People, Ideas, Milestones: A Scientometric Study of Computational Thinking

Abstract

The momentum around computational thinking (CT) has kindled a rising wave of research initiatives and scholarly contributions seeking to capitalize on the opportunities that CT could bring. A number of literature reviews have showed a vibrant community of practitioners and a growing number of publications. However, the history and evolution of the emerging research topic, the milestone publications that have shaped its directions, and timeline of the important developments may be better told through a quantitative, scientometric narrative. This article presents a bibliometric analysis of the drivers of the CT topic, as well as its main themes of research, international collaborations, influential authors, seminal publications, and how authors and publications have influenced one another. The metadata of 1874 documents were retrieved from the Scopus database using the keyword "computational thinking". The results show that CT research has been US-centric from the start, and continues to be dominated by US researchers both in volume and impact. International collaboration is relatively low, but clusters of joint research are found between, for example, a number of Nordic countries, lusophone-and hispanophone countries, and central European countries. The results show that CT features the computing's traditional tripartite disciplinary structure (design, modeling, and theory), a distinct emphasis on programming, and a strong pedagog-ical and educational backdrop including constructionism, self-efficacy, motivation, and teacher training.

Full text

People, Ideas, Milestones: A Scientometric Study of Computational Thinking
MOHAMMED SAQR,
School of Computing, University of Eastern Finland, Finland and School of Electrical
Engineering and Computer Science (EECS), Human Centered Technology, Media Technology and Interaction Design,
KTH Royal Institute of Technology, Stockholm, Sweden
KWOK NG,
School of Educational Science and Psychology, University of Eastern Finland, Finland and Department
of Physical Education and Sport Sciences, University of Limerick, Ireland
SOLOMON SUNDAY OYELERE,
Department of Computer Science, Electrical and Space Engineering, Luleå
University of Technology, Skellefteå , Sweden
MATTI TEDRE, School of Computing, University of Eastern Finland, Finland
The momentum around computational thinking (CT) has kindled a rising wave of research initiatives and scholarly contributions
seeking to capitalize on the opportunities that CT could bring. A number of literature reviews have showed a vibrant community of
practitioners and a growing number of publications. However, the history and evolution of the emerging research topic, the milestone
publications that have shaped its directions, and timeline of the important developments may be better told through a quantitative,
scientometric narrative. This article presents a bibliometric analysis of the drivers of the CT topic, as well as its main themes of
research, international collaborations, inuential authors, seminal publications, and how authors and publications have inuenced one
another. The metadata of 1874 documents were retrieved from the Scopus database using the keyword “computational thinking”. The
results show that CT research has been US-centric from the start, and continues to be dominated by US researchers both in volume
and impact. International collaboration is relatively low, but clusters of joint research are found between, for example, a number of
Nordic countries, lusophone- and hispanophone countries, and central European countries. The results show that CT features the
computing’s traditional tripartite disciplinary structure (design, modeling, and theory), a distinct emphasis on programming, and a
strong pedagogical and educational backdrop including constructionism, self-ecacy, motivation, and teacher training.
CCS Concepts: •Social and professional topics →Computational thinking.
Additional Key Words and Phrases: Computational thinking, bibliometric research, history, scientometrics, literature review, computing
education research, computer science education
ACM Reference Format:
Mohammed Saqr, Kwok Ng, Solomon Sunday Oyelere, and Matti Tedre. 2018. People, Ideas, Milestones: A Scientometric Study of
Computational Thinking. 1, 1 (January 2018), 17 pages. https://doi.org/10.1145/1122445.1122456
Authors’ addresses: Mohammed Saqr, School of Computing, University of Eastern Finland, Finland and, School of Electrical Engineering and Computer
Science (EECS), Human Centered Technology, Media Technology and Interaction Design, KTH Royal Institute of Technology, Stockholm, Sweden; Kwok
Ng, School of Educational Science and Psychology, University of Eastern Finland, Finland and , Department of Physical Education and Sport Sciences,
University of Limerick, Ireland; Solomon Sunday Oyelere, Department of Computer Science, Electrical and Space Engineering, Luleå University of
Technology, Skellefteå , Sweden; Matti Tedre, School of Computing, University of Eastern Finland, Finland.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not
made or distributed for prot or commercial advantage and that copies bear this notice and the full citation on the rst page. Copyrights for components
of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to
redistribute to lists, requires prior specic permission and/or a fee. Request permissions from permissions@acm.org.
©2018 Association for Computing Machinery.
Manuscript submitted to ACM
Manuscript submitted to ACM 1

2 Saqr, Ng, Oyelere, and Tedre
1 INTRODUCTION
Since the modern use of the older phrase “computational thinking” (CT) was introduced in 2006 (
68
; see also
48
), a great
deal has been written about the potential and importance of computational thinking as a set of mental tools, skills, and
attitudes [e.g.,
6
,
19
,
27
,
38
]. Numerous denitions for the term have been proposed, with years of work invested in
describing CT, its manifestations in dierent elds, and ways to teach it at dierent levels of education [
22
,
25
,
40
].
Stories of the historical evolution and emergence of the concept have been told [
19
]. Major organizations, including
CSTA (Computer Science Teachers’ Association), CAS (computing at school of the British Computing Society), and
ACARA (Australian Curriculum, Assessment, and Reporting Authority) have developed their own frameworks for
bringing CT to schools [60].
Although no consensus has formed over the denition of CT, a frequently used quote is Aho’s (2011) description of
CT as “the thought processes involved in formulating problems so their solutions can be represented as computational
steps and algorithms.” Numerous competing denitions exist, and the term itself has been rephrased and reformulated in
many ways, including computational doing [
5
], computational making [
61
], computational action [
62
], computational
uency [
51
], proceduracy [
63
], and computational participation [
31
]. It was predated by older similar ideas like
algorithmic thinking [
32
] and procedural thinking [
59
]. Much of the CT discussions have been centered around how
to teach computing in K–12 setting [
27
], but also in higher education and in dierent disciplines. With all these
competing viewpoints to computational thinking, CT has been in the midst of continuous back-and-forth of arguments,
counter-arguments, new proposals, criticism, reviews, and analyses.
In addition to the excited rush to develop new means for teaching and evaluating CT, critical voices are many, too. The
CT movement has been criticized for ignoring important lessons in computing education research [
49
,
60
,
64
], for either
too much programming or too little programming [
3
,
10
,
11
,
35
,
39
,
41
], for ambivalence between CT and computer
science (CS) [
24
,
44
], and for ambiguity on how to measure progress in CT [
16
]. It has been suggested that there is a
spectrum of views on CT, where at the one end are people who argue that one can learn CT without programming or
rigorous models of computation and that computers are not necessary; at the other end are people who argue that
computers and programming are quintessential for CT [11, 16].
A number of excellent literature reviews and analyses of computational thinking [
22
,
25
,
38
,
40
,
57
] reveal a rich body
of CT literature since 2006. This article presents an alternative viewpoint: A quantitative, bibliometric analysis of the
CT literature. By analyzing article metadata in Elsevier’s abstract and citation database Scopus, the article establishes
milestone articles, key authors, regional dierences, historical growth, bibliometric trends, and a number of timelines
for the emergence of CT. The article is aimed at giving the reader an overview of when and where computational
thinking developed into the ourishing topic of investigation it is today; who and what are the key publications, most
active people, and research communities in the eld; as well as frequency and conceptual structure of keywords in CT.
2 METHODOLOGY
The data were retrieved on January 2, 2020, from Elsevier’s Scopus abstract and citation database, which contains
roughly 70 million records of peer-reviewed articles. The Bibliometrix R library [
2
] was used for analysis. The search
term “computational thinking” was used for document retrieval, and only articles in English were included in order to
enable keyword comparison. The search resulted in 1874 documents from 595 dierent source titles (typically journals,
books, and proceedings) (Table 1). The data set included 3779 authors, and among the data set 295 articles were single
authored.
Manuscript submitted to ACM

