DISCOVER, READ & SHARE
The Institutional Repository is a digital hub for scholarly communication and knowledge dissemination which serves as a vital resource for researchers, students and the wider academic community. Content comprises of research publications, theses, conference proceedings and much more.
Recent Submissions
Behavioral Analysis Using Machine Learning Algorithms in Online Proctored Assessments; The Potential of Mobile Proctoring in E Assessments: A Review
(COMPILER, 2026-10) Oganda Bartholomew Mogoi1; John Kamau; Raymond Ongus
The rapid expansion of online learning has intensified the need for
secure and reliable examination systems. While artificial intelligence
(AI) and machine learning (ML) are increasingly employed to
safeguard integrity through behavioral analysis, current proctoring
solutions remain largely desktop-centric and poorly adapted to mobile
environments, where smartphones dominate access. These
limitations—spanning heterogeneous hardware, unstable lighting and
connectivity, algorithmic bias, privacy concerns, and limited
explainability—undermine fairness and trust in misconduct detection.
This study reviews the potential of mobile-based proctoring systems
that leverage ML for real-time behavioral analysis in e-assessments. A
systematic search across Scopus, IEEE Xplore, SpringerLink, and
Google Scholar identified 200 publications from 2015–2025. After
applying inclusion and exclusion criteria, 30 peer-reviewed empirical
studies were analyzed in depth. Findings reveal that ML-driven
behavioral analytics enhance accuracy and fairness by detecting
anomalies such as gaze aversion, facial emotion changes, and
environmental inconsistencies. However, research gaps persist in
mobile optimization, ethical safeguards, privacy protection, and bias
mitigation. The review concludes that hybrid mobile frameworks
integrating multimodal behavioral features and adaptive ML models
are essential to strengthen proctoring accuracy while ensuring
accessibility and privacy. Future research should prioritize context
aware and explainable AI to foster equitable and trustworthy online
assessment environments.
INFLUENCE OF PHYSICAL FACILITIES ON INCLUSION OF LEARNERS WITH PHYSICAL DISABILITIES IN PUBLIC PRIMARY SCHOOLS IN KISII CENTRAL SUB-COUNTY, KENYA
(Mount Kenya University, 2023-05-09) Sikini,Mary Bosibori; Kimani,Cecilia; Kamau, Joyce
The purpose of this study was to investigate the influence of physical facilities on the
inclusion of learners with physical disabilities in Kisii Central Sub-County, Kenya. The
research adopted the Social Model of Disabilities theory as its foundational framework. It
employed a mixed-method approach and concentrated on 124 schools with 124 headteachers,
1949 teachers, and 92 learners with physical disabilities as its participants. The sample
participants were selected by utilizing the Krejcie and Morgan (1970) table. The schools and
learners with physical disabilities were selected through a simple random sampling technique,
while teachers and headteachers were chosen through a stratified random sampling approach.
Questionnaires and interview schedules served as the primary data collection tools, while a
focus group discussion guide was utilized for engaging learners with physical disabilities.
Additionally, an observation checklist was employed to assess school facilities. The piloting
of instrument was conducted in a single school. The instrument's reliability was evaluated in
the research study through a test-retest analysis. The researcher employed a blend of
qualitative and quantitative methodologies for gathering data. The findings revealed mixed
perceptions among participants regarding the effectiveness and implementation consistency
of inclusive classroom practices such as accessible seating arrangements and adaptive
outdoor equipment. The findings underscore the critical importance of physical facilities and
in promoting inclusive educational environments. These findings suggest that investing in
adequate infrastructure and providing accessible learning materials are pivotal in enhancing
educational outcomes and inclusivity for students with diverse needs. Recommendations
included fostering collaborative efforts among stakeholders to enhance the implementation
and effectiveness of inclusive practices, thereby creating more supportive environments for
learners with physical disabilities.
Strategic Leadership and Incident Response Time Among Law Enforcement Agencies in Juba City, South Sudan
(Mount Kenya University, 2026-01) Ekidor,Paul; Kiprono,Wilson; Wekesa,Peter
The purpose of this study was to investigate the dynamics of strategic leadership on incident
response time among law enforcement agencies in Juba City, South Sudan. The contention of
the article is that incident response time is widely recognized as a critical indicator of law
enforcement effectiveness, particularly within fragile and post-conflict contexts where
institutional capacity and public trust remain limited. In South Sudan, persistent insecurity
and weak governance structures have heightened the demand for timely and effective law
enforcement responses, especially in urban centers such as Juba City. This paper examines
the influence of strategic leadership on incident response time among law enforcement
agencies in Juba City, South Sudan. Anchored in strategic leadership theory and the
organizational performance literature, the study posits that leadership practices including
strategic planning, coordination, resource allocation, and decision-making play a decisive
role in shaping operational responsiveness. Employing an exploratory qualitative approach
informed by a review of relevant empirical studies and contextual evidence, the paper finds
that deficiencies in strategic leadership contribute to delayed response times, fragmented
operations, and ineffective interagency coordination. Conversely, effective strategic
leadership is associated with improved preparedness, clearer command structures, and
enhanced operational agility. The study concludes that leadership-centered reforms are
essential for strengthening incident response capabilities, improving public safety, and
rebuilding public trust in law enforcement institutions in South Sudan.
