Briefing No. 1 - Lockdown Learning

Context

The COVID-19 pandemic and the restrictions that followed dramatically changed the experiences of schooling for young people in England. A national suspension of in-person schooling began in March 2020 and was intermittent until March 2021, covering two school years, with continuing disruption beyond this point. While some pupils, including those with ‘key worker’ parents, were eligible to attend school during these periods, the vast majority embarked on an unprecedented programme of remote learning. The speed and scale of these changes put immense pressure on schools, teachers, pupils and parents, with parents becoming the main facilitators of learning during the initial school closure periods, particularly for younger children.

A variety of studies conducted during 2020 and 2021 examined experiences of remote learning, including the type of learning on offer from schools,¹ whether pupils had access to suitable devices,² and the level of support pupils received for learning at home.³ The shift had an instant impact on learning, with one study finding that the average time spent on learning decreased for primary and post-primary pupils in the UK from 6.3 hours per day pre-lockdown to 4.47 hours during the first lockdown.⁴

Disparities in home learning also became clear, particularly for those from lower socio-economic backgrounds. Schools with more deprived intakes reported that they were less likely to offer ‘live’ online classes, as well as having higher levels of pupils lacking suitable devices for such learning, during both periods in which in-person schooling was suspended.⁵ Differences in time pupils spent on learning during the pandemic according to socio-economic background were also reported in various studies. Pupils from advantaged backgrounds spent more time per day on learning with 44% of middle-class pupils spending 4 or more hours on learning per day compared to 33% of working-class pupils;⁶ pupils with graduate parents worked 0.2 more days per week during the first lockdown than those without; and pupils eligible for free school meals (FSM)⁷ spent 0.29 fewer days per week than non-FSM eligible pupils.⁸ The impact of socio-economic background on time spent studying was mediated both by resources provided by school (online lessons) and at home (devices and dedicated study space.⁹ According to several studies, males completed less schoolwork and learning activities during school closures than female pupils.¹⁰ However, there was little evidence of differences by ethnicity in time spent learning.¹¹

As a result of differences in experiences, an emerging body of evidence is showing that pupils from disadvantaged backgrounds faced greater negative consequences from school closures. 53% of teachers in the most deprived schools reported their belief that pupils were at least 4 months behind in curriculum learning compared to 15% of teachers in the least deprived schools.¹² Pupils with university graduate parents were 14 percentage points more likely to receive higher Centre Assessed Grades (CAGs) than their calculated grades compared to pupils with non-graduate parents, even when accounting for prior attainment at Key Stage 2 and 4 and demographic characteristics.¹³ Furthermore, the first national evidence on the disadvantage attainment gap at the end of primary school showed that a decade of progress in narrowing the gap had reversed between 2019 and 2022.¹⁴

Remote and in-school learning during lockdowns

In-person attendance

The COSMO cohort were in Year 10 when the pandemic first hit. Suspension of in-person schooling in the first national lockdown lasted from 20 March 2020 until summer, with a partial reopening for secondary schools from 15 June for crucial year groups, including this one. Schools fully reopened at the beginning of Year 11, and remained open during the second lockdown in October 2020, but in-person schooling was suspended again during the third national lockdown, from 4 January 2021 until 8 March.

Most of the cohort did not attend school during either of the periods of school closures. Attendance rose from 14% in the first lockdown to 17% in lockdown 3. The biggest increase was in the category reporting ‘other’ reasons for school attendance, with the criteria widening, including lack of access to home learning. 23% of those without access to a device attended school in lockdown 1, compared to 29% in lockdown 3.

Boys were more likely than girls to attend school in-person during both lockdowns. Those at state comprehensive schools were more likely to attend in-person than those at other school types. Among comprehensive schools, those with the most deprived intakes saw higher attendance rates (for example, 23% at the most deprived schools during lockdown 3, compared to 14% at the least deprived).

This is borne out when looking at household characteristics, with children of working-class parents more likely to attend than those with professional and managerial parents. Single parent families and those in social housing were also more likely to attend.

A significant minority of young people (27%) returned to in-person schooling early when some schools reopened in summer 2020, but are not considered lockdown attendees in other analyses here. More advantaged schools were more likely to reopen during this period, with attendance highest at grammar schools (33%), and comprehensives with the least deprived intakes (35%), compared to 24% of comprehensives with the most deprived intakes.

