Learning Technologies and Learning Resources in Computing Science Higher Education: Exploring Access, Adoption, and Effectiveness in relation to Regional Context

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Difficulty in tracking OER usage as it often happens in private spaces such as a classroom, hence it goes unreported.

“One problem highlighted in the literature is the difficulty in tracking OER usage as this often happens in private spaces, such as a classroom [6], hence it goes unreported.”

• cites [6] [6] Tita Beaven. 2018. 'Dark reuse': an empirical study of teachers' OER engagement. Open Praxis (Dec 2018). https://doi.org/10.5944/openpraxis.10.4.889 - sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing).

CS students and educators may be expected to be among early technology adopters.

“Computing science (CS) students and educators may be expected to be among early technology adopters [21].”

• cites [21] [21] Sharlett Gillard, Denice Bailey, and Ernest Nolan. 2008. Ten Reasons for IT Educators to be Early Adopters of IT Innovations. Journal of Information Technology Education: Research 7, 1 (January 2008), 21-33. https://www.learntechlib.org/p/ 111369 - sourcePartly supported

  The Introduction explicitly expects a higher percentage of IT educators to be early adopters. Section 8 says that most of the authors’ better students try new technological gadgets, but it does not establish the same expectation for CS students generally. Page numbers are not available in the supplied passages. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing).

Fischer et al. observed a clear impact on affordability: having access to free textbooks encourages community college students to take one more course, accelerating their progress towards graduation.

“They observed a clear impact on affordability: having access to free textbooks encourage community college students to take one more course, accelerating their progress towards graduation.”

• cites [18] [18] Lane Fischer, John Hilton, T. Robinson, and David Wiley. 2015. A multiinstitutional study of the impact of open textbook adoption on the learning outcomes of post-secondary students. Journal of Computing in Higher Education 27 (12 2015), 159-172. https://doi.org/10.1007/s12528-015-9101-x - sourcePartly supported

  The Discussion reports that students using OER enrolled in significantly more credits the following semester and suggests textbook savings as a possible reason, particularly in community-college settings. The Enrollment intensity in winter semester section reports a higher adjusted mean credit load, but the paper does not establish that free textbooks caused students to take one more course or graduate sooner. (LLM, conf=0.88)

  Experimental — accuracy not yet measured

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Fischer et al. found mixed impact of OER textbooks on course completion and course grades, which they attribute to access.

“They found mixed impact on course completion and course grades, which they attribute to access: some students in the traditional course may have not purchased the textbook.”

• cites [18] [18] Lane Fischer, John Hilton, T. Robinson, and David Wiley. 2015. A multiinstitutional study of the impact of open textbook adoption on the learning outcomes of post-secondary students. Journal of Computing in Higher Education 27 (12 2015), 159-172. https://doi.org/10.1007/s12528-015-9101-x - sourceSupported

May combine two claims.

Our scoping literature review highlights the CS Education community's primary focus on Open Educational Resources [19, 20], followed by Learning Preferences [16] and Technology Adoption [35].

“Our scoping literature review highlights the CS Education community's primary focus on Open Educational Resources [19, 20], followed by Learning Preferences [16] and Technology Adoption [35].”

• cites [19] [19] Max Fowler, David H. Smith IV, Binglin Chen, and Craig Zilles. 2023. "I Don't Gamble To Make My Livelihood": Understanding the Incentives For, Needs Of, and Motivations Surrounding Open Educational Resources in Computing. In Proceedings of the 2023 ACM Conference on International Computing Education Research V.1 (New York, NY, USA). ACM, 430-443. https://doi.org/10.1145/ 3568813.3600136 - sourceSupported
• cites [20] [20] Alicia García-Holgado, Francisco José García-Peñalvo, Colin de la Higuera, Antonio Teixeira, Ulf-Daniel Ehlers, James Bruton, Fabio Nascimbeni, Natalia Padilla Zea, and Daniel Burgos. 2020. Promoting Open Education Through Gamification in Higher Education: the OpenGame project. In Eighth International Conference on Technological Ecosystems for Enhancing Multiculturality (New York, NY, USA). ACM, 399-404. https://doi.org/10.1145/3434780.3436688 - sourceSupported
• cites [16] [16] Javier Escobar-Avila, Deborah Venuti, Massimiliano Di Penta, and Sonia Haiduc. 2019. A Survey on Online Learning Preferences for Computer Science and Programming. In 2019 IEEE/ACM 41st International Conference on Software Engineering: Software Engineering Education and Training (ICSE-SEET) . IEEE, 170-181. https://doi.org/10.1109/ICSE-SEET.2019.00026 - sourceNot checked — no full text of the source
• cites [35] [35] Alga Rosetta, Meilia Agatha Priska, Erlinda Muslim, and Muhammad Rafi. 2020. Evaluating the Success of E-Learning Systems and Strategy Creation: The Perspective of Students in Universitas Indonesia. In ACM International Conference Proceeding Series . ACM, 11-17. https://doi.org/10.1145/3439147.3439155 - sourceNot checked — no full text of the source

Classified as probably not a substantial claim.

Literature confirms the need for adoption of open education resources as replacements for items whose cost are normally paid by the students, such as textbooks.

“This is confirmed by literature, highlighting the need for adoption of open education resources as replacements for items whose cost are normally paid by the students, such as textbooks [3].”

