This guest blog post introduces a new series of country reports from scholars renowned for their contribution to national and international learning analytics research. Professor Xiaoqing Gu of East China Normal University, Shanghai, gives her perspectives on this fast developing field of interest.
Everyone is talking about big data, at the same time we academics are talking about learning analytics. I guess this is not only the case in China but also worldwide. From an educational research perspective, we try to correct the misunderstanding of big data, maintaining they are actually not big data, they are just mass data. At the same time, we ourselves are frequently put right by others that our learning analytics usually are not “real learning analytics”. Currently, it seems that learning analytics is a rather idiosyncratic phenomenon; if you dare to promote a specific point of view you must be prepared for criticism.
My team and I are now doing a study with a twofold aim: First, we want to come up with an innovative and ICT rich learning design in support of teaching and learning; second, we want to use learning analytics to support educational diagnosis and learning improvement. So, we use learning analytics in a broad way, which includes the analysis of data both from online learning environments and blended learning. This may not be ‘genuine learning analytics’ to those who believe that it should only be confined to data from online learning.
We believe that the purpose of the learning analytics is to obtain in-depth understanding of students’ learning behaviours and performance in order to foster data-driven education improvement. So it follows that the data created by students’ learning behaviours and performances can be in different shape and can be tracked in different ways.
Here is what my team is doing in the name of learning analytics: we cover the range of learning analytics across K12 to higher education and from school-based to online learning, while we collect data for the purpose of analysis automatically via the e-learning platform as well as semi-automatically via data collecting tools.
In the context of K12, students’ behaviours both inside and outside the classroom are considered as the source of analysis, to understand how these behaviours are related to their learning performance. For example, in one of our studies, students’ classroom behaviours are measured by employing carefully designed iPad software and using iPad as a data collecting tool, allowing the teachers to catch the key performance behaviours in the classroom. By setting the “attention” rules in the system, the classroom teachers are informed when specific behaviours occur that indicates a key performance of students.
In the context of higher education, we focus the data collection from an e-learning and MOOCs platform. For example, in one of the studies, we use learning analytics methods to provide timely intervention to students who are learning in the Sakai learning platform. A Learning Analytics Intervention Model (LAIM) was established with 14 variables correlated to students’ performance.
In the informal learning context, we are carrying out a learning analytics study in a large-scale online teacher training project, which is a normal practice for secondary teachers in China nowadays. In this case the purpose of the learning analytics is to address the quality issues, which motivated the development of learning analytics in the first place.
This is our understanding of learning analytics and our practice so far in this field. We are prepared to discuss, learn and contribute.
View Twitter conversations and metrics using [Topsy]