School provision of remote learning

Provision of live online lessons from schools reported by young people increased from lockdown 1 (66%) to lockdown 3 (87%). While almost all of the independent sector was providing live lessons in lockdown 1, lockdown 3 saw the gap between state and private narrow. However, in turn, gaps opened up within the state sector. As Figure 3 shows, there were virtually no differences in the rate of live lessons during the first lockdown between state schools. However, by lockdown 3 grammar schools and comprehensive schools with the least deprived intakes offered live lessons at a similar rate to the private sector, while those state schools with more deprived intakes remained behind.

Students also reported more pastoral contact with teachers in lockdown 3 compared to lockdown 1, with the number of pupils having contact with a teacher outside of class once a week or more rising from 45% to 53%. Students reporting that they had problems accessing help from teachers accordingly decreased from 21% to 16%.

Pupils at independent schools were most likely to be in touch with a teacher outside of class at least once a week during both periods of lockdown, with state schools catching up slightly in lockdown 3. Nonetheless, comprehensive pupils were more than twice as likely to report difficulties in access to support from teachers compared to independent schools (17% vs. 8% in lockdown 3).

While pupils at the comprehensive schools with the most deprived intakes were actually most likely to receive regular contact in lockdown 1 – potentially due to the intensive non-teaching support provided by some schools to low-income families – less deprived schools caught up in lockdown 3.

Young carers had a greater need for support during lockdowns, yet they had a similar frequency of teacher contact to non-carers. As such, they were more likely to report that they had problems accessing support. Looking at household characteristics, there were small gaps in teacher contact during lockdown 1 by social class and by parental education, however, these gaps widened in lockdown 3, likely due to the school-level changes described above.

Time spent doing schoolwork

While around half of young people worked 5 days in a typical week during lockdown, a substantial number were working less than this (46% in lockdown 1 and 35% in lockdown 3). Most dramatically, 9% reported not doing any schoolwork in a typical week in lockdown 1, reducing somewhat to 6% in lockdown 3.

Of those who worked at least one day a week, the majority worked at least 3 hours per day: 60% in lockdown 1, rising to 75% in lockdown 3.

Figure 7 reports estimates of the typical hours young people spent doing schoolwork in a 5-day week during each lockdown by background characteristics. While the number of hours increased in all groups between lockdown 1 and lockdown 3 (from 13.3 hours to 16.8 hours), there were differences between groups. Girls tended to work longer hours than boys, reflecting existing evidence. Independent school pupils tended to work longer than those in state schools (both grammar schools and comprehensives), while among state comprehensive schools, those attending schools with the least deprived intakes worked the most hours.

There was little variation by ethnic group in either lockdown. However, young people who reported that they had caring responsibilities worked fewer hours than those who didn’t. Looking at household characteristics, those with professional and managerial parents worked longer hours (over 15 hours in lockdown 1) than those with routine/ manual parents (12 hours). Those whose parents had partners and those whose parents owned their own home worked longer hours than those in single parent households or socially renting households. This may have reflected easier home environments in which young people could work (for example more space).

Looking particularly at those who weren’t attending school in person, how were they engaging with online learning? The proportion of schools not offering remote live lessons plummeted between the two periods of school closure, with an according rise in the number of online classes taken by home learners: the number of pupils attending five or more such classes per day rising from 15% to 29%

However, patterns across groups differed between lockdowns. In lockdown 1 independent schools were far ahead, owing to their higher levels of provision. State grammar schools almost caught up in lockdown 3, with state comprehensive schools also increasing attendance, but not at the same rate. There were no clear patterns in lockdown 1 among state comprehensive schools, however during lockdown 3, online attendance at the schools with least deprived intakes improved substantially more than at those with the most deprived.

This is reflected in patterns by household characteristics, too, with Figure 10 showing that there were few differences in the number of young people attending three or more online classes per day in lockdown 1, but a gap opening up in lockdown 3. This adds to a consistent pattern of established advantages being consolidated in the second period of school closures.

Access to devices for home learning

For those learning at home, having a suitable device, such as a laptop or tablet, was key to being able to access and engage with remote learning content being provided by schools, including online lessons. While most young people of this age had such a device, this was not universal. 86% of young people reported having a ‘suitable device’ in lockdown 1, rising to 90% by lockdown 3. Those using a mobile phone dropped from 11% to 8%, but those with no access at all was unchanged at 3% in lockdown 1 and 3.

