Artificial intelligence (AI) can be used to improve pupil attendance at schools, according to a new study by researchers at De Montfort University Leicester (DMU).
Experts from the Institute of AI at DMU analysed Milton Keynes-based Willen Primary School’s attendance data using AI models to identify patterns relating to pupils who were frequently absent.
Willen Primary School in Milton Keynes
They found that Monday morning was the most common time for absenteeism at the school, with potential reasons including separation anxiety, taking extended weekend breaks and a lack of motivation to attend school after time off.
To improve attendance, as it is so closely linked with pupil performance and outcomes in later life, the researchers developed a two-pronged approach.
Firstly, they increased the frequency and value of rewards given for full attendance, holding monthly raffles with prizes for those children who had not missed a day of school during that month.
Secondly, they introduced a ‘Monday Matters’ initiative, giving pupils fun activities to look forward to on their first day back after the weekend.
Thanks to these measures, the school achieved the required national average attendance of 96% for the first time in four years. They also saw a huge improvement in persistent absenteeism, which improved by more than 55% compared with a year ago.
Dr Raymond Moodley, a visiting researcher at DMU’s Institute of AI, was leading the project along with his colleagues Professor Francisco Chiclana, Dr Fabio Caraffini, and Dr Mario Gongora.
Dr Raymond Moodley is a visiting researcher at DMU
Dr Moodley said: “By using AI, we were able to pinpoint the problem areas for attendance at Willen Primary School, which in this case was Monday mornings.
“One of the key changes we wanted to make was to increase the incentives on offer for pupils who have a 100% attendance record over a shorter timeframe. Schools typically tend to reward children on an ‘all or nothing’ basis by only recognising pupils who have full attendance for the entire year.
“However, our approach sets shorter-term goals for the children which makes it more achievable, and if they fail to achieve full attendance in one month, then they can always try again the following month.”
Research shows 25% of Britain could have avoided lockdown during COVID-19 crisis
Find out more about studying Artificial Intelligence at DMU
Research project that aims to advance neuroscience gets go ahead for €150 million final phase
Pupils that are persistently absent are typically shown to have the weakest performance at school. Persistent absenteeism is also higher for those children that receive free school meals.
Carrie Matthews, headteacher at Willen Primary School, said: “The novel approach provided by the team at the Institute of AI allowed us to improve our understanding of our attendance and put impactful plans in place to reduce absenteeism. The results of our collaboration have been fantastic!”
Dr Moodley and his fellow researchers are now developing an easy-to-use AI-enabled absenteeism diagnostic tool that will eventually be available for schools across the UK to download.
Schools will be able to upload their own attendance data from their attendance recording systems and gain insights into their attendance using the AI models.
In 2018/19, the North East and Yorkshire and Humber regions had the highest rates of overall absence at 5%, while the region with the highest persistent absence rate was the North East at 12.1%.
“Attendance is just one parameter measured by Ofsted when it comes to a school’s overall performance,” continued Dr Moodley.
“We also want to look at other factors that impact pupil performance and outcomes in later life. These include motivations for choosing subjects – particularly girls in STEM (Science, Technology, Engineering, and Mathematics) – as well as behavioural management, and optimising the learning environment (classroom layout, timetable design etc.)
“AI is tremendously powerful, and we want to harness this power to maximise the overall performance of every child in every school.”
Dr Moodley and his colleagues, who are members of a research interest group called RiSE (Research in Societal Enhancement), are currently engaged in a number of AI-led projects in areas including crime prevention, agricultural sustainability, alternative solutions to managing pandemics, and medicine.
For more information about the Institute of AI at DMU visit www.dmu.ac.uk/research/centres-institutes/iai
Posted on Tuesday 28th July 2020