International Journal of Biosafety, Biosecurity & Bioscience Innovations
E-ISSN: 3027-0235
Volume 2&3, No. 1, 2025
Pages 41-56
DOI: 10.36108/ijbbb/5202.320.0130
Assessment of Genetically Modified Crop Containment Strategies in Rural Agricultural Settings: A Case Study Approach with Machine Learning Insights
* NGOR James Terna1, NWABUEZE Ijeoma Maria2, EKWERE Uwem Akpan3, EGU Margaret Maneju2, YUSUF Hajara Oyiza4
1Agricultural Biotechnology Department, National Biotechnology Research and Development Agency, Along Musa Yar’adua Expressway, Airport Road, Lugbe, Abuja
2Bioresources Development Centre Abuja, National Biotechnology Research and Development Agency, Along Musa Yar’adua Expressway, Airport Road, Lugbe, Abuja
3Bioresources Development Centre Ikot Ekpene, National Biotechnology Research and Development Agency, Along Musa Yar’adua Expressway, Airport Road, Lugbe, Abuja
4 Bio entrepreneurship and Consultancy Services Department, National Biotechnology Research and Development Agency, Umaru Musa Yar’adua Expressway, Lugbe, Abuja – Nigeria
*Corresponding Author- Email: jimngorterna@gmail.com Orcid No: https://orcid.org/0000-0001-7321-2651
Abstract
This study assesses genetically modified (GM) crop containment strategies among smallholder farmers in Osun State, Nigeria in terms of awareness, adoption practices, perception, and barriers. Quantitative cross-sectional approach was adopted with structured questionnaire for data collection. Descriptive analysis showed a predominately male (58.8%), middle-aged (55.9% 35-44 years), with more than 10 years’ experience (52.9%). Non-normal data (Shapiro-wilk p < .05) used non-parametric and machine learning (ML) methods to test hypothesis. The results indicate that socio-demographic traits (education, experience, and occupation) had a significant effect on awareness (F = 6.453, p <.05), which was confirmed using the Random Forest (0.920 accuracy). A weak negative correlation between awareness and practices (Spearman’s rho -0.042, p = 0.416), but Gradient Boosting made a good prediction of practices (MSE 0.047, accuracy 0.867). Perceptions of containment effectiveness mediate practices and perceptions of risks, behaviors and perceived barriers and challenges, with a significant total effect (0.179, p< .05). Barriers and challenges have a significant (Spearman’s rho = 0.498, p < .05; Theil-Sen slope 0.216, MSE 0.218; SVR MSE 0.236) impact on perceptions of effectiveness of containment. In conclusion, this study highlights the importance of ML in understanding complex agricultural dynamics and a knowledge-practice gap, perception mediated risk relationships, with socio-demographic effects that revealed the need for tailored training programs in addressing occupation-specific needs, improved perception-based risk management and the reduction of barriers in promoting sustainable GM crop containment adoption in rural settings for development and food security.
Keywords: Containment Strategies, Food security, Genetically Modified Crops, Machine Learning, Rural Agriculture, Smallholder Farmers
*
