Research Papers
Permanent URI for this communityhttps://repository.tmu.ac.ke/handle/123456789/169
The research papers are organized by faculties
Browse
15 results
Search Results
Item Age- and sex- specific differences in haematological abnormalities associated with HBB gene polymorphisms among patients infected with Plasmodium falciparum(Cellular, Molecular and Biomedical Reports, 2026) Kosiyo, Paul Mboya; Ayodo, George; Munde, Elly OchiengAge and sex influence haematological abnormalities in Plasmodium falciparum malaria, but their link to Beta sub-unit genes (HBB) gene polymorphisms is unclear. This study examined these factors in infected patients. In this analytical cross-sectional study of 234 participants in Kenya (90 infected with malaria, 144 non-infected; 20 with HbSS, 214 with HbAS/HbAA), individuals aged ≥6 months with fever >37.5°C were recruited. Participants were stratified by infection status, HBB genotype, and age groups (6–59, 60–120, >120 months). P. falciparum was detected via microscopy, hematological parameters with Celtac G MEK-9100K, and HBB genotyping by real-time PCR. Statistical tests compared proportions, laboratory features, and associations between age/sex and hematological abnormalities. Haemoglobin (Hb) was lower (p=0.002), while Mean cell volume (MCV) (p=0.036) and platelet count (p=0.046) were higher in those infected and have carriage of HbSS relative to those infected and have carriage of HbAS/HbAA, respectively. White blood cell count (WBC) was comparable between the two groups with p=0.076. Participants of age >120 months had increased odds of anaemia as shown by low Hb [OR=13.362, 95% CI=2.742-65.119, p=0.001], low HCT [OR= 3.132, 95% CI= 1.783-24.694, p= 0.002] and low RBC [OR=7.888, 95% CI=1.691-36.791, p=0.009]. Additionally, older patients had reduced risk of agranulocytosis [OR=0.208, 95% CI=0.056-0.778, p=0.002]. With respect to sex, platelets were higher (p=0.007) in infected females relative to infected males; none of the abnormalities was associated with sex within the context of this study's population and sample size. HBB polymorphisms modulate haematological abnormalities and older age (>120 months) is strongly associated with increased risk of anaemia, which is more pronounced in P. falciparum infection. Anaemia and related abnormalities are most evident in HbSS carriers compared to HbAS/HbAA, while sex shows no association with these haematological changes.Item Predictive analysis of criminal behavior based on age and height: A case study of inmate demographic in homabay prison kenya(International Journal of Professional Business Review, 2025-08-15) Okelo, Stephen LucasObjective: This study explored whether attributes such as —age and height—can help anticipate criminal behaviour among inmates in Homa Bay County, Kenya. It aimed to assess how effective age and height contributes to crime prediction. Theoretical Framework: The study was guided by criminology theories linking demographic variables to crime typologies, particularly the age–crime curve (Farrington, 2003; Ngugi & Mwangi, 2020) and studies associating physical traits such as height with aggressive crimes (Adeyemi & Olatunji, 2019). These frameworks supported the hypothesis that age and height jointly influence the likelihood and type of criminal activity. Conceptual Framework: The study was anchored on hypothesized relationships between inmate demographic variables (age and height) and the type of crime committed, as well as how these variables can be used to build predictive models for criminal behavior. Method: a desktop review and quantitative data analysis design was adopted. Target population include 290 records of inmate prisoners from Homa Bay prison. Data collection tools included a record of reported crimes. Quantitative data were analyzed using descriptive statistics and visualizations to examine how offences are distributed across age and height groups. Results and Discussion: the study showed a positive relationship between age and height, whereby age proved to be a stronger predictor, with offences peaking around age 28, while height plays a secondary role, showing modest associations with violent crimes at certain ranges (notably around 6.8 feet). Research Implications: The results provide valuable insights for law enforcement profiling, youth-focused preventive interventions, and rehabilitation planning in Kenya. Predictive models incorporating age and height could help reduce mistaken identity, strengthen criminal investigations, and guide correctional programming. Originality/Value: This study contributes original empirical evidence from a developing county, addressing gaps in criminal identification and supporting strategic decision-making in crime prediction and rehabilitation measures.Item Hydrology-Based Evaluation of Irrigation and Cropping Practices in Relation to Malaria Transmission in Western Kenya(2026-07-13) Jiang, Ai-Ling; Lee, Ming-Chieh; Atieli, Harrysone; Yan, Guiyun; Hsu, Kuo-linIrrigation is widely linked to increased malaria vector population and transmission in rural Africa, yet it remains essential for improving food security. Balancing agricultural productivity with public health therefore requires approaches that consider malaria risk. In this study, we integrated a hydrologic model with an agent-based malaria model to evaluate the impacts of cropping practices, irrigation strategies, and human and mosquito mobility on malaria transmission in Homa Bay, Kenya. We examined scenarios with single crops (either staple or cash crops) and mixed cropping systems, alongside irrigation methods including flood, intermittent, and drip