• cites [3] [3] David Annand and Tilly Jensen. 2017. Incentivizing the Production and Use of Open Educational Resources in Higher Education Institutions. The International Review of Research in Open and Distributed Learning 18, 4 (Jun. 2017). https: //doi.org/10.19173/irrodl.v18i4.3009 - sourceSupported

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

195 participants (65%) identified as men, 91 (30%) as women, 5 (1.6%) as non-binary or preferred to self-identify, and 10 (3.3%) preferred not to answer.

“195 participants (65%) identified as men, 91 (30%) identified as women, 5 participants identified as Non-binary or preferred to self-identify (1.6%). 10 participants (3.3%) indicated they prefer not to answer this question.”

Probably not a substantial claim: may not pass S4 (based).

Limited studies consider shifting needs and preferences with emerging resources and technologies.

“Limited studies consider shifting needs and preferences with emerging resources and technologies.”

May combine two claims.

252 participants (84%) reported no accessibility needs, 21 (6%) preferred not to disclose, 16 (5.3%) reported accessibility needs but not registered, and 12 (4%) reported disability needs and registered.

“252 participants (84%) reported no accessibility needs. 21 participants (6%) preferred not to disclose them. 16 participants (5.3%) reported accessibility needs but not being registered with a disability center. 12 participants (4%) reported disability needs and being registered with a disability center.”

May combine two claims.

Online forums are used at least occasionally by 49% and frequently by 14%.

“Practice and collaboration tools show moderate uptake, coding platforms 62% (21% frequently), collaborative tools 61% (24% frequently), educational apps/software 56% (20% frequently), and online forums 49% (14% frequently).”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The initial search extracted 46 papers covering three main topics: OER, learning preferences, and technology adoption.

“46 papers were extracted from this search, covering three main topics: OER, learning preferences and technology adoption.”

May combine two claims.

Educational apps/software are used at least occasionally by 56% and frequently by 20%.

“Practice and collaboration tools show moderate uptake, coding platforms 62% (21% frequently), collaborative tools 61% (24% frequently), educational apps/software 56% (20% frequently), and online forums 49% (14% frequently).”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

In total, 386 students participated in the survey, but only 301 completed the demographic information.

“In total, 386 students participated in the survey, but only 301 completed the demographic information as reported below.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

The scoping literature review comprised 51 papers, of which 29 are cited in the report.

“Therefore, our literature review comprised 51 papers, of which 29 are cited in the report.”

Probably not a substantial claim: may not pass S3 (load-bearing).

265 participants (88%) indicated undergraduate as their current level of study.

“265 participants (88%) indicated undergraduate as their current level of study, 15 participants (5%) indicated Master degree and 7 (2.3%) indicated PhD. 14 participants (4.6%) indicated certificate or other.”

May combine two claims.

82% of surveyed students and 45% of surveyed educators are from African countries.

“82% of surveyed students and 45% of surveyed educators are from African countries”

• Also stated in: § 1 Introduction · p. 2 · plain

  “Our survey participants, predominantly from African countries which represented 82% of our student respondents and 45% of our educator respondents”

• Also stated in: § 4.1 Student Surveys: Responses · p. 5 · hedged

  “250 participants (83%) indicated being based in Africa”

Probably not a substantial claim: may not pass S3 (load-bearing).

Time is cited by 46%.

“The most frequently cited barrier is cost, 54%, followed by time, 46% and lack of access, 39%.”

May combine two claims.

Personal computers were the most accessible resource and infrastructure (power and internet) was the least accessible.

“Figure 1.(a) summarizes the ease of access, with personal computers being the most accessible resource and infrastructure (power and internet) being the least accessible.”

May combine two claims.

YouTube/video tutorials are used at least occasionally by ≈93% and frequently by 56%.

“Generative AI tools and YouTube/video tutorials dominate use of digital tools for learning, with 96% and ≈ 93% using them at least occasionally, and 75% and 56% using them frequently (4+ times).”

May combine two claims.

The survey subgroups were Africa (65%), outside of Africa (13%), and unspecified location (22%).

“we split student answers into three subgroups: Africa (65%), outside of Africa (13%) and unspecified location (22%).”

• AssumptionRests on an assumption: Student-reported location is an accurate indicator of the regional context relevant to infrastructure access.

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

232 respondents (77%) identified as black, 20 (6%) as white, 27 (9%) as other racial identities, 14 (4.7%) preferred not to answer, and 8 (2.7%) preferred to self-identify.

“232 respondents (77%) identified as black, 20 respondents (6%) as white. 27 participants identified as either Arab, Chinese, Filipino, Indigenous/Aboriginal, Japanese, Korean, Latin American, Mixed, Bi-Racial, Multi-Racial, South Asian, and Southeast Asian. 14 participants (4.7%) preferred not to answer. 8 participants (2.7%) preferred to self-identify.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Learning curve is cited by 31%.

“Other notable hurdles include the learning curve, 31%, system requirements, 28%, and data privacy, 23%.”

May combine two claims.

Survey respondents reported learning curve, finding quality resources, data privacy, and ethical concerns as challenges.

“Other challenges include learning curve, finding quality resources, data privacy, and ethical concerns.”

May combine two claims.

Social media explainer content is used at least occasionally by 63% and frequently by 31%.

