Recent advancement of information and communication technologies (ICT) promotes its use on educational settings. ICT plays important roles in improving education and learning environments, using learning logs that stored in information technologies. This is one of important advantages of ICT use in educational settings. “Learning analytics” aims to contribute to improve education and learning environments, based on the analytics results from various viewpoints including learning logs analytics. The number of papers and projects about learning analytics research is increasing and increasing, due to its feature that contribute to improve education and learning environments. This paper introduces the recent trends of learning analytics research, in particular, focusing on “Journal of Learning Analytics”, and shows future direction of learning analytics research.

In this paper, we examine the relationship between Institutional Research, or IR for educational planning and Learning Analytics through the intermediation of Educational Engineering. First, we review the study field and practices of IR from the perspective of Educational Engineering. Then, the implication of Learning Analytics for Educational Engineering is discussed. As a result, it is suggested that the objects and methods of Educational Engineering include those of IR and Learning Analytics because Educational Engineering attempts to solve problems in education and its environment by engineered approaches. This means that Educational Engineering has a potential to coordinate IR and Learning Analytics. Similarly, we point out the need of active interdisciplinary effort at the Educational Engineering side to expand a constructive relationship with these emerging academic disciplines.

The next government curriculum guideline, which will be put into effect in sequence from 2020, will emphasize the meaning for carrying out “subjective, interactive, and deep learning (active learning)” and the importance of improving classes through primary/secondary education. Moreover, in high school/university articulation reforms, the university entrance examination system is about to change to many-sided and comprehensive evaluation and judgement. It is thought that the success of these reforms will require the support of learning analytics, which is to accumulate student learning records (e-portfolios) through their learning process, analyze the records as educational data, and visualize the results on a dashboard. Therefore, in this paper, we summarized the current state and problems of learning analytics in primary/secondary education and discuss their perspective.

With the development of information and communication technologies (ICT), ICT has been used in various situations in education. Thus, enormous amounts of data in educational and learning activities are rapidly accumulating regardless of inside or outside the class. This paper introduces the research on learning analytics for the purpose of accumulation and analysis of education and learning process data. In addition, as an example, this paper describes some researches on Learning Analytics at Kyushu University for educational improvement and learning support.

This study aimed to analyze nine years’ worth of sixty thousand free descriptions of course evaluation questionnaires, which originally appeared to be overly difficult to handle in terms of extracting information that had been administered to a certain University Students using a topic model based on Latent Dirichlet Allocation (LDA). We extracted 170 topics and labeled them based on a series of steps, and then checked the validity. Consequently, an adequate level of validity was observed, revealing a comprehensible classification by topic model that fit the human senses. This study further offered analyses linking information about the group of courses to which each free description belonged, and visualized the ratio of topics in each course group, or features of topic distribution of each course group compared to the entire spectrum using cross tabulation. Such analyses are expected to be applied in the fields of Institutional Research (IR) and Learning Analytics (LA) in future studies.

The goal of this research is the development of an efficient method for evaluating the worksheets students submit for each class. To begin with, we scanned the hand-written worksheets students submit at the end of every class, turning them into quantifiable data. Next, we focused on the file size of the quantified worksheets and carried out an analysis using the following steps. First, observing the different file sizes, we separated the students into multiple groups. Second, observing changes to the file size, we conducted a cluster analysis. Third, based on the results of the above analysis, we examined their relation to learning outcomes. Here, learning outcomes constitute: (1) self-evaluation of skills and abilities acquired in the class, (2) self-evaluation of whether or not the class was helpful, (3) grades received for end-of-semester reports. As a result of conducting a quantifiable analysis of the above, we ascertained that the group comprising students with large file sizes tended towards higher learning outcomes compared to other groups. Conversely, we were unable to discover a significant gap between the small and average-sized file groups. Thus, it is possible to evaluate learning outcomes even when using a simple indicator such as file size. We anticipate the findings of this research have the potential for further applicability.

Recent education is required to provide learners the opportunity to acquire the attitude to voluntarily find problems, solve these problems with utilizing acquired knowledge and information and communication technologies(ICTs)and collaborating with others. To provide such educational environment for learners, one of the authors introduced pair programming as a peer learning method and flip teaching in computer simulation practices. The flip teaching consists of video lessons of basic programming practices and pair programming practices in in-class tasks based on the contents of the video lessons. In this study, the authors investigated learning activity of students in this class by analyzing access logs and learning records on a learning management system and questionnaire for the students to determine guidelines for improving the learning environment. The results suggested that deep understanding of basis of programming through video lessons, activities to attempt to deeply understand computer simulation practices in class, and collaboration with other students whose programming skills are similar to the students promotes deep understanding of learning content.

While the field related to educational data analytics such as learning analytics, the importance of institutional research is recognized from the viewpoint of the quality assurance of education and management assistance. In order to do organizational analysis and to support according for individuals, integrating such fields will become increasingly important. In this paper, as a framework to utilize the method of learning analytics for institutional research and student support, we propose a method to model the process of students' learning state about enrollment using Bayesian network. And its applicability is investigated based on the results of some numerical simulation.

We examined the relationship between the academic performance and student’s learning attitude or university admissions system, by analyzing approximately 10,000 student’s learning log data on "Laboratory Exercise in Natural Sciences" for first year university students. The data analysis revealed the following:（1）Every year, a certain proportion student gets a failing grade, and the proportion differs among admissions system. （2）The academic performance and learning attitude are also varied by the admissions system. The better the student’s learning attitude, the academic performance tends to be higher. （3）Score of several reports from the start is relevant to the performance evaluation. Therefore, it suggests the possibility of predicting the final grade evaluation at the initial stage of the lecture.

The necessity of teaching skill improvement by promoting ICT use has often been emphasized. Under this study, we analyzed the learning record data accumulated by teachers, and, through adopting the teacher's reflection that objectify instruction, approached to cultivate Education ability of the career counseling. Specifically, in the high school using cloud platform for 304 日本教育工学会論文誌（Jpn．J．Educ．Technol.） education, we analyzed free form text data about students by performing text mining analysis. In addition, we interviewed the teachers, and showed "the know-how" of the instruction in the career counseling. As a result, through the results of the analysis, teachers confirmed the effect of information sharing and problems in career counseling, and it was clarified that teacher's reflection based on the analysis leads to teaching skill improvement of school from organization viewpoints.

In 2012, the world's top universities started to offer Massive Open Online Courses (MOOCs). The trend became rapidly spread as a global online education platform, and it immediately provoked social interests as research theme. While researches were mostly published as practice reports and discussion papers to look out for future possibilities at the beginning, empirical research has progressed in recent years, by incorporating Learning Analytics. In this article, by reviewing previous studies on MOOC with Learning Analytics, we examined the research findings and discussed issues and challenges for future research.