Kodama, Wa a u; Mo gan, Pe e J.; Azhgaliye a, Dina
Wo king Pape
Do emo e lea ning ools educe lea ning loss du ing
school closu e? Expe ience om Cen al Asia and he
Caucasus du ing he COVID-19 pandemic
ADBI Wo king Pape , No. 1450
P o ided in Coope a ion wi h:
Asian De elopmen Bank Ins i u e (ADBI), Tokyo
Sugges ed Ci a ion: Kodama, Wa a u; Mo gan, Pe e J.; Azhgaliye a, Dina (2024) : Do emo e lea ning
ools educe lea ning loss du ing school closu e? Expe ience om Cen al Asia and he Caucasus
du ing he COVID-19 pandemic, ADBI Wo king Pape , No. 1450, Asian De elopmen Bank Ins i u e
(ADBI), Tokyo,
h ps://doi.o g/10.56506/ROAZ2346
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/301955
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by-nc-nd/3.0/igo/
ADBI Wo king Pape Se ies
DO REMOTE LEARNING TOOLS REDUCE
LEARNING LOSS DURING SCHOOL
CLOSURE? EXPERIENCE FROM
CENTRAL ASIA AND THE CAUCASUS
DURING THE COVID-19 PANDEMIC
Wa a u Kodama, Pe e Mo gan,
and Dina Azhgaliye a
No. 1450
May 2024
Asian De elopmen Bank Ins i u e
The Wo king Pape se ies is a con inua ion o he o me ly named Discussion Pape se ies;
he numbe ing o he pape s con inued wi hou in e up ion o change. ADBI’s wo king
pape s e lec ini ial ideas on a opic and a e pos ed online o discussion. Some wo king
pape s may de elop in o o he o ms o publica ion.
Sugges ed ci a ion:
Kodama, W., P. Mo gan, and D. Azhgaliye a. 2024. Do Remo e Lea ning Tools Reduce
Lea ning Loss du ing School Closu e? Expe ience om Cen al Asia and he Caucasus
Du ing he COVID-19 Pandemic. ADBI Wo king Pape 1450. Tokyo: Asian De elopmen
Bank Ins i u e. A ailable: h ps://doi.o g/10.56506/ROAZ2346
Please con ac he au ho s o in o ma ion abou his pape .
Email: [email p o ec ed]
Wa a u Kodama is a esea ch associa e, Pe e Mo gan is senio consul ing economis
and ad iso o he dean, and Dina Azhgaliye a is a senio esea ch ellow, all a he
Asian De elopmen Bank Ins i u e, Tokyo, Japan.
The iews exp essed in his pape a e he iews o he au ho and do no necessa ily
e lec he iews o policies o ADBI, ADB, i s Boa d o Di ec o s, o he go e nmen s
hey ep esen . ADBI does no gua an ee he accu acy o he da a included in his pape
and accep s no esponsibili y o any consequences o hei use. Te minology used may
no necessa ily be consis en wi h ADB o icial e ms.
Discussion pape s a e subjec o o mal e ision and co ec ion be o e hey a e inalized
and conside ed published.
The au ho s a e g a e ul o Daniel C. Su yada ma and pa icipan s a he ADBI
Con e ence on Inc easing he Resilience o Educa ion Sys ems in Asia and he Paci ic
o hei aluable commen s.
Asian De elopmen Bank Ins i u e
Kasumigaseki Building, 8 h Floo
3-2-5 Kasumigaseki, Chiyoda-ku
Tokyo 100-6008, Japan
Tel: +81-3-3593-5500
Fax: +81-3-3593-5571
URL: www.adbi.o g
E-mail: [email p o ec ed]
© 2024 Asian De elopmen Bank Ins i u e
ADBI Wo king Pape 1450 W. Kodama e al.
Abs ac
School closu es ha e a nega i e impac on child en’s educa ional ou comes. Du ing
he COVID-19 pandemic, Cen al Asia and he Caucasus in oduced online classes and
TV p og ams wi h a iew o p e en ing lea ning loss. Based on a household su ey in
nine coun ies (A ghanis an, Aze baijan, Geo gia, Kazakhs an, he Ky gyz Republic,
Mongolia, Pakis an, Tajikis an, and Uzbekis an), his is one o he i s egional s udies o
documen lea ning loss and he e icacy o emo e lea ning ools in Cen al Asia and he
Caucasus. Some 79% o households el ha hei child en’s lea ning p og ess was slowe
han in in-pe son schooling, including 39% who pe cei ed ei he “ e y li le p og ess” o “no
p og ess.” Ou econome ic analysis inds ha online classes educed pe cei ed lea ning
loss du ing school closu es, bu his does no hold o TV p og ams. In addi ion, adi ional
pape lessons we e also e ec i e in educing pe cei ed lea ning loss bu only o sho e
school closu e du a ions. The usage o online classes is s ongly co ela ed wi h household
income and loca ion (u ban s. u al) as well as in e ne connec i i y. Combined usage o
TV p og ams and pape lessons may be an e ec i e al e na i e o online lessons du ing
longe closu e pe iods.
