Journal of Education, 2024

Issue 96, http://journals.ukzn.ac.za/index.php/joe doi: http://dx.doi.org/10.17159/2520-9868/i96a08

The promise and peril of Coding & Robotics education in

South Africa: A scoping review of teacher preparation and
generative artificial intelligence’s potential for delivering
equity

Mashite Tshidi
Division of Science and Technology, Wits School of Education, University of the Witwatersrand, Johannesburg,
South Africa
[email protected]
https://orcid.org/0000-0002-1288-1576

Alton Dewa
Division of Science and Technology, Wits School of Education, University of the Witwatersrand, Johannesburg,
South Africa
[email protected]
https://orcid.org/0000-0002-0776-4447

(Received: 31 March 2024; accepted: 14 August 2024)

Abstract
Integrating the Coding & Robotics (C&R) subject in South African schools signifies the nation’s commitment to
Fourth Industrial Revolution preparedness. However, challenges like inadequate teacher preparation and limited
technological infrastructure must be addressed to ensure equity. Although Generative Artificial Intelligence
(GenAI) may not address the infrastructural deficiencies directly, in this scoping review we examine its
potential to complement existing resources and support teachers in delivering C&R instruction. Following
Arksey and O’Malley’s framework, we conducted a systematic literature search in numerous databases,
followed by a screening procedure: 10 of the 61 eligible papers satisfied the inclusion criteria. Our findings
reveal that GenAI can optimise C&R teacher development through personalised learning, content generation,
feedback on teaching methods, and fostering collaboration with other teachers. Despite its potential, issues
including equity, ethical concerns, technological fluency gaps, and overreliance on GenAI tools, must be
navigated to enhance equitable C&R instruction and prepare learners for the digital era.

Keywords: Coding & Robotics, GenAI, scoping review, teacher preparation, equity

Online ISSN 2520-9868 Print ISSN 0259-479X

 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 141

Introduction
Despite rapidly advancing technology and a global knowledge-based economy, educational
systems continually adapt to meet 21st-century demands. Numerous countries undergo
educational reforms driven by economic, social, and cultural objectives to prepare individuals
for success in a digitally integrated world (Reimers, 2021). Countries have implemented these
reforms to maintain a competitive edge economically, culturally, and socially in a
technology-driven global landscape in recognising the importance of cultivating active
creators rather than passive consumers of technology, and the role of individuals as agents of
cultural change (Storte et al., 2019). The convergence of economic, social, and cultural
factors propels educational systems towards mapping a terrain in which individuals can thrive
in a technological landscape.

In South Africa, a pivotal initiative of this reform was the introduction of a new subject in
2021 called “Coding & Robotics” (C&R) across Grades R–9 by the Department of Basic
Education (DBE). This initiative aligns with the nation’s commitment to the Fourth Industrial
Revolution (4IR) to prevent learners from being left behind (Ogegbo & Ramnarain, 2022).
The 4IR era in South Africa represents a transformative period marked by the convergence of
social, digital, and economic realms driven by technological advancements (Arendse, 2023),
thereby necessitating a paradigm shift in education, like introducing C&R subjects that aim to
equip learners with problem-solving abilities, foster critical thinking, encourage collaborative
and creative skills, and prepare them to thrive in a digital information-driven era (Department
of Basic Education, 2021).

However, the piloting of C&R across Grades R–9 hinges not only on the infrastructure and
curriculum design but also on the training and assistance provided to the teachers. Teachers
have a significant impact on how learners perceive programming and their ability to master it
(Bano et al., 2023). Still, they frequently face challenges in obtaining the expertise and
developing the abilities needed to teach C&R effectively (Hadad et al., 2021). The inadequate
technological infrastructure in many South African schools, with nearly 16,000 of them
lacking dedicated computer labs, further exacerbates these challenges (Greyling, 2023). This
infrastructural deficiency limits the hands-on learning component with digital tools integral to
C&R. C&R provides learners with an engaging and dynamic learning environment by
allowing them to assemble and programme robots thus creating an enjoyable learning
experience (Castro et al., 2018). The ramifications of this deficiency are far-reaching since
the practical aspects of the subject may remain unrealised, hindering the development of
technical skills necessary for navigating the digitally integrated world (Storte et al., 2019).

Compounding this challenge is the limited access to comprehensive and ongoing professional
development programs designed for teaching C&R. Although there is a possibility that many
teachers are (un)aware of the advantages of C&R and are not equipped to teach it well, this is
primarily down to the dearth of specialised training programs provided by educational
institutions and an inadequate level of existing professional development initiatives (Schina
et al., 2021). This challenge stems from barriers hindering sustainable professional
142 Journal of Education, No. 96, 2024

development in C&R education, including the lack of collaboration between higher education
institutions and schools, thereby constraining the impact of professional development
programs on transforming teaching practices (Ni et al., 2023). For instance, while South
Africa’s DBE has long offered Information Technology (IT) as a subject for secondary
learners in Grades 10–12 to instil computing fundamentals, including coding and problem-
solving skills (Department of Basic Education, 2011), many C&R teachers may lack
extensive practical coding experience given their university education or IT teaching
backgrounds. Moreover, implementing short programs without sustained support fails to
support teachers with the requisite continuous learning and growth essential for proficient
coding instruction. This underscores the lack of focus on discipline-specific pedagogical
content knowledge (PCK) in professional development endeavours (Ni et al., 2023), often
resulting in teachers being insufficiently prepared and lacking the specialised skills and
understanding required for effective C&R instruction (Changpetch et al., 2022).

Given these challenges, innovative solutions are needed to bridge the gap between subject
requirements and teacher capacity (Gorrah & Papers, 2024; Schina et al., 2021). GenAI’s
capacity to generate content in the formats of text, photos, music, videos, and 3D models by
detecting patterns and information from written requests provides a potential solution to these
challenges (Chiu, 2023). This is made possible by GenAI’s wide range of applications,
including OpenAI (ChatGPT, GPT-4), Amazon (CodeWhisperer), and Google (Alpha-Code,
Gemini), among many others that have gained attention recently and are used to personalise
learning experiences, automate administrative tasks, and enhance teachers’ and learners’
abilities in programming classrooms (Becker et al., 2023). While GenAI may not address the
physical infrastructural deficiencies, it can complement existing resources and support
teachers in delivering C&R instruction (Greyling, 2023).

