The improvement of an e-learning system for library classification based on the analysis of incorrect answers given by students

Abstract

The purpose of the study described here is to develop an e-learning system called NDC-trainer to be used for library classification training. This system gives LIS students the ability to practice creating proper book classification numbers by considering a wide range of information, including content descriptions or book outlines available on the web. With this approach, students respond to various questions stored in the system. These questions are displayed using bibliographic data from the National Diet Library Database. If the classification number created by a student is incorrect, the system gives an immediate indicator and displays comments based on patterns in the student’s errors. In this paper, we describe the exercise of constructing a classification number by combining the class number for a topic with an entry from the auxiliary tables of NDC. A group of twenty junior and senior students in the Library and Information Science Program used the NDC-trainer for practice and responded to 31 questions. According to their comments, the system is usable and even enjoyable. Three distinctive features—having a large number of sample books, displaying links and rich content, and showing references immediately—are especially acclaimed. They also pointed out weaknesses in the current system and provide a roadmap for continuing improvements.