives of LP 21, the SGD course is based on the AP CS Principles course of the College
Board (CollegeBoard, 2017a). The mapping of the SGD strategy onto the CS part of the LP 21 was
straightforward. The three main LP 21 CS topics (data, algorithms, and systems) were identified as a
subset of the 7 big ideas found in the AP CS Principles (CSP) framework: creativity, abstraction, data,
algorithms, programming, the Internet, and global impact. The 14-week course covered a new big CSP
idea every two weeks. The AP CSP course is an introduction to the fundamentals of computing with
a focus on how computers drive the world. Through the seven big ideas combined with the basics of
computing, students learn to analyze data and create technologies that have a practical impact. They
also gain a broader understanding of how information technology affects people and society as a whole
(CollegeBoard, 2017b).
The SGD course covered 14 weeks of classes (two-hour sessions per week). Each week’s session was
divided into two parts: a theory part and an activity part. The theory part provided knowledge and com-
petencies regarding one of the seven CSP ideas. Each CSP idea was mapped onto a two-week block so
that different aspects of each of the seven CSP ideas could be covered and its relevance to the professional
function of primary school teachers could be demonstrated. For the activity part the students focused on
creating their own projects using CT Tools. Students learned to program through a project first approach,
supported by scaffolding methods and using mainly the CT Tool AgentCubes (Repenning et al., 2012).
After being introduced to AgentCubes in their first session, the students started using it to program im-
mediately. Initially they created simple Frogger like games using step-by-step instructions, but as the
course progressed, they gradually moved on to program more advanced games and STEM simulations
while experimenting without the use of instructions. Other tools, such as Scratch and Processing were
also introduced. In the second half of the course the activity part became even more self-guided since the
students were working on their own final projects, which they had to develop in groups using CT Tools
or object-oriented programming languages. Through this hands-on work using CT Tools, the students
were able to learn how to write easy programs and understand how to think computationally.
The second course is focused on computer science didactics enabling future primary level teachers to
plan, conduct and evaluate CS lessons in an interdisciplinary way. Interdisciplinarity is at the core of this
course for a number of reasons. Conceptually speaking, we wanted CT to be the explicative connection
between CS and other disciplines similar to the way suggested by Martin (2018). By “explicative” we
refer to Papert’s goal of CT to forge ideas that are explicative in the sense that computation can make
difficult concepts more accessible and more powerful (1996). For instance, creating an ecosystem
simulation should not only be a great opportunity to learn how to program but should also result in a
learning process explaining the intricate causal connections of ecosystems. Practically speaking, interdis-
ciplinarity, for elementary school teachers who are not subject teachers but instead teach most subjects
including languages, math, art, science, and perhaps even gym, is a life saver because it allows them to
teach important concepts without the need for a separate course. This is key because most Swiss cantons
do not provide separate lessons for CS, these classes need to be implemented into other subjects. The
LP 21 mandates that CS is being covered in school but does not require schools to provide separate CS
courses. Independent of the reasons–may they be of a more conceptual or practical nature–the situation
at the elementary school level lends itself almost ideally for an interdisciplinary approach.
To experience interdisciplinarity students had to create so called projectlets connecting CS explica-
tively with subject areas through CT. Figure 5 summarizes the 14-week CS didactics course consisting
of the design and implementation of three projectlets (described in detail in sections 3.2.1 to 3.2.3).
Each projectlet connects CS with a subject area such as science, math, art or music. The students need
to create interdisciplinary projectlet tutorials teaching CS as well as subject area competencies. Scaffold-
ing regarding interdisciplinary teaching is gradually fading. The first projectlet provides predefined CS
and the subject concepts. Students produce a video tutorial which their peers will evaluate by building a
project following the tutorial. By the time of the third projectlet scaffolding is almost completely faded.
Students need to define CS and subject concepts and develop their own interdisciplinary connections.
The projectlets are graded according to the potential to teach CS, teach the subject area and the quality
of establishing interdisciplinary connections.
The course’s philosophy is heavily based on Papert’s constructionism theory and power principle
(1996). Tying back to the vision of the Polymath, e.g., the meta-competency to develop new digital
competencies, the power principle is invoked to engage students in authentic interdisciplinary learning.
