http://41.89.164.27:80 The University of Eldoret digital repository system captures, stores, indexes, preserves, and distributes digital research material. 2026-06-08T16:55:27Z http://41.89.164.27:8080/xmlui/handle/123456789/2754 ANALYSING THE POTENTIAL OF CARBON FARMING FOR INCOME IMPROVEMENT AND CLIMATE RESILIENCE AMONG SMALLHOLDER FARMERS IN BARINGO COUNTY, KENYA. KIPROP, JOHN Smallholder farmers in Kenya face the dual challenge of declining soil fertility and increasing vulnerability to climate change, both of which constrain agricultural productivity and household income. However, sustainable carbon farming presents an opportunity to enhance income generation through carbon trade while improving household incomes. Despite its potential, limited awareness and access to climate information hinder its full realization. This study addressed this practical problem by quantitatively examining how socio-economic factor influencing climate information for carbon farming, potential of agroforestry in generating carbon incomes, efficiency of water harvesting and management practices towards drought resilience, effect of minimal land tillage for better crop productivity and improved household livelihoods influence carbon-farming among smallholder farmers in Eldama Ravine and Baringo Central Sub- counties, Baringo county, Kenya. The study adopted a correlation design and utilized a mixed-methods approach combining structured household surveys (n = 374), field observations, and key-informant interviews. Quantitative data were analyzed using multiple linear regression with heteroskedasticity standard errors, supported by descriptive and diagnostic tests for classical OLS assumptions. The dependent variable Carbon Farming was constructed from four measurable indicators: access to climate information, perceived carbon income, efficiency of water harvesting and management practices towards drought resilience, and effect of minimal land tillage for better crop productivity and improved household livelihoods. Empirical results demonstrated strong model performance and significant predictive capacity across all objective-specific models. The pooled regression model achieved an adjusted R² = 0.608 and F = 189.875 (p = 0.000), confirming that approximately 61% of variation in carbon-farming could be explained by the combined effects of the independent variables. Socio-economic variables such as Gender (β = 0.014, p = 0.048), (β = 0.021, p = 0.058), and household income (β = 0.062, p = 0.007) emerged as significant determinants of climate information for carbon farming, suggesting that human capital and resource endowments substantially shape farmers’ ability to adopt and sustain carbon-farming. Agroforestry, Indigenous drought tolerant agroforestry trees, Exotic drought tolerant agroforestry trees, Fruit farming and Fodder trees statistically significantly predicted Carbon incomes with R 2 = 0.545, while Water harvesting and management practices, Spring restoration and protection, Riparian restoration and protection, and Roof water harvesting statistically . R 2 = 0.681, p-values for all variable pairs exceed 0.05 indicating that the assumption of homoscedasticity was met. These findings confirm that practical, low-cost interventions yield measurable carbon and productivity benefits. The aggregated analysis further estimated an average perceived carbon income of 56.98 (KSh. 7,407.4) per month, underscoring the economic potential of verified carbon-credit participation for rural livelihoods. In conclusion, carbon farming offers a viable income-generating opportunity for smallholder farmers in Baringo County, but its success depends on improved climate information access and targeted awareness efforts. The study recommends capacity building on agroforestry and carbon farming, distribution of tree seedlings to farmers, and sensitization programs on carbon trading. Additionally, gender-sensitive approaches should be integrated into agricultural extension services to bridge the gender gap in climate information access and empower women in decision-making. Strengthening climate information dissemination channels and enhancing extension services will be crucial in optimizing the benefits of carbon farming and improving household incomes in the region. 