People, Ideas, Milestones: A Scientometric Study of Computational Thinking 3
Table 1. Basic information about the bibliometric data set.
Data set statistics
Documents 1874
Distinct source titles (journals, books, etc.) 595
Keywords Plus (ID) 5005
Author Keywords (DE) 3138
Period 2006–2019
Mean citations per document 6.828
Document types
Journal article 473
Journal article (in press) 6
Book chapter 71
Conference paper 1288
Editorial 11
Review 25
Authorship data
Authors 3779
Author appearances 5914
Authors of single-authored documents 295
Authors of multi-authored documents 3484
Single-authored documents 348
Mean documents per author 0.496
Mean authors per document 2.02
Mean co-authors per document 3.16
Mean collaboration index (CI) 2.28
The analysis was limited to the period from 2006-2019, as there were only three occurrences before 2006, none of them
relevant. Journal articles, conference papers, reviews, book chapters, editorials, and viewpoints were included. While
editorials are typically not peer reviewed, they were included in the data set because some of the most inuential writings
on the topic—including Wing’s (2006) essay—have been published as non-peer-reviewed editorials or viewpoints. Book
chapters were included, and a number of conference proceedings are classied as edited books. The data were cleaned,
duplicates removed, author names, journals, and conferences with dierent spellings were checked and xed. The
analysis included descriptive statistics of document count, document metadata, authors, sources, article types, and
other measures. As “country” is not a standard publication attribute in the database, countries of authors were extracted
from the rst author’s rst aliations. The Collaboration Index (CI) was computed as the ratio of the sum of authors of
multi-authored articles to total multi-authored articles (Table 1). Annual Percentage Growth Rate was calculated as
average percentage increase in the number of articles over a period of a year.
For keywords analysis, four types of keywords were included in the study: author keywords (keywords provided
by the author), categorized keywords (duplicates and similarities aggregated), keywords extracted from titles, and
keywords extracted from abstracts. Using the four types of keywords helps to triangulate the conceptual structure
of CT publications, and to overcome the shortcomings of author-provided keywords. Of the four types of keywords,
only author keywords data were cleaned; for instance keywords in plural were combined with singular (“primary
school” / “primary schools”), similar terms and abbreviations were combined (“computer science” / “CS”, and “K12” /
“K-12” / “K–12”). The abstract keywords were extracted from the retrieved articles using the term extraction method of
Manuscript submitted to ACM

4 Saqr, Ng, Oyelere, and Tedre
Bibliometrix R library [
2
]; that method includes deletion of stop words and application of a stemming algorithm that
reduces words to their stem. A keyword co-occurrence network was constructed to map the relationships between
dierent concepts within each document. For readability of relationships and labels, the network sizes were limited to
the 30 most frequent keywords. The analysis of keywords evolution used the cumulative frequency of keywords across
the years.
The keywords were clustered using network community detection (a “community of keywords” is a group of keywords
that occur frequently together) using the Louvain modularity algorithm, which has been shown to be computationally
ecient with very good clustering results [
13
]. Each keyword community was uniquely colored in the network plot.
To understand the structure of collaborating networks, the Louvain modularity algorithm for community detection
was also applied to cluster countries that collaborate frequently together. Each community of countries were uniquely
colored in the network plot.
The study also used historiography mapping, a popular bibliometric method developed by [
23
] to map the chrono-
logical direct citation network within a set of papers. The method maps the most relevant papers by ranking them by
their citations within the examined data set—known as local citation score (LCS)—as opposed to the papers’ overall
citations—known as global citation score (GCS). Papers with high local citation scores are considered more relevant to
the examined topic, and they show higher on the graph. For each author in the data set, the h-index and its variants
(m-quotient and g-index) were calculated (within the reviewed papers); h-index is a metric that ranks the impact of an
author through the frequency and citation count of the most cited publications. A country collaboration network was
created from documents with two or more authors aliated with institutions in dierent countries.
3 RESULTS
The results of this study show a timeline similar to descriptions of the new, post-2006 wave of CT presented in research
literature elsewhere [e.g.,
19
,
27
]. The phrase “computational thinking” entered the common computing education
parlance through Wing’s (2006) essay in ACM’s premier magazine Communications of the ACM (CACM). Initially, the
number of CT-related publications grew slowly: less than 13% of all the articles in the data set were published in the rst
half of the studied period (2006–2012). Over that period of time there were concerted and well funded eorts to bring
CT into the everyday consciousness and vocabulary in computing elds as well as outside computing. For instance, in
the US, Wing successfully leveraged her position as the 2007–2010 head of Computer and Information Science and
Engineering (CISE) of the US National Science Foundation (NSF), which stated that winning CPATH funding from NSF
required projects to demonstrate how CT is incorporated within the project (
4
, pp.46,111;
30
). In 2008 CISE (NSF) asked
the US National Research Council (NRC) to organize two workshops on computational thinking—one for discussing
what the newly introduced phrase meant for the workshop participants [
45
, p.7], and one for discussing educational and
pedagogical questions related to CT [
46
]. Those workshops gathered together many computing pioneers and leading
gures in computing education. The two workshop reports paint an excellent image of the challenges in reaching
consensus over CT and ultimately over computing’s disciplinary identity, and the reports reect well the next decade of
CT discussions.
Despite the lack of consensus over what CT is, the eorts started to bear fruit. The growth of CT-related publications
accelerated after 2012, and 87% of the titles were published in the second half of the study period, between 2013–2019.
The annual output of CT-related articles accelerated rapidly over the second half, reaching 430 articles in 2019 with an
annual percentage growth rate (AGR) of 61.2%. Figure 1 illustrates the trends of annual publishing activity on CT in the
most productive countries.
Manuscript submitted to ACM

