In early 2020, parents from 22 Bristol schools whose children had been regularly absent received a text message. Sent by the school, the short message notified them of how many days of school their child had missed, reminding them the new term approaching represented an opportunity to improve attendance. The response to the effort to tackle absenteeism was striking: the proportion of students keeping good attendance records increased by four percentage points to 95 per cent.
The experiment relied on what is known as ‘nudge theory’: the idea that an individual’s choices can be affected by the way their environment is shaped. It is an approach, already shaping various areas of public policy, that a growing body of evidence and expert opinion suggests could tackle absenteeism, according to Lal Chadeesingh, an education specialist at the Behavioural Insights Team.
A 2018 study from Harvard University, for instance, found that sending letters to parents telling them how many days of school their child had missed, while stating the importance of school attendance and their ability to influence it, reduced chronic absenteeism (missing 18 or more days of school) by more than 10 per cent.
Two years on from the Bristol trial, absenteeism remains a problem. From state to private schools, grammar to foundation, a significant number of students are missing lessons. It’s a pronounced problem that has clear negative consequences. It’s also getting worse.
Indeed, SIMS’ report, Engaging on Absenteeism, which was conducted in August this year, found that almost 60 per cent of UK teachers think absenteeism has worsened since the pandemic. These are not isolated findings: the Department for Education’s latest figures for October reveal that persistent absence has increased by more than 10 per cent when compared with 2018/19.
The difficulty is spotting those at risk
Absence has no single cause. Dr Raymond Moodley, an expert on the application of AI to education, states that absence is often, at root, a social problem. Children who refuse school or play truant either do so because they consider other activities to be more important, or to avoid negative elements of the school day, such as embarrassment or anxiety. The SIMS report, meanwhile, found that a significant cause of absence was parents withdrawing their children from school, be it for a holiday or religious festival.
Whatever the cause, there is a common solution to the problem: improved parental engagement. Indeed, SIMS’ research found that 60 per cent of UK teachers believe that greater parental engagement is the key to tackling absenteeism. What is needed is the ability to identify those at risk of absence – or those already falling into a pattern – and to apply nudge techniques in early interventions. The difficulty, though, is spotting those at risk.
Identifying patterns of absence is a tricky task. For instance, it is easy enough to spot the child who is regularly absent on a Monday, but what about the student who is regularly absent at particular times of day, or every third week? When attendance involves multiple teachers across different subjects, patterns can elude even the most eagle-eyed.
Registers provide a treasure trove of information, but it’s very difficult for a human – especially time-pressed teachers – to quickly spot patterns. Algorithmic systems, already being developed, can quickly and accurately comb through oceans of data to identify those children regularly absent and those who are displaying the warning signs. They can even allow schools to predict when absence might increase: a video game release, for instance, or the start of the summer weather.
This allows interventions to be staged early and to great effect. And it’s not just a case of using AI to identify those at risk of absence. Rather, algorithmic systems can also be deployed to analyse communication to establish which form – text, letter or email – is most likely to generate a response.
Edtech looks set to explode in the years ahead: research from Arizton suggests the market value could grow significantly over the next five years. The upshot is that, in a digital world, it is perhaps inevitable that AI systems will be applied to tackle absence. When combined with nudge theory, we could positively affect attendance.
Your thoughts