Data Insight: From school absences to crime involvement

This Data Insight by ADR UK fellow Dr David Buil Gil and Prof Ken Pease explores the relationship between school absenteeism and subsequent patterns of criminal activity. It addresses the question ‘What are the long-term effects of school absenteeism on crime at different stages of life?’ Specifically, it examines whether authorised and unauthorised absences are associated with future cautions and convictions.

Read publication

To investigate this, we used linked data from the Department for Education (DfE) and the Ministry of Justice (MoJ), covering 4,926,067 pupils born between 1990/91 and 1997/98 in England. School absences data (2005–2013) and cautions and convictions (2005–2022) are analysed using descriptive statistics and survival analysis.

Our findings reveal that pupils with higher levels of recorded absences are more likely to be involved in cautions and convictions later in life. The association is particularly strong for unauthorised absences, which increase the risk of first-time offending by 0.5% per session missed. This corresponds to a 9% higher risk of cautions and convictions for pupils with 20 unauthorised absences and a 26% higher risk for those with 50. Authorised absences also show a slightly smaller but still substantial and significant effect. These patterns persist over the life course, with risk rising consistently at higher levels of absence.

What we found

Authorised absences were nearly universal, recorded for 99.2% of pupils (4,887,321 out of 4,926,067) (Table 1). Unauthorised absences were also widespread, with 66.4% (3,271,283 pupils) having at least one unauthorised session. On average, pupils had 97.5 authorised absences (median = 61) and 19.3 unauthorised absences (median = 3). The maximum recorded for a single pupil exceeded 1,500 authorised and 1,400 unauthorised absences. Overall, 717,495 pupils (14.6%) had at least one recorded caution or conviction by age 31. The mean number of cautions and convictions was 1.6; the median was 0; the maximum was over 350.

Figure 1 shows the proportion of pupils with cautions and convictions by absence quintile.

• By age 15, 2% of pupils in the lowest authorised absence quintile had a caution or conviction (19,630 out of 985,214), compared with nearly 15% in the highest (151,361 out of 985,213).
• By age 30, these figures were around 7% (68,560) and 27% (270,682), respectively.

The pattern is even more pronounced for unauthorised absences:

• By age 18, 4% of pupils in the lowest quintile (42,609) had a caution or conviction, versus 27% in the highest (270,590).

By age 30, these figures rose to 7% (72,401) and 33% (333,151), respectively.

The Cox proportional hazards model shows that each additional unauthorised absence increases the risk of first-time cautions or convictions by 0.5% (hazard ratio [HR] = 1.0046, p < 0.001). This equates to a 9% higher risk for pupils with 20 unauthorised absences and 26% for those with 50. Authorised absences also raise the risk, with a 0.4% increase per session (p < 0.001). The same 0.4% increase applies when combining both authorised and unauthorised absences.

Figure 2 presents survival curves showing the probability of no recorded crime over time by level of unauthorised absences. After 16 years:

• 92% of pupils with no unauthorised absences remained crime-free (1,547,741 out of 1,681,818)
• 89% for those with 1–5 absences (1,154,695 of 1,300,851)
• 85% for 6–10 (438,576 of 516,613)
• 81% for 11–20 (397,283 of 487,701)
• 78% for 21–30 (195,288 of 251,038)
• 74% for 31–50 (187,438 of 253,472)
• 66% for over 50 absences (288,174 of 434,574)

Similar trends, though less pronounced, are observed for authorised absences.

Why it matters

Understanding the long-term implications of school absenteeism is vital for designing more effective early intervention policies. This research offers evidence that higher levels of school absences, particularly unauthorised absences, are strongly associated with increased risk of later criminal involvement, even after many years. These findings underscore the importance of viewing absenteeism not just as an educational issue, but as a cross-sectoral concern with implications for criminal justice, social care, and community safety.

The public value of this research lies in its potential to support early identification of risk and inform targeted interventions. By identifying patterns that link absenteeism with future offending, policymakers and practitioners can better allocate resources to prevent long-term harm. This is especially relevant for supporting vulnerable groups, reducing educational and social inequalities, and informing cross-departmental strategies that span education, justice, and public health.