Interdisciplinary understanding cannot be reduced to a collection of facts about the disciplines involved
but needs to be actively experienced through a process, i.e., CT, attempting to connect the disciplines
by building computational artifacts. This kind of constructionism (Papert and Harel, 1991) learning
happens, according to Papert, when students create models to understand the world around them. Con-
structionism is built upon Piaget’s notion of constructivism (Piaget 1970) and has been influenced by
experiential learning ideas. Constructionism argues that learning is most effective when the learners are
actively making tangible artifacts in the real world. It follows a project-based learning approach where
students make connections between different areas of knowledge and ideas through scaffolding facilitated
by the teacher (Papert and Harel, 1991). Along the same lines Papert’s power principle argues that the
natural way of learning is by doing first (Papert, 1996). The Projectlet idea even goes one step further
by not only involving learners in building artifacts for themselves but by employing these artifacts as
Figure 6. Examples of predefined projects, a simple 2d simulation of the diffusion process (left), a 2d
bacteria simulation to visualize exponential growth (middle), and a 2d simulation of an hourglass with
simple model of gravitation and friction
Figure 5. A representation of the 14-week CS didactics course consisting of the design and implementa-
tion of three projectlets
objects to think with (Turkle 2007) that are shared and discussed with other students in order to provide
didactic feedback. The following sections describe how the projectlets gradually developed students’
understanding of interdisciplinarity in three stages.
The first projectlet stage was heavily scaffolded by providing predefined CS and subject area concepts,
including references to the LP 21 curriculum (subjects: CS, natural science). Students, working in pairs,
had to select a simulation from a shortlist (Figure 6: How do smells travel? How do cells replicate? How
does a sand clock work?).
Then students had to figure out how to connect CS concepts to subject concepts through CT. Employ-
ing a flipped classroom model, students were asked to program these simulations during class and were
asked to create a video tutorial enabling others to rebuild a similar simulation using these steps at home:
• Program the simulation. All simulations we simple enough that they could be built in just 3 IF/
THEN AgentCubes rules. The students were not provided the solutions. They had to develop their
own program.
• Create a video tutorial aimed at elementary school kids teaching how to build the project.
• Use tutorials from other groups to build their project.
• Reflect on how well tutorial worked. Pairs creating tutorial could observe pairs using their tutori-
als and had to write meta-reflections interpreting the reflections of their users.
Not surprisingly, many of the stage #1 projectlets ended up as IKEA furniture assembly instruction-
like tutorials providing the necessary level of details to replicate the simulations by programming them
but not explicitly explaining pertinent interdisciplinary concepts.
In order to better understand interdisciplinarity, students had to develop their own project ideas including
CT scaffolding to connect CS with other disciplines. The stage #2 projectlet enabled students to create
their own projects within a given discipline such as music. They were provided a minimal introduction
on how to use relevant CS functionality, e.g., how to use the MIDI music commands in AgentCubes,
to connect CS with music through CT. Projectlet ideas could connect, for instance, CS with music by
building a digital instrument, with language by creating a digital story, with arts by generating paint-
ing algorithms, with biology by building a simulation of an ecosystem, with science by building an
avalanche-simulation, or with agriculture by building a forest-fire simulation.
In addition to creating their own project idea students had to think explicitly about scaffolding CS and
discipline knowledge in their tutorials. Students learned how to make so called Zones of Proximal Flow
(ZPF) tutorials including explicit navigation structures for differentiation. That is, users of ZPF tutorials
are presented with a choice of WHAT and HOW slides providing different degrees of scaffolding based
on the theory of Zones of Proximal Flow described in detailed in section 3.1.1 (Basawapatna et al. 2013).
WHAT slides provide minimalistic description of what needs to be done aimed at experienced users.
HOW slides provide detailed, step-by-step instructions typically consisting of short videos. Instead of
a screencast-video-tutorial, students were asked to build these interactive tutorials following the idea of
the ZPF. A ZPF tutorial offers navigation buttons matching the colors of ZPF zones shown in Figure 3.
Level 1 (green) only explains the tasks and offers little help of how to fulfill this task (for example create
a new project and name it Ecosystem). The user then has to decide whether he or she needs more support
in fulfilling this task. The ZPF-tutorial offers different paths through this slideshow. For example, a ZPF
tutorial can have 3 different levels of support. Green: task only, yellow: further help needed, red: lost.
Students could decide by themselves what they would offer on which stage of the tutorial: Screenshot,
Video-Screencast, Project in the given state, etc. (Figure 7). We speculated that creating ZPF tutorials
would help to make interdisciplinary concepts more visible to students. Instead of degrading into “just
the steps” IKEA tutorials, these tutorials would force their designers to think more explicitly about what
K-12 students already knew and how to leverage this knowledge through differentiated scaffolding.