Keywo ds: emo e lea ning, lea ning loss, online class, COVID-19 pandemic, Cen al Asia
JEL Classi ica ion: I20, I21, I25
ADBI Wo king Pape 1450 W. Kodama e al.
Con en s
1. INTRODUCTION .......................................................................................................... 1
2. DATA AND CONTEXT ................................................................................................. 2
2.1 Da a .................................................................................................................. 2
2.2 Con ex ............................................................................................................. 3
3. STYLIZED FACTS ....................................................................................................... 5
4. ECONOMETRIC ANALYSIS ........................................................................................ 7
4.1 De e minan s o he Usage o Dis ance Lea ning Tools ................................... 7
4.2 Remo e Lea ning Tools and Pe cei ed Lea ning Loss .................................... 9
4.3 E ec i eness o Mul iple Dis ance Lea ning Modes ....................................... 11
4.4 He e ogenei y ................................................................................................. 13
5. CONCLUSION AND POLICY RECOMMENDATIONS .............................................. 14
REFERENCES ...................................................................................................................... 15
APPENDIX ............................................................................................................................. 17
ADBI Wo king Pape 1450 W. Kodama e al.
1
1. INTRODUCTION
School closu es ha e a nega i e impac on child en's educa ional ou comes, which
could e en ually lead o a educ ion in he u u e p oduc i i y and income o bo h
indi iduals and coun ies (Hanushek and Woessmann 2020). Du ing he COVID-19
pandemic, la ge and unequal e ec s o school closu e on lea ning p og ess we e ound
(Agos inelli e al. 2022; Engzell, F ey, and Ve hagen 2021; Haele mans e al. 2022;
Lichand e al. 2022; Mosco iz and E ans 2022). Pa inos (2023) ind ha one
addi ional week o school closu e inc eases he deg ee o lea ning loss by 1% o
a s anda d de ia ion (SD). Alasino e al. (2024) ind ha s uden pe o mance is
0.2–0.3 SD lowe a e school closu es in Mexico, equi alen o 0.66–1.05 yea s o
schooling. Du ing school closu es, he ansi ion o online classes/lea ning was
obse ed wo ldwide because in-pe son classes we e no a ailable. In addi ion o online
classes, many Asian coun ies in oduced TV and/o adio p og ams as an addi ional
emo e lea ning ool because in e ne connec ions we e no necessa ily a ailable in
some a eas (UNESCO, UNICEF, and Wo ld Bank 2021). Due o e o s o ensu e
con inui y in lea ning wi h emo e lea ning ools in Uzbekis an, one s udy sugges s ha
he e we e no lea ning losses du ing school closu es (Iqbal and Pa inos 2023).
Howe e , e idence om o he Asian coun ies sugges s he e we e delays in lea ning
p og ess du ing he school closu e pe iod (ADB 2022; Maddawin e al. 2024).
Expe iences o school closu es in Cen al Asia and he Caucasus p o ide impo an
lessons o building esilien educa ion sys ems. The egion expe ienced pe haps he
la ges e ec o school closu es on s uden s’ lea ning p og ess due o ela i ely longe
pe iods o closu e (UNESCO 2021). A he same ime, o p e en lea ning losses, he
go e nmen s and in e na ional pa ne s p o ided no only online classes bu also o he
emo e lea ning ools such as TV, pape lessons, and adio p og ams. Assessing he
lea ning losses du ing he school closu e pe iods and he e icacy o hese emo e
lea ning ools in educing such losses may help p e en hem du ing u u e school
closu es, which could be due no only o pandemics bu also o o he shocks, including
ai pollu ion and na u al disas e s. I is impo an o lea n om he expe ience o he
COVID-19 pandemic o p o ide e idence-based policy ecommenda ions. Thus a ,
howe e , he e idence is e y limi ed in Cen al Asia and he Caucasus, and his is one
o he i s egional s udies o documen lea ning losses and he e icacy o emo e
lea ning ools du ing he pandemic.
