Learning loss since lockdown One of the most enduring
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
CentrePiece Autumn 2021
Young people in education
have lost significant amounts
of time in school during the
pandemic – and this carries
risks of lifetime scars.
Lee Elliot Major, Andrew
Eyles and Stephen Machin
reveal the scale of learning
loss across the UK nations
and assess what it means for
future life opportunities.
Learning loss
since lockdown
O
ne of the most enduring lockdown and school closures during the
generational impacts of the pandemic. We consider the impacts on
pandemic concerns the learning pupils in primary and secondary schools –
lost by children and young and we compare and contrast the scale and
people. Missing out on education from range of losses across the UK nations.
which they would otherwise have The calculations are made by
benefited is likely to have profound effects triangulating data from different sources
on life outcomes for what we have called to produce robust estimates of educational The extent
the Covid generation. There are genuine
concerns that the pandemic will
loss. Considering how these are likely to
affect human capital formation in later life
of schooling
exacerbate existing inequalities and reduce allows us to estimate the likely increase in disruption
future levels of social mobility – the
capacity of young people to transcend
intergenerational persistence for young
generations growing up in the wake of
during lockdown
their background. the pandemic. has been
In our latest research, we appraise
how children’s exposure to formal learning
Our analysis is based on a number of
separate sources of data. We document
unprecedented
has evolved over the different periods of the different school opening and closure in its scale
14
CentrePiece Autumn 2021
dates in the four nations during the been unprecedented in its scale. It differed days (with a further five days for teacher
pandemic. We use data from official across the four nations because education development). This equates to more than
attendance statistics published by each is devolved across the UK with variation half of school days being missed.
nation. Alongside these, we use data in national policies between England, Figure 1 plots weekly attendance data
from parental responses gathered in Scotland, Wales and Northern Ireland. separately for the four nations in each
successive waves from a nationally These include historic differences in term school week since 23 March 2020, taken
representative longitudinal study called dates, and decisions about when to re- from official published attendance figures.
Understanding Society. open schools. The patterns in school attendance
Combining figures from 23 March reflect the distinctive school closure and
School days missed 2020 to 23 March 2021, we find that the opening dates in the four nations, but
Reviewing the key school dates and school following days were lost due to school also policies concerning attendance of
attendance data permits a calculation of closures over the calendar year: 110 days vulnerable pupils and pupils of key workers
the maximum total numbers of classroom (England); 119 days (Scotland); 124 days during school closures.
days missed by pupils across the different (Wales); and 119 days (Northern Ireland). Attendance was very low in the period
nations. The extent of schooling disruption These figures all compare to a full calendar stretching from 23 March 2020 until the
between March 2020 and April 2021 has year during normal times of 190 classroom summer break and in the spring term
Figure 1:
School attendance, four nations
England Northern Ireland
Pupils attending state-funded schools Pupils attending state-funded schools
100 100
80 80
60 60
Percentage
Percentage
40 40
20 20
0 0
0
0
20
20 0
20
Ja 20
22 eb 021
M 20 1
19 ar 2 1
17 pr 1
ay 1
21
0
20
20
2 0
D 020
20
22 Feb 021
M 20 1
19 ar 2 1
17 pr 1
ay 1
21
02
02
ov 2
22 b 2
2
A 02
M 202
02
ov 2
22 eb 02
2
A 02
M 202
20
N 0
11 t 20
Fe 20
20
20
20
N 0
20
20
2
l2
2 t2
2
2
2 t2
2
F 2
ar
p
n
ar
n
ug
ec
n
c
Ju
c
c
O
Se
Ju
Ja
O
O
M
M
F
A
13
12
11
11
29
12
7
8
8
7
23
23
24
Scotland Wales
Pupils attending state-funded schools Pupils attending state-funded schools
100 100
80 80
The poorest
60 60
pupils in England
Percentage
Percentage
experienced
40 40
lower learning
20 20
loss than the most
affluent pupils in
0 0 Scotland, Wales
and Northern
0
20
20
20 0
20
20 0
21
M 2021
ay 1
21
0
0
20
2020
20
n 0
22 Feb 1
M 20 1
12 ar 21
17 r 2 1
21
M 021
02
ct 02
n 2
17 pr 02
02
02
Ja 02
2
22 b 2
p 2
20
20
Ja 20
20
20
ov 0
20
Fe 20
A 20
20
2
O t2
A r2
2
l2
N 2
4 c2
Ireland
ar
n
ug
4 ec
ar
p
2 Oct
ay
Ju
19 Oc
12 Ma
e
Ju
Se
M
M
D
D
A
13
22
5
19
7
8
21
14
23
29
23
17
15CentrePiece Autumn 2021
of 2021. Attendance rates in England Overall, learning capacity due to the 61 days in England; 61 days in Northern
during school closures were around twice closure of schools during the first lockdown Ireland; 64 days in Scotland; and 66 days
the level of attendance rates elsewhere in 2020 fell sharply for all four nations. in Wales.