“Social media explainer content is also common (63% at least occasionally; 31% frequently).”

Probably not a substantial claim: may not pass S3 (load-bearing).

The literature discovered using our scoping review search terms included very little mention of AI.

“The literature discovered using our scoping review search terms included very little mention of AI.”

May combine two claims.

Collaborative tools are used at least occasionally by 61% and frequently by 24%.

“Practice and collaboration tools show moderate uptake, coding platforms 62% (21% frequently), collaborative tools 61% (24% frequently), educational apps/software 56% (20% frequently), and online forums 49% (14% frequently).”

May combine two claims.

Generative AI tools are used at least occasionally by 96% and frequently by 75%.

“Generative AI tools and YouTube/video tutorials dominate use of digital tools for learning, with 96% and ≈ 93% using them at least occasionally, and 75% and 56% using them frequently (4+ times).”

May combine two claims.

Access to a Personal Computer was rated as the easiest resource to obtain, achieving 79% device penetration among the student body.

“Access to essential hardware, specifically a Personal Computer, was rated as the easiest resource to obtain, achieving 79% device penetration among the student body.”

May combine two claims.

VR/AR remains niche, with 12% reporting regular access.

“VR/AR remains niche, with 12% reporting regular access.”

May combine two claims.

Coding platforms are used at least occasionally by 62% and frequently by 21%.

“Practice and collaboration tools show moderate uptake, coding platforms 62% (21% frequently), collaborative tools 61% (24% frequently), educational apps/software 56% (20% frequently), and online forums 49% (14% frequently).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Instructor suggestion was 5%.

“Fun/engagement contributes 10%, while peer recommendation, 6% and instructor suggestion, 5% are less prominent, alongside trying tools when struggling with content, 5%.”

May combine two claims.

AR/VR tools see limited use: 20% at least occasionally, 56% never.

“More specialised resources see limited use: simulations/virtual labs 24% at least occasionally and AR/VR tools 20%, with 46% and 56% never using them, respectively.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Emerging hardware technologies were mentioned in 5.4% of answers.

“Among the answers mentioning emerging technology themes, answers mentioning AI-centered technologies showed the highest frequency (21%), followed by Immersive technologies (7.8%), and emerging hardware technologies (5.4%).”

May combine two claims.

Simulations/virtual labs see limited use: 24% at least occasionally, 46% never.

“More specialised resources see limited use: simulations/virtual labs 24% at least occasionally and AR/VR tools 20%, with 46% and 56% never using them, respectively.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Design/interactivity/features was 22%.

“Respondents indicate quality (32%), whether it is in design, interactivity, and features (22%) or in content quality (12%) as the most frequent reason.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Data privacy is cited by 23%.

“Other notable hurdles include the learning curve, 31%, system requirements, 28%, and data privacy, 23%.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Learning curve was 12%.

“Learning curve (12%), cost (6%) and setup time (2%).”

Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

African students exhibit higher usage compared to their counterparts.

“African students exhibit higher usage compared to their counterparts. They also rely more on social media content.”

• AssumptionRests on an assumption: Student-reported location is an accurate indicator of the regional context relevant to infrastructure access.

The quote only partly states this claim.

Negative answers prevalence was 48.7% of answers to SSQ4.

“An interesting finding is the prevalence of negative answers at 48.7% of the answers provided to this open-ended question.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Peer recommendation was 6%.

“Fun/engagement contributes 10%, while peer recommendation, 6% and instructor suggestion, 5% are less prominent, alongside trying tools when struggling with content, 5%.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Other was 2% and none of the above was 2%.

“Only small shares selected other (2%) or none of the above (2%).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Content quality was 12%.

“Respondents indicate quality (32%), whether it is in design, interactivity, and features (22%) or in content quality (12%) as the most frequent reason.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Timing was 3%.

“Hardware and infrastructure requirements (7%), ethical concerns (5%) including those about spread of misinformation, and timing (3%) reporting on limited time to allocate to using the technologies were the following themes.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Cost concern was 10%.

“The themes that appeared with the highest frequency in the responses were high learning curve in using new technologies (14%) followed by cost (10%) of these technologies, and privacy concerns (9%).”

May combine two claims.

GenAI and video tutorials are the most effective tools.

“Overall, all scores for SSQ15 and SSQ21 are positive, indicating most tools are perceived as helpful and somewhat effective, with genAI and video tutorials being the most effective, while AR/VR was the least effective.”

May combine two claims.

The most commonly assigned resources are YouTube/video tutorials and generative AI tools, with 63% of students reporting at least occasional use for each.

“The most commonly assigned resources are YouTube/video tutorials and generative AI tools, with 63% of students reporting at least occasional use for each, frequently: 21-31%.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Hardware and infrastructure requirements were 7%.

“Hardware and infrastructure requirements (7%), ethical concerns (5%) including those about spread of misinformation, and timing (3%) reporting on limited time to allocate to using the technologies were the following themes.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Requirements were mentioned by 10.6%.

“The following themes were cost (18.1%) and requirements (10.6%) indicating lack of enough resources to enable learners to explore and use all online learning tools.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Learning curve was 9%.

“Cost (19%), safety including the security of software and the privacy of handling data (11%), and learning curve (9%) were the most frequent themes in the responses.”