People, Ideas, Milestones: A Scientometric Study of Computational Thinking 5
0 30 60 90 120
2010 2015
Year
Frequency
Country USA
CHINA
ITALY
SPAIN
UNITED KINGDOM
BRAZIL
KOREA
CANADA
GREECE
AUSTRALIA
Fig. 1. The annual number of publications on CT over the period 2006–2019 in the most productive countries.
The types of the most cited publications on CT (Table 2) resemble the types of computing education research (CER)
literature in general [cf.
58
], and the list in Table 2 includes all the main types of CER literature: essays, reviews, position
papers, and empirical research reports. Around two thirds of the publications (68.7%) were published as conference
papers, roughly a quarter (25.2%) were journal articles, and approximately 6% were book chapters, editorials, and review
papers. The most common identiable source title was the Lecture Notes in Computer Science series (80 publications),
followed by proceedings of the Frontiers in Education conference (75 publications) and proceedings of the ASEE Annual
Conference (54 publications).
The most inuential CT publication in the Scopus dataset, with 1782 citations, was Wing’s (2006) essay that introduced
the modern meaning of the phrase “computational thinking,” and that started the CT bandwagon rolling. Wing wrote her
essay at a time when enrollments in computing were seriously declining, wanting to make computing a more inspiring
career choice for the youth [
47
, p.86]. The timing of Wing’s article was opportune, and she was able to articulate the
needs of the 21st century in a way that resonated well with what many educators had witnessed themselves, too [
60
].
The top-cited contributions to CT became popular for dierent reasons and they served dierent purposes. The
second highest-cited paper was a 2013 review and analysis of CT by [
25
] that was published in a prominent education
journal Educational Researcher. At a time when CT was already gathering considerable interest, the paper became
popular for successfully explaining to the journal’s primary readership, educators, what CT is, why it had rapidly
gathered so much interest, and what were the currently pertinent and upcoming research foci around the topic. The
Manuscript submitted to ACM

6 Saqr, Ng, Oyelere, and Tedre
Table 2. Top 15 most cited CT publications in the Scopus database.
Authors Source title GCS LCS
68 Communications of the ACM 1782 935
25 Educational Researcher 499 168
69 Philosophical Transactions of the Royal Society A 395 239
40 Computers in Human Behavior 249 -
9 Computers & Education 178 81
66 Journal of Science Education and Technology 158 105
55 Education and Information Technologies 146 86
26 Communications of the ACM 143 -
50 Proceedings of ACM SIGCSE’10 115 73
67 Proceedings of ACM SIGCSE’12 113 78
14 Communications of the ACM 96 14
52 Computers & Education 90 -
70 ACM Transactions on Computing Education 86 68
64 Education and Information Technologies 81 57
28 Journalism 69 -
GCS = Global Citation Score
LCS = Local Citation Score
third highest-cited paper was Wing’s (2008) position paper that responded to the dire need for clarifying what CT is.
In her paper, which was presented to the 2008 British Royal Society discussion meeting, Wing elaborated her earlier
description of CT that was gradually gaining momentum.
The fourth most cited paper addressed the debate concerning the relationship between programming and CT in
K–12 education. That paper was a 2014 review of empirical research on learning CT through programming, by [
40
],
and published in Computers in Human Behavior. Many organizations and authors had skirted around the subject of the
role of programming in learning CT to an extent that was later criticized for “losing of the original denition of CT”
[
36
], while people seldom recognized that for many authors the goal of CT was “to bring programming back into the
classroom” [
31
], which Lye and Koh’s review made clear. Their review was widely read, and it was followed by years of
lively debates about the role and relevance of programming in computational thinking [e.g.,
3
,
10
,
31
,
35
,
39
,
44
,
57
].
Both Grover and Pea’s (2013) and Lye and Koh’s (2014) reviews noted an abundance of discussions over denitions of
CT and urged CT researchers for more empirical, intervention-based research.
The full list of top fteen most cited papers (Table 2) contains essays and viewpoints on CT [
14
,
26
,
68
], theoretical
contributions or position papers that dene CT or add to it in one way or another [
28
,
50
,
64
,
66
,
69
], literature reviews
[
25
,
40
], and empirical papers reporting on CT initiatives [
9
,
52
,
55
,
67
,
70
]. Many of the top-cited works gathered
considerable interest also outside the elds of computing and computing education research, as witnessed by their
high global citation scores (Table 2). But what caught the attention outside computing education circles was slightly
dierent from what became popular within CT circles: A number of articles with high local citation score (citations
within the examined data set) had a relatively low global citation score; examples include Lee et al.’s (2011) article in the
computing education magazine ACM Inroads (LCS=114), Aho’s (2011) description of CT (LCS=55), Lu and Fletcher’s
(2009) position paper on CT and programming (LCS=47), Hambrusch et al.’s (2009) CT course description, and Seiter
and Foreman’s (2013) assessment framework.
Manuscript submitted to ACM

People, Ideas, Milestones: A Scientometric Study of Computational Thinking 7
WING JM, 2006
WING JM, 2008
LU JJ, 2009
HAMBRUSCH S, 2009
REPENNING A, 2010
LEE I, 2011
AHO AV, 2012
WERNER L, 2012
SENGUPTA P, 2013
GROVER S, 2013
YADAV A, 2014
BERS MU, 2014
VOOGT J, 2015
WEINTROP D, 2016
SHUTE VJ, 2017
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
Fig. 2. A historical direct citation network of fieen most influential papers within the field (highest local citation scores).
Many inuential CT publications have cited relatively little other CT literature. A historical direct citation network
(Figure 2) shows the relationships between CT articles that have been central for dening the topic. The citation network
of inuential papers within the topic (highest local citation scores) in Figure 2 shows that those papers have exerted
only limited inuence on each other. The number of literature references in the articles in Figure 2 varies greatly from
roughly a dozen in many articles to nearly a hundred or more (55, 66).
3.1 Keyword Analysis
Based on the keyword sections of CT papers, programming and K–12 education were the dominant themes in CT
studies. The author-selected keyword “computational thinking” and its variants (such as computational thinking skills
and computational thinking patterns) appeared in the keywords section of 1155 (61.6%) papers. Programming and
coding related terms were the next most frequent, appearing in 510 (27.2%) of the publications. Education-related terms,
such as teaching, learning, and pedagogy appeared in 359 (19.1%) of the papers. Table 3 shows the top ten keywords
under keywords given by the authors, categorized keywords, keywords extracted from the article titles, and keywords
extracted from the abstracts. The last two (occurrences in title and occurrences in abstract) count each occurrence
separately: For instance, if an abstract mentions the keyword “programming” thrice, that counts as three occurrences.
In keyword association analysis, three keywords “CT,” “programming,” and “education” formed a triangle, signifying
a strong association between the themes. Figure 3 uses keyword association networks to visualize co-occurrences
between keywords. The network in Figure 3 demonstrates the conceptual structure in the analyzed set of CT papers and
the relational dimensions between concepts in that data set. That keyword association network was constructed from
the categorized keyword list where the keyword “computational thinking” was removed (it did most often co-occur
with the keywords “programming” and “education”).
After removing the “CT” keyword for readability, the co-occurrence network in Figure 3 sheds light on the nature of
CT as a distinctively K–12 educational initiative. A dense cluster of school concepts formed around two key concepts
Manuscript submitted to ACM

