Over the years, large record of student data exists in Institutions of higher learning because students graduate from these institutions yearly. This has necessitated the need to explore those data to discover some patterns and relationships existing in them and as well make strategic decisions for a better education system. This paper focused on developing an enhanced Hybrid data mining model to mine students’ progress and performance for knowledge discovery and for decision making purposes. This was implemented using Clustering Algorithms (k-means and k-representative) as a tool in data mining. The hybrid model was efficient to cluster the mixed dataset of student’s records. Object-Oriented and Analysis Design Methodology was adopted for the design, Java programming and NetBeans Integrated Development Environment (IDE) was used for the interfaces and Matlab editor respectively. The result shows that the proposed hybrid clustering algorithm improved K-means clustering algorithm for optimal solution and efficient clustering of mixed data sets with 99% performance and clustering error of 0.0025. Academic planners should implement the result of this study and as well make strategic decisions for a better education system.