2025-01-01T00:00:00Z http://41.89.164.27:8080/xmlui/handle/123456789/2753 MACROINVERTEBRATE ASSEMBLAGES AND BODY SIZE DISTRIBUTION IN RESPONSE TO LAND-USE CHANGES IN HEADWATER STREAMS OF THE SONDU-MIRIU RIVER BASIN, KENYA KULUO, GIDEON Assemblages of aquatic macroinvertebrates have spatial and temporal variations in structure in response to environmental changes of their habitats, such as streams and rivers. Taxonomic approaches for monitoring the ecological status of aquatic ecosystems using macroinvertebrate assemblages face several limitations, creating a need for ataxonomic methods that are valid, disturbance-sensitive, and cost-effective for freshwater monitoring. This study evaluated the influence of seasonality and land-use changes on macroinvertebrate assemblage structure and water physico-chemical parameters of headwater streams of the Sondu-Miriu River basin, Kenya. Additionally, the study evaluated the utility of macroinvertebrate size-spectrum metrics and abundance-biomass comparison (ABC) curves as ataxonomic methods of evaluating land-use influence on the river’s ecological condition. Macroinvertebrate taxonomic abundance (individuals/m2), wet weight (mg), and physico-chemical variables were measured during the wet and dry seasons in March and September 2024, respectively, from 24 headwater streams distributed across four land-use types: natural forest (NF), tea and tree plantations (TTP), smallholder tea (SHT), and smallholder agriculture (SHA). Results based on water quality showed turbidity, total suspended solids (TSS), total dissolved solids (TDS), electrical conductivity (EC), particulate organic matter (POM), and nitrate (NO3 - -N) identified streams in NF as least disturbed and in SHA as most disturbed; TTP and SHT streams were intermediate in water quality. Taxon richness, diversity, and Ephemeroptera, Plecoptera, and Trichoptera (EPT) indices indicated land-use specific influences on assemblage structure. Redundancy analysis (RDA) showed thatsmall-bodied taxa (<8 mg wet weight) were associated with high disturbance indicators, while large-bodied taxa (>32 mg) were associated with low disturbance indicators and sites. Slopes (λ) of normalized abundance- and biomass-based size-spectrum (a measure of trophic transfer efficiency) deviated from theoretical steady-state conditions (λ=-2.0 and -1.0, respectively), indicating that the sites were disturbed, but the slopes did not vary significantly between land uses or seasons, suggesting size-spectrum slopes had low sensitivity to land-use-based changes in water quality. In contrast, size-spectrum midpoint heights (a measure of ecosystem production) differed significantly between sites, highest at SHT and NF and lowest at SHA and TTP streams. Spectrum midpoint heights were, therefore, more responsive to disturbance than slopes, highlighting their potential as indicators of land-use influence on the Afrotropical streams. Although the ABC curves indicated undisturbed conditions for all sites, Warwick’s (W) statistics revealed subtle differences among streams, suggesting variation in local-scale ecological conditions. This study demonstrated that catchment land use significantly influences water quality and macroinvertebrate assemblages in headwater streams of the Sondu-Miriu River basin. Traditional metrics (%EPT and diversity indices) are useful indicators of land use-based disturbance, while size-spectrum midpoint heights are potentially useful ataxonomic indicators of disturbance in the studied streams. It is recommended to integrate both community indices and functional metrics (especially midpoint height) into stream biomonitoring as complementary method for evaluating the ecological status of headwater streams. Management efforts should prioritize riparian buffers and nutrient/sediment control in SHA areas of the basin, while conserving less-impacted NF sites as reference areas. Long-term and broader spatial-scale studies are needed to validate the stability of size-spectrum metrics and ABC curves as rapid tools for monitoring the ecological status of headwater streams in response to anthropogenic influences. 