Young people having no device or just a mobile phone was socio-economically patterned. Almost a quarter (24%) of those attending state comprehensives with the most deprived intakes did not have access to a suitable device in lockdown 1, compared to 6% at the least deprived state schools, and less than 2% at independent schools. While the proportion of young people without a suitable device decreased between lockdowns, in the most deprived comprehensive schools, 19% of pupils still lacked a computer or tablet for learning by lockdown 3.

Looking at household characteristics, just 7% of those with parents in professional and managerial jobs lacked a device in lockdown 1, compared to over 20% whose parents had routine/manual jobs or who did not work. There were also differences by housing tenure, parental education, and two-parent versus single-parent homes.

But lacking a device was only one part of the problem. Lots of families needed to share devices between children, or between children and homeworking parents. 13% of young people reported needing to share devices in lockdown 1, dropping to 9% in lockdown 3. Again, there were patterns by background characteristics. 13% of those at state comprehensive schools reported issues due to having to share devices, compared to 4% at independent schools.

9% of those whose parents have professional/managerial occupations had such issues in lockdown 1, compared to over 15% of those whose parents have routine/manual occupations. Household size was also a factor. In lockdown 1, 6% of those in one/two-person households reported problems due to having to share devices, compared to 22% of those in a household with six or more. Sharing devices also differed by ethnic background. In lockdown 1, 11% of those in White households had to share devices, compared to 19% in Asian households and 23% in Black households.¹⁵

Supply of devices

One of the reasons for the improvement in access to devices between lockdowns 1 and 3 was schools providing these (including based on government programmes to support this). Of those without a suitable device at the beginning of lockdown 1, 39% had received one by the end of that lockdown period. Similarly, of those who lacked a suitable device at the beginning of lockdown 3, 43% of those had received a device by the end of that period. Overall, 47% of those needing devices at the beginning of the pandemic had received one by the end of the school closures period.

27% of pupils reported receiving a device from their school in lockdown 1, and 30% in lockdown 3. At the state comprehensive schools with the most deprived intakes, 54% of pupils had received devices from their school by lockdown 3. While grammar schools (11%) and the least deprived comprehensive schools (12%) were least likely to supply their pupils with devices, around 18% of pupils in independent schools received one. Those at schools who are part of a multi-academy trust (MAT) were more likely to receive a device than those at standalone academies, potentially driven by MAT-wide programmes of provision – for example the Academies Enterprise Trust¹⁶ – but those at local authority-maintained schools were more likely still.

In each lockdown, around a quarter of young people reported regular problems accessing the internet. However, this was largely uncorrelated with socio-economic factors. This is perhaps due to countervailing patterns of increased problems due to lack of financial resources (such as relying on a mobile phone connection rather than fixed line broadband), and increased problems due to greater intensity of more connection-intensive forms of online learning (such as live lessons). For example, in both lockdowns those at independent schools were most likely to report internet problems, likely due to their higher rate of live online lessons.

Support at home

Parental support

With most young people learning at home during these periods, parental support became more important than usual. Given pre-existing inequalities in parents’ capacity to support learning – owing to their own experiences with education as well as the time they could dedicate to provide this support – there was a danger this could widen existing attainment gaps. Overall, 58% of parents reported that they were confident in supporting their child’s home learning, with 25% disagreeing. 30% of parents without a degree did not feel confident compared to 24% of those who had one. There were also differences by ethnicity, with 29% of White parents reporting they weren’t confident, compared to 17% of Asian parents and 14% of Black parents.

35% of parents in lockdown 1 and 33% of parents in lockdown 3 said that they helped their child’s learning more than once a week. Those who were confident reported helping more often: 41% of confident parents helped more than once a week, compared to 31% of those not confident, and 24% of those who were unsure.

Nonetheless, young people whose parents had a lower occupational status job, without a graduate parent, or in single parent households were in fact more likely to receive regular support, potentially reflecting the additional needs of these young people. This is consistent with previous findings.¹⁷

Another important aspect of the home learning environment is having a quiet place to study. 17% of young people in lockdown 1 reported they did not have a quiet place to work, and 16% in lockdown 3. The minimal change may reflect that this is a difficult factor to change in a matter of months.

