This document analyzes the impact of school closures during the COVID-19 pandemic on learning loss and inequalities using data from the UK Understanding Society study. It finds that while schools provided more schoolwork and online/offline lessons during the second closure period, inequalities persisted between high and low socioeconomic status families. The transition to remote learning exacerbated existing gaps, though single parents working from home saw some reduction. Recommendations include keeping schools open if possible, increasing targeted support for disadvantaged students through additional funding, tutoring and internet access, and engaging schools directly to identify and help those most in need.

1. LEARNING LOSS AND LEARNING
INEQUALITIES DURING THE COVID-19
PANDEMIC: AN ANALYSIS USING THE UK
UNDERSTANDING SOCIETY DATA
Southampton Education School,
Leadership, Educational Effectiveness & Policy (LEEP) research centre
Dr Nicola Pensiero
Professor Anthony Kelly
Dr Christian Bokhove

2. Background
https://www.chamberofcommerce.org/findaphoto/search?q=empty+classroom
Wikimedia commons: https://commons.wikimedia.org/wiki/File:Student_condition_in_covid.jpg
https://www.chamberofcommerce.org/

3. Background
• School closures because of Covid-19
• Schools were closed from late March
2020 to the start of June 2020
• Students transitioned to distance
learning again in January 2021
• The abrupt transition to home-
schooling may have had severe
consequences for students across the
UK
• These consequences may have been
more severe for socio-economically
disadvantaged students
• Hence, the transition to distance
schooling is likely to exacerbate
inequalities by socio-economic group

4. Schoolwork during the school closures
• During the first school closure from late March 2020 to the start of June 2020, students spent fewer hours
doing schoolwork than they would have done in normal circumstances (Andrew et al. 2020, Bayrakdar &
Guveli 2020, ONS 2020, Pensiero et al. 2020)
• Most schoolwork consisted of offline asynchronous activities rather than live sessions (Green 2020,
Pensiero et al. 2020)
• The reduction in schoolwork and learning was more severe among disadvantaged (low SES) students
(Andrew et al. 2020, Bayrakdar & Guveli 2020, Pensiero et al. 2020). Similar effects were found in in the
Netherlands (Bol 2020) and in Denmark (Jæger & Blaabæk 2020)
• Student engagement with schoolwork was lower in more deprived schools, among students eligible for free
school meals (FSM) and among those with limited access to computers and study space at home (Lucas et
al. 2020)
• UK children from families in the top income quintile spent on average 1.3 hours per day longer on
educational activities during the first closure period than those in the bottom income quintile (5.8 hours per
day compared to 4.5 hours per day, Andrew et al. 2020)
• In the UK, between April/May 2020 when schools were closed, and June/July 2020 when schools
reopened, the total learning time fell for both primary and secondary school students (Cattan et al. 2021)
• Our study is unique in that it looks for the first time at schoolwork during both school closures for the same
students and their families

5. The consequences of losing school
Losing hours of school has negative effects on students’
cognitive abilities, national exam results, finishing upper
secondary, entering university, finding a job, future
earnings, health, and life-prospects of their children
• Summer vacation closure is associated with months of progress lost
(Cooper et al., 1996).
• Strikes during primary school reduces earnings of males and females
by age 40 in Argentina (Jaume & Willén, 2019)
• Extra days of school instruction raises scores on (crystallised)
intelligence tests. An additional year of schooling increases
intelligence test scores by 20% (Carlsson et al., 2015).
• Fewer number of weeks of schooling (from 37 to 24) in Germany led
to increased grade repetition in primary school and to fewer students
attending higher secondary school tracks (Pischke, 2007).