We collec ed household su ey da a om 10 membe coun ies o he Cen al Asia
Regional Economic Coope a ion P og am (CAREC), excluding he People’s Republic
o China (PRC). Since he e we e no school closu es in Tu kmenis an, o his s udy we
use da a om nine coun ies, namely A ghanis an, Aze baijan, Geo gia, Kazakhs an,
he Ky gyz Republic, Mongolia, Pakis an, Tajikis an, and Uzbekis an. Ou su ey da a
include pa en s’ pe cei ed p og ess o hei child en’s s udies (“pe cei ed lea ning
p og ess o loss”) and child en’s ac ual usage o emo e lea ning ools. The esul s
sugges ha 79% o he households el ha hei child en’s lea ning p og ess was
slowe han in in-pe son schooling, including 39% who pe cei ed ei he “ e y li le
p og ess” o “no p og ess,” despi e he go e nmen s’ e o s o minimize lea ning loss
wi h emo e lea ning ools. Ou econome ic analysis sugges s ha online classes,
whose usage s ongly co ela es wi h household income and loca ion (u ban s. u al),
we e e ec i e in educing pe cei ed lea ning loss du ing school closu es, bu his does
no hold o TV p og ams. T adi ional pape lessons we e also commonly used, bu
hey we e e ec i e in educing lea ning losses only o sho e school closu e pe iods.
We also ind ha a combined usage o TV p og ams and pape lessons seemed
e ec i e unde longe closu e pe iods.
ADBI Wo king Pape 1450 W. Kodama e al.
2
This s udy con ibu es o he li e a u e in wo ways. Fi s , i in es iga es he
e ec i eness o emo e lea ning, including ools like online classes, pape lessons, and
TV p og ams. P e ious s udies sugges ha , i used p ope ly, online classes may be as
e ec i e as adi ional in-pe son classes in e ms o lea ning ou comes (Swan 2003).
Some ecen s udies sugges ha online classes we e o some ex en e ec i e o
lea ning p og ess du ing school closu es (Awal 2023; Roman and Plopeanu 2021).
Howe e online classes in Cen al Asia and he Caucasus (pa icula ly in u al a eas)
we e challenged by he lack o access o a good-quali y in e ne connec ion, as well as
access o de ices (e.g., compu e s, able s, sma phones). Ins ead, o line ools such
as TV p og ams and pape lessons we e p o ided o child en who could no access
online classes, which could help o mi iga e he nega i e impac . Ou esul s sugges
ha , when combined wi h pape lessons, TV p og ams may mi iga e he nega i e
e ec s o school closu e on lea ning p og ess, al hough he e ec s a e smalle han
hose o online classes.
Second, his is one o he i s egional s udies o documen lea ning losses in Cen al
Asia and he Caucasus du ing he pandemic. P e ious s udies on lea ning losses
mainly ocus on high-income coun ies such as he Uni ed S a es, he Ne he lands, and
Ge many, while some ocus on middle- and low-income coun ies like Bangladesh,
B azil, Indonesia, and Kenya (c ., Pa inos, Vegas, and Ca e -Rau 2023). Maddawin
e al. (2024) ecen ly epo s lea ning losses in Sou heas Asian coun ies, ye
e idence om Cen al Asia and he Caucasus is hus a limi ed despi e he longe
du a ion o school closu es in he egion. Al hough we we e no able o implemen
objec i e measu es o lea ning losses like es sco es, ou su ey da a e lec
esponden s’ ac ual esponses o go e nmen s’ policies on school closu es and emo e
lea ning. We add unique e idence on pa en s’ pe cei ed lea ning p og ess in Cen al
Asia and he Caucasus o he li e a u e.
The emainde o his pape is o ganized as ollows. Sec ion 2 desc ibes he da a and
con ex o his s udy. Sec ion 3 p esen s desc ip i e e idence. Sec ion 4 in es iga es
he e ec i eness o emo e lea ning ools and p esen s he esul s o econome ic
analysis. Sec ion 5 concludes and p o ides policy ecommenda ions.
2. DATA AND CONTEXT
2.1 Da a
Following he COVID-19 pandemic and school closu es in 2020 and 2021, we
conduc ed household su eys in nine CAREC membe coun ies (excluding he PRC
and Tu kmenis an) om Sep embe o Oc obe 2022 wi h a iew o unde s anding
he impac s o he pandemic on households and child en’s educa ional p og ess.1
Compu e -assis ed elephone in e iews (CATIs) we e ca ied ou on andomly
selec ed ep esen a i e samples o mo e han 1,000 households om each coun y, as
ace- o- ace in e iews we e no p ac ical unde he pandemic condi ions. Fo each
coun y, we con i m ha he egional and household income (quan ile) dis ibu ions
a e in line wi h he na ional s a is ics. The ques ionnai e co e ed socioeconomic
s a us, employmen , gende and educa ion o he household head, household income
and expendi u e, and child en’s educa ion (i any) du ing he ime o he pandemic.