irrigation. The modelling framework captures spatiotemporal dynamics of mosquito breeding habitats driven by crop water demands and irrigation scheduling. Results show that staple crops such as rice required approximately 12% more irrigation than vegetables but were associated with 3-7% lower transmission metrics. This suggests that the timing and spatial distribution of water application are more important than total irrigation volume in determining larval habitat conditions. Combining intermittent and drip irrigation reduced malaria prevalence by 44% relative to flood irrigation, as it limited the extent and persistence of larval habitats. Water-efficient irrigation strategies also concentrated transmission into smaller areas, making targeted interventions more feasible. Under these more heterogeneous risk patterns, human movement and vector dispersal played a critical role in shaping malaria prevalence. Overall, the integrated modelling approach provides a framework for evidence based agricultural planning that meets crop water demands while reducing malaria risk.Item Anopheles larval ecology and physicochemical characterization of larval habitats in Dire Dawa: an area colonized by Anopheles stephensi in Eastern Ethiopia(BioMed Central, 2026-08-01) Abiy, Ephrem; Degefa, Teshome; Balkew, Meshesha; Merga, Hailu; Zewdu, Deneke; Ayala, Alamayo; Atieli, Harrysone; Lee, Ming-Chieh; Zhou, Guofa; Yan, Guiyun; Yewhalaw, DelenasawBackground: Understanding mosquito larval ecology is crucial for effective vector control strategies. While much is known about Anopheles larval habitat distribution, the impact of landscape (urban vs. rural) on Anopheles stephensi larval distribution and population dynamics in Ethiopia remains unclear. Methods: A longitudinal study was conducted from February 2023 to December 2024 in aquatic habitats in urban, peri-urban, and rural areas of Dire Dawa city administration, eastern Ethiopia. Mosquito larvae and pupae were collected and reared in a field insectary until adult emergence. The female Anopheles were morphologically identified and further confirmed via polymerase chain reaction (PCR). The physicochemical parameters of the larval habitats were measured via a HANNA multiparameter water probe. Statistical analyses included general linear models, ANOVA, logistic regression, paired sample t tests, chi-square tests, principal component analysis and correlation analyses to assess larval presence, density, and distribution in relation to water physicochemical parameters across the three ecological settings. Results: A total of 23,526 larvae and 1,808 pupae of Anopheles mosquitoes were collected from 856 man-made habitats (bricks, plastic sheets, steel drums, tire tracks, canal ditches, and barrels) and 53 natural habitats (river edges, animal hoof prints , ponds, and swamps) in urban, periurban, and rural areas. Uncovered cemented cisterns were the main human-made larval habitats; river edges were the main natural habitats. Anopheles larvae were absent in the steel drums and plastic barrels at the rural sites. Anopheles larval density significantly differed across the different ecological settings (F=7.8, df =1, 908, p=0.005) and among the different larval habitat types (F= 326.2, df = 1, 907, p< 0.001), land use/surrounding environment (χ2 = 32.6, df = 13, p = 0.002), vegetation (χ2 = 27.1, df =12, p=0.008), shade coverage (χ2 = 25.9, df = 15, p=0.039), substrate type (χ2 = 34.1, df = 6, p < 0.001), and competitor/predator presence (χ2 = 14.6, df = 1, p = 0.001). However, larval presence was not associated with the presence of larval interventions (χ2 = 1.3, df = 1, p = 0.250). Water pH (r = 0.26, n =102, p < 0.050) and water pressure (r = 0.21, n= 102, p < 0.050) were the only water physicochemical parameters that were positively correlated with An. stephensi presence. Conclusion: An. stephensi was dominant in urban areas but was also present in rural areas. Existing larval source management methods do not prevent Anopheles from occurring in many habitats, indicating an urgent need for improved strategies.Item Reduced bio-efficacy of aged PermaNet® 3.0 Nets against local pyrethroid-resistant Anopheles arabiensis in Western Kenya(Frontiers in Tropical Diseases, 2026-01-30) Machani, Maxwell G.; Zhou, Guofa; Oyweri, Job; Nzioki, Irene; Githure, John; Atieli, Harrysone; Wang, Chloe; Zhong, Daibin; Lee, Ming-Chieh; Afrane, Yaw A.; Yan, GuiyunBackground: Long-lasting insecticidal nets (LLINs) are central to malaria control and designed to protect for up to three years; however, increasing pyrethroid resistance undermines their effectiveness. Piperonyl butoxide (PBO) LLINs, designed to enhance pyrethroid efficacy, are recommended in high-resistance areas, yet their long-term operational bio-efficacy remains unclear. This study evaluated the killing efficacy of PBO LLINs over three years of field use compared to standard LLINs and monitored insecticide resistance in local Anopheles gambiae s.l. populations. Methods: During a trial in Muhoroni, western Kenya (2021–2024), standard and PBO LLINs were collected at 6, 18, and 36 months of use. Residual bio-efficacy was assessed using WHO cone bioassays against a susceptible Anopheles gambiae s.s. Kisumu strain and field An. gambiae s.l. populations. WHO tube and bottle assays determined insecticide resistance, while synergist assay assessed metabolic resistance. Quantitative polymerase chain reaction detected target-site mutations (kdr-1014 F/S and Ace-1 G119S). Results: Molecular identification confirmed all An. gambiae s.l. were Anopheles arabiensis, showing increasing pyrethroid resistance, with deltamethrin mortality