8 Saqr, Ng, Oyelere, and Tedre
Table 3. Keyword frequencies in CT literature. The laer two analyses (title and abstract) count each occurrence within the title and
abstract.
Author keywords Categorized keywords Occurrences in title Occurrences in abstract
Keyword N Keyword N Keyword N Keyword N
CT 1142 CT 1155 CT 897 CT 4207
Programming 147 Programming 510 Programming 312 Students 3381
CSE 113 Education 359 Learning 296 Programming 2012
Education 93 STEM 137 Education 229 Learning 1918
Scratch 87 Games 116 Students 188 Education 1323
Problem-solving 75 K12 108 Teaching 181 Skills 1272
Coding 63 CS 84 School 164 Study 1016
STEM 57 Robotics 79 CS 135 Computing 1012
Assessment 53 Problem-solving 75 Skills 120 CS 972
K12 51 Assessment 53 Computational 118 Teaching 942
CT = Computational Thinking
CSE = Computer Science Education
CS = Computer Science
































Fig. 3. Associations between keywords in the data set, with the keyword “computational thinking” removed. (Circle size indicates
keyword count in the sample; line thickness indicates co-occurrence frequency; colors indicate a “community” or cluster of keywords).
Manuscript submitted to ACM

People, Ideas, Milestones: A Scientometric Study of Computational Thinking 9
Table 4. Most active contributors to computational thinking body of knowledge, sorted by number of Scopus-indexed papers published
#Author A. N h Earliest Cites Most cited paper
1 Repenning, A. US 33 13 2009 544 50
2 Biswas, G. US 23 7 2012 276 55
3 Basu, S. US 20 7 2012 285 55
Robles, G. ES 20 10 2015 208 43
5 Fronza, I. IT 17 3 2015 52 20
Grover, S. US 17 6 2013 688 25
7 Moreno-León, J. ES 16 9 2015 193 43
Sengupta, P. US 16 7 2012 257 55
9 Dagien˙
e, V. LT 14 7 2011 164 41
Koh, K. H. US 14 10 2009 320 33
Soh, L.-K. US 14 6 2009 135 56
12 Basawapatna, A. US 13 9 2009 255 7
García-Peñalvo, F. J. ES 13 6 2016 124 21
Wilensky, U. US 13 7 2014 238 66
15 Settle, A. US 12 6 2009 186 41
Shell, D. F. US 12 6 2009 135 56
Wiebe, E. US 12 2 2008 21 42
A. = Country of aliation in the most cited article, N= Number of articles, h= h-index.
“programming” and “education,” along with its curricular integration keyword “STEM.” Those three were surrounded
by the keywords K–12, children, and higher education, as well as primary, middle, and high school. Computing’s
traditional disciplinary dimensions [
17
] were all present: design (through keywords like creativity, robotics, technology,
and engineering), modeling (simulation, abstraction, STEM), and theory (algorithm/algorithmic thinking, problem-
solving, computation/computing, informatics). The education side of computing education research featured strongly
in keywords like constructionism, self-ecacy, assessment, motivation, teacher training, games, and collaboration, as
well as gender.
3.2 Author Analysis
Whereas Wing’s 2006 essay holds top position in terms of citations, other researchers have picked up top positions in
terms of number of CT publications. Table 4 shows the fteen most active contributors to CT in terms of number of
papers (there were several authors with 12 publications, who all are listed, making the list 17 names long). The list of
authors is very US-centric: Twelve people on the list of seventeen authors were aliated with US-based institutions at
the time of their most cited publication. All the most productive CT authors in Table 4 have published twelve or more
Scopus-indexed papers on CT. Table 4 presents the most active authors, the number of articles they have co-authored,
their h-indexes within this data set, the year of their rst CT-related entry in Scopus, the total number of citations to
their CT-related articles, and their most cited papers.
The list of most active authors on CT by volume (Table 4) partly overlaps with the list of most cited CT publications
(Table 2), but the dierences between those tables show that inuence is not necessarily about volume. Some of the
most inuential papers were written by people with only a few CT-related publications. Examples include Wing, whose
two essays (2006, 2008) hold top positions in terms of citations but who has since published rarely on the topic; [
9
],
famous for their construction-based robotics activities around CT; [
26
], who has actively advocated strengthening the
Manuscript submitted to ACM

10 Saqr, Ng, Oyelere, and Tedre
KINNEBREW JS
WONG GKW
BARNES T
WIEBE E
SHELL DF
SETTLE A
WILENSKY U
GARC ́
IA-PEN ̃
ALVO FJ
BASAWAPATNA A
SOH LK
KOH KH
DAGIENE V
SENGUPTA P
MORENO-LE ́
ON J.
GROVER S
FRONZA I
ROBLES G
BASU S
BISWAS G
REPENNING A
2008 2010 2012 2014 2016 2018
Year
Author
N.Articles
1
2
3
4
5
6
7
TC per Year
0
20
40
60
Fig. 4. Timeline of publishing activity of the most productive CT authors (circle size indicates number of articles each year and circle
color indicates citations to that person’s articles published in that year).
link between CT initiatives and existing work in the eld of computing education research, and [
67
] with their work
on measuring CT. Other examples include [
14
] who has been the voice of computing as a discipline for decades [e.g.,
17
,
18
] but whose major contributions to computing’s disciplinary self-understanding have been framed in traditional
computing vocabulary, and only some of them have used CT vocabulary (e.g.
14
,
15
,
19
). It is noteworthy that some of the
most prolic authors have only become involved in CT in 2015 or 2016: Figure 4 shows a timeline of publishing activity
of the most productive CT authors). Furthermore, whereas each of the authors in Table 4 have gathered recognition in
the eld with some well-cited papers on CT, the h-indexes in Table 4 show that much work on CT also gets ignored.
For instance, h-index 3 means that the author has three CT papers cited at least thrice, and h-index 4 would require
four of the author’s CT papers each cited at least four times.
As the list of most active contributors to CT suggests, almost all of those authors belong to active research teams or
researcher networks focused on CT-related topics. Many authors in Table 4 share the same most-cited paper, and two
especially inuential clusters emerge from author network analysis, shown in Figure 5: A relatively tight network with
A. Repenning as the most active paper author [e.g.
34
,
50
], and a larger network where S. Basu and S. Biswas have been
the most active paper authors in their co-author network [e.g.,
55
]. Figure 5 shows co-authorships between fty most
active CT authors by the number of publications.
3.3 Country Analysis
Research output was also analyzed at the country scale in order to give an overview of geographical spread of CT
research activity. The data set was analyzed for the country of the rst author, who typically (but not always) is
Manuscript submitted to ACM