The quote only partly states this claim.

Online forums/communities feature regularly with 37% at least occasional use.

“Collaborative tools, 57% and coding practice platforms, 54% also feature regularly, followed by educational apps/software, 49% and online forums/communities, 37%.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Other reasons were selected by 7%.

“smaller share selected other reasons, 7%, indicating additional context-specific constraints.”

May combine two claims.

22.3% of students showed high alignment with ρ ≥ 0.7; 4.35% with ρ ≥ 0.9.

“Among students with sufficient non-missing data (184 of 386), the median withinstudent Spearman correlation was moderate ( ˜ 𝜌 = 0 . 429, IQR = 0 . 580), although a subset of students showed high alignment (22.3% with 𝜌 ≥ 0 . 7; 4.35% with 𝜌 ≥ 0 . 9).”

• AssumptionThe authors state this as an assumption: Instructor integration frequency can be treated as a proxy for in-class/coursework exposure, and perceived effectiveness as a proxy for perceived learning value.

Probably not a substantial claim: may not pass S3 (load-bearing).

Quality was 7%.

“The respondents also mentioned quality (7%) of the technology, and ethics (2%) such as applying their moral compass to the impact of the technology, and the human rights abuses and environmental damages that the new technology may cause.”

Probably not a substantial claim: may not pass S3 (load-bearing).

System requirements are cited by 28%.

“Other notable hurdles include the learning curve, 31%, system requirements, 28%, and data privacy, 23%.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Quality of resources was mentioned by 7.3%.

“Quality of resources (7.3%) was the next frequent category, discussing the challenging experiences respondents had with emerging technologies.”

May combine two claims.

AR/VR drew the lowest helpfulness, 33%, and the highest not applicable/not sure responses, 23% combined.

“AR/VR drew the lowest helpfulness, 33% and the highest not applicable/not sure responses, 23% combined.”

The quote only partly states this claim.

Immersive technologies were mentioned in 7.8% of answers.

“Among the answers mentioning emerging technology themes, answers mentioning AI-centered technologies showed the highest frequency (21%), followed by Immersive technologies (7.8%), and emerging hardware technologies (5.4%).”

May combine two claims.

Documentation, coding platforms, and interactive textbooks fall within the high integration-high effectiveness quadrant.

“Documentation (integration mean = 1 . 62, effectiveness mean = 3 . 63), coding platforms (integration mean = 1 . 42, effectiveness mean = 3 . 75), and interactive textbooks (integration mean = 1 . 33, effectiveness mean = 3 . 56) also fall within this quadrant.”

• AssumptionThe authors state this as an assumption: Instructor integration frequency can be treated as a proxy for in-class/coursework exposure, and perceived effectiveness as a proxy for perceived learning value.

Probably not a substantial claim: may not pass S3 (load-bearing).

Ethical concerns were 5%.

“Hardware and infrastructure requirements (7%), ethical concerns (5%) including those about spread of misinformation, and timing (3%) reporting on limited time to allocate to using the technologies were the following themes.”

May combine two claims.

Use cases mentioned included aiding learning in general (7.3%), clarification (5.2%), and interaction (2.3%).

“The use cases mentioned included aiding their learning in general (7.3%), helping them in getting clarification through methods such as simplifying concepts (5.2%), and interactions with the technology, possibly making their learning more engaging (2.3%).”

May combine two claims.

Generative AI tools and YouTube/video tutorials are widely used both inside and outside the classroom.

“Asreported in section 4.2, Generative AI tools and YouTube/video tutorials are widely used both inside and outside the classroom.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Ethics was 2%.

“The respondents also mentioned quality (7%) of the technology, and ethics (2%) such as applying their moral compass to the impact of the technology, and the human rights abuses and environmental damages that the new technology may cause.”

May combine two claims.

Students rated AI chatbots as most helpful, ≈80% helpful or extremely helpful.

“Regarding perceived helpfulness of instructor-integrated tools (SSQ15), students rated AI chatbots as most helpful, ≈ 80% helpful or extremely helpful, alongside YouTube/video tutorials, 75%.”

May combine two claims. The quote only partly states this claim.

These issues emerged strongly as themes (internet and requirements in SSQ5) and quantitative analysis discovered pronounced regional differences.

“These issues emerged strongly as themes (”

• AssumptionRests on an assumption: Student-reported location is an accurate indicator of the regional context relevant to infrastructure access.

Probably not a substantial claim: may not pass S3 (load-bearing).

3.1% of answers not fitting into any theme were categorized as other.

“Answers are categorized into six major themes, and 3.1% of answers not fitting into any of the themes were categorized as other.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Cost was mentioned by 18.1%.

“The following themes were cost (18.1%) and requirements (10.6%) indicating lack of enough resources to enable learners to explore and use all online learning tools.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Cost was 6%.

“Learning curve (12%), cost (6%) and setup time (2%).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Setup time was 2%.

“Learning curve (12%), cost (6%) and setup time (2%).”

Probably not a substantial claim: may not pass S3 (load-bearing).

23% of respondents indicated they did not have concerns.

“SSQ12 explores concerns students have in using new digital learning tools. 23% of respondents indicated they did not have concerns.”

May combine two claims.

Constraints are often described as routine rather than exceptional, with little expectation that institutions will intervene to resolve access-related barriers.