There was also a socio-economic gradient to this, with 20% of those whose parents had routine/manual occupations reporting this issue compared to 13% of those with parents in professional/managerial jobs. 23% of those in social housing reported not having a quiet place to work, compared to 21% in private rented accommodation and 13% of those in owner occupied homes. Larger households were also less likely to have a quiet space: 14% of young people in one/two-person households reported this issue, compared to 27% of those in households of six or more. 24% of Black and 22% of Asian pupils faced this issue, compared to 16% of White pupils, however this was largely due to differences in household size across ethnic groups.

Extra support

Parental support for learning can also take the form of purchasing external help, in the form of private tutoring. At the outset of the pandemic there were widespread fears that such tutoring could contribute to exacerbating inequalities in learning between those who could and couldn’t afford to pay.¹⁸ Overall, 10% of parents reported paying for extra tutoring between March 2020 and August 2021. While this wasn’t substantially different from the 11% who reported using tutors in the school year up until the pandemic, this obscures a change in the population receiving tuition. 71% of those using private tutors during the pandemic hadn’t done so in the year pre-pandemic. Socio-economic patterns remained consistent: 14% of those whose parents have professional or managerial occupations received tuition compared to 6% of those with parents in routine/manual jobs, both before and during the pandemic. There were changes in patterns of private tutoring by ethnicity, however. Rates stayed at 9% for White children, but rates in Asian families dropped from 21% to 13%, and Black families from 19% to 17%.

Figure 16 shows how tutoring levels fluctuated in years 10 and 11 for this cohort of young people. At the onset of the pandemic there was a – perhaps surprising – immediate drop in the use of private tutoring compared to before the pandemic. This may simply reflect disruption of existing tutoring relationships by the start of the first national lockdown. After the start of pandemic restrictions, around 5% of parents paid for tutoring in lockdown 1, rising to 7% in lockdown 3, which then stayed consistent in the spring of Year 11 when schools opened again. Few parents used tutors in the holiday period after year 10 or 11.

As well as tutoring, some families spent money on other items to support their child’s learning during the pandemic, including laptops, tablets, and other equipment, as well as supplies that the child may have otherwise received at school. While most parents (54%) didn’t report any additional spending, 17% of parents reported spending £300 extra or more on their child. Unsurprisingly, professional/ managerial households, those who owned their house and two-parent households were all more likely to report spending larger amounts.

Barriers to learning and effects

The barriers to learning during the pandemic that young people reported facing are summarised in Figure 18. In every category (apart from parental confidence where we are restricted to an overall measure so no observable improvement is possible), the situation improved between the two periods of school closures. The biggest improvement was in a lack of access to live lessons, which more than halved between lockdowns.

Figure 19 summarises the socio-economic differences in the perceived barriers to learning in lockdown faced by young people. Almost every barrier has a socio-economic gradient between those whose parents have professional or managerial jobs and those with routine/manual jobs. The pattern is somewhat different for parental confidence, with those in intermediate occupations most likely to have concerns about confidence in supporting home learning, and is least marked in perceptions of internet problems (as discussed above).

But what impact did these barriers actually have on pupils’ engagement with education during this period? Figure 20 shows the associations between barriers faced during lockdowns and the number of hours the young person reported spending doing schoolwork during lockdowns. All do, indeed, appear to have been important barriers to young people being able to continue with their studies during the pandemic’s disruption. The difference in time spent doing schoolwork is most pronounced between those with a laptop or tablet compared to those without a device for remote learning. In most cases, those facing the barrier worked fewer hours than those who didn’t, with a notable exception being those who reported internet problems. Multivariate analysis conducted supports lacking a device or only having access to a mobile phone as the strongest influences on time spent engaging with schoolwork, controlling for background characteristics.

Conclusions and policy implications

• Changes in patterns between lockdowns demonstrate the ability of better-off groups to adapt to crises and maintain their advantaged position. While overall levels of learning increased by the third lockdown, and support from schools and the government targeted the poorest, across several measures inequalities still widened, particularly within the state school sector.

• Much of the focus of attention of learning loss has been on younger children. However, our findings show that 15- and 16-yearolds also faced significant barriers to learning during the pandemic. We also find that socio-economically disadvantaged groups faced greater barriers and lost more learning time than others. However, targeted Pupil Premium funding, and the Recovery Premium¹⁹ catch up funding, ends after Year 11. Educational disadvantage does not end at age 16, so targeted funding should not end there either. The Sutton Trust, along with the Social Mobility Commission and other groups have campaigned for Pupil Premium funding to be extended to those in 16-19 provision,²⁰ and given the level of disruption, this is more important than ever, both for the COSMO cohort, and the cohorts that will follow them.

