In Indiana, US, the Purdue University implemented a so- called Signals system in 2007 to enhance students' suc- cess on
The good example - Learning analytics
Reducing drop-outs at Purdue University
In brief
In Indiana, US, the Purdue University implemented a socalled Signals system in 2007 to enhance students’ success on their road to graduation. The signals aim to help students understand how they can improve their studies in an early phase and hereby avoid a potential drop-out.
The Signal system is a “traffic light” indicator showing how at risk each student may be. It provides early warnings of their study status:
Data becomes “traffic light” The Signal system mines data from e.g. the SIS (student information system) which contains student data such as their performance, prior academic history and their characteristics like the students’ age or points attempted. Then, the system produces a “traffic light” indicator showing how at risk each student may be. Red indicates a high likelihood of being unsuccessful, yellow potential problems, and green a high likelihood of success. The teacher now has the opportunity to intervene and inform the students about their specific signal colour. Results After two years, the results of the Signal system were clear. For students who had been informed of their traffic signal, their grades had improved, and they asked for help 30% more often than those who did not receive their traffic signals. Since the students became aware of their risk level, they tended to change behaviour resulting in improvements to their performance. In general, the students seemed more engaged in their studies and attended more additional tutoring than usually. Also, the signals improved the communication between students and teachers. Critics While considered helpful, the Signal system contains some problematic areas such as maintaining the privacy of teaching performance and ensuring the validity of the data. Also, the Signal-warnings often came too late to be effective and were not specific enough. Therefore, the approach did not help students to change their behaviours adequately. Some staff were concerned to overdo the sending of positive or negative feedback and particularly to give false signals too early, which experiments with the Signal system has not yet given a clear answer on.
• Red indicates a high likelihood of being unsuccessful • Yellow is potential problems • Green shows a high likelihood of success
The Signal system incorporate dynamic data on performance and behaviour. ‘Real-time’ data are considered essential in being able to predict academic success. The signals reduced the students’ feelings of just being “another student”. The students find the traffic lights informative and helpful in changing their behaviour but ask for more detailed information on how to improve their studies.
Contact Pernille Stenkil Hansen 6550 1076
[email protected] SDU Centre for Teaching and Learning Campusvej 55, 5230 Odense M Entrance H by the SDU Swimming Pool - Directory March 2017
SDU CENTRE FOR TEACHING AND LEARNING