declining from 96.3% (2021) to 22.7% (2024) while remaining susceptible to pirimiphos-methyl and clothianidin. PBO pre-exposure restored deltamethrin mortality from 22.7% to 98.9%. The frequency of 1014F increased from 0.09 to 0.17 and 1014S from 0.04 to 0.06, with no Ace-1 mutations detected. Standard LLINs retained >80% efficacy against the susceptible strain for 18 months but were below threshold against field mosquitoes even when new. PBO LLINs were effective at baseline against field populations but declined sharply, with mortality dropping to 24% by 6 months. Overall, both net types exhibited a marked decline in killing efficacy over time against field mosquitoes, with mortality falling < 20% within six months. Conclusion: Anopheles arabiensis showed increasing pyrethroid resistance, driven largely by metabolic mechanisms. Standard LLINs showed suboptimal killing efficacy against field populations, while PBO LLINs achieved high baseline efficacy but declined significantly by six months. These findings indicate that PBO LLINs can improve protection against resistant vectors but may be insufficient in high-resistance areas, underscoring the need for alternative non-pyrethroid interventions, strategic deployment, and revised LLIN re-distribution cycles aligned with their functional lifespan.Item Online Perceived Risk’s Influence on Consumer Purchase Decisions of Households in Nairobi City County, Kenya(International Journal of Research and Scientific Innovation (IJRSI), 2025-07-02) Linsey Wanjiku Waweru1; Stephen Ntuara Kiriinya2; Elyjoy Micheni3; Hellen Kabue4The study assessed the influence of online shopping dimensions: convenience, product presentation and perceived risks in regard to the purchase decisions of households in Nairobi City County, Kenya. The study also investigated the effect of data security and internet access as the moderating and intervening variables, respectively. The study was based on positivist philosophy and employed both descriptive and explanatory research designs. It was anchored on the Theory of Reasoned Action, Uses and Gratification Theory and Howard-Sheth Model. In determining sample sizes, the study adopted the Roscoe’s rule of thumb. The study further used a mixed sampling design where cluster sampling was applied to come up with the households of the eight (8) constituencies in Nairobi City County. Simple random sampling was also used to select households from the respective clusters. To identify the respondents, purposeful sampling technique was used to pick the household member who most frequently shopped online. The study employed structured questionnaires with both openended and close-ended questions to ensure the researcher’s biasness did not interfere with the data collection process. Descriptive statistics, like cross-tabulations, frequency distributions, means and standard deviations were used to summarize and make sense of the data which described all the variables of the study. The Statistical Package for Social Sciences (SPSS) Version 27 was used for data analysis. Inferential statistics were employed to draw conclusions about the entire population by looking only at a sample of the populationItem Towards Integration of AI Driven Predictive Model in identifying At-Risk Students in Online Learning Platforms(African Journal of Computing, Data Science and Informatics (AJCDS, 2025-06) 1Dr. Salmon Oliech Owidi; 2Prof. Elyjoy M. MicheniThe rapid growth of online learning environments presents both opportunities and challenges in student engagement and performance. Predictive modeling using artificial intelligence (AI) has emerged as a promising tool to identify atrisk students and personalize learning experiences. However, the influence ofvarious factors, such as engagement metrics, interventions, and demographicvariables, on student success in online education remains underexplored. This study aims to examine the impact of engagement metrics and targeted interventions, as well as explore the role of demographic variables in predictive modeling within online learning environments. This study employed a simple random sampling method, collecting data from 200 students enrolled in an online learning environment. Data on engagement, interventions, demographic variables, and academic performance were analyzed using descriptive and inferential statistical methods. The findings highlight the importance of engagement metrics in predicting academic success and the effectiveness of personalized interventions such as feedback and tutoring. Integrating personalized interventions can further support at-risk students, creating a more inclusive and effective educational experience. This study underscores the need for continuous development of AI models to better serve diverse student populations and improve educational outcomes.Item The Dark Side of Generative AI in Higher Education: Challenges, Ethical Concerns, and Implications(East African Journal of Arts and Social Sciences, 2025-09-23) Salmon Oliech Owidi1*; Joanne Nabwire Lyanda2; Elijah Owuor Okono2; Elyjoy Muthoni Micheni1This paper presents a meta-analysis of literature examining the dark side of generative artificial intelligence (GenAI) in higher education, with a focus on the ethical, pedagogical, and institutional challenges arising from its rapid integration. Drawing from a comprehensive review of 300 peer-reviewed articles, reports, and policy documents published between 2018 and 2024, the study synthesises key concerns across academic integrity, algorithmic bias, data privacy, faculty roles, and policy development. Sources were systematically retrieved from major academic