The final projectlet of stage 3 opened the range of disciplines and allowed students to come up with their
own ideas of how to create an interdisciplinary activity but they would also have to think much more
about the added value of connecting CS to some other discipline (Figure 8). Specifically, they would have
to think about the previously mentioned notion of explicativeness. Most simulations serve as examples
of added value because–very much in the spirit of Paper’s definition of CT to forge explicative idea–
simulations can help to understand powerful disciplinary ideas through the process of programming.
Programming quizzes, on the other hand, does not add significant value. Quizzes can be interesting in
terms of computer science by offering insight into programming techniques to interact with users, and
they may also be interesting in terms of a discipline, e.g., they could be used to drill and kill (Jorgenson
and Vanosdall, 2002) facts about chemistry. However, unlike explicative connections they do not provide
value adding insights into the discipline. In contrast to programming a chemical simulation, the process
of programming a chemistry quiz is unlikely to contribute to an improved chemistry understanding
beyond the facts captured in the questions/answer pairs of the quiz.
Creating, running, changing and analyzing a simulation is likely to result in new insights even for
the author of the simulation (Figure 9). Simulations carry the potential to raise questions and enables
Figure 7. An example of a ZPF tutorial offering 3 different levels of support
students to practice and improve their ability of CT. If a simulation does not work as expected it is either
due to an error of abstraction (idea how a phenomena works is incorrect or partly incorrect) or due to a
programming error, as the computer only executes given instructions. The action of formulating rules
(programming) causes the simulation-agents to behave accordingly. If the observed behavior is not as
intended, a student has the power, to change the rules accordingly and can directly observe the conse-
quences of these changes in his/her simulation. Simulations carry the potential to raise questions and
enables students to practice and improve their ability of CT. If a simulation does not work as expected it
is either due to an error of abstraction (idea how a phenomena works is incorrect or partly incorrect) or
due to a programming error, as the computer only executes given instructions. The action of formulat-
ing rules (programming) causes the simulation-agents to behave accordingly. If the observed behavior
is not as intended, a student has the power, to change the rules accordingly and can directly observe the
consequences of these changes in his/her simulation.
The notions of added value resulting from a compelling connection between computer science and
disciplines turned out to be difficult to convey. On the one hand, students started to appreciate the idea
that simulations can be more explicative than quizzes. On the other hand, however, many felt that con-
ceptualizing simulations was more complex and perhaps out of reach for elementary school children.
Figure 8. The added value of connecting CS to some other discipline through CT
Figure 9. An example of a viral infection simulation developed by a group of students as projectlet #3
Part of this perception was rooted in students’ misguided intuition that true value of simulations is only
reached when the simulation approximates reality as close as possible. This, in turn, is hard to achieve
resulting in simulations with large number of agents and complex causal interactions. Students had to be
reminded of the very basic simulations they had been creating in stage #1, e.g., the perfume simulation
can be built with a single IF/THEN rule.
Since the SGD courses are the first of their kind, the teaching was accompanied by a research project,
which documented and collected data. The first course was taken by approximately 600 students (pre-
service teachers) mostly from the first and third semester. On September 17, 2017, in 4 states, 7 instruc-
tors began to teach 26 CS courses with approximately 25 pre-service teachers each. During this time
a variety of methods from questionnaires to in-depth interviews were used to collect data. During the
first course, we ran three questionnaires: one at the beginning, one in the middle and one at the end of
the course. The questionnaires were consisted by background, self-efficacy, CS attitude and CS knowl-
edge/skills questions and used a five-point Likert scale (1 = strongly disagree --- 5 = strongly agree).
The middle questionnaire was focused more on the assessment of the tools used in the class, while the
end questionnaire included a course evaluation. In order to assess the skills and knowledge gained by
the students during the course, we repeated a number of questions from the initial questionnaire in the
final and calculated the effect sizes for these questions. During the second course we run two identical
questionnaires one in the beginning and one at the end of the course. The questionnaires resembled the
questionnaire from the first course. Additionally, for the purposes of teaching around 600 students and
coordinate seven instructors, the teaching team used a teaching/learning platform where each instruc-
tor could communicate with the students and the other instructors online. The platform contained the
teaching material and homework assignments while it gave the possibility to develop wikis, groups,
discussion forums etc. This gave us access to a plethora of qualitative data from projects to homework
answers etc., and the ability to ask specific open-ended questions.
Our students made a very unique and interesting research subject. As mentioned above the first course
was taken by approximately 600 pre-service primary level teachers (students) mostly coming from the
first and third semester. From these students 541 completed a pre-course questionnaire. The answers to
these questionnaires give us a pretty good picture about who these students were. Our students were in
their majority female (74%). Their answers to the initial questionnaire show that they are not gamers but
have a rather positive attitude towards CS while they indicated a rather positive belief in their abilities
regarding technology and CS related tasks. In more detail our students’ answers showed that while they
believe that they have average computer skills, they have literally no skills in programming. The students
also showed indifference about whether they would like to learn programming or develop video games.