• The roll-out of devices provided by central government and schools themselves was significant, with 30% of all Year 11s being provided with a device by the end of lockdown 3. Nonetheless, over half of those who lacked a laptop or tablet at home at the outset of the pandemic, still hadn’t received one by this point. Having no device, or relying only on a mobile phone, had the strongest association with the amount of time young people reported spending on schoolwork during lockdowns. Regardless of the future likelihood of lockdowns, it seems likely that online learning is here to stay. Provision of devices for learning for those who can’t afford it themselves should be put on a long-term footing. This could take the form of guaranteeing every pupil eligible for the Pupil Premium access to a laptop or tablet for learning if they need it.

• Learning when schools reopened after the national lockdown periods in 2021 is explored in the accompanying Education Recovery and Catch Up Briefing. The extent to which lockdown conditions affected Teacher Assessed Grades at the end of Year 11, using data linked to the National Pupil Database, will also be investigated in a forthcoming COSMO briefing.

Citing this briefing

Cullinane, C., Anders, J., De Gennaro, A., Early, E., Holt-White, E., Montacute, R., Shao, X., & Yarde, J. (2022). Wave 1 Initial Findings – Lockdown Learning. COVID Social Mobility & Opportunities (COSMO) study  Briefing No. 1. London: UCL Centre for Education Policy and Equalising Opportunities & Sutton Trust. Available at: https://cosmostudy.uk/publications/lockdown-learning

Sample and methods

The data for this briefing come from Wave 1 of the COVID Social Mobility & Opportunities (COSMO) study. COSMO is based on a probability sample drawn from the Department for Education’s National Pupil Database (plus additional recruitment from pupils at private schools), with clustering within schools (for practicality reasons) and oversampling of certain groups using stratification. 

 Our analysis in this briefing is primarily based on descriptive statistics reporting averages, distributions and differences between groups. Analyses use weights to take into account the over-sampling inherent in the study design, as well as initial non-response by young people and, where relevant, their parents. Differences are only highlighted where these are found to be statistically significant at the p<0.05 level. Any statistical inference testing reported also accounts for the clustering and stratification in the study design.

While our full sample of young people has N=12,828, the parents of participants were not as likely to respond, reducing analyses involving parents to at most N=9,330. As noted above, young person and parental non-response have been modelled separately, with different weights to ensure (insofar as is possible) representativeness of our analysis sample to the intended population. Item-level nonresponse also results in some further variation to the analysis sample, which is minimised within analyses to ensure consistency. Analyses of some groups, for example those who attended special schools or who identify as non-binary/in another way, have not been able to be reported due to small sample sizes.

Aspects of the analysis use administrative data from the Department for Education (DfE)’s National Pupil Database (NPD), where consent was gained for this linkage (73% of young people), with additional weighting carried out to ensure (insofar as is possible) representativeness of analysis using linked administrative data. This work was produced using statistical data from the DfE processed in the Office for National Statistics’ (ONS) Secure Research Service (SRS). The use of the DfE statistical data in this work does not imply the endorsement of the DfE or ONS in relation to the interpretation or analysis of the statistical data. This work uses research datasets, which may not exactly reproduce National Statistics aggregates.

About the COVID Social Mobility & Opportunities (COSMO) study

The COVID Social Mobility & Opportunities (COSMO) study is a new national cohort study generating high-quality evidence about how the COVID-19 pandemic has affected socio-economic inequalities in life chances, both in terms of short- and long-term effects on education, wellbeing, and career outcomes. A representative sample of young people in England who were in Year 11 in the 2020/2021 academic year were invited to take part in the survey, with the aim of following them as they progress through the final stages of education and into the labour market. A sample of more than 13,000 cohort members was recruited in Wave 1.

This work was supported by UK Research and Innovation Economic and Social Research Council as part of their COVID-19 response fund [grant number ES/W001756/1]. COSMO is a collaboration between the UCL Centre for Education Policy & Equalising Opportunities (CEPEO), the Sutton Trust, and the UCL Centre for Longitudinal Studies (CLS). Our principal fieldwork partner is Kantar Public.

Researchers can access data from Wave 1 of the study through the UK Data Service.²¹