databases, including Scopus, Web of Science, ERIC, JSTOR, IEEE Xplore, and Google Scholar. The literature was selected based on inclusion criteria that emphasised peer-reviewed status, relevance to higher education, and a clear discussion of the ethical or practical challenges of GenAI deployment. Exclusion criteria involved studies focused solely on K–12 education, regions outside of Africa without relevance to the African higher education context or works that concentrated purely on the technical development of AI without exploring its pedagogical, ethical, or institutional implications within educational environments. The meta-analysis employed a thematic coding approach, identifying recurrent concerns in areas such as accountability, fairness, accessibility, assessment validity, and institutional readiness. The appropriate and improper uses of ChatGPT and related technologies in academic work, including assignments, research, and tests, should be clearly defined in these rules. The study highlights the necessity of structured training programs for teachers and students in addition to policy creation. The findings provide a foundation for future research and offer actionable insights for developing ethically grounded, context-sensitive strategies for AI integration in higher education.Item Group Career Counselling and its Influence on Students’ Career Choices in Homa Bay County, Kenya(INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS), 2025-09-04) Linet Akinyi Elisha; Joab M. Kinzi, Phd; Lilian C. Kimaiyo, PhdThe society expect students to make career choices that not only make them employable but also suitable by the end of their schooling. According to Kenya National Bureau of Statistics (2024), 75% of the unemployed youth did not know what career would suit them by the time they left school. Making career choices is known to be one of the most difficult decisions. This means that students require guidance and counselling in career failure to which they may end up in unsuitable career with lifelong frustrations. Effort have been made by the school career counsellors to manage career guidance and counselling, however, many school leavers continued continued to display lack of career awareness in kind of jobs that would suit them after school. The study objective was to investigate the influence of both peer and professional led group career counselling on career choice among secondary school students in Homa Bay County, Kenya. This was a quantitative study that adopted correlational research design. Form three students were targeted and a sample of 379 participants obtained using proportional sampling. Data was collected using a 5-point scaled questionnaire validated by experts and whose Cronbach's alpha calculated reliability was .78. The analysis included calculation of Means,percentages, correlation and regression. It was found that group counselling has statistically significant influence on career choice with correlation (r) of 0.50 and (R2) of 0.25. This was interpreted to mean the relationship was positive and moderate and the (R2) value, meant that group counselling strategy influenced career choice by 25%. In conclusion there is a significant positive relationship between group career counselling and career choice. Regression analysis revealed influence of group career counselling on career choice. This suggest that improving counselling strategies also improves student career choices.Item The Influence of Individual Career Counselling Strategy on Career Choices among Students in Secondary Schools in Homa Bay County, Kenya(INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS), 2025-07-21) Linet Akinyi Elisha; Joab M. Kinzi PhD; Lilian C. Kimaiyo PhDThe Kenya government has a policy for career guidance and counselling which is meant to help provide a robust and professional career guidance and counselling to students. The policy aims at creating adequate human resource and reducing the mismatch between career choices and demand for labour. This is the rationale behind provision of career guidance services to students in schools. Most Secondary school counsellors have adapted individual career counselling strategy to help adhere to the policy. Despite the effort made by the the guidance counsellors to manage career guidance and counselling in secondary school, many school leavers continue to display lack of awareness in kind of career that would be suitable for them after school. The objective of this study was to investigate the influence of the individual career counselling strategy used on career choice among secondary school students in Homa Bay County, Kenya. The study employed ex-post-facto research design and sampled 372 form three students. The Hollands theory of career decision guided the study. Data was collected using questionnaires. The instruments were validated by experts and reliability of .78 obtained. The analysis was by calculating percentages, correlation and regression. Results findings were that individual counselling strategy was frequently used at 47 % in terms of percentages. It was also established that individual counselling strategy had statistically significant influence on career choice with Pearson correlation coefficient (r) of 0.65, which was a strong positive relationship. The coefficient determinant (R²) obtain was 0.4225 which was meant that individual counselling strategy influenced career choice by 42.25%. It was concluded that individual counselling significantly influenced career decision among students in homabay county. This also increased the chance of a student making right decision on career choices by almost 50 percent.