A model for new linkages for prior learning assessment

The purpose of this paper is twofold: first the paper aims to sketch the theoretical basis for the use of electronic portfolios for prior learning assessment; second it endeavours to introduce latent semantic analysis (LSA) as a powerful method for the computation of semantic similarity between texts and a basis for a new observation link for prior learning assessment.

A short literature review about e‐assessment was conducted with the result that none of the reviews included new and innovative methods for the assessment of open responses and narrative of learners. On a theoretical basis the connection between e‐portfolio research and research about prior learning assessment is explained based on existing literature. After that, LSA is introduced and several examples from similar educational applications are provided. A model for prior learning assessment on the basis of LSA is presented. A case study at the Open University of The Netherlands is presented and preliminary results are discussed.

A first inspection of the results shows that the similarity measurement that is produced by the system can differentiate between learners who sent in different material and between the learning activities and chapters.

The paper is original because it combines research from natural language processing with very practical educational problems in higher education and technology‐enhanced learning. For faculty members the presented model and technology can help them in the assessment phase in an APL procedure. In addition, the presented model offers a dynamic method for reasoning about prior knowledge in adaptive e‐learning systems.