“Notably, such constraints are often described as routine rather than exceptional, with little expectation that institutions will intervene to resolve access-related barriers.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Efficiency was 24%.

“Regarding motivations for trying new tools, curiosity is the leading driver, 42%, followed by efficiency, 24%.”

May combine two claims.

A middle tier includes coding practice platforms (59%), collaborative tools (49%), and educational apps/software (46%).

“A middle tier includes coding practice platforms, 59%, collaborative tools, 49%, and educational apps/software, 46%.”

May combine two claims.

A Friedman test over the nine tools listed in SEQ7 (complete cases n = 92) showed strong evidence that use frequency differed by tool, χ2(8)=216.96, p<0.001, with a moderate effect size (Kendall's W=0.295).

“A Friedman test over the nine tools listed in SEQ7 (complete cases 𝑛 = 92) showed strong evidence that use frequency differed by tool, 𝜒 2 ( 8 ) = 216 . 96, 𝑝 &lt; 0 . 001, with a moderate effect size (Kendall's 𝑊 = 0 . 295).”

May combine two claims.

Educators reported the most frequent engagement with Generative AI tools and YouTube/video tutorials, alongside collaborative tools (median = 3 for each).

“Educators reported the most frequent engagement with Generative AI tools and YouTube/video tutorials, alongside collaborative tools (median = 3 for each; corresponding to sometimes -often ).”

May combine two claims.

Online forum communities are used frequently by 10% of African students compared to 50% of OofA students.

“online forum communities are used frequently by 10% of African students compared to 50% of OofA students.”

May combine two claims.

The geographic spread is more balanced for educators, with only 31% being based in Africa, while one third did not specify the location.

“We should note the geographic spread is more balanced for educators, with only 31% being based in Africa, while one third of them did not specify the location.”

May combine two claims.

Spearman correlations between each tool's use frequency and its perceived preparation effectiveness were generally positive, and remained significant for seven of nine tools after Holm adjustment, with effect sizes ranging from small to large (ρ ≈ 0.29 to 0.60).

“Associations were generally positive, and remained significant for seven of nine tools after Holm adjustment, with effect sizes ranging from small to large ( 𝜌 ≈ 0 . 29 to 0 . 60).”

May combine two claims.

Lack of access increased from 34% when used by themselves to 40% when used in the classroom; system requirements increased from 25% to 33% and data privacy from 16% to 30%.

“Lack of access increased from 34% when used by themselves to 40% when used in the classroom. Similarly, system requirements increase from 25% to 33% and data privacy from 16% to 30%.”

May combine two claims.

In both SEQ12 and SEQ13, the most frequent responses were cost and time, both at 62%, followed by learning curve (43%).

“In both questions, the most frequent responses 2 were cost and time, both at 62% , followed by learning curve (43%).”

May combine two claims.

In SEQ5, educators also worried about equitable access (7.6%) and other impacts of new technologies on learning and assessment.

“Educators also worried about equitable access(7.6%) and other impacts of new technologies on learning and assessment.”

May combine two claims.

7.5% of educators report they can often access a personal device, while 5% have limited access to any kind of device.

“7.5% of educators report they can often access a personal device, while 5% have limited access to any kind of device.”

May combine two claims.

More formal or institutional supports were less frequent: university workshops 17%, mentorship or learning communities 13%, library or academic success staff 8%, IT helpdesk 6%, orientation/onboarding 5%, and pop-up support desks 3%.

“More formal or institutional supports were less frequent: university workshops, 17%, mentorship or learning communities, 13%, library or academic success staff, 8%, IT helpdesk, 6%, orientation/onboarding, 5%, and pop-up support desks, 3%.”

May combine two claims.

Integration and perceived effectiveness vary with location.

“Looking at the quantitative analysis of section 4.2, we can see both integration and perceived effectiveness varies with location.”

May combine two claims.

A minority of students reported receiving no support (7%) or other forms (2%).

“A minority reported receiving no support, 7% or other forms, 2%.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Only 84 educator participants filled the demographic information reported.

“Only 84 participants filled the demographic information reported below.”

May combine two claims.

Access to reliable home internet (often or all times) is 75%, with 69% having access to high-speed internet connection at home.

“Access to reliable home internet (often or all times) is 75%, with 69% having access to high-speed internet connection at home.”

May combine two claims.

25% of students are neutral and 6% disagree or strongly disagree that they prefer learning new material outside class using technology.

“About two-thirds of respondents, 67% agree or strongly agree that they prefer learning new material outside class using technology, while 25% are neutral and 6% disagree or strongly disagree, strongly agree 26%, agree 40%, neutral 25%, disagree 5%, and strongly disagree 2%.”

May combine two claims.

Everyday digital tools used often by educators include generative AI tools 70%, online video platforms 58%, LMS 56%, and educational software/apps 42%.

“Everyday digital tools used often include generative AI tools 70%, online video platforms 58%, LMS 56% and educational software/apps, 42%, while usage of VR/AR tools is .”

May combine two claims.

Educators identified insufficient institutional support, described as licensing restrictions (9%), limited or inadequate computing infrastructures (11%), and a lack of technical or pedagogical training opportunities (5%), as examples of institutional-level barriers to adoption.