Against this backdrop, in this scoping review, we investigate how GenAI can address
teachers’ challenges in implementing C&R instruction in South Africa. The study is
predicated on the idea that effective integration of C&R instruction requires well-prepared
teachers. Adequate teacher preparation significantly affects the development of learners’
digital abilities, especially their coding skills (Chiu, 2023). Thus, in this review, we seek to
understand how GenAI can mitigate PCK gaps, minimise the adverse impacts of inadequate
technological infrastructure, and promote long-term learning and growth among teachers
overseeing C&R instruction. To achieve these objectives, the following research questions
guided the review:

1. How can GenAI address teachers’ challenges in acquiring the necessary pedagogical
skills and programming knowledge for effective C&R instruction?
2. What are GenAI’s potential implications for teacher preparation and support in C&R
education?

To answer these research questions, in the rest of the review, we build on the introduced need
for teacher preparation for C&R in South African schools. We begin the discussion in the
next section with a historical overview of programming education in South Africa through
the key policies and initiatives that have influenced it. This includes attending to teachers’
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 143

challenges in acquiring the necessary knowledge and skills to teach C&R effectively. In the
following section, our focus shifts to the methodological approach to the study. Moving
forward, we then focus on the results and discussion of potential applications of GenAI in
C&R education and navigate the tensions associated with its integration from the reviewed
studies. Finally, we offer a summary and a call to action for further research.

Foundations of programming education in South African

schools
Formal programming education in South Africa has a complex history that is influenced by
the legacy of apartheid. Before 1994, racial segregation resulted in a stark disparity in
resources and instructional quality between white and Black schools (Fleisch, 2002). White
schools received far more resources and provided a wider range of academic subjects,
whereas Black schools had lower academic expectations and emphasised practical skills for
blue-collar occupations (Fiske & Ladd, 2004). The unequal distribution of funding and
resources resulted in overcrowding in classrooms, poor infrastructure, and insufficiently
qualified teachers in schools serving Black learners, thus perpetuating attainment
discrepancies (Fleisch, 2002). This disparity hindered the education of Black learners in
computer science and impeded the development of a diverse IT workforce (Galpin &
Sanders, 2007).

Post-apartheid efforts were made to address these inequalities. A major step in eradicating the
legacy of apartheid education was the adoption of Curriculum 2005 (C2005) in 1998. C2005
focused on learner-centred and outcomes-based education (OBE) but faced challenges
because of teachers’ inadequate training (Fiske & Ladd, 2004). Factored in here, was the
butterfly effect of apartheid’s legacy that left many schools, particularly those in historically
disadvantaged communities, with inadequate resources and (technological) infrastructure to
fully implement the C2005 philosophy in their programmes (Fleisch, 2002). In 2002, C2005
was revised and became the Revised National Curriculum Statement (RNCS). With an
emphasis on eight learning areas, including Technology, the RNCS stressed principles like
social justice, fairness, and development and encouraged the growth of creative, critical, and
problem-solving individuals to provide the groundwork for computer science education and
also sought to improve teacher training and resources (Department of Education, 2002;
Galpin & Sanders, 2007).

The 2004 introduction of RNCS was followed by its 2007 revision of the National
Curriculum Statement (NCS). NCS’s extension focus on integrating technology into the
classroom led to the introduction of new subjects such as Computer Applications Technology
(CAT) and Information Technology (IT) in Grades 10–12 under the Curriculum Assessment
Policy Statement (CAPS) (Department of Basic Education, 2011). These subjects were
endorsed by two ICT-focused white paper policies that were produced by the South African
Government Gazette with other plans and strategies, such as the 2016 National Integrated
ICT Policy White Paper, 2017 National e-Strategy, and 2020–24 Basic Education Strategic
Plan supporting their implementation to create capacity, establish ICT norms and standards,
144 Journal of Education, No. 96, 2024

promote fairness, and provide access to ICT infrastructure (Department of Education, 2002,
2004). The 2004 White Paper on e-Education encourages teachers to use ICT to improve
teaching and learning across the board in the education and training system to guarantee ICT
competence for all learners in the nation (Department of Education, 2004). While the
introduction of CAT and IT represented progress in the attainment of these goals, literature
on these fields highlighted challenges that contributed to the digital divide and limited
foundational digital skills development since they were targeted at the senior grades in
secondary school (Koorsse et al., 2015; Mentz et al., 2012).

Effective computer programming requires integrating critical thinking, problem-solving, and

science-based reasoning with program design and development. Teachers unprepared for this
multifaceted approach struggle to balance these demands (Ogegbo & Ramnarain, 2022). The
focus of IT is often on how teachers can support learners in learning how to solve problems
by providing procedural information rather than prioritising the development of critical
thinking and problem-solving abilities. This is reflected in the objectives of the subject;
learners are required to demonstrate proficiency in using appropriate techniques and
procedures to plan solutions and devise algorithms for problem-solving using suitable
techniques and tools (Department of Basic Education 2011). This approach may limit
learners’ ability to adapt to new programming challenges and languages by concentrating on
predefined methods rather than fostering creative problem-solving skills (Prather et al.,
2023).

Unequal resource distribution exacerbates these challenges. The DBE mandates computer
labs and internet access for IT classes (Department of Education, 2011). However, many
public schools, particularly in rural and historically disadvantaged areas, still lack the
necessary infrastructure (Munje & Jita, 2020). This digital divide restricts learners’ access to
computers and hampers their ability to acquire basic programming skills.

The Department of Basic Education’s 2021 initiative to make C&R a compulsory subject for
Grades R–9 aims to address these challenges by equipping learners with fundamental ICT
skills, including problem-solving, critical thinking, collaboration, and innovation (Moraiti et
al., 2022). C&R is intended to be taught alongside core subjects such as reading, writing, and
mathematics. However, many South African learners lack basic skills in these foundational
subjects. This has ignited ongoing debate on whether addressing the lack of these basic skills
should precede the immediate challenges associated with the C&R subject (Geldenhuys &
Fataar, 2021). An examination of learners’ proficiencies in areas such as writing, reading, and
mathematics indicates the need to integrate computational thinking that encompasses
problem-solving capabilities, system design acumen, and a nuanced understanding of human
behaviour (Moraiti et al., 2022). This aspect also bolsters acquiring and retaining STEM
knowledge and skills. Therefore, the endorsement of computational thinking in early
exposure to C&R can enhance learners’ STEM knowledge by piquing their curiosity, helping
them apply existing concepts, and inspiring the development of new ones (Bano et al., 2023).
This technique can help to add interest and enjoyment to this learning exercise. Research has
indicated that learners may learn computer programming, science, maths, and problem-
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 145

solving techniques with robots (Prather et al., 2023). Given that C&R is enjoyable,
introducing this in primary and secondary schools while fostering learners’ creativity and
logical and language abilities is a great initiative. However, the efficacy of integrating
computational thinking into standard teaching practices of C&R depends on the ability of
teaching staff to meet the demands of the subject (Schina et al., 2021).

