Off Bungoma-Chwele Road

sgs@kibu.ac.ke

+254721589365

Dr. Robert Kati

Office Hours: Monday–Friday

8:00 AM – 5:00 PM

An Integrated Framework for Web Api and Linked Data to Improve Data Integration in Academic Digital Repositories

Student’s Name:
Kirui Kibet Leonard

Supervisors:
1. Dr. Roselida Maroko Ongare
2. Dr. Bostley Muyembe Asenahabi

Doctor of Philosophy in Information Technology

ABSTRACT

The exponential digitization of academic content has led to the widespread establishment of institutional repositories. However, in many developing contexts, challenges related to data accessibility and interoperability persist. These limitations hinder the effective integration and utilization of repository systems. This study investigated the role of Web Application Programming Interfaces (Web APIs) and Linked Data technologies in enhancing data integration performance (DIP) within academic digital repositories. Performance was measured through the key dimensions of accessibility and interoperability. The study was guided by four objectives. First, it examined the relationship between Web APIs/Linked Data and data accessibility. Second, it analyzed the combined effect of these technologies on repository performance. Third, it evaluated their influence on data interoperability. Finally, it developed and validated an integrated framework for enhanced data management. Anchored in a pragmatic research paradigm, the study adopted a mixed-methods approach with a descriptive correlational design. Data were collected from two Kenyan public universities using cluster sampling for repository users (n = 80) and purposive sampling for key informants. Instrument reliability was confirmed (Cronbach’s α > 0.70). Data analysis employed Spearman’s and Pearson’s correlation, regression analysis, principal component analysis, and factor analysis. The analysis demonstrated a statistically significant and positive overall impact of Web API and Linked Data technologies on DIP. This was evidenced by consistent, significant correlations across key metrics. Specifically, the technologies had a moderate positive effect on data accessibility (ρ = 0.367, *p* = 0.002) and a significant association with data interoperability. Furthermore, factor analysis validated the study’s theoretical model, explaining over 70% of the total variance and confirming the core constructs essential for integration. These validated findings provided the empirical foundation for developing the Integrated Digital Repository Ecosystem Framework (IDREF). The framework comprises three weighted sub-frameworks: Web APIs and Linked Data (23.2%), Data Accessibility (41.1%), and Data Interoperability (35.7%). Expert validation yielded a high average rating (M = 4.458, SD = 0.39), confirming the framework’s robustness and applicability. The study concludes that Web APIs and Linked Data significantly enhance repository functionality. However, their full impact depends on institutional technical readiness and capacity. It recommends phased implementation of the IDREF framework, targeted training programs to build API and Linked Data competencies, and adoption of hybrid interoperability strategies. Furthermore, the integration of performance metrics such as the DIP index is encouraged to support continuous improvement.