People, Ideas, Milestones: A Scientometric Study of Computational Thinking 11
REPENNING A
BASU S
BISWAS G
GROVER S
WILENSKY U
WIEBE E
ROBLES G
BARNES T
SOH LK
WEINTROP D
SENGUPTA P
SHELL DF
KOH KH
MORENOLEON J
FRONZA I
BASAWAPATNA A
LYTLE N
HORN M
DONG Y
CATET V
DAGIENE V
RUTSTEIN D
SETTLE A
KINNEBREW JS
BOYER KE
LEE I
HARTEVELD C
LUI D
KAFAI YB
SNOW E
LONATI V
JONA K
VERGARA CE
BRIEDIS D BUCH N
STICKLEN J
PAQUETTE L
DRESEN C
FRAZIER K
MALCHIODI D
ROMN-GONZLEZ M
MOTT BW
NICKERSON H
YADAV A MANNILA L
MONGA M
MORPURGO A
TISSENBAUM M
SHELDON J
WONG GKW
Fig. 5. Co-author network of CT-related articles in Scopus database. (circle size indicates paper authorship, colors indicate distinct
communities of researchers who frequently collaborate together)
considered the lead author of a study (however, as the number of multi-country publications was small compared to the
number of single-country publications, the results are roughly in the right order of magnitude even if that assumption
does not hold). By that measure, the data set contained papers with lead authors from 64 countries. The article metadata
were then analyzed for the countries of the rest of the authors. Table 5 ranks ten most active countries by the number of
papers from each country, and reports the number of papers, the percentage of papers from each country, the number
of single-country (SCP) and multiple-country publications (MCP), number of citations to papers from that country, and
mean citations per paper for each country.
The United States dominates CT-related research output—although inclusion of publications in other languages
(which were excluded from this study) could slightly change that balance. NSF’s CT-focused eorts [
4
] seem to have
worked, as 42% of all CT-related articles had a US-based lead author, a great majority of those were single-country
publications (only US-based institutions were involved), and nine out of fteen most inuential CT papers in Figure
2 thanked NSF support in their acknowledgments section. The next nine countries combined produced fewer CT
publications than the US did. That dierence is not explainable by larger overall research output of the US: The
proportion of US research output in CT-related topics is more than twice the proportion of US research output in all
elds (17.3% of world research output) and three times that of US research output in computer science (13.4% of the
Manuscript submitted to ACM

12 Saqr, Ng, Oyelere, and Tedre
Table 5. Publications by first author’s country (2006–2019)
Country Papers Freq. SCP MCP Cites Mean cites
USA 556 41.65% 521 35 7771 13.98
China 81 6.07% 71 10 142 1.75
Spain 64 4.79% 48 16 399 6.23
Italy 59 4.42% 45 14 173 2.93
South Korea 53 3.97% 53 0 43 0.81
United Kingdom 47 3.52% 36 11 321 6.83
Brazil 46 3.45% 37 9 108 2.35
India 26 1.95% 25 1 71 2.73
Germany 25 1.87% 20 5 39 1.56
Canada 22 1.65% 16 6 111 5.05
SCP = Single-country publication
MCP = Multiple-country publication
world CS output) (2018 Scopus data). US-based publications also gathered many more citations (7771) than the others
did (1407 citations between the nine countries in Table 5) with a massive lead in mean citations per publication (nearly
14 citations per article). Although many prominent CT-related initiatives, such as Dagien
˙
e’s Bebras Challenge [
12
]
and Bell’s CS Unplugged [
8
] come from outside the US, the bulk of CT research comes from US-based initiatives and
institutions, citing mostly US-based research.
CT publications are distinctively national not only in the US but also often elsewhere, with limited international
collaboration—although that varies greatly between countries. In some countries one in four or ve publications were
done in international collaboration; the most international studies were published by lead authors from Spain (where
25.0% of papers were multiple-country publications), Italy (23.7%), UK (23.4%), and Brazil (19.6%). Papers with lead
researchers from the US (6.3%), India (3.9%), and South Korea (0%) had notably less international collaboration. CT
research activity has rapidly grown in some of the countries included, and the most productive countries (US excluded)
have produced between 70% and 97% of their CT research output in the last two years (2018–2019). One perspective
to the disproportionate amount of US-based research on CT is that by the time other countries jumped on the CT
bandwagon, US-based researchers had already produced hundreds of papers on the topic.
Network analysis of international collaboration networks, based on countries of all paper authors, revealed large and
diverse collaboration groups that still followed some regional and linguistic divisions. Figure 6 shows a network analysis
of country-level collaborations: In the gure, circle size indicates the number of papers per country, line thickness
indicates number of co-authored articles, and circle color indicates clusters of countries from which authors frequently
collaborate together. The United States had the most collaborating country partners (38), followed by the UK (24), Spain
(23), Italy (19), Finland (17), and Australia (15).
Countries were clustered using community detection, which showed Spain, Italy, and Portugal forming one community
with South American countries Chile, Mexico, Brazil, Colombia, and Argentina. Another cluster was formed by three
Nordic countries Sweden, Finland, and Denmark, and another by Czech Republic, Hungary, Slovenia, and Germany. The
largest cluster was formed around the US and Canada with a number of Asian and Middle Eastern countries around it.
Despite the relatively national nature of CT publications in the US, the US still had the highest betweenness centrality,
highlighting its importance in bridging links between other countries; Spain scored second on that measure.
Manuscript submitted to ACM

