We have developed the hands-on science educational program, mainly of scientific experiments for elementary school children of forth to fifth grade and their parents. The name is “Oyako de Science: Learning about Science with Parents (in short: OYAKO)”. With OYAKO, a child will learn about the subject “light” for three weeks. The program is composed of: 1) science experiment workshop which child and parent take part in, 2) experiments at home by child alone, 3) experiments at home by child and parent together. In OYAKO, cell phones are used as learning tools. When Children do experiments, they can grasp the procedure of science experiment and after that they can take a short quiz on the cell phones. Also the cell phone is considered as tool to pull out “the engagement of parents in their child’s learning.” Cell phone provides some information about what, when children learn, how well they understand and what kind of advise parents should give to their parents by e-mail. We found that with the participation of parents and thus their engagement in their child’s education, the understanding of the child on concepts improved and that changes occurred in their naive concepts of light.

In this paper, we propose a method of constructing a system assisting a teacher in evaluating learners’ programs by comparing them with an algorithm represented by using abstract data structures. In our previous work, we constructed a system which can evaluate programs by comparing them with an algorithm represented by using operations on data structures provided by a specific programming language. In this paper, we discuss how to improve the system to be able to deal with algorithms represented by using abstract data structures. We propose a method of representing such algorithms and comparing them with learners’ programs. We also examine whether our system can help teachers in actual education experimentally.

The objective of this paper is to describe a design method for instruction in computer programming. One of the goals of instruction is to help students develop their own organization of the concepts associated with a domain so they can apply these ideas to new problems. In computer programming domain, the most effective method is developing some real software to solve the familiar problem. Students can learn a lot of knowledge and skills of program comprehension and system design, developing real software and their own organization of the concepts at a time. I designed and practiced “Recycling Learning Circle” for instruction, utilizing visual programming languages owing to its easy designing for graphical user interface, adopting “Learning by teaching” method, “CSCL” and “Archives”. I report the effectiveness of this method from the reaction of the students in this class and many software output.

In this study, a functionality of freely browsing opinions of other members was added to CSCL system KNITS. In adding the function to the system, two versions were created to represent the relative importance of flexibility versus required cognitive resource in user-interface. Type A seeks greater flexibility, allowing users to choose, read, and rate other users’ opinions without restriction. Type B reduces cognitive load of participants by automatically selecting four neighbors’ opinions in the virtual network. Participants were university students enrolled in an introductory psychology course. Preliminary practice used the original system, the first practice, used Type A. and the second practice used Type B, with 82, 71. and 76 users respectively. Participants’ evaluation revealed that Type B was considered to be easier to use and required less effort than Type A. The results confirmed our claim that it is important to consider cognitive load of participants in designing CSCL.

In learning science, it is sometime difficult to sample subjects randomly from population and to execute experimental study with random assignment. In this paper I introduce a “missing data framework” to deal with bias due to nonrandom assignment/sampling. I review Propensity score adjustment method using covariate information that is useful to achieve sufficient internal and external validity. It is also shown that we can deal with Data fusion problem using the same framework.