This qualitative study explores university students’ perceptions, ethical justifications, and trust dynamics regarding the use of generative AI tools like ChatGPT and DeepSeek in online examinations. Through phenomenological analysis of semi-structured interviews and focus groups with 20 students across diverse disciplines, the research reveals four key themes: (1) the reframing of AI as a “collaborator” rather than a cheating tool, (2) a paradox of high reliance despite awareness of AI’s limitations, (3) moral disengagement strategies to justify use, and (4) preferences for online exams due to convenience despite ethical dilemmas. Findings contrast with institutional definitions of academic misconduct, demonstrating how students rationalize AI use through critiques of traditional assessment methods. The study contributes to academic integrity literature by adapting neutralization theory to AI contexts and highlights the need for assessment reforms that address both technological realities and pedagogical values. Practical implications suggest moving beyond detection-focused approaches toward AI-integrated assessments and targeted digital literacy training. Limitations include sample specificity to one institution, suggesting need for cross-cultural replication.
Leather production is an important contributor to economic growth, but its reliance on chromium salts for tanning poses environmental and health concerns. This study evaluated Pterocarpus angolensis (DC.) sawdust as a sustainable source of vegetable tannins for eco-friendly leather tanning. Tannins were extracted using microwave-assisted extraction (MAE) optimized by response surface methodology (RSM) with a Box–Behnken design, considering microwave power, extraction time, and solid-to-solvent ratio. The extracts were characterized for total phenolic and flavonoid contents and functional groups using FT-IR spectroscopy. Their collagen crosslinking ability was assessed by differential scanning calorimetry (DSC) and the 2,4,6-trinitrobenzenesulfonic acid (TNBS) assay. Leather was tanned using the optimized extract, and its physical-mechanical properties were compared with mimosa-tanned leather. The optimized extraction conditions (280 W, 10 g/mL solid-to-solvent ratio, and 12 min) produced the highest yield (7.2 g). The extract contained 235.15 µg GAE/mL total phenolics and 45.4 µg CE/mL total flavonoids. DSC and TNBS analyzes confirmed effective collagen crosslinking, with a denaturation temperature of 70 °C and high amino-group participation at pH 5. Leather tanned with P. angolensis extract showed a shrinkage temperature of 78.8 °C, tensile strength of 12.09 N/mm², tear strength of 26.24 N/mm, elongation of 46%, flex endurance beyond 20,000 cycles without damage, and color fastness of 3/5 on the gray scale. These properties were comparable to mimosa-tanned leather and met ISO/IULTCS standards for vegetable-tanned leather, demonstrating that P. angolensis sawdust is a promising renewable tanning material for sustainable leather production.
Background: Cardiovascular disease (CVD) is the leading cause of death globally and accounts for approximately 13% of all deaths in Kenya. Despite this burden, few statistical models accounting for the clustered structure of longitudinal public health data have been applied to CVD risk assessment in Kenya. Conventional regression approaches that ignore within-cluster correlation yield underestimated standard errors and invalid inferences. Methods: We analysed longitudinal binary data on hypertension from the SCALE UP study (2012–2014) and the Nairobi Urban Health and Demographic Surveillance System (NUHDSS, 2002–2015) among adults aged 35 years or older in two urban informal settlements in Nairobi, Kenya. Two complementary modelling frameworks were compared: (i) a hierarchical generalised linear mixed model (GLMM), providing subject-specific estimates via individual-level random effects and (ii) the marginally specified logistic-normal model of Heagerty, which parameterises the marginal mean directly while retaining a latent normal random-effects structure to characterise within-subject dependence. Results: Hypertension prevalence declined from 75.7% at baseline (N=622) to 60.4% at endline (N=414; p<0.001). Both models identified the same six significant predictors at the 5% level: time, dehydration, blood glucose, body swelling, waist circumference, and hip circumference. The GLMM random-effects variance was σˆ2=3.322, indicating substantial between-subject heterogeneity. The marginal model yielded a conditional-scale variance of 4.358, reflecting the additional variability required to induce the specified marginal logistic form. The marginal model consistently produced narrower confidence intervals and attenuated odds ratios relative to the GLMM, consistent with a population-averaged interpretation. Conclusion: Both models identified central adiposity, measured by waist circumference (marginal OR=1.57, 95% CI: 1.18–2.09), as a key modifiable risk factor for hypertension, while greater hip circumference was inversely associated (OR=0.61, 95% CI: 0.46–0.80). The marginal model provides more stable population-averaged estimates suited to public health planning, whereas the GLMM captures individual-level heterogeneity useful for clinical risk stratification. A dual-model approach is recommended for comprehensive CVD risk modelling in similar resource-limited settings.