People, Ideas, Milestones: A Scientometric Study of Computational Thinking 13
USA
ITALY
CANADA
SPAIN
FINLAND
AUSTRALIA
IRELAND
GERMANY
UNITED KINGDOM
SWEDEN
INDIA
NETHERLANDS
MEXICO
NORWAY
BULGARIA
SWITZERLAND
CHILE
FRANCE
LITHUANIA
ROMANIA
CZECH REPUBLIC HUNGARY
SLOVENIA
NIGERIA
MALAYSIA
CHINA SINGAPORE
NEW ZEALAND
COLOMBIA
BRAZIL
TURKEY
DENMARK
HONG KONG
ISRAEL
JAPAN
TAIWAN
SAUDI ARABIA
PHILIPPINES
EGYPT
AZERBAIJAN
CYPRUS
NAMIBIA
QATAR
SOUTH AFRICA
DOMINICAN REPUBLIC
PARAGUAY
ZIMBABWE
GREECE
BELGIUM
PORTUGAL
MOROCCO
ARGENTINA
PANAMA
ECUADOR
CUBA
KOREA
MACEDONIA
LATVIA
SLOVAKIA
BAHRAIN
ESTONIA
GUATEMALA
THAILAND
CROATIA
Fig. 6. Network of international collaborations in CT-related articles (countries who frequently collaborate together have a similar
color, circle size indicates papers per country; line thickness indicates number of co-authored articles).
4 DISCUSSION
CT literature inherits some characteristics from the broader computing literature. Two of those characteristics—legitimate
venues for publishing and types of research published—have been a source of lively debates in computing. Firstly,
CT literature inherits eld-specic publishing preferences from computing as a discipline, which has traditionally
focused on conferences stronger than other elds do [
65
]. With two thirds of the publications, conference papers were
the most common form of publishing CT literature, and one in four publications were journal articles. But journal
articles got cited much more (mean 10.5 citations per journal article) than conference papers did (mean 3.8 citations
per paper)—surpassed only by review articles (mean 41.9 citations per article). Secondly, CT literature inherits the
diversity of paper types from computing education research (CER), where researchers have for more than three decades
discussed the need for more empirical research, and where the shift towards empirical research has been considered a
sign of maturing of the eld [
58
]. CER literature has traditionally been a mixture of position papers, essays, course
reports, empirical research, literature reviews, and systems papers [see, e.g.
58
], and the computational thinking body
of literature shares that characteristic, as witnessed by the diverse nature of CT’s most cited publications.
CT research lacks paradigmatic exemplars and strong consensus on methods, theories, and what signicant results
look like. A set of inuential CT articles can clearly be identied, but instead of forming a grand narrative, they are
Manuscript submitted to ACM

14 Saqr, Ng, Oyelere, and Tedre
inuential for very dierent reasons and intended for very dierent audiences: Some became inuential for their
persuasive argument, some for reporting innovative CT initiatives, and some for explaining CT to audiences outside
computing. In terms of citing patterns, the key inuential articles in the rst half of the studied period (Figure 2,
2006–2012) cited other CT entries sparsely, and their lists of references were short in general, too (their bibliographies
contained 13.6 mean cited items per article). In the second half of the studied period, the CT research community saw
an increased consensus of signicant CT literature, and the period saw many more citations to CT-specic literature
and much longer lists of references—also in articles that were not literature reviews.
The most inuential authors were not necessarily the most prolic: There are only a handful of people who are both
on the list of authors of most-cited articles in CT and at the same time on the list of most active contributors to CT
in terms of publication numbers. But although some opinion leaders’ contributions to CT and CER body of literature
are limited to a small number of position papers, the same people can at the same time be known for their inuential,
massive scale CT advocacy work with transformative impact.
Computational thinking has been a relatively US-centric movement from the start. The rst, slowly growing wave of
CT publications (2006–2012) (Figure 1) was heavily US-based. The US institutions dominated the second wave, too
(2013–2019), but institutions from a growing number of countries joined the second wave. Still, twelve of the seventeen
most prolic CT authors were aliated with US-based institutions in their most cited publications, 42% of all CT papers
had lead authors from US-based institutions, and US-led publications were cited over ve times more than the next nine
countries combined. NSF’s role in the early years of CT cannot be downplayed: By requiring all CPATH-funded projects
to address the term “computational thinking,” NSF funding ensured that some of the leading computing education
research groups in the US framed their work in CT vocabulary. Funding from the NSF was acknowledged in nine out of
fteen most inuential papers in Figure 2.
Other countries started to join the movement in the second half of the investigation period (2013–2019), but the US
remains by far the largest producer of CT-related research (Figure 1), with disproportionately high share of papers
compared to the country’s share of research in computer science as well as in all elds of research. US has some
multi-country publications (6.3%), but CT research in many countries is much more international than US-based
research is. The expansion of CT research outside the US alleviates risks such as cultural bias, diminished applicability,
and losing richness of ideas and approaches from other educational and intellectual contexts. There are relatively active
clusters of international collaboration centered around, for example, Spain, Italy, and Brazil, but their absolute number
of multi-country publications is relatively small—around ten or so per country. There may be various reasons to the
very low citation numbers to South Korean (0.81 mean cites) and Chinese (1.75 mean cites) CT publications, such as the
dierences between countries’ educational systems that complicates collaboration as well as applicability of many K–12
CT projects. Very little CT research originated in Africa, a home to 1.2 billion people, although as this study was limited
to publications in English, any major publishing outlets in other languages were not accounted for. To what extent a
predominantly western body of K–12 CT literature can cater to the needs of education systems on all continents and all
countries remains an open question.
[
31
] wrote that CT is about programming—a “battle cry for coding in K–12 education,” yet the role of programming
in K–12 CT was heavily debated in the beginning. These results conrm Kafai’s view, and they also conrm that
much of CT literature is about “CT for beginners” and how to teach programming-related concepts to children [
19
].
Keyword analysis and keyword co-occurrence network analysis associated computational thinking strongest with
programming-related keywords and with a variety of education-related keywords each of which point at distinct
branches of CT-related research (such as K–12, STEM, assessment, and robotics). Assessment, which has often been
Manuscript submitted to ACM