in the UK, driven by the opening of all Learning losses were particularly large for
schools in England, and (to a lesser extent) Scotland and Wales, where on average Educational inequality
government policy permitting vulnerable pupils missed out on around two-thirds of We find that children and young people
children and children of key workers the lessons they would normally receive. from more affluent backgrounds – whether
to attend school. Different approaches In England and Northern Ireland, the measured by parental income or type of
were adopted across the four nations; for learning loss was lower, but still of sizable schooling – received more instruction time
example, all schools in England remained magnitude, at 57% and 59% respectively. during the first lockdown.
open, whereas other parts of the UK opted The daily estimates of learning loss can Figure 2 shows how learning losses
for a hub model in which fewer schools be combined with the attendance numbers differed across the family income
were kept open. Attendance meanwhile in Figure 1 to obtain estimates of lost days distribution for the three periods under
was high in the autumn term, when all of schooling in each school term across the study. Due to relatively small sample sizes
nations re-opened the school doors. four nations. Multiplying the daily learning for Northern Ireland, Scotland and Wales,
loss by the length of term in days produces their data are pooled together so that
Rates of classroom and estimates of losses for the summer term of the data contrast England with the rest of
home learning 2019/20, the 2020/21 autumn term and the UK.
These data relate only to classroom the 2020/21 spring term. As the focus is on Two findings are apparent. First,
learning, not learning occurring at home. days lost since the first lockdown occurred there were significant divides in learning
With varying degrees of effectiveness, on 23 March 2020, the extra ten days of loss across the year of the pandemic.
schools provided online lessons and the 2019/20 spring term are added to the Pupils from the bottom fifth of incomes
expectations of what children should cover length of the 2019/20 summer term. experienced higher learning loss than those
at home during school closures. Home This produces the following estimates from the top fifth. In England, for example,
learning experiences differed enormously for overall days lost, presented in Figure 2. during the 2021 school closures, the
depending on the availability of a quiet We estimate the following overall losses: poorest pupils missed out on a third of their
place to study, internet connectivity and
resources provided by schools.
Differences in home learning can
be assessed using microdata from the
dedicated Covid waves released as part
Figure 2:
of Understanding Society, which asked
Learning losses, four nations
parents a battery of questions about home
schooling in the two closure periods. From
these responses, we are able to estimate Average learning loss
daily learning including online lessons
during the pandemic. England Northern Ireland Scotland Wales
Total days 61 61 64 66
of which:
Summer term 2020 40 36 39 40
Autumn term 2020 8 5 7 8
Spring term 2021 13 21 18 18
Inequality of learning loss
April 2000 November 2020 January 2021
England Rest of UK England Rest of UK England Rest of UK
1 2 3 4 5 6
Bottom 20% 60.9 71.4 13.4 8.3 34.9 47.4
Recovery Middle 60%
Top 20%
56.4
47.5
62.7
59.7
11.6
5.9
9.1
6.7
27
24.4
38.3
38.6
programmes
will need to be Sample size 2,952 651 1,973 402 1,910 386
substantial to
address significant Notes: Top panel shows average learning days lost during the first year of the pandemic. Panel
above shows the percentage loss in learning experienced by pupils from the bottom fifth of
learning loss incomes, the middle 60% of incomes and the top fifth of incomes.