In terms of CS, they think that it is a rather difficult subject but do not think that it is boring. They think
that teaching CS through the development of video games is a rather good idea, while finally they think
that CS is rather important for their career as primary teachers and important for their students’ educa-
tion. In terms of self-efficacy, our students rated quite high, appearing confident with handling and using
computers and technology as teaching tools even if they had no previous experience using them. In the
following sections we present in more details some of the findings.
Following the widely perception that CS is hard and boring we asked our students in the beginning and
the end of the course to rate from 1 to 5 (Likert scale) how difficult and how boring they find CS. The
students rated CS as difficult but not as boring. In particularly in the case of difficulty the Cohen’s d
(Cohen 1988) was found -0.20 indicating a small negative effect, meaning that after the end of the course
the students found CS less difficult (Figure 10). The effect size for the “I think CS is boring” question
was found not significant.
We wanted to find out how high do our students rate the importance of CS for their future profession as
primary level teachers. The students rated above average the importance of CS for their future profes-
sion, however they rated that importance lower at the end of the course. Cohen’s d for this particular
question was calculated as -0.49 which indicates a negative middle effect size. As a result, our students
found that CS is less important for their future after the end of the course (Figure 11).
We also wanted to find out how important our students rate CS for the future of their future students.
Even though they rated such importance high, they found such importance lower at the end of the course
as they did in the beginning. Cohen’s d for the specific question was calculated as -0.31 indicating a
Figure 10. Percentages of the answer to the question: I think that CS is difficult, before and after the
first CS course
negative small effect size. Our students found CS less important for the future of their own students after
the end of the course (Figure 12).
With regards to our students’ skills we asked our students both in the beginning and at the end of the
course how good they are with computers. Both in the beginning and at the end of the course the stu-
dents rated their abilities as medium, however they rated their abilities higher at the end. Cohen’s d was
calculated as 0.58, indicating a middle effect size. This result indicates that our students are better with
computers after taking the course (Figure13).
Figure 11. Percentages of the answer to the question: I believe that CS is important for my
Figure 12. I believe that CS is important for the future of my students profession as primary level teacher,
before and after the first CS course
We asked our students to rate their ability to program in both the beginning and the end of the course.
In the beginning of the course 77.9% of the students rated their ability to program as practically non-
existent (1 in Likert scale). This number drops to 5.3% at the end of the course. Cohen’s d for this
Figure 14. Percentages of the answer to the question: I am good with computers, before and after the
first CS course
Figure 13. Percentages of the answer to the question: I believe that CS is important for the future of my
students, before and after the first CS course
particular question was calculated to 2.05 which indicates a large effect size. The figure below displays
the percentages of the answers before and after the course (Figure 14) (Lamprou and Repenning 2018).
In order to understand the role of CT as means to connect CS with other subjects, our students had to
understand CT. In October, after about one and a half months into the course we asked the students to
answer the open-ended question: What is CT for you? We received 447 replies which we grouped ac-
cording to seven major categories/themes. The themes were developed based on the answers given, while
the responses were categorized as correct or wrong based on the definition of CT on which the course
was based on. The seven themes were: 1. Thinking like a computer, that included answers describing
CT as thinking like a computer (n=94); 2. Programming/ CS/ Computer work, that included answers
strongly connecting CT to the ability to program or to apply CS principles (n=72); 3. Problem division,
that included answers identifying CT as a process that involves breaking a problem into smaller less
complicated parts (n=66); 4. Thinking with the computer, that included answers indicating the combi-
nation of human abilities with computer affordances (n=64); 5. Problem solving, that included answers
identifying CT as being a problem solving process, without much specification (n=60); 6. CT process/
AAA, that included answers describing the CT process along the lines of the CT definition that was
taught in the class (n=51); 7. Other, that included answers stating ignorance / answers that describing
CT as a way of thinking / answers consisting of correct points but written in a wrong order or context
(n=40). 25.7% (115 answers) of the students gave an answer that is along the lines of the CT definition
that was taught in the class. From this, 14.3% (64 answers) were answers that identified CT as “thinking
with the computer” (denken mit dem Computer) while 11.4% (51 answers) described a process along the
lines of Wing’s definition or the AAA process (Figure 2). 14.8% (66 answers) described CT as a process
of dividing a problem to smaller easier parts in order to solve it, while 13.4% (60 answers) referred to
Figure 15. Percentages of the answer to the question: I know how to program, before and after the first
CS course
Figure 16. An example of storytelling project developed by a group of students as projectlet #3
Figure 16. Categorized students’ answers to the question “What is CT for you?”