People, Ideas, Milestones: A Scientometric Study of Computational Thinking 15
identied as a main weakness of CT initiatives [
16
], is among the top ten keywords in CT publications. The keyword
co-occurrence network further featured elements of computing’s traditional tripartite disciplinary structure—design,
modeling, and theory—and it also showed a number of prominent pedagogical and educational features of CT, including
constructionism, self-ecacy, motivation, and teacher training, for example.
As CT continues to develop as a research topic, it faces pressure to keep aligned with and connected to computing
education research and its prevailing trends, including developing stronger theoretical grounding and empirically
established knowledge base. What remains to be seen is to what extent the ongoing extension of computing from the
classical “good old rule-driven programming” to data-driven automation changes CT and CER. There is a growing
recognition of the need to better understand how people, including children in K–12 education, learn articial intelligence
and machine learning, yet that shift is not yet well visible in CT literature. And in order to full its original promise to
better prepare learners for a world driven by algorithms, there still is work to be done on extending CT beyond the
currently popular groups—K–12 learners and computer science students—into educating people like doctors, economists,
lawyers, and all other groups whose jobs are likely to be aected [53].
REFERENCES
[1] Aho, A. V. (2011). Ubiquity symposium: Computation and computational thinking. Ubiquity, 2011(January).
[2] Aria, M. and Cuccurullo, C. (2017). Bibliometrix: An R-tool for comprehensive science mapping analysis. Journal of Informetrics, 11(4):959–975.
[3]
Armoni, M. (2016). Computer science, computational thinking, programming, coding: The anomalies of transitivity in k–12 computer science
education. ACM Inroads, 7(4):24–27.
[4] Aspray, W. (2016). Participation in Computing: The National Science Foundation’s Expansionary Programs. Springer, Switzerland.
[5]
Barr, V. (2016). Disciplinary thinking, computational doing: Promoting interdisciplinary computing while transforming computer science enrollments.
ACM Inroads, 7(2):48–57.
[6]
Barr, V. and Stephenson, C. (2011). Bringing computational thinking to K-12: What is involved and what is the role of the computer science education
community? ACM Inroads, 2(1):48–54.
[7]
Basawapatna, A., Koh, K. H., Repenning, A., Webb, D. C., and Marshall, K. S. (2011). Recognizing computational thinking patterns. In Proceedings of
the 42nd ACM Technical Symposium on Computer Science Education, SIGCSE ’11, pages 245–250, New York, NY, USA. ACM.
[8] Bell, T., Alexander, J., Freeman, I., and Grimley, M. (2009). Computer science unplugged: School students doing real computing without computers.
The New Zealand Journal of Applied Computing and Information Technology, 13(1):20–29.
[9]
Bers, M. U., Flannery, L., Kazako, E. R., and Sullivan, A. (2014). Computational thinking and tinkering: Exploration of an early childhood robotics
curriculum. Computers & Education, 72:145–157.
[10] Connor, R., Cutts, Q., and Robertson, J. (2017). Keeping the machinery in computing education. Communications of the ACM, 60(11):26–28.
[11]
Curzon, P., Bell, T., Waite, J., and Dorling, M. (2019). Computational thinking. In Fincher, S. A. and Robins, A. V., editors, The Cambridge Handbook
of Computing Education Research, pages 513–546. Cambridge University Press.
[12]
Dagien
˙
e, V. and Sentance, S. (2016). It’s computational thinking! Bebras tasks in the curriculum. In Brodnik, A. and Tort, F., editors, Informatics in
Schools: Improvement of Informatics Knowledge and Perception, pages 28–39, Cham. Springer International Publishing.
[13]
De Meo, P., Ferrara, E., Fiumara, G., and Provetti, A. (2011). Generalized Louvain method for community detection in large networks. In Proceedings
of the 11th International Conference on Intelligent Systems Design and Applications, pages 88–93.
[14] Denning, P. J. (2009). Beyond computational thinking. Communications of the ACM, 52(6):28–30.
[15] Denning, P. J. (2017a). Computational thinking in science. American Scientist, 105(1):13–17.
[16] Denning, P. J. (2017b). Remaining trouble spots with computational thinking. Communications of the ACM, 60(6):33–39.
[17]
Denning, P. J., Comer, D. E., Gries, D., Mulder, M. C., Tucker, A., Turner, A. J., and Young, P. R. (1989). Computing as a discipline. Communications of
the ACM, 32(1):9–23.
[18] Denning, P. J. and Martell, C. H. (2015). Great Principles of Computing. The MIT Press, Cambridge, MA, USA.
[19] Denning, P. J. and Tedre, M. (2019). Computational Thinking. The MIT Press, Cambridge, MA, USA.
[20]
Fronza, I., El Ioini, N., and Corral, L. (2015). Students want to create apps: Leveraging computational thinking to teach mobile software development.
In Proceedings of the 16th Annual Conference on Information Technology Education, SIGITE ’15, pages 21–26, New York, NY, USA. ACM.
[21]
García-Peñalvo, F. J. (2016). A brief introduction to TACCLE 3—coding European project. In 2016 International Symposium on Computers in Education
(SIIE), pages 1–4.
[22]
García-Peñalvo, F. J., Reimann, D., Tuul, M., Rees, A., and Jormanainen, I. (2016). An overview of the most relevant literature on coding and
computational thinking with emphasis on the relevant issues for teachers. Technical report, TACCLE3 Consortium, Belgium.
Manuscript submitted to ACM