“For example, educators identified insufficient institutional support, described as licensing restrictions, 9%, limited or inadequate computing infrastructures, 11%, and a lack of technical or pedagogical training opportunities, 5% as examples of institutional-level barriers to adoption.”

May combine two claims.

SSQ2 reports ease of access to five key resources: personal computers are the most accessible resource, necessary software is the least accessible, and infrastructure (power and internet) sits in the middle.

“SSQ2 reports ease of access to the five key resources as shown in Figure 4.(a) Personal computers being the most accessible resource and necessary software being the least accessible, while infrastructure (power and internet) sits in the middle.”

May combine two claims.

Social media/influencer content and AR/VR tools ranked lowest for preparation support, consistent with their low average use in the sample.

“Social media/influencer content and AR/VR tools ranked lowest for preparation support, consistent with their low average use in the sample.”

May combine two claims.

The two main challenges for tool adoption, regardless of location, are cost (54%) and required time to use the new tool (46%).

“The two main challenges for tool adoption, regardless of location, are cost (54%) and required time to use the new tool (46%).”

• AssumptionRests on an assumption: Student-reported location is an accurate indicator of the regional context relevant to infrastructure access.

May combine two claims.

70% of educators report full-time access to a personal laptop or tablet, while 38% have a school-provided laptop, often having both.

“70% of educators report full-time access to a personal laptop or tablet, while 38% have a school-provided laptop, often having both.”

May combine two claims.

Other themes that emerged strongly among students included quality of software, safety (including data privacy), and a wide range of ethical concerns including human rights and the ethics of generative AI.

“Other themes that emerged strongly included quality of software, safety (including data privacy) and a wide range of ethical concerns including human rights and the ethics of generative AI.”

May combine two claims.

In SEQ5, the following themes were fit (10.1%), overheads (10%), and requirements (9%), indicating lack of time/resources to enable educators to explore and use all online learning tools.

“The following themes were fit (10.1%), overheads (10%) and requirements (9%) indicating lack of time/ resources to enable educators to explore and use all online learning tools.”

Probably not a substantial claim: may not pass S3 (load-bearing).

The educator survey received 119 responses.

“Our educator survey received 119 responses.”

May combine two claims.

AI chatbots and YouTube/video tutorials lead, with 80% and 77% rating them effective or extremely effective.

“AI chatbots and YouTube/video tutorials lead, with 80% and 77% rating them effective or extremely effective.”

May combine two claims.

Forums/communities (39%) and social media explainer content (38%) are viewed as moderately effective.

“Forums/communities, 39% and social media explainer content, 38% are viewed as moderately effective, while simulations/virtual labs, 34% and AR/VR, 32% draw the lowest effectiveness ratings and the highest neutral responses 28% each.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

Coding practice platforms have similar usage overall, but much heavier presence inside the classroom for African students.

“Coding practice platforms have similar usage overall, but much heavier presence inside the classroom for African students, which is also the case for genAI tools.”

May combine two claims.

Among students, privacy is considered a challenge by students outside of Africa, while lack of access is a key issue for the other two subgroups.

“As shown in Figure 2, privacy is considered a challenge by students outside of Africa, while lack of access is a key use for the other two subgroups.”

May combine two claims.

In regards to educators' intentions of trying new digital tools, AI chatbots are the most sought after, with 76% indicating likely/very likely, compared to VR/AR tools with 33% likely/very likely and Tutor robot with 31% likely/very likely.

“In regards to educator's intentions of trying new digital tools, AI chatbots are the most sought after, with 76% indicating likely/very likely, compared to VR/AR tools with 33% likely/very likely and lastly Tutor robot with 31% likely/very likely.”

May combine two claims.

Students most often relied on peer help (56%), online how-to videos or documentation (51%), and figuring things out on their own (50%).

“On support for using learning tools, students most often relied on peer help, 56%, online how-to videos or documentation, 51%, and figuring things out on their own, 50%.”

May combine two claims.

Social media content is used by 30% of African students compared to 16% of OofA students.

“social media content, which 30% of African students (despite indicating later that it was not that helpful or effective) compared to 16% for OofA students.”

May combine two claims.

Generative AI tools and collaborative tools were rated most effective for preparing teaching (both median = 4), followed by YouTube/video tutorials.

“Generative AI tools and collaborative tools were rated most effective (both median = 4), followed by YouTube/video tutorials.”

May combine two claims.

On seeking help for tech challenges, a majority search for help on their own (51%), followed by turning to peers or classmates (23%) and instructors/TAs (11%).

“Onseeking help for tech challenges, a majority search for help on their own, 51%, followed by turning to peers or classmates, 23% and instructors/TAs, 11%.”

May combine two claims.

Fewer students rely on online forums or communities (5%), librarians or academic support staff (3%), or campus IT helpdesks (2%).

“Fewer rely on online forums or communities, 5%, librarians or academic support staff, 3%, or campus IT helpdesks, 2%.”

May combine two claims.

Correlations for collaborative tools and coding practice platforms were weaker and did not remain significant after correction.

“Correlations for collaborative tools and coding practice platforms were weaker and did not remain significant after correction.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Only a small minority of students usually do not ask for help (2%).

“Only a small minority usually do not ask for help, 2%.”

May combine two claims.

Among educator answers mentioning emerging technology themes, AI-centered technologies showed the highest frequency (32%), followed by other technologies (13.4%) and immersive technologies (11.8%).