Integrating computational thinking into C&R teaching requires addressing teachers’

misconceptions and limited knowledge. Research indicates that teachers often lack
engagement with C&R concepts because of insufficient training (Changpetch et al., 2022).
The effectiveness of C&R activities is diminished if prior training is lacking (Castro et al.,
2018). Furthermore, research indicates that teachers’ attitudes significantly influence how
well educational technology is used (Litlhakanyane, 2022; Schina et al., 2021). Teachers tend
to be less motivated to conduct tasks if they lack competence in the subject matter. Although
brief exposure to educational technology courses and tools has demonstrated the potential to
augment teachers’ understanding of C&R (Boz & Allexsaht-Snider, 2022), comprehensive
training programs covering basic programming concepts are still needed. Addressing these
challenges is essential for the success of the C&R initiative. Hence, in this scoping review,
we examine the promising solutions that GenAI offers to bridge this gap.

Methodology
Our scoping review aimed to explore the promises and perils related to integrating GenAI in
teacher preparation for C&R education. For this review, we used a systematic strategy to
search, select, and synthesise existing literature. We adhered to the framework established by
Arksey and O’Malley (2005). This review consists of four steps beyond the identified
research questions that include (a) identifying relevant studies, (b) study selection, (c)
charting the data, and (d) collating, summarising, and reporting results. The adopted rigorous
transparency method we adopted allows the search strategy to be replicated and enhances the
validity of the research findings (Fan et al., 2022).

Identifying relevant studies

We undertook a comprehensive literature review published from 2022 to May 2024. Figure
1—Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)
diagrams, provided by Moher et al. (2009), detail the procedure used to synthesise the
published peer-reviewed literature. We carried out the literature search using many databases,
including the Education Resources Information Center (ERIC), Google Scholar, IEEE
Xplore, and Scopus. Examples of the used search strings focused on the concept of “GenAI”
and its applications in “teacher preparation” and “coding and robotics”, including derivative
terms like “artificial intelligence,” “machine learning,” and “educational technology.”
Outcome-related keywords included “pedagogical content knowledge,” “technological
infrastructure,” “teacher training,” and “C&R instruction.” These terms and keywords were
found in published works on GenAI, C&R education and teacher preparation.
146 Journal of Education, No. 96, 2024

Figure 1
Flow diagram for data searched (Moher et al., 2009).

Study selection

We screened over 221 abstracts using Covidence, a systematic review management system
enabling individuals to evaluate abstracts for inclusion. The following criteria were used to
determine study inclusion: (a) studies focused on using GenAI in educational contexts,
c (b)
studies on teacher preparation and professional development for C&R education, (c) studies
published between 2022 and 2024, and (e) studies examining GenAI applications and
computational thinking in teacher training. Dissertation studies were omitted. When we
concluded that the abstract satisfied the inclusion requirements, we moved on to the full-text
full
review stage of this article.

We evaluated more than 200 publications in their entirety. Following the analysis, we used
specific grounds for exclusion
clusion in the following sequence when publications were eliminated
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 147

during the full-text review: (a) studies outside the scope of teacher preparation or GenAI, (b)
preparation for C&R education, (c) articles not available in full text, (d) non-peer-reviewed
sources, and (e) inadequate rigour in the methodology. This priority allows coding exclusions
when articles may have been omitted for various reasons (Fan et al., 2022). Subsequently, we
searched all included publications manually, searched through reference lists, and identified
recent publications by using the Times Cited count feature of the Web of Science database.

Charting the data

Table 1 describes the resulting 10 studies, organised by the data extracted, including author(s)
and year of publication, study design and methodology, population and sample size, context
and setting, key findings related to GenAI and teacher preparation, and implications for
policy and practice.
148 Journal of Education, No. 96, 2024

Table 1
Summary of data from included studies.

Author(s) Study Population/s Context Findings related to GenAI tools and Implications for policy and practice
and year of design/methodology ample size and setting teacher preparation
publication
Becker et al. Position paper N/A Schools Simplifying problem descriptions can GenAI-generated code can include
(2023) improve teachers’ ability to communicate biases, demanding human oversight
algorithmic challenges accurately, enhancing to ensure safe and ethical school use.
GenAI tools’ success in producing accurate Policymakers must establish
answers. guidelines for the ethical use of
Code-generating tools may expose teachers GenAI tools in the classroom.
to various problem-solving techniques, thus Policy should be developed to
aiding their coding efficiency and guarantee that AI technologies
approaches. facilitate inclusive education.
Teachers may use GenAI tools, including Promote creating a wide range of
Codex, to help learners understand instructional materials using GenAI
programming concepts by creating new tools.
programming assignments and explanations.
Copilot can assist teachers in overcoming
writer’s block by generating initial code and
encouraging rewriting, refactoring, and
debugging.

Bringula Qualitative Teachers and University ChatGPT can create functional programming Teachers need to verify the accuracy
(2024) learners codes with annotated comments. and relevance of generated content.
ChatGPT can generate content for teaching- Regulations need to govern
learning activities and act as a virtual tutor. ChatGPT’s use in educational
Excessive use of ChatGPT may hinder contexts.
teachers’ use of varied instructional materials
and critical thinking.
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 149

Author(s) Study Population/s Context Findings related to GenAI tools and Implications for policy and practice
and year of design/methodology ample size and setting teacher preparation
publication
Changpetch, Mixed-methods 151 University GenAI tools can create interactive learning Policies should support Python
et al. (2022) preservice environments, like the WARiable game, inclusion in the teacher education
teachers helping pre-service teachers grasp abstract curriculum because of its efficacy in
programming concepts through real-world reducing programming
simulations and immediate feedback. misconceptions.
Professional development programs
should equip pre-service and in-
service teachers with tools, including
GenAI, for teaching programming
languages.
Teachers should assess learners’
understanding regularly to identify
and correct misconceptions early.
Teachers should use GenAI tools that
offer early feedback to assist learners
in understanding and correcting their
misconceptions.
150 Journal of Education, No. 96, 2024

Author(s) Study Population/s Context Findings related to GenAI tools and Implications for policy and practice
and year of design/methodology ample size and setting teacher preparation
publication
Chiu et al. Qualitative 88 teachers School GenAI tools, notably ChatGPT and Professional development programs
(2023) Midjourney, help teachers view learning should foster a learn-it-all mentality
objectives differently. These tools emphasise to improve facilitation skills and
the need for digital, media, and information interdisciplinary teaching.
literacy, critical thinking, AI literacy, and the Schools should implement
development of learners’ basic abilities. interdisciplinary teaching strategies to
Incorporating ChatGPT and Midjourney into prepare students for the workforce
classrooms requires teacher development in a and future education in the AI era.
GenAI-driven environment to improve Schools must create new assessment
curriculum leadership, AI literacy, techniques, such as evaluating AI
facilitating abilities, and interdisciplinary literacy, inquiry-based learning, and
teaching. critical thinking.
GenAI tools necessitate rethinking
assessment methods in focusing on formative
and summative approaches that assess digital
literacy and critical thinking.

