AI evidence extraction

Engineering Students' Perceptions and Use of Generative Artificial Intelligence

Johannes L. Jooste, Karin Wolff, J. Joubert · Computer Applications in Engineering Education · 2025

AI-generated evidence extraction, verified across multiple analytical personas. Not a substitute for the peer-reviewed original.

7/10

Relevance

0/4

Quality (LMQS)

Evidence

Citations

1.78

FWCI

Top 10%

Impact

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.1002/cae.70064

Methodology & findings

Study design

Survey-based empirical study collecting quantitative and qualitative data from engineering students regarding their perceptions and use of generative AI applications

Main result

The study found that "the results reveal widespread adoption of GAI applications, with preferences regarding appropriate use and regulation approaches. Students generally support GAI as a learning aid while also expressing concerns about its use in formal assessments." Additionally, "only 1% supporting a complete ban on GAI use" and "many students planning to continue using these technologies in their future careers."

Reports effect sizes.

Research paradigm

Pragmatist/Mixed-methods (primarily quantitative survey with qualitative interpretation)

Author conclusions

The authors conclude that "This study provides insights for engineering educators working to balance innovation with academic integrity, highlighting the need for collaborative approaches between students and educators in engineering education to maximise GAI benefits while mitigating limitations."

Risk of bias

Selection bias: Survey participants were self-selected from a single South African university; Response bias: Participants' self-reported perceptions of GAI use may not reflect actual usage patterns; Geographic bias: Limited to one university in South Africa, reducing generalizability; Temporal bias: Study conducted during a specific period when GAI adoption was evolving; Selection bias: Sample limited to single South African university, may not be representative of global engineering student population; Self-selection bias: Respondents who chose to complete survey may differ from non-respondents in their GAI attitudes; Recall bias: Survey relies on self-reported perceptions and use patterns; Social desirability bias: Students may underreport unethical GAI use due to concerns about academic integrity; Selection bias: Limited to a single South African university context, may not generalize to other institutions or countries; Self-report bias: Survey-based data relies on students' self-reported perceptions and behaviors; Temporal bias: Study captures perceptions at a specific point as GAI technologies are rapidly evolving

Open questions raised

Future research should explore effective integration strategies and discipline-specific best practices as GAI capabilities continue to evolve. The authors identify a need for collaborative approaches between students and educators and emphasize the importance of structured education about ethical GAI use.
"Future research should explore effective integration strategies and discipline-specific best practices as GAI capabilities continue to evolve."
"Future research should explore effective integration strategies and discipline‐specific best practices as GAI capabilities continue to evolve."

Data: not_statedCode: not_statedExtracted from: pdfAgreement 63%

Explore related topics

Responsible AI Use AI Governance Survey