2025-01-01T00:00:00Z http://41.89.164.27:8080/xmlui/handle/123456789/2752 THE IMPACT OF THE &#39;SKILLING UGANDA’ PROGRAMME ON YOUTH EMPLOYMENT IN EASTERN UGANDA. KIPYEKO, KENNETH This study examined the impact of the ‘Skilling Uganda’ programme on youth employment in eastern Uganda. According to the 2021 National Labor Force Survey, youth unemployment stood at 16.5%, higher than 2020’s 13.3% and the national average of 9.2%. Youth unemployment has economic, social, and political consequences, prompting calls for empirical evaluation of interventions like the “Skilling Uganda” programme. Guided by Human Capital Theory and General Systems Theory, the study aimed to: (1) establish the employment status of programme graduates, (2) determine their employability, (3) identify employment-related challenges, and (4) explore institutional challenges in sustaining the programme. Using a mixed-methods design with a pragmatic paradigm, 208 respondents were sampled from a target of 500 through purposive, simple random, and stratified sampling. Data were collected via questionnaires and interviews, then analyzed using SPSS. Findings revealed that 14.4% of graduates were unemployed, and those employed among the respondents faced low or inconsistent earnings, indicating persistent underemployment. Significant skill gaps were noted in digital literacy (51.4%) and leadership (64.0%), with respondents attributing these skills to personal initiative rather than the training. Key employment challenges included low capital (76%), technological limitations (39.9%), lack of employability skills (27.9%), and weak social networks (27.9%). Institutional challenges included budget constraints, limited focus on employability skills in curricula, and weak industry linkages, particularly in securing partnerships with the private sector. The study concludes that while the programme has improved access to work, the quality and sustainability of employment remain limited by skill gaps, inadequate capital, weak industry linkages, and low technological proficiency. It recommends enhancing employability training, strengthening post-training support and financing, building stronger industry linkages, and expanding technological training. 2025-01-01T00:00:00Z http://41.89.164.27:8080/xmlui/handle/123456789/2751 CEFFECT OF MODULAR SYLLABUS IMPLEMENTATION STRATEGIES ON PERFORMANCE OF TECHNICAL COURSES IN TECHNICAL VOCATIONAL EDUCATION TRAINING IN NAIROBI COUNTY, KENYA KIMEMIA, JOSEPH Quality education and training are essential for national progress, with Technical Education and Vocational Training institutions focusing on continuous improvement aligned with Sustainable Development Goals. However, rising unemployment and inadequate skills among Kenyan Technical Education and Vocational Training graduates, due to poor modular strategy implementation, threaten youth livelihoods. In this regard, the main objective of this study was to investigate the effect of modular syllabus implementation strategies on the performance of technical courses in Technical Vocational Education Training in Nairobi County, Kenya. The specific objectives of the study were to: Determine the effect of modular partnership strategy on the performance of technical courses in Technical Vocational Education Training in Nairobi County, Kenya; assess the effect of modular digitalization strategy on the performance of technical courses in Technical Vocational Education Training in Nairobi County, Kenya; and establish the effect of modular assessment strategy on the performance of technical courses in Technical Vocational Education Training in Nairobi County, Kenya. Based on Constructivist Theory, this study employed a descriptive survey method targeting 5,633 participants, including 5,204 engineering students, 418 engineering tutors, and 11 principals from 11 Technical Education and Vocational Training institutions in Nairobi County that implemented modular syllabi. The sample size included 359 individuals: 11 principals, 26 tutors, and 322 students. Principals were selected through purposive sampling, while stratified random sampling was used for tutors and students. The instruments’ reliability was confirmed through a pilot study, with Cronbach’s alpha coefficients of 0.817, 0.752 and 0.814 for modular partnership, digitalization and assessment strategy, respectively, indicating their reliability. The data was analyzed using SPSS version 28.0, where quantitative data underwent descriptive and inferential statistical analysis, and results were presented in tables. Qualitative data was analyzed for themes and sub-themes, and findings were illustrated with quotations. The study found significant positive effects of modular partnership (B1 = 0.403, p = 0.000), digitalization (B2 = 0.432, p = 0.000), and assessment strategies (B3 = 0.216, p = 0.000) on the performance of technical courses in Technical Education and Vocational Training. Moreover, modular digitalization strategy (t = 12.263) had the greatest impact on the performance of technical courses in Technical Education and Vocational Training, followed by modular partnership strategy (t = 11.915), and lastly modular assessment strategy (t = 6.794). The study concludes that modular partnership, digitalization and assessment strategies significantly affect the performance of technical courses in Technical Education and Vocational Training in Nairobi County, Kenya. 