Why Has School Attendance Declined and What Can Schools Do?

Adam Jang-Jones, Ministry of Education

Every day matters

In 2019, the Ministry of Education’s normal yearly monitoring of Term 2 attendance data threw up a surprise. School attendance, edging down since 2015, had dropped substantially. In Term 2 of 2015, 70% of students had been ‘regular attendees’. (This is the Ministry’s main measure of attendance, and is defined as attending more than 90% of possible half-days in the term.) In 2018, that figure had dipped to 64%. Then in 2019, it nose-dived again to 58%. This trend is the same just about everywhere: all regions, all deciles, all school types and all year levels show declines.

At the same time, the latest international survey data from PISA, released in 2018, shows that New Zealand has a comparatively high rate of truancy (where ‘truancy’ is defined as ‘skipping’ a whole school day in the past two full weeks of school): 29% of New Zealand 15-year-olds skip school (where the average in OECD member countries is 21%), and this percentage has been rising steadily since 2012.

A follow-up analysis linking students’ NCEA attainment with their prior attendance data found that even from Year 4 onward, the relationship between attendance and attainment is effectively a straight-line. Essentially, every day matters. Only a small number of absences (about 1.5 days per term) is related to no change in expected attainment, and every additional absence is associated with lower attainment. Of course, Covid-19 has been affecting attendance, but that’s another story.

In terms of how this should translate into actions by schools, we first need to understand:

(1) the drivers of the decline,

(2) the common drivers of non-attendance, and

(3) what good attendance practices should look like in schools.

Drivers of the decline

Although all year levels showed declines in regular attendees, simple analysis shows that the largest decline was in primary school years. During primary years, a child’s attendance is far more reliant on the parent than in later years, suggesting some shift in parental attitudes towards schooling or attendance. Moreover, much of the decline in Term 2 of 2019 was due to a rise in absences recorded by schools as ‘sickness’, despite no unusual disease outbreaks at the time.

Common drivers of non-attendance

A detailed analysis of 2018 PISA data from New Zealand’s 15-year-old students was carried out to understand drivers of non-attendance, using a multivariate regression analysis on the rich array of factors that PISA asks students about. PISA doesn’t include every possible factor affecting a student, though. Submissions to a recent Select Committee inquiry into attendance noted factors that do have proxies in the PISA data (such as a lack of household resources, or poor relationships with teachers, or poor support from whānau) and those that don’t (such as the availability of jobs that don’t need a formal qualification, or complex social and home-life issues affecting some families).

Importantly, some students from resource-poor backgrounds or experiencing low support still attend well, so this analysis isolates the relative importance on skipping behaviours related to each factor in the PISA data while controlling for all the other factors in the model. The analysis found that in New Zealand the strongest predictors of truancy across the board are:

  • lower academic performance – i.e. higher performers are less likely to skip school. This aligns with other research on truancy related to academic struggles; that is, students who feel they are struggling to make adequate progress are more likely to skip school.
  • low valuing of the outcomes of school – i.e. not seeing schooling as important for getting into university or getting a job.

Also associated with high truancy are those students who experience:

  • bullying behaviours – This too aligns with wider research on truancy. Bullying is common in New Zealand compared to most other countries, and PISA data shows bullying crept even higher between 2015 and 2018, accompanied by a decline in students feeling ‘safe’ at school.
  • perceived discrimination by teachers – a small proportion of students experience this.
  • low levels of parental support – a small proportion of students also experience this.

Interesting (though weaker) findings include:

  • higher screen time outside school (for leisure) is related to skipping school.
  • a less disciplined classroom environment is linked to late arrival (though not skipping). PISA data suggests that New Zealand classrooms have been gradually getting noisier and more disorderly.

Similar results were produced from a re-analysis using the slightly different factors available in the 2015 PISA data. Likewise, re-analysis of the 2018 PISA data with those students’ actual Term 2 attendance data matched on found that the main predictive factors of poor attendance (rather than truancy) are:

  • lower academic performance
  • students in low-decile students and Māori students (even after controlling for students’ socioeconomic background and skills, suggestive of a community norms effect or clustering effect)
  • higher screen time outside school
  • a less disciplined classroom environment

Good attendance practices in schools

Scanning international research on attendance interventions, it’s clear there’s no single silver bullet. Addressing non-attendance and truancy requires a holistic approach to disengagement that includes weaving together good-practice attendance monitoring with timely follow-ups with whānau, along with longer-term changes to school practices.

At the level of school practices, it is important to address foundational issues in the school environment. For instance, ensuring a safe environment where the school is clear and consistent in its messaging against bullying or discrimination and deals with these behaviours swiftly and effectively is crucial. Setting clear expectations about classroom behaviour and consequences for poor behaviour means that classroom environments are conducive to focused concentration, thinking and learning, and this helps attendance rates. School leadership also need to ensure teaching stays up to date with relevant research so that it is high impact.

At the level of monitoring, good practices include notifying whānau quickly that their child has been absent (e.g. by automated text message). If a student has been absent more than, say, 5 days in a term (which is 10% of a typical term), good practice is to explore with whānau the underlying reasons for the non-attendance, and to make clear the central importance of attendance for learning and later qualifications and outcomes. Formulating and agreeing on an attendance improvement plan targeted to fit a student’s individual and family circumstances, both within and outside the school, is also useful.

Schools should also take heart that most students who have fallen out of the ‘regular attendees’ category are those who have slipped just a few days below, noting that 90% attendance is equivalent to missing 5 whole days in a typical term. Logically therefore, schools should quickly reap the benefits from focusing first on students who are attending ‘somewhat irregularly’ (e.g. slightly more than 5 days’ absence in the term) and help them improve their attendance by a few extra days in the term. To read more on this topic, see the Ministry of Education’s recently released Attendance and Engagement Strategy.


Adam Jang-Jones is a team leader in the Ministry of Education’s Evidence, Data and Knowledge group. He was previously the National Project Manager for the OECD’s PISA research programme. His research interests include student wellbeing, non-attendance, reading literacy, inquiry learning and environmental education, with a focus on exploring complex relationships within large-scale quantitative datasets and triangulating sources of evidence towards developing actionable insights.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s