International Journal of Professional Practice

Institutional Factors Affecting Quality of Documentation of Nursing Care in Selected County Referral Hospitals in Kenya

Anne Mukuna — 2025-07-10

Nurses are responsible for continuous patient care. The proof of care activities is through documentation. Several studies have shown serious shortcomings in nursing care documentation. This study sought to determine the institutional factors that affect the quality of nursing care documentation. It was a descriptive survey, carried out in three County Referral Hospitals; Isiolo, Nyeri and Nyandarua in Kenya. The target population was nurse managers, and nurses in the selected hospitals, and patient case files in the medical surgical units of the sampled hospitals. Multistage technique was used to sample 88 nurses, 6 nurse managers, and 158 patient case files. Data was collected using a questionnaire and key informant guide. Themes and content analysis were used for qualitative data, while quantitative data were analyzed using regression analysis with SPSS (version 26.0). Findings were presented using frequency tables and charts. The results revealed only one-third (35.4%) of nursing care documentation practices were done well. Factors identified to influence nursing care documentation include existence of standard operating procedures on nursing care documentation, a high patient load per shift, and institutional culture on nursing documentation. Regression analysis demonstrated a positive relationship between the institutional factors, presence of SoPs, and institutional culture, with bivariate logistic regression scores of 1.335, 1.133, and 1.026 respectively. This association was not statistically significant, pointing to existence of confounding factors. This implies that improvement efforts must be made to identify and address other key process determinants. The study concludes that improvement of nursing care documentation practice requires identification and address of the multiple factors that affect the nursing practice. Health facility management is recommended to ensure that nurses have access to nursing documentation practices SOPs; to organize CPD sessions on nursing documentation; to build a positive culture on nursing documentation practices; and to adhere to the recommended nurse-patient ratios.

The Influence of Clan Culture on Knowledge Sharing among Commercial Banks’ Employees in Nakuru City, Kenya

Daniel Ongeri Kerandi — 2025-07-10

Knowledge is considered a critical resource that today's organizations can use for competitive advantage. To create organizational value, knowledge should be shared across members of an organization. Despite knowledge being a vital resource, many commercial banks face challenges in sharing their knowledge resources effectively. Organizational culture is both the primary barrier to and the source of empowerment for information sharing. This study aimed to investigate the influence of clan culture on knowledge sharing among employees of selected commercial banks in Nakuru City, Kenya. The Competing Values Framework, and Social Exchange Theory were used to anchor the study. Quantitative research approach was adopted with a positivist worldview of deductive testing. Explanatory and cross-sectional survey design was used, targeting 28 commercial banks in Nakuru City. Based on CBK classification, three large, two medium, and five small banks were purposively selected. Stratified random sampling was then used to select a sample of 178 respondents, who responded to questionnaires administered to them. Both descriptive and inferential statistics were conducted using SPSS version 27.0. Spearman’s rank correlation showed a significant positive relationship between clan culture and knowledge sharing. Binary logistic regression further revealed that clan culture accounts for 56.4% of the variance in knowledge sharing (Nagelkerke R² = 0.564). The findings revealed that clan culture is dominant among employees in commercial banks. Its positive influence on knowledge sharing underscores the need for bank managers to foster a knowledge-sharing culture to enhance organizational learning and competitive advantage. This study's results contribute to the body of knowledge on knowledge sharing in Kenyan commercial banks and associated contexts. Additionally, the study findings will enable managers of commercial banks to learn how to foster a culture of knowledge sharing inside their institutions in order to add value to their organizations and gain a competitive edge in the banking sector.

Development of a Machine Learning-Based Model In Detecting Fake News

Esther Kinkosi Tomba — 2025-07-10

Information through social media and other news outlets made detecting fake news crucial for individuals. The Pew Research Centre conducted surveys in the U.S.A to examine how adults consume news via social media, aiming to understand the behaviours and demographics of those relying on such platforms. This study addressed a critical gap in traditional fake news detection methods, which mainly used manual approaches and lacked advanced machine learning or AI techniques. Traditional methods are insufficient to handle the complexity, and contextual manipulation, where accurate information is presented misleadingly. To overcome these limitations, the study developed a ML Based model for detecting fake news, by analysing article content, and identifying patterns of misinformation. It employed advanced natural language processing techniques and supervised learning algorithms such as Decision Trees with 99.67% of accuracy, Logistic Regression with 99.13%, and Random Forest with 99.15%. Methods like Tokenization and TF-IDF were used to train the model using the ISO Fake news dataset. This dataset included real news from Reuters.com and fake news from unreliable sources flagged by PolitiFact and Wikipedia. Additional labelled datasets like LIAR and FakeNewsNet, along with newly gathered data, were used to supplement the training. Model performance was assessed using accuracy, precision, recall and F1-Score, all achieving 99.67%, demonstrating superior detection capabilities. The research contributed to ML by advancing NLP Techniques and improving fake news detection models. The study recommends future researchers, engineers and all those involved in developing machine learning systems to enhance further effectiveness should expand datasets and including diverse languages, applying deep learning models like RNN, CNN, and Transformers, (e.g., BERT, ROBERTa) for better contextual analysis, and establishing benchmarks using real-world case studies.