2025-01-01T00:00:00Z http://41.89.164.27:8080/xmlui/handle/123456789/2750 THE EFFECT OF EBB-AND-FLOW TECHNOLOGY, SUBSTRATE TYPES AND SALT-TOLERANT CROP (Apium graveolens L.) ON NUTRIENT REMOVAL FROM A BREWERY EFFLUENT OBADO, ELIZABETH Brewing industries face economic and environmental challenges of water use, energy consumption, and effluent disposal. The limited water availability and strict effluent discharge regulations in the South African Brewery justify wastewater treatment and recycling. Constructed wetland (CW) design and operation have been optimised for sustainable wastewater treatment. An ebb-and-flow technology is a CW design that creates aerobic and anaerobic conditions essential for wastewater treatment. South African Ibhayi Brewery effluent is treated onsite using a commercial-scale anaerobic digester (AD) and activated sludge units. However, the anaerobically digested effluent has high treatment costs and does not meet the environmental discharge standards. Therefore, the post-AD effluent is further treated using a low-cost CW technology. This study investigated the effect of ebb-and-flow CW operation, substrate types and celery plant growth on nutrient removal from a post-AD brewery effluent. Two experiments were conducted; Experiment 1 examined the effect of retention times (10, 20 and 40 minutes) and gravel sizes (7, 13, and 19 mm) on nutrient removal. The retention times (RT) and gravel sizes were allocated as treatment combinations and replicated thrice in a completely randomised design. In the second experiment, an ebb-and-flow CW was used to investigate the effect of media types planted with a celery crop on nutrient removal. Eight media (clay pebbles, clay bricks, sand, bioballs, recycled plastic, and gravel sizes), either alone or mixed, were tested. A 2:1:1 ratio of each media, pine bark and granular activated carbon, respectively, was used for mixed treatments. A uniform celery planting density of eight seedlings per m 2 and a 12-minute effluent retention time were used. Results for experiment 1 indicated significant differences at p≤ 0.05 between RT and gravel size treatment combinations on nutrient removal. The 10-minute RT at 19-mm gravel achieved the highest reduction in chemical oxygen demand (COD) of 8.2 %, 7.7 % ammonia-N and 38 % total inorganic nitrogen removal. The 40-minute RT at 7-mm gravel had the highest nitrate-N removal of 18.6 %. Orthophosphate removal was below 5 % in all treatments. The peak removal for ammonia and total inorganic nitrogen was after 8 and 10 weeks, respectively, suggesting that the efficiency of the ebb-and-flow design on nutrient removal is time-dependent. The aerobic and anaerobic conditions of ebb-and-flow operation, gravel surface area and effluent retention time influence nutrient transformation and removal. Results for Experiment 2 indicated significant differences between unmixed and mixed media on nutrient removal (p ≤0.05). The unmixed media of clay pebbles had the highest mean reduction chemical oxygen demand of 7.5% and 8.1% ammonia-N removal. Mixing resulted in better overall mean removal efficiency of nitrite- N (7.6%), nitrate-N (15.3%), total inorganic nitrogen (29.9%), orthophosphate (12.7%) and plant biomass (11158.5 gm -2 ) than unmixed media. Media porosity, surface area, chemical composition and celery plant growth enhanced nutrient removal through synergistic interactions. Experiment 1 recommends a short effluent retention time of 10 minutes on 19-mm gravel for improved nutrient removal in the ebb-and-flow system design. Experiment 2 recommends unmixed clay pebbles, bioballs, and 19-mm gravel forimproved ammonia-N removal and mixed media for multiple pollutant removal and celery productivity in the ebb-and-flow constructed wetland. 