Follow us
Publications Commentary Research People Events News Resources and Videos About IFS
Home Publications September return to school offers a chance to level the playing field

September return to school offers a chance to level the playing field


While most things have changed in 2020, the end of the summer will once again see students preparing to go back to school. But this year’s return to school will be unlike any that has come before; for many students, it will have been more than five months since they last attended school in person.

Reopening schools has been contentious, but the Department for Education in England – partly motivated by research showing the challenges from home learning and the growing inequalities it has brought – has signalled its determination for all students to return come September. It will also reintroduce fines for parents whose children are absent without good cause.

This will be a decisive shift from a period in which schools were open for some year groups, some of the time, with some families choosing to attend while others stayed home.

In this observation, we use data from an online survey of parents with school-aged children – funded by the Nuffield Foundation and collected during June and July 2020 – to document the patchwork of in-person schooling that children had before the summer. We also explore parents’ concerns about sending their children back to school at the end of the last term.

We find that:

  • During June and July, schools offered a patchwork of provision. While two-thirds of parents report their child’s primary school was offering full-day classes and nearly half of primary schools were open five days a week, the majority of parents reported their child’s secondary school was open only one day each week and most offered only half-day lessons.
  • When school attendance was optional, there were large inequalities in who chose to send their children back. Among families whose child had the option of going back to school in June and July, 80% of the richest third of parents sent their child to school, compared to only 64% of the poorest third of parents. This could reflect different perceptions of the relative risks and benefits of returning to the classroom.
  • Among parents who were not willing to send their child back, the most common concern was the perceived health risk to their child, followed by health risks to other family members. Both government and schools should offer parents information and reassurance that the September return to school can be managed safely.
  • Many parents also felt that their child could learn just as well at home. While this might have been true in June and July, when schools were splitting their resources between in-person and online education, a universal return to the classroom is an opportunity for schools to put more focus on the classroom. This is also an opportunity to start to unwind some of the substantial inequalities that have deepened during the past months of home learning.

School attendance was far from universal in June and July

During the lockdown period, schools were asked to stay open for vulnerable children and those in key worker families only. Statistics from the Department for Education show that most schools – around 80% - did so; but they were virtually empty, with fewer than 3% of pupils attending school on any given day.

Since the start of June, some children have also had the option to go back to their classrooms. The government provided guidance on which year groups to prioritise: Reception, Year 1 and Year 6 from the start of June, followed by Years 10 and 12 two weeks later. But otherwise schools had near-total flexibility over how much in-person schooling they provided, what schedule it happened on, and what sort of teaching they delivered at school.

Figure 1 shows that, by the end of June, most schools – 90% of primary schools and 75% of secondary schools – were open to students in these target year groups on Mondays through Thursdays (around 10% fewer schools opened on Fridays). This does not necessarily mean that 10% of schools were not open at all; schools were able to offer different schedules.

But even though most schools were open, their classrooms were noticeably emptier. Only around a third of primary school students in the target year groups, and just one in eight students in Years 10 and 12, attended school on the average schoolday in July.

Figure 1.1 Share of schools open and share of pupils in targeted years groups attending since 1 June 2020

Note: Targeted years at primary school are Reception, Year 1 and Year 6. At secondary school they are Years 10 and 12. Share of schools open refers to the share of schools open to pupils in a target year group (not just vulnerable children and children in key worker families).

Source: Department for Education, “Attendance in education and early years settings during the coronavirus (COVID-19) outbreak.” Accessed 11 August 2020.

A patchwork of in-person provision led to unequal access to education

While the national statistics show that most schools opened to some students at least some of the time, in practice schools offered very different provision in June and July.

We conducted a survey of 1,280 parents with children in Years 1 through 10 between the 26th of June and the 27th of July, asking a series of question about the schooling provision they had been offered in June and July. This is a follow-up to an earlier survey we did in May (find out more here), with the data re-weighted to be representative of English parents. Importantly, since our survey covered a month-long period in which school provision and attendance were increasing, the figures we present here are averages across the month, rather than estimates of the situation at the end of term in mid-July.

In our survey, we find 83% of parents say that their child’s school had reopened to at least some year groups – broadly the same across primary and secondary school parents, and broadly consistent with official statistics.

But while the share of schools that had opened was very similar for primary and secondary students, the provision that schools offered was not. Two-thirds of primary schools were offering full-day classes, compared to only 39% of secondary schools. Figure 2 shows that, compared to secondary schools, primary schools offered more school days as well as longer ones: nearly half of primary schools were open five days a week. The majority of secondary schools were open only one day each week. Primary schools were also slightly more likely to be open every week (80% compared to 75%), while 12% of secondary schools opened less often than every other week.