CT as a problem-solving process. Finally, 9% (40 answers) were classified as “other” clustering the
remaining responses (Figure 15) (Lamprou and Repenning 2018).
As mentioned above the research project accompanied the second course as well, but the analysis of the
data collected is still pending. Big part of what we wanted to explore through the research is the ability
of our students to use interdisciplinarity and connect CS to other subjects through CT. A first indicator
to evaluate this is the students’ projects. All students were able to create projectlets, and many of the
projects did exhibit high level of interdisciplinarity were CT was successfully used to forge “explica-
tive” ideas. Figure 9 in section 3.2.3 presents an example of a project where value is added both to CS
and other subjects/ disciplines through CT. This specific virus epidemic simulation offers “explicative”
connections between CS and math, CS and social and natural sciences (MNG), and CS and shop. It also
raises questions and enables students to practice and improve their ability of CT. On the other side of
the spectrum of projects, projects where ideas were not “explicative” forged through CT, are projects
such as quizzes or storytelling. Figure 16 shows a storytelling project. The project is developed in order
to connect CS with languages and CS with NMG. Telling a story with the help of the computer can be
easily linked to any other subject or discipline. However, the act of programming a story does not allow
or enable a deeper understanding of domain-specific knowledge, as a simulation can do.
The large effect size for the perception of programming skills suggests that the students (pre-service
teachers) think that they successfully learned how to program. Our students, despite not having any
programming experience and moderate interest to learn to program, felt after taking the first course,
that game design equipped them with the ability to program. They also felt that they know how to use
a computer better. However, it is less clear what they actually learnt with respect to CT. The disparity
between teaching programming and teaching CT is consistent with Duncan’s findings (Duncan and
Bell 2015). Even though the correctness of some of these answers is debatable, we are only considering
categories 3, 4, and 6 to be “right” in the sense that they were the intended learning outcomes of the
course. Looking through this lens, only 40.5% of the answers correct. However, the large percentage of
false answers (59.5%) does appear to suggest either instructional difficulties or tenacious misconceptions
that are hard to change. Thinking like a Computer, that represented 21% of the answers surprised us,
as the course tried to explicitly stress the notion of CT as a synergistic combination of human abilities
with computer affordances (Figure 2 center). Do computers really think? Should people actually think
the way computers operate, and, if so, at what level, e.g., machine language, would this kind of think-
ing take place? The answers categorized as Programming/CS/Computer work, that represented 16.1%
of the answers, suggest that CT is an activity connected to programming, or to the application of CS
principles. While this is true, it is not clear how inclusive these kinds of statements are. Is any form
of programming or application of CS principles always necessarily a strong manifestation of CT? For
the most part, these kinds of answers reflect our own blurred comprehensions of what CT is or is not.
Finally, while CT does include some problem-solving components it is more than just problem solving,
an answer given by 13.4% of the students. For instance, not every kind of problem solving includes the
expression of programs (Lamprou and Repenning 2018).
In contrast to our beliefs and previous research, our students did not rate CS as boring as we expected.
The fact that our students found CS less difficult at the end of our course is very encouraging because
it suggests that non-self-selected students are not necessarily thinking negatively about learning and
teaching CS. Our research showed that the value of the course was still unclear for some pre-service
teachers. Even though the students rated high the importance of the CS course for their profession and
for their future students in the beginning of the course they rated this importance lower at the end of the
course. We assume that this might be because of the uncertainty connected to how the subject will be
taught in the primary school. In the first course, our students learned a lot about CS but nothing about
how to teach it. Not having any experience from their own time as students, there is a possibility that
after the course, they felt that they will not know how to teach the subject. Uncertainty is also added
by the vague way the LP 21 will be implemented. Will CS education be feasible without taking away
time from other courses, which traditionally have a place in the curriculum? Will interdisciplinarity be
possible given the luck of experience with interdisciplinary teaching?
The second part of the course focused in equipping the pre-service teachers to teach CT and CS in
an interdisciplinary manner. Analysis of the data collected is still pending, but first impressions show
that the students had a positive learning experience. Even though not all projects were characterized by
a high level of added value and explicativeness, all students were able to create projects that connected
CS to other subjects / disciplines through CT to a degree.