16CentrePiece Autumn 2021
learning (34.9%) while the richest pupils Conclusions This article summarises ‘Learning Loss
missed out on a quarter of their learning Our research indicates that any recovery since Lockdown: Variation across the Home
(24.4%). programmes rolled out by the respective Nations’ by Lee Elliot Major, Andrew Eyles
Second, the poor-rich gap in learning UK governments will need to be substantial and Stephen Machin, CEP Covid-19 Analysis
loss varied across nations. These differences to address significant learning losses No. 23 (https://cep.lse.ac.uk/pubs/download/
are significant: during the 2021 school suffered by pupils across the four nations. cepcovid-19-023.pdf).
closures, the poorest pupils in England They will also need to be highly targeted
experienced lower learning loss than the to help disadvantaged children and Lee Elliot Major is professor of social
most affluent pupils in Scotland, Wales and young people who, on average, missed mobility at the University of Exeter and a
Northern Ireland (34.9% compared with out on more education than their more CEP research associate. Andrew Eyles is
38.6%, respectively). privileged peers. a research economist in CEP’s education
Several studies have confirmed We have previously advocated extra and skills programme. Stephen Machin is
that learning losses suffered during the targeted one-to-one or small group professor of economics at LSE and director
pandemic are manifested in stark gaps tutoring as one credible policy response, of CEP.
in actual exam results between children which has been subsequently taken up by
from poorer backgrounds and their more the government in England.
privileged counterparts. Other research Another suggestion is to extend school Further reading
shows a robust positive relationship time or at least ensure minimum school
between hours of study and attainment. days across all schools. We find that a Lee Elliot Major, Andrew Eyles and Stephen
Considering the likely consequences narrow majority – 53% – of our survey Machin (2021) ‘Unequal Learning and Labour
of increasing educational inequalities participants believe that extending school Market Losses in the Crisis: Consequences
for earnings and employment in the time would be an appropriate policy for Social Mobility’, CEP Discussion Paper No.
labour market, a likely consequence response to the learning losses. But we 1748 (https://cep.lse.ac.uk/pubs/download/
for this generation will be a significant did not detail how this extended school dp1748.pdf).
decline in social mobility levels (Elliot day would be used. The level of agreement
Major et al, 2021). is largely invariant to the hypothetical
learning losses that we present
Policy matters to participants.
Finally, we explore the extent to which It remains unclear what level of
government policies affected nation- Covid-induced absences will continue
specific gaps in learning loss. We do this by with schools now back for the autumn
asking what would have happened under term of 2021/22. Governments across
counterfactual scenarios where nations the four nations should be aware that
had the same policies and/or learning policy interventions – encouraging key
losses. The counterfactuals reveal that both pupils to attend schools during closures,
education policy decisions enacted under reorganising school term dates or opening
the pandemic and historical differences in schools early – can all have a significant
term dates explain differential degrees of impact in reducing learning losses.
learning loss across the home nations. These are important considerations given
England, for example, benefitted the lifetime scarring likely to occur for the
from higher attendance rates during the Covid generation.
pandemic, both during normal term time
and during partial school closures by urging
vulnerable pupils and children of key
workers to attend. Ministers also decided
to open the country’s schools to all pupils
at an earlier date during the 2020/21
spring term. Scotland, on the other hand,
benefitted from earlier scheduled school
summer holidays during the summer term
of 2019/20.
17You can also read