Denny et al. Mixed-methods Teachers School GenAI presents challenges and opportunities Integrating GenAI tools for teaching
(2024) in C&R education, requiring refining and learning requires teachers to
pedagogical strategies to focus on new skill understand their advantages and
sets. disadvantages.
GenAI tools can solve typical programming Developing assessment procedures
challenges. Therefore, teachers need to that consider the existence of GenAI
rethink assessment strategies. tools is essential.
Pedagogical strategies could emphasise
teaching strategic problem decomposition
and accurate prompt specification for GenAI
tools to students.
Teachers must guide students in using these
tools responsibly.
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 151

Author(s) Study Population/s Context Findings related to GenAI tools and Implications for policy and practice
and year of design/methodology ample size and setting teacher preparation
publication
Ni et al. Systematic review Professional Learning Communities (PLCs) Implementing online, hybrid and in-
(2023) encourage teamwork and reduce isolation person PLCs can meet diverse teacher
among teachers. needs.
PLCs assist teachers inside and Policies should involve
outside professional development initiatives administrators and stakeholders to
through collaboration and resource sharing. support C&R education.

Prather et al. Mixed methods 22 teachers Schools Existing computer programs could include Teachers should acknowledge the
(2023) unstated covert learning objectives, such as presence and potential use of GenAI
problem decomposition, code reading and tools, whether they allow or restrict
tracing. Teachers must identify and them.
incorporate these covert objectives into They must also ensure that learning
lesson design and assessments to better objectives, activities, and assessments
match intended results. have been matched constructively and
Reducing the focus on programming syntax consider how GenAI tools can
with GenAI tools may improve retention influence these components.
rates. GenAI tools can assist in creating
instructional materials, but teachers
should be aware of potential biases in
the outputs.
Teachers can use GenAI to help
learners understand GenAI-generated
code and provide additional practice
opportunities.
152 Journal of Education, No. 96, 2024

Author(s) Study Population/s Context Findings related to GenAI tools and Implications for policy and practice
and year of design/methodology ample size and setting teacher preparation
publication
Sun et al. Quasi-experimental 82 learners College The results indicated that ChatGPT-enabled Teachers must understand the pros
(2024) design learning delivered personalised feedback that and cons of integrating ChatGPT into
aided programming learning. lessons. While it offers personalised
Following the intervention, learners had feedback, it can also produce errors.
better attitudes towards and intentions about Maximising its pros and reducing any
using ChatGPT because they thought it was possible problems requires strategy
helpful and straightforward. and execution.
Teachers should create pedagogical
strategies that use GenAI tools. This
involves explaining how to apply
ChatGPT in programming tasks.
Teacher preparation programs should
include training on using GenAI
tools, covering pedagogical and
technical aspects.

Wieser et al. Mixed-methods 120 learners University Teachers must modify assessment procedures Schools may need to adjust academic
(2023) to account for ChatGPT and other GenAI integrity policies to address the
tools, possibly including assessments to challenges GenAI poses.
better evaluate learners’ comprehension and Policies should guide using ChatGPT
abilities. in assignments, ensuring learners’
While ChatGPT can detect errors and suggest work reflects their understanding and
alternatives, teachers should remain the abilities.
primary assessors for consistency and
precision in grading.
To optimise ChatGPT’s role in personalised
instruction, teachers must plan carefully and
define learning objectives before using it as a
supplementary tutor.
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 153

Author(s) Study Population/s Context Findings related to GenAI tools and Implications for policy and practice
and year of design/methodology ample size and setting teacher preparation
publication
Yilmaz and Mixed-methods 41 learners University ChatGPT allows individuals to practice Teachers should incorporate GenAI
Yilmaz coding and debug code. tools into their teaching practices to
(2023) It can generate solutions to problems but may supplement learning.
offer inaccurate responses if questions are Assignments should require
misinterpreted. algorithmic thinking and integrate
Overreliance on ChatGPT could impede the GenAI responses into broader
development of algorithmic thinking in projects.
teachers and learners. Exploring the versatility of GenAI
tools can help tailor teaching
strategies for different programming
classes and age groups.
Teachers should discuss the ethical
considerations of using GenAI with
learners, underscoring the need to
verify the information produced by
the tools and recognise any potential
biases in its outputs.
154 Journal of Education, No. 96, 2024

Reporting results and discussion

This scoping review maps out the body of research on the potential of integrating GenAI to
address challenges with teacher preparation in C&R instruction. We performed thematic
analysis to organise and categorise the included studies based on key themes linked to the
research questions, including teacher preparation, the potential application of GenAI tools,
and navigating the potential perils are discussed in the subsequent sections.

The potential application of GenAI tools

Research question 1

• How can GenAI address teachers’ challenges in acquiring the necessary pedagogical
skills and programming knowledge for effective C&R instruction?

The introduction of C&R as a compulsory subject in South African schools presents an

opportunity, but as noted earlier, its implementation hinges on a well-equipped teaching
force. Recent AI developments, especially GenAI, have emerged as a potential solution to
this challenge. Many GenAI tools can comprehend and produce source codes and instructions
written in plain language, primarily because of large language models (LLMs) (Wieser et al.,
2023). The reviewed studies indicate the potential of GenAI to address the challenges
teachers encounter in the programming paradigms used in C&R—text-based and block-based
programming. The C&R subject uses pre-existing programming environments and tools, like
Scratch, Microsoft Micro: bit, Arduino, Code Combat, and MakeCode, to make learning
programming efficient and fun (Yilmaz & Yilmaz, 2023). LLMs leverage natural language
processing and machine learning technologies to support these programming tools through
user-friendly interfaces, integration with many programming languages, comprehensive
search capabilities, concise explanations with useful examples, and personalised learning
experiences (Sun et al., 2024; Yilmaz & Yilmaz, 2023).