6. The consequences of losing school
• Engzell, Frey and Verhagen (2020) - primary school closure during the lockdown in the
Netherlands using a dataset representative of the Dutch primary school student
population, comprising 350,000 students (15% of total)
• In Dutch primary schools, standardized national learning tests are held twice a year. In
January-February and May-June. In 2020, the tests occurred exactly before and after
school closure, which lasted 8 weeks starting on March 16
• The authors compared the progress in test results of 2020 with that of previous years
and found that the school closure was associated with a reduction in scores on math,
spelling and reading tests by three percentage points, equal to 0.08 standard
deviations.
• 8 weeks of closure corresponds to the attainment loss that would occur with a 20%
reduction in the duration of the school year. This means that students made little or no
progress whilst learning from home. The test score reduction is 55% larger for children
from lower educated families
The Netherlands are an interesting case because:
1. Is one of the countries with the highest penetration rate of broadband, and where the
government and local administrations have paid great attention to the computerization
of schools
2. School closures were one of the shortest

7. Research objectives
We use data from Understanding Society, the UK Household
Longitudinal Study to analyse:
• Whether (and to what extent) the prevalence of schoolwork
changed between the first and the second school closure
periods. We focus on four outcomes: time spent doing work
provided by the school; provision of online school lessons;
provision of offline school lessons; and amount of parental
support
• Whether (and to what extent) the gap in schoolwork between
the least disadvantaged (high socio-economic status, SES) and
most disadvantaged (low SES) families changed between the
first and the second school closures

8. Methodology
• Data
• Covid-19 dataset from April 2020 and January 2021 waves of the
Understanding Society study (N 5300), which surveyed all
household members and included a module on schooling
undertaken by parents.
• Also included data from previous Understanding Society waves.
• Variables
• Dependent:
pri/sec hrs schoolwork, hrs support, offline lessons, online lessons
• Independent:
parental occupation (NS-SEC 3 cat), working patterns & family
structure, birth order, computer, gender, age, family size,
nationality, region
• Analytical approach
• Mixed Models in Stata 16 / interval regression

9. General findings (good news)
• Schools provided more schoolwork in January 2021 than in April 2020
• The percentage of children attending schools that provided online lessons each day
went up from 40% to 78% in primary and from 42% to 92% in secondary
• The percentage of children attending schools that provided offline lessons each day
went up from 91% to 93% in primary and from 79% to 88% in secondary
• Hours of schoolwork:
• went up from 2.3 hours in April 2020 (2 offline and 0.6 online lessons) to 3.3 hours (2.8 offline and 0.8
online lessons) in January 2021 for primary students
• from 2.6 hours in April 2020 (2 offline and 1.3 online lessons) to 4 in January 2021 (2.1 offline and 2.4
online lessons) for secondary students
• Hours per day of support from adults
• 2 in April 2020 and 2.4 in January 2021 in primary
• 0.7 in April 2020 and 0.8 in January 2021 in secondary

10. Hours of schoolwork and computer
availability
Computer availability
Primary April 2020 January 2021
Percent (N) Percent (N)
Yes, child's
own 29 43
Yes, shared 62 51
No 4 4
No
schoolwork 4 2
Total 100 (1336) 100 (665)
Secondary April 2020 January 2022
Yes, child's
own 50 76
Yes, shared 35 20
No 2 1
No
schoolwork 13 3
Total 100 (1622) 100 (1020)

11. Hours of schoolwork and parents’ working
patters
Secondary students

12. Hours of schoolwork by region

13. Inequalities by socio-economic groups

14. Did inequalities change?
• Primary school children of single parents who worked from
home were able to reduce the gap in schoolwork done
compared to the most advantaged socio-economic group, but
generally, inequalities between socio-economic groups in the
uptake of schoolwork remained stable between the two school
closure periods

15. Why did schoolwork improve?

16. Conclusions
• The provision of schoolwork improved in both primary and secondary
schools in the second closure period (January 2021 through February
2021) compared to the first school closure period (from late March 2020
to the start of June 2020). The number of offline and online lessons per
day increased and this led to a larger volume of schoolwork being done.
• The transition to distance schooling has exacerbated inequalities by
socio-economic status
• The effect of parental occupation on schoolwork was amplified when
combined with student access to computers, family circumstances and
parental working patterns
• Single parents who work from home reduced their disadvantage, but
generally inequalities remained stable
• The increase in schoolwork provision can be explained by the improved
provision of lessons, by greater availability of computers and by the fact
that families were better prepared for the second school closure and
could engage more with the schoolwork provided