“Among the answers mentioning emerging technology themes, answers mentioning AI-centered technologies showed the highest frequency (32%), followed by other technologies (13.4%), and immersive technologies (11.8%).”

Probably not a substantial claim: may not pass S3 (load-bearing).

Answers highlighted cost of subscriptions and institutional costs.

“Answers highlighted cost of subscriptions and institutional costs”

May combine two claims.

Simulations/virtual labs (34%) and AR/VR (32%) draw the lowest effectiveness ratings and the highest neutral responses (28% each).

“Forums/communities, 39% and social media explainer content, 38% are viewed as moderately effective, while simulations/virtual labs, 34% and AR/VR, 32% draw the lowest effectiveness ratings and the highest neutral responses 28% each.”

May combine two claims.

Alignment with learning goals (11.4%) and ability to draw data-driven insights about the tool's impact on student learning served as influential indicators for discontinued use.

“Educators shared that these student experiences, combined with how well the tool aligned with the learning goals, 11.4%, as well as the ability to draw data-driven insights about the tool's impact on student learning, served as influential indicators for discontinued use.”

May combine two claims.

Of educators with limited access, often(sometimes), 8(4) were based in Africa and 7(2) in unreported locations.

“Of those with limited access, often(sometimes), 8(4) were based in Africa and 7(2) in unreported locations.”

May combine two claims.

Visualization and demonstration tools were the most frequently coded theme (18%) and were praised for simplifying complex computing or mathematical concepts.

“The most frequently coded themes included visualization and demonstration tools, 18%, which were praised for simplifying complex computing or mathematical concepts, and video or multimedia resources, 15%, valued for making theoretical material more approachable.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Issues related to the suitability of course content were mentioned by 4% of respondents.

“Smaller subgroups of instructors also shared concerns with technical and pedagogical support, 5%, as well as issues related to the suitability of course content, 4%.”

Probably not a substantial claim: may not pass S3 (load-bearing).

3% said they usually do not ask for help.

“14% indicated Campus IT helpdesk or tech support as their usual go to place, 7% indicated online forums or communities (e.g., Stack Overflow, Reddit, Discord), 4% librarian or academic support staff, and 3$ said they usually do not ask for help.”

May combine two claims.

Subscription and licensing considerations were mentioned by 11% and 9% of respondents respectively.

“Other prominent barriers included subscription and licensing considerations, 11% and 9% respectively, as well as power and infrastructure reliability, 9%.”

May combine two claims.

Verbal feedback (54%), course effectiveness surveys (35%), and informal student preferences (29%) are popular ways respondents collect and integrate student preferences or feedback when choosing learning tools.

“Verbal feedback, 54%, course effectiveness surveys, 35%, and informal student preferences, 29% are popular ways respondents collect and integrate student preferences or feedback when choosing learning tools.”

May combine two claims.

In SEQ25, respondents indicate quality (30%), set-up and learning curve (25%), and access (17%) as the most frequent sources of frustration.

“Respondents indicate quality (30%), set-up and learning curve (25%) and access (17%) as the most frequent sources of frustation.”

May combine two claims.

Some student discontent stemmed from accessibility challenges as well as cumbersome navigation or interface designs.

“Some of this student discontent stemmed from accessibility challenges as well as cumbersome navigation or interface designs.”

May combine two claims.

Cost was the primary reason to stop using a tool (35.7%), followed by data privacy concerns (14.3%) and inadequate institutional infrastructure (11.4%).

“These patterns were also reinforced as instructors also described cost as the primary reason to stop using a tool, 35.7% followed by data privacy concerns, 14.3% and inadequate institutional infrastructure, 11.4%.”

Probably not a substantial claim: may not pass S3 (load-bearing).

Orientation or onboarding sessions were a support type at 14%.

“The other types of support included workshops or tutorials provided by the university, 39%, mentorship or learning communities, 26%, library or academic success center staff, 19%, IT helpdesk or technical support, 18%, and orientation or onboarding sessions, 14%, respectively.”

May combine two claims.

12.6% of educators missed full-time access: 7.5% reported accessing often a personal laptop (7) or desktop (2), while 5% reported only occasional access to a device.

“Still, 12.6% missed on full-time access: 7.5% of educators reported accessing often a personal laptop (7) or desktop (2) while 5% reported only occasional access to a device.”

May combine two claims.

Curiosity, fun and engagement, and efficiency were the top reasons respondents indicated for trying a new tool, with 27%, 22%, and 20% of responses respectively.

“Curiosity, fun and engagement, and efficiency were the top reasons respondents indicated for trying a new tool for their classroom, with 27%, 22%, and 20% of the responses.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

In SEQ25, other themes include Fit to purpose (12%) and cost (5%).

“Other themes include Fit to purpose (12%) and cost (5%).”

Probably not a substantial claim: may not pass S3 (load-bearing).

In SEQ14, educators mention the need to seek formal approval or budget support prior to adoption (9%).

“They also mention the need to seek formal approval or budget support prior to adoption (9%).”

May combine two claims.

Educators identified insufficient institutional supports described as licensing restrictions (9.6%), limited or inadequate computing infrastructures (11.4%), and a lack of technical or pedagogical training opportunities (5.8%) as examples of institution-level barriers to adoption.