The introduction of mandatory CSed courses into the pre-service teacher college curriculums is an
important first step towards crossing the CSed chasm by providing K-12 CSed for all teachers and all
students (Phase 3 figure 1). Teaching mandatory CSed courses to an audience consisting of pre-service
teachers provided us with new research insights because of their profile that is significantly different from
typical in-service teachers, e.g., 75% women, no previous CS knowledge and no personal pre-existing
interest to take CS courses. The course also raised many new questions and offered surprises. Regard-
less of unexpected surprises, we strongly believe that the course outcome was excellent. The large effect
size calculated from the question “I know how to program” shows that our students think they learnt
how to program after taking our course. However, it is less clear what they learnt with regards to CT.
Our students’ answers to the open-ended question: “what CT is” suggested that some still had a limited
understanding of CT or at least were not able to verbally express it.
In terms of the attitude of our students we were surprised to find that they do not view CS as some-
thing boring. We were also surprised with the positive attitude of students throughout the course. When
comparing pre/post course data what surprised us most was the drop in perceived relevance of CS for
their profession and for their students. We can only speculate regarding this drop. Before the course many
may have believed that computer science was about the use of applications, such as MS Office, which
they believed to be important for all students. When the course shifted students’ perception of computer
science to prominently include programming they may have lowered their judgment of relevancy simply
because they not have been sold on the idea that elementary school kids should learn to program. Fewer
conclusions can be reached regarding the second course because at time of this writing the data analy-
sis is not complete. Students were able to create interdisciplinary projects and accompanying tutorials.
Also, students rated the course highly. The projects still need to be analyzed with respect to value added
through interdisciplinarity. The main conclusion of this first iteration of these courses is that in spite of
significant difference in audience profiles, with proper scaffolding, successful in-service teacher profes-
sional development can be turned into successful pre-service teacher education.
An, S. A., Ma, T., & Capraro, M. M. (2011). Preservice teachers’ beliefs and attitude about teaching and
learning mathematics through music: An intervention study. School Science and Mathematics, 111(5),
236–248. doi:10.1111/j.1949-8594.2011.00082.x
Anna, L., Repenning, A., & Escherle, N. A. (2017). The Solothurn Project — Bringing Computer Science
Education to Primary Schools in Switzerland. Proceedings of the 22nd Annual Conference on Innovation
and Technology in Computer Science Education (ITiCSE 2017). Academic Press.
Basawapatna, A., Repenning, A., Koh, K. H., & Nickerson, H. (2013). The Zones of Proximal Flow:
Guiding Students Through A Space Of Computational Thinking Skills and Challenges. In Proceedings
of the International Computing Education Research (ICER 2013). Academic Press. Retrieved from
https://bjc.berkeley.edu/curriculum/
Bradshaw, P. & Woollard, J. (2012). Computing at school: an emergent community of practice for a
reemergent subject. Open University.
Burkhart, H, Guggisberg, M., Kuhnt, B., Lehmann, M., & Peter, J. (2013). EFI-CH Abschlussbericht.
C. S. Teachers Association. (n.d.). Retrieved from https://www.csteachers.org/
Chettiparamb, A. (2015). Interdisciplinarity: A Literature Review. The Interdisciplinary Teaching and
Learning Group.
Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Hillsdale, NJ: Law-
rence Erlbaum Associates.
CollegeBoard. (2017a). AP Computer Science Principles. The College Board. Retrieved from https://
apstudent.collegeboard.org/apcourse/ap-computer-science-principles/course-details
CollegeBoard. (2017b). AP Computer Science Principles. College Board.
Crick, T., & Sentance, S. (2011). Computing at school: stimulating computing education in the UK.
Proceedings of the 11th Koli Calling International Conference on Computing Education Research. Aca-
demic Press. 10.1145/2094131.2094158
Csikszentmihalyi, M. (1997). Flow and the psychology of discovery and invention. New York: Harp-
erPerennial.
Darling-Hammond, L., & Bransford, J. (2007). Preparing teachers for a changing world: What teachers
should learn and be able to do. John Wiley & Sons.
Denning, P. J. (2017). Remaining trouble spots with computational thinking. Communications of the
ACM, 60(6), 33–39. doi:10.1145/2998438
Döbeli Honegger, B. & Hielscher, M. (2017). Vom Lehrplan zur LehrerInnenbildung-Erste Erfahrun-
gen mit obligatorischer Informatikdidaktik für angehende Schweizer PrimarlehrerInnen. Informatische
Bildung zum Verstehen und Gestalten der digitalen Welt.
