The Comparison of Learning Analytics Tools
Lingling MENG,Xiaoqing GU,Ze Ll
DOI: https://doi.org/10.3969/j.issn.1007-2179.2014.04.009
2014-01-01
Abstract:In recent years, with the rapid development of a smart learning environment, massive, rich, diverse, and heterogeneous data are increasing amazingly. In education field, students’ interests, preferences, activities, learning process information, such as the interaction with learning platform, as well as their implicit feedback to the e-learning platform, can all be recorded and traced. How to effectively make use of these data has drawn great concern. The da-ta of a single person seems to be chaotic, but with the data accumulating to a certain extent, it will be presented in an order. There are strong or weak relations among the data. For example, what are the characteristics of students in dif-ferent region or countries? What are the characteristics of learning behavior in different ages? What are the learning habits of different students? Which courses are needed urgently for a successful career? For a special course, which u-nits are needed for review? Which units are needed to be emphasized? Which students encounter difficulties and need help? Therefore there are amazing insights behind the data. If we extract the rules or determine the relationships among data, tremendous value will be created. Therefore, learning analysis techniques arise. According to the Horizon Report 2011 in the New Media Consortiums Horizon Project, learning analytics technol-ogy will become a hot topic in the next few years. It will contribute to improving the learning process and make the learning more intelligent. As we can imagine, the analytics tools play an important role in the process of learning an-alytics. Good tools can make the research process more effective. Many analytics tools have been developed. For example, Nvivo, Atlas. ti can be used to annotate the text and multimedia content. Gephi, JUNG, Guess can be used to analyze learning networks, and SPSS can analyze user data statistics. However, a key issue is how to choose the appropriate tool because different tools will have different func-tions. Previous studies conducted analysis on the analytic tools, which are fragmented and focused on the comparison of learning network analytics tools. Different from previous works, this paper categorizes the learning analytics tools from multiple perspectives, and compares the tools from the point of developer, application environment, users, the supported data format, visualization, advantages, disadvantages, and so on. Twenty-four tools are discussed in de-tail in order to provide proper technical support for learning analytics and help to make education more intelligent and effectively.