2025-01-01T00:00:00Z http://41.89.164.27:8080/xmlui/handle/123456789/2749 MACHINE LEARNING BASED CERVICAL CANCER DETECTION MODEL IN WESTERN KENYA MURERE, JOHN Cervical cancer is the leading cause of cancer-related deaths among Kenyan women, with approximately 3,200 deaths reported annually, driven mainly by low screening uptake (16%) and late diagnosis. The aim of this study was to develop a machine learning based model that would enhance the detection of cervical cancer in Western Kenya, a region that has limited healthcare resources. This study used a cross-sectional study design where data from 968 women were collected, including information on demographics, reproduction, and clinical characteristics. Data was collected from health facilities. The study showed that 93.7% (n = 907) had no biopsy-confirmed abnormalities, while 6.3% (n = 61) had abnormalities. There were five machine learning models (Logistic Regression, Random Forest, Decision Tree, Support Vector Machine, and Artificial Neural Network) that were trained on 70% of the data (training set) and tested on 30% of the data (testing set). The random forest model achieved the highest accuracy (94.33%) and specificity (98.37%), which outperformed the other models and traditional methods like Human papilloma virus (HPV) testing (70-80% specificity) and Pap smear (>90% specificity) for confirming negative cancer cases. The logistic regression model had the highest sensitivity of 70% which was comparable to the Pap-smear method (60-95% sensitivity), but it was lower than the HPV testing, with a sensitivity greater than 90% which makes it suitable for initial cervical cancer screening. The Pap smear results and use of hormonal contraceptives emerged as the key significant predictors of cervical cancer, which supports targeted screening strategies. The findings from this study confirmed there was a significant difference in model performance with partial superiority over existing methods and the influence of key cervical cancer risk factors. The combined approach of using a random forest model for confirmation and logistic regression for screening could optimize cervical cancer screening further in the resource-constrained Western setting. This study has underscored the potential that machine learning has in addressing cervical cancer disparities in Western Kenya, with implications for both public and private health interventions and future research work. 2025-01-01T00:00:00Z http://41.89.164.27:8080/xmlui/handle/123456789/2748 GENETIC DIVERSITY, CROSS PATHOGENICITY AND CONTROL OF BACTERIAL BLIGHT OF COWPEA USING BACILLUS SP. AND SELECTED BOTANICALS KIRAREI, EZRA Cowpea (Vigna unguiculata (L.) Walp.) is an economically important crop cultivated both for domestic and commercial purposes. It is one of the most resilient crops suited to arid and semi-arid areas. Cowpea production is affected by several diseases caused by phytopathogens. Bacterial blight caused by Xanthomonas axonopodis pv. vignicola has been reported in many cowpea producing areas causing reduction in both quality and quantity of the harvestable leaves and grains. However, in Kenya, the disease occurrence status has not been exhaustively documented, similar to the causal pathogen characterization and management using biological control agents and botanicals. This research conducted field surveys in farms from six counties representing different zones to analyse the disease occurrence levels. The identification of the pathogen was done by morphological and biochemical features as well as by analysis of 16S rDNA and its genetic diversity was determined using inter-simple sequence repeat markers. The pathogen cross pathogenicity was tested by inoculating other legumes (beans, soya bean, green gram, garden peas and lentils) with the most virulent isolate. Dual culture and inverted plate techniques were used to test bio-efficacy of the biological control agents and botanical extracts. Xanthomonas axonopodis pv. vignicola was confirmed as the causal pathogen in 48 farms out of 80 farms. The mean disease incidence was 44.89% across the sampled sites, being highest in Kakamega County (50.49%) and the least in Uasin Gishu County (33.57%). Xanthomonas axonopodis pv. vignicola isolates displayed slight variance in morphological and cultural characteristics on nutrient agar. The biochemical tests and analysis by blasting of sequence from 16S rDNA region confirmed Xanthomonas axonopodis pv. vignicola as the causal agent. The polymorphic information content ranged from 0.2384 to 0.4486, indicating genetic variations which was strongly correlated with the differences within populations. The pathogen, infection was observed in all the cowpea varieties tested and cross infection on soya bean and lentils. The percentage disease severity ranged between 25.83% to 51.67%, which was significantly higher in cowpea varieties. The bio-efficacy of selected