Drawing on teachers’ existing proficiencies across diverse programming paradigms,

emerging findings from reviewed studies indicate that GenAI tools can evaluate a teacher’s
proficiency, understanding of computer science principles, and desired objectives relevant to
each program’s learning paradigm. This enables GenAI to assess teachers’ basic
understanding of computational concepts and recommend materials that focus on areas in
which the teacher requires the most support and allow them to work at their own pace (Sun et
al., 2024; Wieser et al., 2023). This personalised learning path could be curated in a block-
based learning platform like Scratch. Similar to a study by Govender and Govender (2021)
that focused on pre-service teachers from a South African university and their experiences
with physical computer programming using Scratch for Arduino, Scratch was found to be an
ideal platform to introduce programming concepts because it provides an informal and
beginner-friendly approach. This learning route could be augmented by tutorials and practice
tasks, resulting in proficiency in basic programming concepts before advancing to the text-
based paradigm (Yilmaz & Yilmaz, 2023). In this thread, an LLM such as ChatGPT could
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 155

serve as a personal tutor, modifying examples to help teachers practise a particular coding
concept and offering step-by-step solutions to learn at their own pace (Wieser et al., 2023).

This personalised method enables teachers to avoid reiterating sections of instructional

material that they have previously mastered. Govender and Govender (2021) indicated that a
rapid and efficient transfer to text-based coding should occur once the basic coding concepts
have been covered in block-based programming. When it comes to block-based settings, a
novice teacher may get too accustomed to them and be reluctant to transition to text-based
coding. The text-based programming paradigm requires technical expertise, and an
understanding of the basic concepts required to resolve programming-related challenges
(Wieser et al., 2023). Becoming proficient in various text-based programming languages with
numerous features is the goal of learning to code. To achieve this goal, GenAI can create
learning pathways that concentrate on text-based languages, such as Arduino, frequently
utilised in the subject’s C&R senior phase strand. These paths could include interactive
lessons covering syntax, coding conventions, and debugging techniques (Yilmaz & Yilmaz,
2023). The findings highlight that GenAI-powered learning platforms provide tasks focused
on problem-solving and accurate code writing, helping users improve their coding abilities
(Becker et al., 2023). With the aid of these resources, teachers may enhance their
programming abilities and advance at their own speed while receiving personalised
instruction and immediate feedback (Sun et al., 2024).

GenAI’s capacity to offer timely and personalised feedback on lesson plans, teaching
methods, and coding exercises across various programming paradigms was also found to
address the deficiencies in these areas. Research demonstrates that LLMs like ChatGPT and
Codex offer interactive features that can benefit teachers. These can function as
conversational agents responding to inquiries, explaining programming terminology,
providing relevant content, and even assisting with debugging code (Bringula, 2024). This
interactive process nurtures teachers to accept suggestions and feedback on their lesson plans
and teaching methods thus helping them to improve their programming practice and identify
potential areas for refinement (Sun et al., 2024; Yilmaz & Yilmaz, 2023).

Denny et al. (2024) conducted a study focusing on the potential of LLMs to assist teachers in
creating programming exercises and code explanations. Using Codex, exercises based on
input prompts with contextual themes and programming principles were generated for
coding. The results indicated that most created exercises featured executable example
answers and test cases. Although not all sample solutions passed the tests, the test suites
provided complete statement coverage when they did. In terms of code explanations, Codex
produced detailed explanations of the given code because of the input. An analysis showed
that around 70% of the explanations for individual lines were accurate, and 90% of the
explanations covered every section of the code. The results are similar to other studies, i.e.
Prather et al. (2023), that suggest that LLMs can reduce the workload of C&R teachers by
creating vast repositories of diverse learning resources and support materials.
156 Journal of Education, No. 96, 2024

In a further review of a study by Prather et al. (2023) on using LLMs in C&R classrooms, the
findings suggest that 60% of teachers use ChatGPT in these settings. One key application
involves generating code examples for learners as starting points for modification and
analysis. These examples have a dual purpose: they help teachers understand the applicability
of topics and common coding patterns and streamline the process of creating engaging
programming activities for learners to practice their coding skills. This technique allowed
teachers to analyse efficiently the relevance of topics to their C&R curriculum, identify
common coding patterns, and provide opportunities for learners to improve their coding
abilities. To optimise the benefits of GenAI for C&R teachers, it is essential to promote
collaboration through PLCs.

Teacher preparation
Research question 2

• What are GenAI’s potential implications for teacher preparation and support in C&R
education?

The findings revealed that teacher training programs for C&R often face challenges because
of geographically dispersed locations. The lack of training options in the areas where C&R
teachers teach contributed to their sense of isolation (Ni et al., 2023). The isolation of C&R
teachers highlights the necessity of providing training that encourages teacher network-
building and collaboration. The concept of a PLC emphasises a potential intervention during
which teachers could collaborate to improve each other’s learning. With the prevalence of
digital technologies, such as smartphones, C&R teachers could transcend their geographical
limitations by participating in opportunities to join digital PLCs with individuals with similar
educational interests (Department of Basic Education, 2018). Participants in these PLCs
examine lessons, data, outcomes, and strategies to improve their teaching as part of a
continuous improvement cycle. This method of inquiry not only improves learner
achievement but also improves teacher satisfaction (Ogegbo & Ramnarain, 2022). Effective
PLCs promote enhanced teacher self-efficacy and teamwork, minimising teacher isolation
(Ni et al., 2023). To address the issues faced by C&R teachers, building PLCs with GenAI
could represent an essential step in advancing C&R education in grades R–9.

The PLCs may operate at various levels and can vary significantly in the domain of teaching
C&R with each programming paradigm. Examples include digital PLCs such as social media,
blogs, and social website groups that can facilitate discussions dedicated to sharing practices
and resources and building peer support networks among C&R teachers from different
schools (Department of Basic Education, 2018). While the current digital platforms provide a
foundation for discussions, GenAI may enhance these digital PLCs. Through real-time
analysis of discussion data (Chiu, 2023), GenAI tools draw attention to areas in which
teachers require support and provide suggestions and feedback on their teaching (Bringula,
2024). Given the significance of offering teacher development opportunities tailored to
curriculum requirements (Hadad et al., 2021), GenAI, for instance, can analyse C&R lesson
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 157

plans and materials that are shared in the PLC and provide input on how well they meet
curriculum requirements for block-based and text-based programming paradigms and how
effective they are. These GenAI features can transform digital PLCs into dynamic, supportive
environments for professional development and collaboration among C&R teachers. While
this section highlighted the immense potential of introducing technologies like GenAI for
C&R, its potential challenges and tensions must be considered for its adaptation.