“For example, educators identified insufficient institutional supports described as licensing restrictions, 9.6%, limited or inadequate computing infrastructures, 11.4%, and a lack of technical or pedagogical training opportunities, 5.8% as examples of institution-level barriers to adoption.”

Probably not a substantial claim: may not pass S3 (load-bearing).

IT helpdesk or technical support was a support type at 18%.

“The other types of support included workshops or tutorials provided by the university, 39%, mentorship or learning communities, 26%, library or academic success center staff, 19%, IT helpdesk or technical support, 18%, and orientation or onboarding sessions, 14%, respectively.”

May combine two claims.

26% of respondents believe the support they received was extremely useful, 31% rated it as very useful, 23% as somewhat useful, and 9% reported not useful.

“26% of respondents believe the support they received was extremely useful, 31% rated it as very useful, 23% as somewhat useful, and 9% reported not useful.”

May combine two claims.

In SEQ15, respondents also mentioned safety (14%), obsolescence (9%), and lack of access (6%).

“The respondents also mentioned safety (14%), obsolence (9%) and lack of access (6%).”

May combine two claims.

Overall, respondents rated reliable Internet, electricity or power, online learning platforms, necessary software, and personal computer in increasing ease of access, respectively.

“Overall, respondents rated reliable Internet, electricity or power, online learning platforms, necessary software, and personal computer in increasing ease of access, respectively.”

Probably not a substantial claim: may not pass S3 (load-bearing).

7% of respondents indicated other types of support including authors of Github tools, and colleagues.

“7% of respondents indicated other types of support including authors of Github tools, and colleagues”

May combine two claims.

African educators struggle with access to electricity, but seem to have higher access to reliable internet compared to African students.

“As expected, African educators struggle with access to electricity, but seem to have higher access to reliable internet compared to the African students.”

May combine two claims.

The likelihood that an educator would integrate new learning tools was informed by student perceptions of accessibility (20.5%) and by positive student feedback (19.2%).

“Similarly, the likelihood that an educator would integrate new learning tools was also informed by student perceptions of accessibility, 20.5% and by positive student feedback, 19.2%.”

May combine two claims.

Generative AI tools, YouTube or other video tutorials, and Educational apps or software are the most popular tools that educators find effective for preparing them for teaching.

“On the tools that educators find effective for preparing them for teaching, Generative AI tools (e.g., ChatGPT, Claude), YouTube or other video tutorials, and Educational apps or software (e.g., Wolfram Alpha, Grammarly) are the most popular but a deeper look into the qualitative results illustrates a more granular breakdown of tool preferences based on function or tool output.”

Classified as probably not a substantial claim.

The analysis of the survey for this subgroup indicates location plays a significant role in the way some tools are integrated and used.

“The analysis of the survey for this subgroup indicates location plays a significant role in the way some tools are integrated and used.”

May combine two claims.

The most common concerns related to tool ineffectiveness are access issues (19%), UI/UX design issues (14%), and issues with the learning curve (9%).

“The most common concerns related to tool ineffectiveness are access issues, 19%, UI/UX design issues, 14%, and issues with the learning curve, 9%.”

May combine two claims.

Educators described coding practice tools (9%) and collaboration platforms (7%) as enabling hands-on learning and formative feedback.

“Educators also described coding practice tools, 9%, such as Python Tutor or CodingBat, and collaboration platforms, 7% like GitHub and Microsoft Teams, as enabling hands-on learning and formative feedback.”

May combine two claims.

Platform compatibility (9%), paywall and licensing considerations (9%), including a lack of institutional support, and privacy concerns (3%) also explained educator sentiment related to the effectiveness of the tool.

“Some of these themes are further supported, as responses indicated that platform compatibility, 9%, paywall and licensing considerations, 9%, including a lack of institutional support, and privacy concerns, 3% also explained educator sentiment related to the effectiveness of the tool.”

May combine two claims.

Cost was also the most frequently mentioned concern by the participants of our educator survey for trying new technologies for themselves or for integration in their classrooms, followed by time and learning curve.

“Cost was also the most frequently mentioned concern by the participants of our educator survey for trying new technologies for themselves or for integration in their classrooms, followed by time and learning curve.”

May combine two claims.

On our student surveys, highlighting regional experience from African countries, lack of internet access and cost were the most frequently mentioned challenges by the respondent when trying new digital learning tools, followed by time.

“On our student surveys, highlighting regional experience from African countries, lack on internet access and cost were the most frequently mentioned challenges by the respondent when trying new digital learning tools, followed by time.”

Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

On open educational resources, types of contents and the creation processes are explored, and discovery and reuse of such resources represent primary areas of concern in the literature for effective adoption, and assessment [23].

“On open educational resources, types of contents and the creation processes are explored, and discovery and reuse of such resources represent primary areas of concern in the literature for effective adoption, and assessment [23].”

• cites [23] [23] Dalia Hanna, Abdolreza Abhari, and Alexander Ferworn. 2017. Comparing quantitative and comment-based ratings for recommending open educational resources. In Proceedings of the 20th Communications &amp; Networking Symposium (Virginia Beach, Virginia) (CNS '17) . Society for Computer Simulation International, San Diego, CA, USA, Article 4, 10 pages. -Not checked yet

  Not checked in the free run.