bio-agents showed varying levels of percentage inhibition against X. axonopodis pv. vignicola, depending on the method and the duration of exposure. Bacillus subtilis displayed the highest antibacterial activity between 68.33% and 87.79% by dual culture technique and between 38.33% to 71.33% in inverted plate method over seventy-two hours. Bacillus amyloliquefaciens showed a statistically significant antibacterial activity between 45.00% and 76.12% in dual culture technique and between 45.00% and 73.89% by inverted plate method. Cyprofloxacin the antibiotic used for positive check, ethanolic extracts of neem, garlic and ginger inhibited X. axonopodis pv. vignicola by 56.3 mm, 38.5 mm, 30.8 mm and 25.0 mm respectively, but no potency was noted for Salvia nilotica. Genetic variations of Xanthomonas axonopodis pv. vignicola was noted from different regions of Kenya. This study showed that B. subtilis and B. amyloliquefaciens, neem, garlic and ginger extracts are useful biocontrol options in management of X. axonopodis pv. vignicola and therefore can be recommended for integration in the management of this pathogen in cowpea. 2025-01-01T00:00:00Z http://41.89.164.27:8080/xmlui/handle/123456789/2747 THE RELATIONSHIP BETWEEN REMUNERATION AND WORKPLACE PERFORMANCE OF TVET GRADUATES IN UGANDA’S MINERAL WATER PRODUCTION INDUSTRY MWEBAZA, IVAN TVET graduates in Uganda’s mineral water industry are frequently subjected to low remuneration. They are often employed on casual or short-term contracts with minimal job security. This study aimed at investigating the relationship between remuneration and workplace performance of TVET graduates in Uganda’s mineral water production industry. The study objectives were; to assess the relationship between monetary remuneration and workplace performance of TVET graduates in Uganda’s mineral water production industry, to evaluate the relationship between non-monetary remuneration and workplace performance of TVET graduates in Uganda’s mineral water production industry and to establish the workplace performance of TVET graduates in Uganda’s mineral water production industry. A quantitative methodology and a descriptive research design were utilized. The investigation was guided by the Human Capital Theory in conjunction with Maslow&#39;s Hierarchy of Needs. It embraced a positivist research paradigm, concentrating on the gathering of empirical data to uncover causal relationships. The research focused on a demographic of 90 graduates from TVET currently employed in the mineral water production industry. A sample of 73 individuals was chosen for the research. The stratified sampling technique was employed, and quantitative information was gathered through closed-ended surveys. The data gathered was examined through SPSS software. The results indicated a robust positive link between financial incentives and immediate performance. Conversely, non-financial rewards were shown to significantly influence employee engagement and long- term job satisfaction, underscoring their crucial role in boosting workplace morale and loyalty. The research established that a notable correlation is present between both financial and non-financial incentives and the performance levels of employees within Uganda&#39;s mineral water production sector, especially among graduates of TVET. As a result, it was recommended that companies adopt an integrated reward system to improve both employee contentment and overall productivity. 2025-01-01T00:00:00Z http://41.89.164.27:8080/xmlui/handle/123456789/2746 COMPETENCY-BASED TRAINING AND SKILLS ACQUISITION OF AUTOMOTIVE STUDENTS IN TECHNICAL TRAINING INSTITUTES IN THE EASTERN REGION OF UGANDA OKUNG, SAMSON This study examined competence-based training (CBT) and skill acquisition for automotive students in Technical Training Institutes in Eastern Uganda, aiming to determine the influence of availability of instructional resources, infrastructure relevance, human resource capacity, and financial adequacy on skill acquisition. A mixed-methods approach integrated quantitative and qualitative data from three hundred forty-one (341) participants, 288 students, 40 instructors, and 13 administrators, selected via stratified sampling. Data were collected using a combination of research instruments, including questionnaires, interviews, focus group discussions (FGDs), and document analysis. The collected data were analyzed using Minitab (version 2022) to generate both descriptive and inferential statistical