Navigating potential perils of GenAI integration

Although GenAI presents an avenue for C&R teacher preparation, concerns about its
potential impede several challenges. The primary challenge is equity. The implications drawn
from the reviewed studies are that the efficiency of GenAI tools depends on the availability
of technology and reliable internet connections (Becker et al., 2023; Prather et al., 2023).
This has the potential to exacerbate existing inequities between well-resourced and under-
resourced schools. Elsewhere, it has been found that teachers in under-resourced schools
could find it difficult to use GenAI tools, hindering their capacity to benefit from tailored
learning experiences and immediate feedback (Yan et al., 2024).

The uptake of GenAI tools is contingent upon the willingness of teachers to include them in
their C&R teaching and learning process. Although South Africa’s Department of Basic
Education endorses policies such as the Professional Development Framework Digital
Learning that necessitates teachers to integrate digital tools and resources in their teaching
and learning practices (Department of Basic Education, 2018), some teachers may view
GenAI as a challenge to their pedagogical expertise or as a replacement for their role in the
classroom. Findings suggest that teachers concerned about the effects of GenAI may change
their teaching methods in ways that reduce the efficacy of their lessons (Prather et al., 2023).
For instance, they could emphasise assigning students to replicate predefined code that
restricts their ability to think critically and learn, rather than on tasks like programming
exercises that call for these skills (Hadad et al., 2021). Alternatively, they might change the
subject material to exclude topics that call for coding knowledge, making learners less ready
for subsequent grades that require such coding expertise (Prather et al., 2023). Mitigating
these potential tensions necessitates transparent communication regarding the use of GenAI
tools to modify methods of instruction to improve the learning experience while adhering to
the C&R subject requirements (Sun et al., 2024).

GenAI is considered a technological tool so, to use it for C&R education, most teachers who
are unfamiliar with it require assistance (Chiu, 2023). Without assistance, teachers may
inadvertently create underspecified exercises, inaccurate reference answers, or insufficient or
erroneous test cases when they employ GenAI to help create C&R learning resources (Prather
et al., 2023). Thus, it is recommended from the reviewed studies that training in using GenAI
becomes a necessary component of professional development programs so that teachers can
use the tool’s potential to improve their teaching (Chiu, 2023; Ni et al., 2023). An instance
deduced from this is research conducted by Wieser et al. (2023), who assessed ChatGPT’s
programming tutoring capabilities and discovered that its general recommendations were
158 Journal of Education, No. 96, 2024

frequently generic without a clear learning objective. This is where the quality assurance of
GenAI recommendations comes into play.

Findings revealed that rigorous quality assurance processes must be implemented to ensure
the accuracy and effectiveness of material generated by GenAI. Although GPT-4 and other
advanced LLMs have much to offer, they still have some limitations similar to previous
iterations (Sun et al., 2024). Specifically, they can produce inaccurate information, potentially
biased content, and induce logical errors (Chiu, 2023). Thus, caution must be exercised, and
ethical considerations must be at the forefront while one is using LLM outputs, particularly in
high-stakes settings (United Nations Educational, Scientific and Cultural Organization 2023,
2023), including classrooms where C&R learning occurs. Depending on the particular use
case, precautions like peer review or additional information may be required to ensure the
accuracy and dependability of the data generated by the language model (Chiu, 2023; United
Nations Educational, Scientific and Cultural Organization, 2023). These exercises must be
created using a performance metric suitable for the type of tasks that teachers ask GenAI to
perform (United Nations Educational, Scientific and Cultural Organization 2023). The
portion of the generated code that is directly executable (i.e., if the generated code can be run
in a programming language and pass the unit tests) could be used as the code generation data
(Denny et al., 2024; United Nations Educational, Scientific and Cultural Organization, 2023).
Doing this may take numerous prompt recurrences until the intended result is obtained.

From these findings, it is important to note that some errors in the error detection system of
GenAI tools could affect teachers’ interest and capacity to use the feedback offered.
Submitting a learning objective leads to better recommendations for teacher practice,
including step-by-step responses and solution approaches (Wieser et al., 2023). The use of
GenAI in conjunction with skills endorsed for C&R should be prioritised in teacher
development programmes to reduce this risk (Chiu, 2023). Encouraging C&R teachers to
analyse GenAI recommendations, adjust them based on the unique needs of learners, and
reflect on their methods of instruction to ensure GenAI is used to complement teacher
creativity and critical thinking.

Another concern with incorporating GenAI into C&R teaching methods is an over-reliance
on automated assistance with tasks, including lesson planning, feedback generation, and
learning resource development. C&R education supports the development of problem-solving
skills, computational thinking, critical thinking, and creativity (Moraiti et al., 2022). Teachers
may become accustomed quickly to auto-suggested solutions, especially when using GenAI-
powered tools like LLMs, whose coding design can complement various programming
paradigms (Denny et al., 2024). An overreliance on LLMs could lead to a lack of critical
reading of problem statements and a reduced awareness of the computational steps necessary
for problem-solving (Becker et al., 2023). The process of teaching and learning to code has
become challenging and requires cognitive work to be able to retain language syntax (Denny
et al., 2024), think computationally (Moraiti et al., 2022), and understand C&R PCK
knowledge (Ni et al., 2023). Therefore, overreliance on GenAI tools might prevent these
capabilities from being developed.
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 159

Limitations of the study

We understand that this scoping review is based on a limited number of papers, that may not
represent the scope of available research on GenAI’s role in teacher preparation for C&R
instruction. The inclusion criteria were based on studies published in the selected timeframe
(2022–May 2024) and in English, which may have omitted significant research in other
languages given inadequate translation resources. The variation in the calibre and rigour of
the included studies further impacts the reliability of the findings. While GenAI’s ethical and
privacy concerns are acknowledged, they have not been thoroughly investigated and
require further research.