Figure 1.2. Number of days a week that school is open

Socio-economic differences in school attendance mean a partial reopening is no panacea for inequalities

The offerings of individual schools are only one side of the equation when it comes to school attendance. Even when a school offered in-person provision, it was up to individual families to decide whether to take part.

In a previous wave of our survey, conducted in May, we found that 39% of primary school parents and 45% of those whose child is in secondary school would send their child back to school if given the option. These preferences were very unequal; while 50% of parents of secondary aged children in the 20% richest families would have been happy for their child to go back to school, this fell to 31% of those in the poorest fifth.

In practice, based on our data from June and July, we find that two-thirds of children who had been invited to return to school had been back for at least one day in person. And even among those whose schools remained closed to them, attitudes were softening: 59% of parents were willing to send their child to school if given the choice.

However, while the average level of comfort with a return to in-person schooling has risen, the inequalities we documented earlier remain. As shown in Figure 3, 80% of the richest third of parents who had the opportunity to send their child back to school did so, compared to only 64% of the poorest third of parents. Likewise, for those parents whose child had not been asked back, 62% of better-off parents said they would send their child back to school if they had the chance, compared to 53% of the poorest parents.

It is crucial to stress that these numbers are based on the situation at the end of last term, and our survey does find that parents have become more willing to send their children back to school as time goes on. However, the persistent differences between families in different income groups underline that simply opening the school doors to all children equally will not guarantee that children are equally likely to come through them.

Figure 1.3. Willingness to send child back to school

Note: “Have sent child back to school” shows the percentage of parents who have reported sending their child back to school (among the group of children who have been offered the choice of going back). “Would send child back to school” shows the percentage of parents who say they would have sent their child back to school (among the group of children whose school is not yet open to their year group). Poorest, middle and richest thirds refer to the distribution of pre-pandemic equivalised earnings.

Parents’ preferences

Since July, falling case numbers, changes to government advice, or growing evidence that children do not tend to get seriously ill from COVID-19 might have changed parents’ minds and encouraged higher rates of attendance. Even so, at the end of the last school year, around four in ten parents whose child’s school was not yet open were not willing to send their child back to school. Come the start of the new school year, most of these children will be required to go back to the classroom. What can schools and government do to help ease their concerns?

In our survey, we asked parents whose school was open but did not send their child to school what factors influenced their decision. Figures 4 summarises the share of parents who cited each reason, and the share of parents who cited each reason as the main factor.

Unsurprisingly, the most common reason parents are reluctant to send their children back is the perceived health risk to their child, followed by health risks to other family members. While individual children might have pre-existing conditions that place them at higher risk, there is some evidence suggesting that school-aged children have a low risk of getting serious symptoms should they contract COVID-19, and children – at least those of primary school age – are less likely to transmit the disease (Davies et al., 2020; Park et al., 2020).

Figure 1.4. Reasons for not sending child back to school in June/July

In addition to the risks of going back into the classroom, many parents also say that they perceive little benefit from returning to school. Over a third of parents say that their child can learn just as well at home, while others point out that a return to school has little benefit to their families (either because a child’s siblings are still at home, or because the schedule offered by the school does not meet the family’s needs).

All of these factors could have changed over the summer. The latter two reasons should disappear with the return to full-time school for all children in September, and even families saying that their child can learn just as well at home might change their minds. At the time of our survey, some children, especially in better-off families, had extensive programmes of online lessons and teacher support (as well as the resources at home to make use of them). At other schools, in-person provision in June and July consisted of doing distance learning from the school. But a universal return to school could see schools shift their focus away from delivering online learning and back towards education in person, in the classroom.


The partial re-opening of schools from June resulted in patchwork education provision across England, with some children being offered significantly more comprehensive packages than others. Provision in secondary school was particularly patchy, with many older pupils limited to partial days once every few weeks. Policymakers and educators should be wary of placing too much weight on the amount of learning that occurred from June till the end of term.

We find that richer parents are more likely to have sent their child to school already (if the school is open), and more willing to send their child back in the future (if they haven’t yet had the choice). This evidence also suggests that the partial re-opening in the final months of the 2020 school year did little to address the previously identified gaps in learning time between the most and least deprived children.

Taken together, this also means that an optional system, where schools are given substantial flexibility in what they offer and parents have the choice to opt in or out, can lead to very unequal outcomes. However, policymakers and schools should also be sensitive to parents’ concerns ahead of a compulsory return to school. While some of the major concerns will be addressed by a more general return to school, our data clearly show that there is scope to do more to inform and reassure parents about the health risks to their children and their families.