Duncan, C., & Bell, T. (2015). A Pilot Computer Science and Programming Course for Primary School
Students. Proceedings of the Workshop in Primary and Secondary Computing Education. Academic
Press. 10.1145/2818314.2818328
Franklin, D. (2015). Putting the computer science in computing education research. Communications of
the ACM, 58(2), 34–36. doi:10.1145/2700376
Franks, D., Dale, P., Hindmarsh, R., Fellows, C., Buckridge, M., & Cybinski, P. (2007). Interdisciplin-
ary foundations: Reflecting on interdisciplinarity and three decades of teaching and research at Griffith
University, Australia. Studies in Higher Education, 32(2), 167–185. doi:10.1080/03075070701267228
Google Inc. (n.d.). Women Who Choose Computer Science – What Really Matters. The Critical Role
of Encouragement and Exposure.
Grover, S., & Pea, R. (2013). Computational Thinking in K–12 A Review of the State of the Field.
Educational Researcher, 42(1), 38–43. doi:10.3102/0013189X12463051
Grover, S., & Pea, R. (2018). Computational Thinking: A competency whose time has come. In S.
Sentance, S. Carsten, & E. Barendsen (Eds.), Computer Science Education: Perspectives on teaching
and learning. Bloomsbury.
Huett, K. C., & Varga, M. A. (2016). Building Pre-Service Teacher Interest in Computer Science Educa-
tion through Mentoring Experiences. Proceedings of the 47th ACM Technical Symposium on Computing
Science Education. ACM. 10.1145/2839509.2850547
Jorgenson, O., & Vanosdall, R. (2002). The Death of Science?: What We Risk in Our Rush toward Standard-
ized Testing and the Three R’S. Phi Delta Kappan, 83(8), 601–605. doi:10.1177/003172170208300808
Koh, K. H., Basawapatna, A., Bennett, V., & Repenning, A. (2010). Towards the Automatic Recogni-
tion of Computational Thinking for Adaptive Visual Language Learning. In Proceedings of the Confer-
ence on Visual Languages and Human Centric Computing (VL/HCC 2010). Academic Press. 10.1109/
VLHCC.2010.17
Kohlas, J. (2013). informatik@ gymnasium: Ein Entwurf für die Schweiz: NZZ Libro.
Lamprou, A., & Repenning, A. (2018). Teaching How to Teach Computational Thinking. Proceedings
of ITiCSE, Larnaca, Cyprus. Academic Press. doi:10.1145/3197091.3197120
An, S. A., Zhang, M., Tillman, D. A., Lesser, L. M., Siemssen, A., & Tinajero, J. V. (2016). Learning
to Teach Music-Themed Mathematics: An Examination of Preservice Teachers’ Beliefs about Develop-
ing and Implementing Interdisciplinary Mathematics Pedagogy. Mathematics Teacher Education and
Development, 18(1), 20–36.
Martin, F. (2018). Rethinking Computational Thinking. CS Teachers. Retrieved from http://advocate.
csteachers.org/2018/02/17/rethinking-computational-thinking/
Moore, G. A. (1999). Crossing the Chasm: Marketing and Selling High-Tech Products to Mainstream
Customers. New York, NY: HarperBusiness.
Papert, S. (1980). Mindstorms: Children, computers, and powerful ideas. Basic Books, Inc.
Papert, S. (1996). An exploration in the space of mathematics educations. International Journal of
Computers for Mathematical Learning, 1(1), 95–123. doi:10.1007/BF00191473
Papert, S., & Harel, I. (1991). Situating constructionism. Constructionism, 36(2), 1–11.
Parriaux, G., & Pellet, J.-P. (2016). Computer science in the eyes of its teachers in french-speaking Swit-
zerland. Proceedings of the International Conference on Informatics in Schools: Situation, Evolution,
and Perspectives. Academic Press. 10.1007/978-3-319-46747-4_15
Pea, R.D. (1983). Logo Programming and Problem Solving.
Pea, R. D., & Midian Kurland, D. (1984). On the cognitive effects of learning computer programming.
New Ideas in Psychology, 2(2), 137–168. doi:10.1016/0732-118X(84)90018-7
Pharo, E. J., Davison, A., Warr, K., Nursey-Bray, M., Beswick, K., Wapstra, E., & Jones, C. (2012).
Can teacher collaboration overcome barriers to interdisciplinary learning in a disciplinary university?
A case study using climate change. Teaching in Higher Education, 17(5), 497–507. doi:10.1080/1356
2517.2012.658560
Piaget, J. (1970). Piaget’s theory.
Repenning, A. (2016). Transforming “Hard and Boring” into “Accessible and Exciting”. In NordiCHI
2016, Workshop on Cultures of Participation in the Digital Age: From “Have to” to “Want to” Partici-
pate. Academic Press.