Conclusion
In this scoping review, we aimed to provide insight into how GenAI can address teachers’
challenges in acquiring the necessary pedagogical skills and programming knowledge for
effective C&R instruction and its potential implications for teacher preparation and support in
South African schools. The review identified GenAI tools, leveraging LLMs, provide
personalised learning routes to enhance teacher proficiency in block-based and text-based
programming paradigms through personalised instruction and real-time feedback. These tools
can assess teachers’ understanding of computational concepts, create learning resources and
offer step-by-step solutions. Furthermore, the review highlighted the potential of GenAI tools
to assist PLC members. Through digital PLCs, GenAI can facilitate discussions, provide
feedback on lesson plans, and suggest improvements, thus promoting continuous professional
development. However, managing this potential transformation requires addressing perils and
tensions.

In the review, we identified perils regarding equity and a lack of technology fluency among
teachers, as well as a demand for communication and professional development support. We
also noted ethical concerns such as the accuracy and dependability of GenAI-generated
content and the need for rigorous quality assurance processes. To further reduce the risk of an
over-reliance on automated solutions, it is essential to help teachers develop their capacity for
critical thinking and their pedagogical knowledge in tandem with using GenAI. By blending
AI-generated resources with teacher-created content, GenAI can serve as a tool to enhance,
not replace, teacher creativity.

Recognising that GenAI implementation and C&R teacher training are still in the earliest
phases is essential. Further research must be conducted to assess their long-term effectiveness
in improving teacher abilities and learner outcomes. Furthermore, research on developing
ethical practices for responsible GenAI integration in the specific context of C&R education,
including in special needs contexts (Bano et al., 2023), is needed. This would make it
possible to create strategies promoting policies, such as the United Nations Educational,
Scientific and Cultural Organization’s (2023) advocacy for ethical GenAI integration through
teacher training efforts and fair access to technology.
160 Journal of Education, No. 96, 2024

Declaration of interest statement

The authors declare no conflicts of interest.

References
Arendse, A. (2023). Discussion paper on the Fourth Industrial Revolution Youth Expo. Cape
Town: Parliament of the Republic of South Africa.
https://www.parliament.gov.za/storage/app/media/Pages/2023/19-06-
2023_4IR/Discussion_Paper_4IR.pdf

Arksey, H., & O’Malley, L. (2005). Scoping studies: Towards a methodological framework.
International Journal of Social Research Methodology, 8(1), 19–32.
https://doi.org/10.1080/1364557032000119616

Bano, S., Atif, K., & Mehdi, S. A. (2023). Systematic review: Potential effectiveness of
educational robotics for 21st century skills development in young learners. Education
and Information Technologies, 1–19. https://doi.org/10.1007/s10639-023-12233-2

Becker, B. A., Denny, P., Finnie-Ansley, J., Luxton-Reilly, A., Prather, J., & Santos, E. A.
(2023). Programming is hard-or at least it used to be: Educational opportunities and
challenges of AI code generation. Proceedings of the 54th ACM Technical Symposium
on Computer Science Education V. 1 (pp. 500–506). Association for Computing
Machinery. https://doi.org/10.1145/3545945.3569759

Boz, T., & Allexsaht-Snider, M. (2022). How do elementary school teachers learn coding and
robotics? A case study of mediations and conflicts. Education and Information
Technologies, 27(3), 3935–3963. https://doi.org/10.1007/s10639-021-10736-4

Bringula, R. (2024). ChatGPT in a programming course: Benefits and limitations. Frontiers

in Education, 9, 1–6. https://doi.org/10.3389/feduc.2024.1248705

Castro, E., Cecchi, F., Salvini, P., Valente, M., Buselli, E., Menichetti, L., Calvani, A., &
Dario, P. (2018). Design and impact of a teacher training course, and attitude change
concerning educational robotics. International Journal of Social Robotics, 10, 669–
685. https://doi.org/10.1007/s12369-018-0475-6

Changpetch, C., Panjaburee, P., & Srisawasdi, N. (2022). A comparison of pre-service

teachers’ variable misconceptions in various computer-programming preferences:
findings to teacher education course. Journal of Computers in Education, 9(2), 149–
172. https://doi.org/10.1007/s40692-021-00200-0

Chiu, T. K. (2023). The impact of Generative AI (GenAI) on practices, policies and research
direction in education: A case of ChatGPT and Midjourney. Interactive Learning
Environments, 1–17. https://doi.org/10.1080/10494820.2023.2253861
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 161

Denny, P., Prather, J., Becker, B. A., Finnie-Ansley, J., Hellas, A., Leinonen, J., Luxton-
Reily, A., Reeves, B., Antonio Santos, E., & Sarsa, S. (2024). Computing education in
the era of generative AI. Communications of the ACM, 67(2), 56–67.
https://doi.org/10.1145/3624720

Department of Basic Education. (2011). Curriculum and Assessment Policy Statement

Grades 10-12 Information Technology. https://wcedeportal.co.za/eresource/106226

Department of Basic Education. (2018). Professional Development Framework for Digital

Learning: Building educator competencies in facilitating learning with digital tools
and resources.
https://www.education.gov.za/Portals/0/Documents/Publications/Digital%20Learning
%20Framework.pdf?ver=2018-07-09-101748-953

Department of Basic Education. (2021). PROPOSED AMENDMENTS TO THE

CURRICULUM AND ASSESSMENT POLICY STATEMENT (CAPS) TO MAKE
PROVISION FOR CODING AND ROBOTICS GRADES R-9.
https://www.education.gov.za/Portals/0/Documents/Legislation/Call%20for%20Com
ments/draftcodingandroboticscurriculum/Grade%20R-
3%20Coding%20%20and%20Robotics%20Draft%20CAPS%20Final%2019Mar2021
.pdf?ver=2021-03-24-164612-000

Department of Education. (2002). Revised National Curriculum Statement Grades R-9

(Schools). https://wcedonline.westerncape.gov.za/ncs/rncs.pdf

Department of Education. (2004). White Paper on e-Education: Transforming Learning and

Teaching through Information and Communication Technologies (ICTs).
https://www.gov.za/sites/default/files/gcis_document/201409/267341.pdf

Fan, D. B.W. (2022). Advancing literature review methodology through rigour, generativity,
scope and transparency. International Journal of Management Reviews, 24(2), 171–
180. https://doi.org/10.1111/ijmr.12291

Fiske, E. B., & Ladd, H. F. (2004). Elusive equity: Education reform in post-apartheid South
Africa. Brookings Institution Press.

Fleisch, B. (2002). State formation and the origins of Bantu Education. In P. Kallaway (Ed.),
The history of education under apartheid 1948–1994: The doors of learning and
culture shall be opened (pp. 39–55). Maskew Miller Longman.

