This paper proposes a model of predicting EFL learners’ sentence reading time based on learners’ reading ability and linguistic features of a sentence. The reading ability here means the learners’ TOEIC (Test of English for International Communications) scores, and the linguistic features indicate lexical, syntactic and discourse complexities. We use natural language processing technology to automatically extract the linguistic features from the sentences, and construct the model by using the Support Vector Machines as a learning mechanism to combine the learners’ and the linguistic features.   An experiment has shown that the proposed model predicts reading time with an error rate of 18.8%, which is lower than the error rates marked by other models.

This paper describes an environment for mechanics learning by designing situations from formulas. In learning of mechanics of high school, students are usually required to derive mechanics formulas from mechanics situations. However, to derive the situations from the formulas is also important practice in order to realize deeper understanding of mechanics. In this paper, firstly, we have confirmed that students who can derive formulas from situations are not always able to derive situations from formulas through an experiment. Then, the design and implementation of an environment for mechanics learning by designing situations from formulas are explained. In the system, a student is required to design a situation from a formula, and then, derive another formula from the situation. The system diagnoses the derived formulas and situations, and gives the student the results of the diagnosis to assist error corrections. Experimental evaluation of this system is described, too.

In this study, we propose an evidence based approach to improve web-based teaching materials based on access logs of students using the materials. In this approach, the web-based teaching materials are evaluated based on the following data: (1) students' learning pace, (2) their activities of referring help items, and (3) students' questions to teaching assistants. We evaluated the utility of this approach by using the data in university classes for freshmen. As a result, we confirmed the following points: the improvement of the materials based on this approach actually reduced the number of questions to the teaching assistants, and also reduced and the number of slower-paced students.

This paper describes how to redesign a drill-and-practice-typed e-learning system SANNO KNOWLEDGE FIELD (SKF) and its contents in order to bring them into compliance with SCORM1.2/2004 international standard. At first, SKF had been independently developed as ASP-typed WBT and its contents had been also implemented on it with the unique specification. Such specification allowed the system to be most suitable and gave an advantage in not causing any compatibility problem. However, there was difficulty in providing the contents for the other e-learning systems. Redesigning the contents in compliance with the SCORM standard enables them to run on various types of LMSs in worldwide. In this paper, we show the way to adapt the contents to the SCORM standard and challenges in implementing them when the redesigned contents complying with the SCORM standard realizing the same learning mechanism as the existing LMS had.

With the spread of the Internet, information communication technology, primarily the Internet itself, is increasingly being used in education. Under these circumstances, it is necessary to compare learning assistance methods with common indexes, and ascertain their characteristics in order to enable teachers to select Web-based learning assistance methods that correspond to their class goals. This research examined three types of Web-based learning assistance method: drill format Web quizzes (“drills”), Web course materials based on PowerPoint lecture slides (“slides”) and student submission of test questions prepared as assignments to an electronic bulletin board (“bulletin board”). The three types of learning assistance method were then compared by investigating the relationship between the rates of information usage on each learning assistance system with test scores. As a result, it was found that a common characteristic of the methods was that groups with low test scores had low rates of usage of the learning assistance methods, regardless of the learning assistance method. Further, the “drills” and “bulletin board” exhibited a correlation between the rates of usage of the learning assistance and test scores; in many classes, groups with high rates of learning assistance usage had high test scores. However, it was found that “slides” did not exhibit this sort of tendency.

The author discusses the significance of qualitative research methodology in educational technology research. For that purpose, he first describes what qualitative research is, showing its characteristics, process, differences from quantitative research, crucial points for its understanding and practice, generalizability and research evaluation. Then he discusses its significance in educational technology research, in terms of from technology-push to requirement-pull, from research for the research article’s own to effective system development, from product-focused to process-focused, necessity of goal-free evaluation, research on teachers who are the gate keepers of educational technology, and educational research through educational technology research.