results. Quantitative data was processed with descriptive and inferential statistics, while qualitative data underwent thematic analysis. Instrument validity and reliability were ensured through pilot testing and expert validation. Questionnaires showed instructional resources were moderately effective (mean 2.89, SD 1.28), with 75.7% of students using them weekly, but 80.9% face access issues; FGDs and interviews confirmed shortages (47% students, 80% instructors). Infrastructure was slightly adequate (mean 3.24 students, SD 1.24); with 90.6% finding it conducive per questionnaires, yet interviews revealed 80% of instructors noted outdated equipment (60%), supported by maintenance logs. Human resource capacity supported CBT, with 77.8% of students noting methodology alignment (mean 3.56, SD 1.10) via questionnaires, but interviews indicated 70% of instructors reported staffing shortages, verified by training records. Financial support is inadequate (mean 2.84, SD 1.32), with 87.5% affirming practical training impact per questionnaires, though 38.5% of administrators lacked funding mechanisms per interviews and budget reviews. Recommendations, institutes should prioritize the modernization and availability of instructional resources to reflect current industry standards, institutes need to invest in expanding and upgrading training facilities, recruit more qualified instructors and support staff to mitigate current staffing shortages and institutes must advocate for increased budget allocations and explore alternative funding sources 2025-01-01T00:00:00Z http://41.89.164.27:8080/xmlui/handle/123456789/2745 SOIL BIOASSAY AS A DIAGNOSTIC TOOL IN THE DETECTION OF SOIL- BORNE PATHOGENS OF COMMON BEAN (Phaseolus vulgaris) IN WESTERN KENYA NDUATI, GABRIEL Lack of accurate and precise diagnostic tools and methods for detecting soil pathogens and their impacts on crop yields hinders the productivity of various crops, including beans. This highlights the need to link field disease severities to bioassay results to ease not only the accuracy in detection but also ensure the reliability of the predictions of crop performance. The study was set to evaluate the effectiveness of existing soil bioassays for screening, detecting, and quantifying soil-borne pathogens (Fusarium spp., Pythium spp., and root- knot nematode) and predicting impacts on grain yields and severity levels in smallholder fields. The study was conducted in 50 farmer fields in five counties across Western Kenya. Field demonstration trials consisting of four bean varieties planted on the 4 by 4 m treatment plots were established in each of the five sites. At planting, soil samples were collected for physical and biochemical analysis. A subset of the soil samples was bulked at the field level and assessed for bioassays targeting Fusarium, Pythium, and plant parasitic nematodes. The bean varieties were assessed for disease severities, and agronomic indicators recorded. The study employed descriptive statistics, correlation analysis, and stepwise linear regression using R software. Pearson’s correlation coefficient (r) indicated a significant but weak negative relationship between the Fusarium stem assay and vascular browning (r= -0.27, p = 0.0091). In addition, Pythium seed assay had no association with field root rot severity, while the root knot nematode assay showed a marginally significant but weak positive relationship with field galling infestation (r = 0.14, p = 0.05). Also, a strong negative correlation between vascular browning and bean yield (r = -0.28, p< 0.001), and between disease pressure and yield (r = -0.22, p < 0.001) was observed. A stepwise linear regression model revealed that the interaction between Fusarium stem assay infestation and pH influenced vascular browning (p< 0.001) and stand count (p< 0.01). Similarly, the interaction of Pythium seed assay infestation with POXC significantly affected stand count (p< 0.001), and the association between Root knot nematode assay infestation and stand count was influenced by soil clay content (p< 0.001). Apart from the Root knot nematode assay, the soil pathogen bioassays tested here were not effective in predicting the field incidence. However, our findings revealed the ability of the fusarium stem assay to predict other symptoms associated with disease pressure, suggesting that the bioassays result together with soil fertility assessment can led to a more effective prediction of early disease pressure and yield. 2025-01-01T00:00:00Z