Gender and Functional Differentiation in Generative AI Usage among Malaysian Higher Education Student
Saved in:
| Title: | Gender and Functional Differentiation in Generative AI Usage among Malaysian Higher Education Student |
|---|---|
| Language: | English |
| Authors: | Fahd Ali Raza (ORCID |
| Source: | European Journal of STEM Education. 2026 11(1). |
| Availability: | Lectito Journals. Wassenaarseweb 20, 2596 CH, The Hague, The Netherlands. Tel: 31-70-2190600; e-mail: info@lectitojournals.com; Web site: http://www.lectitopublishing.nl |
| Peer Reviewed: | Y |
| Page Count: | 41 |
| Publication Date: | 2026 |
| Document Type: | Journal Articles Reports - Research Tests/Questionnaires |
| Education Level: | Higher Education Postsecondary Education |
| Descriptors: | Gender Differences, Artificial Intelligence, Technology Uses in Education, College Students, Intellectual Disciplines, Foreign Countries, STEM Education, Cultural Differences |
| Geographic Terms: | Malaysia |
| ISSN: | 2468-1954 2468-4368 |
| Abstract: | Despite achieving gender parity in higher education enrolment (60% female), Malaysia faces an emerging digital divide in how students use artificial intelligence. This study examined whether gender predicts task-specific ChatGPT usage patterns among Malaysian students (n = 443), employing latent profile analysis and multinomial logistic regression on CC BY 4.0 licensed Global ChatGPT Student Survey data (October 2024-February 2025). Four distinct usage profiles emerged: Selective Users (14.9%), Moderate Adopters (31.8%, coding-focused), Academic Enthusiasts (33.0%, text-focused), and Comprehensive Users (20.3%). Gender significantly predicted specialized profile membership (χ² = 19.47, p < 0.001). Males concentrated in coding-focused use (OR = 0.48 for females, p = 0.007), females in text-focused use (OR = 1.89, p = .020), while Comprehensive Users exhibited gender parity. Exploratory analyses indicate a tentative pattern of larger gender gaps in technical AI use in STEM fields (33.4 percentage points in Applied Sciences) versus Social Sciences (12.7 points), though small cell sizes (n=3 for female Natural Sciences students) prevent definitive conclusions. Profiles predicted domain-specific skill development with large effects (η² = 0.18-0.33). Findings reveal that equal access masks unequal functional engagement, with implications for gender-segregated occupational pathways that warrant further investigation with larger samples. |
| Abstractor: | As Provided |
| Notes: | https://doi.org/10.17632/nv2343nwsb.2 |
| Entry Date: | 2026 |
| Accession Number: | EJ1505190 |
| Database: | ERIC |
| Abstract: | Despite achieving gender parity in higher education enrolment (60% female), Malaysia faces an emerging digital divide in how students use artificial intelligence. This study examined whether gender predicts task-specific ChatGPT usage patterns among Malaysian students (n = 443), employing latent profile analysis and multinomial logistic regression on CC BY 4.0 licensed Global ChatGPT Student Survey data (October 2024-February 2025). Four distinct usage profiles emerged: Selective Users (14.9%), Moderate Adopters (31.8%, coding-focused), Academic Enthusiasts (33.0%, text-focused), and Comprehensive Users (20.3%). Gender significantly predicted specialized profile membership (χ² = 19.47, p < 0.001). Males concentrated in coding-focused use (OR = 0.48 for females, p = 0.007), females in text-focused use (OR = 1.89, p = .020), while Comprehensive Users exhibited gender parity. Exploratory analyses indicate a tentative pattern of larger gender gaps in technical AI use in STEM fields (33.4 percentage points in Applied Sciences) versus Social Sciences (12.7 points), though small cell sizes (n=3 for female Natural Sciences students) prevent definitive conclusions. Profiles predicted domain-specific skill development with large effects (η² = 0.18-0.33). Findings reveal that equal access masks unequal functional engagement, with implications for gender-segregated occupational pathways that warrant further investigation with larger samples. |
|---|---|
| ISSN: | 2468-1954 2468-4368 |