16 Saqr, Ng, Oyelere, and Tedre
[23] Gareld, E. (2004). Historiographic mapping of knowledge domains literature. Journal of Information Science, 30(2):119–145.
[24] Grover, S. (2018). A tale of two CTs (and a revised timeline for computational thinking). Blog@CACM.
[25] Grover, S. and Pea, R. D. (2013). Computational thinking in K–12: A review of the state of the eld. Educational Researcher, 42(1):38–43.
[26] Guzdial, M. (2008). Education: Paving the way for computational thinking. Communications of the ACM, 51(8):25–27.
[27]
Guzdial, M. (2015). Learner-Centered Design of Computing Education: Research on Computing for Everyone. Synthesis Lectures on Human-Centered
Informatics. Morgan & Claypool, San Rafael, CA, USA.
[28]
Gynnild, A. (2014). Journalism innovation leads to innovation journalism: The impact of computational exploration on changing mindsets. Journalism,
15(6):713–730.
[29]
Hambrusch, S., Homann, C., Korb, J. T., Haugan, M., and Hosking, A. L. (2009). A multidisciplinary approach towards computational thinking for
science majors. In Proceedings of the 40th ACM Technical Symposium on Computer Science Education, SIGCSE ’09, pages 183–187, New York, NY, USA.
ACM.
[30] Hemmendinger, D. (2010). A plea for modesty. ACM Inroads, 1(2):4–7.
[31] Kafai, Y. B. (2016). From computational thinking to computational participation in K–12 education. Communications of the ACM, 59(8):26–27.
[32] Knuth, D. E. (1974). Computer science and its relation to mathematics. American Mathematical Monthly, 81(Apr.1974):323–343.
[33]
Koh, K. H., Basawapatna, A., Bennett, V., and Repenning, A. (2010). Towards the automatic recognition of computational thinking for adaptive
visual language learning. In 2010 IEEE Symposium on Visual Languages and Human-Centric Computing, pages 59–66.
[34]
Koh, K. H., Basawapatna, A., Nickerson, H., and Repenning, A. (2014). Real time assessment of computational thinking. In 2014 IEEE Symposium on
Visual Languages and Human-Centric Computing (VL/HCC), pages 49–52.
[35] Lamprou, A. and Repenning, A. (2017). Computational thinking ≠programming. SI Magazine, 2.
[36] Lee, I. (2016). Reclaiming the roots of CT. CSTA Voice: The Voice of K–12 Computer Science Education and Its Educators, 12(1):3–4.
[37]
Lee, I., Martin, F., Denner, J., Coulter, B., Allan, W., Erickson, J., Malyn-Smith, J., and Werner, L. (2011). Computational thinking for youth in practice.
ACM Inroads, 2(1):32–37.
[38]
Lockwood, J. and Mooney, A. (2017). Computational thinking in education: Where does it t? a systematic literary review. Technical report, National
University of Ireland Maynooth.
[39] Lu, J. J. and Fletcher, G. H. (2009). Thinking about computational thinking. SIGCSE Bulletin, 41(1):260–264.
[40]
Lye, S. Y. and Koh, J. H. L. (2014). Review on teaching and learning of computational thinking through programming: What is next for K–12?
Computers in Human Behavior, 41:51–61.
[41]
Mannila, L., Dagien
˙
e, V., Demo, B., Grgurina, N., Mirolo, C., Rolandsson, L., and Settle, A. (2014). Computational thinking in K–9 education. In
Proceedings of the Working Group Reports of the 2014 on Innovation & Technology in Computer Science Education Conference, ITiCSE-WGR ’14, pages 1–29,
New York, NY, USA. ACM.
[42]
Min, W., Frankosky, M. H., Mott, B. W., Rowe, J. P., Wiebe, E., Boyer, K. E., and Lester, J. C. (2015). Deepstealth: Leveraging deep learning models for
stealth assessment in game-based learning environments. In Conati, C., Heernan, N., Mitrovic, A., and Verdejo, M. F., editors, Articial Intelligence in
Education, pages 277–286, Cham. Springer International Publishing.
[43]
Moreno-León, J. and Robles, G. (2015). Dr. Scratch: A web tool to automatically evaluate scratch projects. In Proceedings of the Workshop in Primary
and Secondary Computing Education, WiPSCE ’15, pages 132–133, New York, NY, USA. ACM.
[44] Nardelli, E. (2019). Do we really need computational thinking? Communications of the ACM, 62(2):32–35.
[45]
National Research Council (2010). Report of a Workshop on the Scope and Nature of Computational Thinking. The National Academies Press,
Washington, DC, USA.
[46]
National Research Council (2011). Report of a Workshop on the Pedagogical Aspects of Computational Thinking. The National Academies Press,
Washington, DC, USA.
[47] Nicolajsen, S. M. (2020). Perspectives on Computational Thinking. IT University of Copenhagen, Copenhagen, Denmark.
[48] Papert, S. (1980). Mindstorms: Children, Computers, and Powerful Ideas. Basic Books, New York, NY, USA.
[49] Pears, A. N. (2019). Developing computational thinking, “fad” or “fundamental”? Constructivist Foundations, 14(3):410–412.
[50]
Repenning, A., Webb, D., and Ioannidou, A. (2010). Scalable game design and the development of a checklist for getting computational thinking into
public schools. In Proceedings of the 41st ACM Technical Symposium on Computer Science Education, SIGCSE ’10, pages 265–269, New York, NY, USA.
ACM.
[51] Resnick, M. (2017). Lifelong Kindergarten: Cultivating Creativity through Projects, Passion, Peers, and Play. The MIT Press, Cambridge, MA, USA.
[52]
Sáez-López, J.-M., Román-González, M., and Vázquez-Cano, E. (2016). Visual programming languages integrated across the curriculum in elementary
school: A two year case study using “scratch” in ve schools. Computers & Education, 97:129–141.
[53]
Saqr, M. and Tedre, M. (2019). Should we teach computational thinking and big data principles to medical students? International journal of health
sciences, 13(4):1–2.
[54]
Seiter, L. and Foreman, B. (2013). Modeling the learning progressions of computational thinking of primary grade students. In Proceedings of the
Ninth Annual International ACM Conference on International Computing Education Research, ICER ’13, pages 59–66, New York, NY, USA. ACM.
[55]
Sengupta, P., Kinnebrew, J. S., Basu, S., Biswas, G., and Clark, D. (2013). Integrating computational thinking with K–12 science education using
agent-based computation: A theoretical framework. Education and Information Technologies, 18(2):351–380.
Manuscript submitted to ACM

People, Ideas, Milestones: A Scientometric Study of Computational Thinking 17
[56]
Shell, D. F. and Soh, L.-K. (2013). Proles of motivated self-regulation in college computer science courses: Dierences in major versus required
non-major courses. Journal of Science Education and Technology, 22(6):899–913.
[57] Shute, V. J., Sun, C., and Asbell-Clarke, J. (2017). Demystifying computational thinking. Educational Research Review, 22:142–158.
[58] Simon (2015). Emergence of Computing Education as a Research Discipline. PhD thesis, Aalto University, Finland.
[59]
Solomon, C. J. (1976). Teaching the computer to add: An example of problem-solving in an anthropomorphic computer culture. Memo 396,
Massachusetts Institute of Technology, AI Laboratory.
[60]
Tedre, M. and Denning, P. J. (2016). The long quest for computational thinking. In Proceedings of the 16th Koli Calling International Conference on
Computing Education Research, Koli Calling ’16, pages 120–129, New York, NY, USA. ACM.
[61] Tenenberg, J. (2018). Computational making. ACM Inroads, 9(1):22–23.
[62]
Tissenbaum, M., Sheldon, J., and Abelson, H. (2019). From computational thinking to computational action. Communications of the ACM, 62(3):34–36.
[63] Vee, A. (2017). Coding Literacy: How Computer Programming is Changing Writing. The MI T Press, Cambridge, MA, USA.
[64]
Voogt, J., Fisser, P., Good, J., Mishra, P., and Yadav, A. (2015). Computational thinking in compulsory education: Towards an agenda for research and
practice. Education and Information Technologies, 20(4):715–728.
[65]
Vrettas, G. and Sanderson, M. (2015). Conferences versus journals in computer science. Journal of the Association for Information Science and
Technology, 66(12):2674–2684.
[66]
Weintrop, D., Beheshti, E., Horn, M., Orton, K., Jona, K., Trouille, L., and Wilensky, U. (2016). Dening computational thinking for mathematics and
science classrooms. Journal of Science Education and Technology, 25(1):127–147.
[67]
Werner, L., Denner, J., Campe, S., and Kawamoto, D. C. (2012). The fairy performance assessment: Measuring computational thinking in middle
school. In Proceedings of the 43rd ACM Technical Symposium on Computer Science Education, SIGCSE ’12, pages 215–220, New York, NY, USA. ACM.
[68] Wing, J. M. (2006). Computational thinking. Communications of the ACM, 49(3):33–35.
[69] Wing, J. M. (2008). Computational thinking and thinking about computing. Philosophical Transactions of the Royal Society A, 36(1881):3717–3725.
[70]
Yadav, A., Mayeld, C., Zhou, N., Hambrusch, S., and Korb, J. T. (2014). Computational thinking in elementary and secondary teacher education.
ACM Transactions on Computing Education, 14(1):5:1–5:16.
Manuscript submitted to ACM