This event has ended. Visit the official site or create your own event on Sched.
Welcome to iCTLT 2016!
Wednesday, March 30 • 3:00pm - 3:30pm
e-Poster Exhibition: 246 Predicting Performance using Smart Data from E-Learning for Timely Intervention

Log in to save this to your schedule, view media, leave feedback and see who's attending!

Teachers have the task of ploughing through data of the class(es) they teach in efforts to create for their pupils a “personalised education program designed to maximise education outcomes.”

“They (teachers) have always been data workers – assessing students’ understanding of the material based on test scores, classroom engagement, quality of homework, etc., with the goal of improving that understanding” (Olavsrud, T).

In Wellington, aside from monitoring pupils’ daily assignments and their class participation rates, teachers often race against time at the end of an exam in each term collating data for Item Analysis purposes for each subject in order to identify questions that pupils have performed weakly on so as to remediate after the exam. The problem with this is that data gathered from such analysis provides information only after the exam. Remediation helps the pupils only in the next exam.

What is needed is a faster and more effective means of identifying weak topics so that intervention can be done earlier to help pupils’ scores prior to the exams in each term. Online resources have largely been an under-utilised method of data gathering for this purpose. According to Putnam & Borko, “…multimedia systems, with their new and flexible ways of representing and connecting information, can enable teachers to explore unfamiliar pedagogical practices and various problems of pedagogy.”

Hence, we did a study with the purpose to find a platform that enables teachers to identify topics that their pupils are weak in so that they can conduct intervention and remediation to improve their pupils’ understanding of these topics before their exam.

As such, the intervention meted by team were using online competitions and personalised e-learning activities to encourage pupils to do more Mathematics practices and measuring monthly participation and proficiency report to monitor the performance of each class and identify pupils’ weak topics.

We thus, Welington uses e-learning (Koobits) results to predict the performance can be used effectively and is a new opportunity for teachers to discover problems earlier and intervene to help students.

To illustrate, in the case of class 4D, the snapshot of e-learning results at the end of March 2015 showed a 90% correlation to the upcoming performance of pupils for SA1 in the next month. Combined with the matrices that will be shared with participants, the prediction accuracy was up to 95%. Do come and join in the presentation to be more enlightened.



Wellington Primary School


Wellington Primary School


Wellington Primary School
avatar for Mr. Roslee bin Jalie

Mr. Roslee bin Jalie

Roslee Bin Jalie is the HOD ICT of Wellington Primary School and has been teaching for twenty years. His work in the area of Flipped Classroom started in early 2012 but he has gone on to share on various local and international platform. He uses a variety of tools along such as the... Read More →


Wellington Primary

Wednesday March 30, 2016 3:00pm - 3:30pm GMT+08
FutureSpace Suntec City Convention Hall