Strategic Leadership and Case Resolution Performance in Police Departments in Juba City, South Sudan
(Mount Kenya University, 2026-01) Ekidor, Paul
The purpose of this study was to investigate the association between strategic leadership and
crime clearance performance, measured through the number of cases resolved per quarter,
within police divisions in Juba City, South Sudan. The contention of the study is that crime
clearance performance remains a central indicator of policing effectiveness, particularly in
post conflict urban environments where institutional capacity, coordination, and leadership
practices directly influence investigative outcomes. A mixed methods explanatory sequential
design was adopted, integrating quantitative survey data from 235 law enforcement personnel
with qualitative insights from key informant interviews. Structured questionnaires measured
perceptions of leadership practices and investigative performance, while interviews provided
contextual explanations of supervisory processes and operational challenges. Descriptive
findings indicated moderately strong clearance performance, with a composite mean score of
3.81. Leadership contribution to case clearance recorded the highest mean of 4.05, suggesting
that supervisory direction and oversight are perceived as central drivers of investigative
productivity. Reliability analysis confirmed strong internal consistency of the measurement
scale, with Cronbach alpha values exceeding 0.79. Pearson correlation analysis revealed
statistically significant positive relationships between case resolution and all leadership
dimensions, with inclusive leadership showing the strongest association (r = .659), followed
by accountable leadership (r = .570), visionary leadership (r = .552), and adaptive leadership
(r = .516). Multiple regression results demonstrated that leadership variables collectively
explained 48.0 % of the variance in case resolution performance (R² = .480, F = 53.141, p <
.001). Inclusive leadership emerged as the strongest independent predictor (β = .462, p <
.001), while visionary leadership also showed a significant positive effect (β = .226, p =
.001). The findings confirm that collaborative leadership practices and clear strategic
direction substantially enhance investigative throughput and case completion within policing
institutions operating under complex operational conditions. The study recommends
strengthening participatory supervisory structures, improving communication of investigative
targets, and institutionalizing routine performance monitoring systems to enhance crime
clearance performance in Juba City.
Multimodal CNN–LSTM Framework for Real-Time Maize Disease Detection
(Mount Kenya University, 2026-06-03) Tonui,Mercy Chepkoech; Kamau , John; Ongus,Raymond Wafula
Maize diseases present a major challenge to agricultural productivity and
food security, particularly in low-resource settings in sub-Saharan Africa.
Timely detection plays an important role in reducing yield losses and
enabling effective farm management. This research introduces and validates
a multimodal machine learning–based system for real-time maize disease
detection in Bomet County, Kenya. The system integrates maize leaf image
data, environmental sensor data, and farmer-reported observations to
develop a hybrid Convolutional Neural Network–Long Short-Term Memory
(CNN–LSTM) model designed to automatically identify and categorize
maize diseases. A mixed-methods research design was adopted, combining
machine learning experiments with surveys and interviews involving
farmers and agricultural officers. The findings revealed that Maize Lethal
Necrosis (MLN) was the most prevalent disease (41%), followed by Gray
Leaf Spot (33%) and Northern Leaf Blight (26%). Environmental variables
such as humidity and temperature demonstrated strong associations with
disease occurrence. The proposed multimodal CNN–LSTM framework
integrates maize leaf images, environmental sensor data, and farmer
observations, achieving an accuracy of 94.2%, which outperforms
conventional image-only CNN models (87.5%) and environmental-data
based LSTM models (81.3%). Additionally, 78% of farmers reported faster
disease diagnosis using the developed system. The findings demonstrate that
the proposed system supports real-time maize disease detection through an
edge-enabled architecture, enabling deployment on mobile devices and
facilitating practical intelligent system integration in agricultural
environments.
Discover
Top 5 Subjects
- Education 643
- Management 109
- Organization 108
- school of education 104
- Educational planning 93
Top 5 Authors