17. Recommendations
• If it is feasible in terms of containment of the virus, it is important that
schools remain open during any further phases of high infection in
order to avoid a further widening of the achievement gap between
socio-economic groups
• Should schools be forced to close again in the event of another
lockdown, inequalities in learning can be remediated by increasing
the provision of lessons, providing students with better access to IT
and by providing academic tutors to compensate for the absence of
parents who cannot work from home
• When providing schoolwork remotely during a school closure, schools
should consider providing guidance and tutoring targeted at: children
who do not have a computer or do not have ready access to one;
children of single parents; and children of routine class parents who
cannot work from home

18. Recommendations
• The government catch-up funding and schemes like the National
Tutoring Programme are a timely response and a laudable effort at
closing the attainment gap between socio-economic groups. From a
quantitative point of view, the total allocation per pupil of £80 (6
additional days of schooling) is likely to be insufficient. Catch-up
tutoring should be of the order of several weeks of schooling, but
qualitatively it is important to target online and offline resources,
tutoring (online and face-to-face), IT hardware and guidance for
parents on using it to those students which our research has identified
as being in greatest need.
• Schools and not central government are in the best position to identify
those students and families most in need and should be provided with
the necessary funding and flexibility to target provision and support
immediately to the most disadvantaged students.

19. Questions and discussion
Dr Nicola Pensiero n.pensiero@soton.ac.uk
Professor Anthony Kelly A.Kelly@soton.ac.uk
Dr Christian Bokhove C.Bokhove@soton.ac.uk
Also see our full
report via this link.

20. Selected references
Booth, A. L., & Kee, H. J. (2009). Birth order matters: the effect of family size and birth order on educational
attainment. Journal of Population Economics, 22(2), 367-397.
Carlsson, M., Dahl, G. B., Öckert, B., & Rooth, D. O. (2015). The effect of schooling on cognitive skills.
Review of Economics and Statistics, 97(3), 533-547.
Cooper, H., Nye, B. & Charlton, K. (1996). The effects of summer scores: A narrative and meta-analytic
review, Review of Educational Research, 66(3): 227–68.
Engzell, P., Frey, A., & Verhagen, M. D. (2020, October 29). Learning Inequality During the Covid-19
Pandemic. https://doi.org/10.31235/osf.io/ve4z7.
Goodman, J. (2014). Flaking out: Student absences and snow days as disruptions of instructional time (No.
w20221). National Bureau of Economic Research
Hampden-Thompson, G. (2009). Are two better than one?: A comparative study of achievement gaps and
family structure. Compare: A Journal of Comparative and International Education, 39(4), 517–537.
Jaume, D., & Willén, A. (2019). The long-run effects of teacher strikes: evidence from Argentina. Journal of
Labor Economics, 37(4), 1097-1139.
Pensiero, N., Kelly, A. & Bokhove, C. (2020). Learning inequalities during the Covid-19 pandemic: how
families cope with home-schooling. University of Southampton research report.
https://doi.org/10.5258/SOTON/P0025
Pischke, J. S. (2007). The impact of length of the school year on student performance and earnings:
Evidence from the German short school years. The Economic Journal, 117(523), 1216-1242.
Sacerdote, B. (2012) When the Saints Go Marching Out: Long-Term Outcomes for Student Evacuees from
Hurricanes Katrina and Rita. American Economic Journal: Applied Economics, 4(1):109-35
Sirin, S. R. (2005). Socioeconomic Status and Academic Achievement: A Meta-Analytic Review of
Research. Review of Educational Research, 75(3), 417–453. https://doi.org/10.3102/00346543075003417