Galpin, V. C., & Sanders, I. D. (2007). Perceptions of computer science at a South African
university. Computers & Education, 49(4), 1330–1356.
https://doi.org/10.1016/j.compedu.2006.02.008
162 Journal of Education, No. 96, 2024

Geldenhuys, C. J., & Fataar, A. (2021). Foundation Phase teachers’ experiences of teaching
the subject, coding, in selected Western Cape schools. South African Journal of
Education, 41(4), 1–9. https://doi.org/10.15700/saje.v41n4a1959

Gorrah, A. & Papers, F. (2024). Implementation of Coding and Robotics in South African
public schools, fostering teachers’ self-directed learning: A scoping review. In C. G.
Bosch (Ed.), Navigating Computer Science Education in the 21st Century (pp. 241–
260). IGI Global.

Govender, R. G., & Govender, D. W. (2021). A physical computing approach to the

introduction of computer programming among a group of pre-service teachers.
African Journal of Research in Mathematics, Science and Technology Education,
25(1), 91–102. https://doi.org/10.1080/18117295.2021.1924440

Greyling, J. (2023). Coding unplugged—A guide to introducing Coding and Robotics to

South African Schools. In J. A. Halberstadt (Ed.), Transforming Entrepreneurship
Education (pp. 155–174). Springer. https://doi.org/10.1007/978-3-031-11578-3_9

Hadad, S., Shamir-Inbal, T., Blau, I., & Leykin, E. (2021). Professional development of code
and robotics teachers through small private online course (SPOC): Teacher centrality
and pedagogical strategies for developing computational thinking of students. Journal
of Educational Computing Research, 59(4), 763–791.
https://doi.org/10.1177/0735633120973432

Koorsse, M., Cilliers, C., & Calitz, A. (2015). Programming assistance tools to support the
learning of IT programming in South African secondary schools. Computers &
Education, 82, 162–178. https://doi.org/10.1016/j.compedu.2014.11.020

Litlhakanyane, T. D. (2022). Pair programming and underrepresented groups in Information

Technology in South Africa. [Doctoral dissertation, North-West University, South
Africa.]

Mentz, E., Bailey, R., Havenga, M., Breed, B., Govender, D., Govender, I., & Dignum, V.
(2012). The diverse educational needs and challenges of Information Technology
teachers in two black rural schools. Perspectives in Education, 30(1), 70–78.
https://hdl.handle.net/10520/EJC87660

Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & PRISMA Group*, T. (2009). Preferred
reporting items for systematic reviews and meta-analyses: the PRISMA statement.
Annals of Internal Medicine, 151(4), 264–269. https://doi.org/10.1136/bmj.b2535

Moraiti, I., Fotoglou, A., & Drigas, A. (2022). Coding with block programming languages in
educational robotics and mobiles, improve problem solving, creativity & critical
thinking skills. International Journal of Interactive Mobile Technologies, 16(20), 59–
78. https://doi.org/10.3991/ijim.v16i20.34247
 Tshidi & Dewa: The promise and peril of coding & robotics education in South Africa 163

Munje, P. N., & Jita, T. (2020). The impact of the lack of ICT resources on teaching and
learning in selected South African primary schools. International Journal of
Learning, Teaching and Educational Research, 19(7), 263–279.
https://doi.org/10.26803/ijlter.19.7.15

Ni, L., Bausch, G., & Benjamin, R. (2023). Computer science teacher professional
development and professional learning communities: A review of the research
literature. Computer Science Education, 33(1), 29–60.
https://doi.org/10.1080/08993408.2021.1993666

Ogegbo, A. A., & Ramnarain, U. (2022). Teachers’ perceptions of and concerns about
integrating computational thinking into science teaching after a professional
development activity. African Journal of Research in Mathematics, Science and
Technology Education, 26(3), 181–191.
https://doi.org/10.1080/18117295.2022.2133739

Prather, J., Denny, P., Leinonen, J., Becker, B. A., Albluwi, I., Craig, M., & Savelka, J.
(2023). The robots are here: Navigating the generative AI revolution in computing
education. In Proceedings of the 2023 Working Group Reports on Innovation and
Technology in Computer Science Education (pp. 108–159). Association for
Computing Machinery. https://doi.org/10.1145/3623762.3633499

Reimers, F. M. (2021). In search of a twenty-first century education renaissance after a

global pandemic. Cambridge, MA: Springer. https://doi.org/10.1007/978-3-030-
57039-2

Schina, D., Esteve-González, V., & Usart, M. (2021). An overview of teacher training
programs in educational robotics: Characteristics, best practices and
recommendations. Education and Information Technologies, 26(3), 2831–2852.
https://doi.org/10.1007/s10639-020-10377-z

Storte, D., Webb, M., Bottino, R., Passey, D., Kalas, I., Bescherer, C., & Fuschek, G. (2019).
Coding, programming and the changing curriculum for computing in schools. Report
of UNESCO/IFIP TC 3 Meeting at OCCE. https://hdl.handle.net/102.100.100/495503

Sun, D., Boudouaia, A., Zhu, C., & Li, Y. (2024). Would ChatGPT-facilitated programming
mode impact college students’ programming behaviors, performances, and
perceptions? An empirical study. International Journal of Educational Technology in
Higher Education, 21(1), 14, 1–22. https://doi.org/10.1186/s41239-024-00446-5

United Nations Educational, Scientific and Cultural Organization. (2023). Guidance for
generative AI in education and research.
https://www.unesco.org/en/articles/guidance-generative-ai-education-and-research
164 Journal of Education, No. 96, 2024

Wieser, M., Schöffmann, K., Stefanics, D., Bollin, A., & Pasterk, S. (2023). Investigating the
role of ChatGPT in supporting text-based programming education for students and
teachers. International Conference on Informatics in Schools: Situation, Evolution,
and Perspectives (pp. 40–53). Springer Nature Switzerland.
https://doi.org/10.1007/978-3-031-44900-0

Yan, L., Martinez-Maldonado, R., & Gasevic, D. (2024). Generative artificial intelligence in
learning analytics: Contextualising opportunities and challenges through the learning
analytics cycle. Proceedings of the 14th Learning Analytics and Knowledge
Conference (pp. 101–111). Association for Computing Machinery.
https://doi.org/10.1145/3636555.3636856

Yilmaz, R., & Yilmaz, F. G. K. (2023). Augmented intelligence in programming learning:

Examining student views on the use of ChatGPT for programming learning.
Computers in Human Behavior: Artificial Humans, 1(2), 1–7.
https://doi.org/10.1016/j.chbah.2023.100005