Repenning, A. (2017). Moving Beyond Syntax: Lessons from 20 Years of Blocks Programing in Agent-
Sheets. Journal of Visual Languages and Sentient Systems, 3(July), 68–89. doi:10.18293/VLSS2017-010
Repenning, A. (2018). Scale or Fail: Moving beyond self-selected computer science education in Swit-
zerland. Communications of the ACM, 61(5), 40–42. doi:10.1145/3199603
Repenning, A., & Ambach, J. (1997). The Agentsheets Behavior Exchange: Supporting Social Behavior
Processing. Proceedings of the CHI 97, Conference on Human Factors in Computing Systems, Extended
Abstracts. Academic Press. 10.1145/1120212.1120231
Repenning, A., Basawapatna, A., & Escherle, N. (2016). Computational Thinking Tools. In Proceedings
of the IEEE Symposium on Visual Languages and Human-Centric Computing. IEEE Press.
Repenning, A., Smith, C., Owen, B., & Repenning, N. (2012). AgentCubes: Enabling 3D Creativity by
Addressing Cognitive and Affective Programming Challenges. Proceedings of the World Conference
on Educational Media and Technology, EdMedia 2012. Academic Press.
Repenning, A., Webb, D. C., Koh, K. H., Nickerson, H., Miller, S. B., Brand, C., ... Repenning, N.
(2015). Scalable Game Design: A Strategy to Bring Systemic Computer Science Education to Schools
through Game Design and Simulation Creation. Transactions on Computing Education, 15(2), 1–31.
doi:10.1145/2700517
Rogers, E. M. (1962). Diffusion of innovations (1st ed.). New York: Free Press of Glencoe.
Roos, M., Obrist, G.S., Wandeler, E., & Mosimann, M. (2014). Implementation des Ergänzungsfachs
Informatik an Schweizer Gymnasien. Academic Press.
Schanzer, E., Fisler, K., Krishnamurthi, S., & Felleisen, M. (2015). Transferring Skills at Solving Word
Problems from Computing to Algebra Through Bootstrap. Proceedings of the 46th ACM Technical
Symposium on Computer Science Education. ACM. 10.1145/2676723.2677238
Seaburn, D. B., Lorenz, A. D., Gunn, W. B., Jr., & Gawinski, B. A. (1996). Models of collaboration: A
guide for mental health professionals working with health care practitioners. Retrieved from https://sgd.
cs.colorado.edu/wiki/Summer_Institutes
Smith, M. (2016). Computer Science For All.
Teo, T. (2008). Pre-service teachers’ attitudes towards computer use: A Singapore survey. Australasian
Journal of Educational Technology, 24(4). doi:10.14742/ajet.1201
Malone, T. W., Laubacher, R. J., & Johns, T. (2011). The age of hyper specialization. Harvard Business
Review, 89(7-8), 56.
The University of Adelaide. (n.d.). CSER Digital Technologies Education.
Turkle, S. (Ed.). (2007). Evocative objects: Things we think with. MIT press.
Vygotsky, L. (1978). Interaction between learning and development. Readings on the development of
children, 23(3), 34-41.
Weinberg, A., & Harding, C. (2004). Interdisciplinary teaching and collaboration in higher education:
A concept whose time has come. Wash. UJL & Pol’y, 14, 15.
Wing, J. M. (2006). Computational Thinking. Communications of the ACM, 49(3), 33–35.
doi:10.1145/1118178.1118215
Wing, J.M. (2014). Computational Thinking Benefits Society. In 40th Anniversary Blog of Social Issues
in Computing.
Yadav, A., Mayfield, C., Zhou, N., Hambrusch, S., & Korb, J.T. (2014). Computational thinking in
elementary and secondary teacher education. ACM Transactions on Computing Education, 14(1), 5.
Yadav, A., Stephenson, C., & Hong, H. (2017). Computational thinking for teacher education. Com-
munications of the ACM, 60(4), 55–62. doi:10.1145/2994591
Digital Polymath: An individual, empowered by digital technology, able to access information from
many different disciplines enabled and skilled to solve problems by drawing and connecting knowledge
from these disciplines.
Computational Thinking Process: A process consisting of problem abstraction, automation through
programming and analysis.
Computational Thinking Tool: A tool supporting the Computational Thinking process by support-
ing the abstraction, automation and analysis.
Computer Science Education Chasm: A difficult to cross gap in adoption of computer science
education between early adopters and early majority.
Renaissance Polymath: An individual, such as Leonardo Da Vinci or Hildegard von Bingen, versed in
many different disciplines to solve problems by drawing and connecting knowledge from these disciplines.
STEM: Science Technology Engineering and Math.