Azhgaliye a e al. (2022) and Kodama e al. ( o hcoming) o u he de ails.
1 Tu kmenis an was excluded om his s udy because i epo ed no school closu es du ing he
COVID-19 pandemic.
ADBI Wo king Pape 1450 W. Kodama e al.
3
Fo hose esponden s who ha e any school-age child en (aged six o 18), we
asked ques ions ela ed o school closu es and he pe cei ed lea ning p og ess o he
school-age child who mos ecen ly had a bi hday. In addi ion o basic in o ma ion on
his child such as g ade le el as o now and be o e he school closu es (Decembe
2019), we asked abou he du a ion o school closu es since Feb ua y 2020, modes
o emo e lea ning (class ins uc ion) hey used du ing he closu es, he pa en s’
pe cep ions o he child’s lea ning p og ess du ing he closu es, and whe he he child
was en olled in school a e eopening.
In his s udy, we es ic he samples o households who had any school-age child en
and expe ienced school closu es. This lea es a subsample o 3,600 households,
equi alen o a ound 40% o he o iginal da ase . Table 1 shows he dis ibu ion o ou
sample. The sample mainly consis s o households wi h elemen a y school and middle
school child en: 36% in G ades 1 o 3, 25% in G ades 4 o 6, and 24% in middle
school. Nea ly hal o he households eside in u ban a eas and he emaining hal li e
in u al a eas.
Table 1: Sample Dis ibu ion: G ade Le el, A ea (Ru al s. U ban), and Coun y
Sha e, %
( es ic ed sample)
Sha e, %
(o iginal sample)
G ade le el (Decembe 2019)
P eschool/kinde ga en
5.20
Elemen a y P1-P3
36.25
Elemen a y P4-P6
25.14
Middle school
23.83
High/ oca ional school
9.58
Loca ion ( u al s. u ban)
Ru al
53.05
53.06
U ban
46.95
46.94
Coun y
A ghanis an
6.75
11.21
Aze baijan
10.33
10.86
Geo gia
8.25
10.86
Kazakhs an
10.33
11.00
Ky gyz Republic
13.12
11.44
Mongolia
12.44
10.97
Pakis an
18.07
11.12
Tajikis an
10.95
10.96
Uzbekis an
9.77
11.58
Obse a ions
3,600
10,207
Sou ce: Au ho s’ es ima es om he su ey.
2.2 Con ex
Du ing he ime o he COVID-19 pandemic, CAREC coun ies implemen ed a ious
measu es, including lockdown and school closu es. Table 2 shows he du a ion o
school closu es in CAREC coun ies based on UNESCO’s da abase. In A ghanis an
and Pakis an, on a e age, schools we e ully closed o mo e han 35 weeks and
pa ially closed o a ound 25 weeks. These closu es may ha e led o lea ning losses
o 0.35 o 0.60 SDs (based on he es ima es by Pa inos (2023)). Aze baijan and
ADBI Wo king Pape 1450 W. Kodama e al.
4
Mongolia also had long closu e pe iods wi h 29 and 24 weeks o ull closu e and
20 and 34 weeks o pa ial closu e, espec i ely. In Kazakhs an, schools we e pa ially
closed o 43 weeks. In con as , Tajikis an had no ull school closu e bu had ou
weeks o pa ial closu e.
Table 2: Du a ion o School Closu es and Types o Dis ance Lea ning
(Feb 2020 – Ma ch 2022)
Du a ion (week)
Modes o Remo e Lea ning
Fully
Closed
Pa ially
Closed
Online
Classes
TV
P og ams
Radio
P og ams
Take-home
Packages
(pape )
A ghanis an
35
26
✔
✔
✔
Aze baijan
29
20
✔
✔
Geo gia
19
16
✔
✔
Kazakhs an
9
43
✔
✔
✔
Ky gyz Rep.
14
13
✔
✔
✔
✔
Mongolia
24
34
✔
✔
Pakis an
37
24
✔
✔
✔
✔
PRC
9
21
✔
✔
✔
Tajikis an
0
4
Uzbekis an
13
1
✔
✔
✔
Sou ces: UNESCO, UNICEF, and Wo ld Bank (2021); UNESCO’s COVID-19 Global Moni o ing Da abase, COVID-19
Educa ion Response. h ps://co id19.uis.unesco.o g/global-moni o ing-school-closu es-co id19/coun y-dashboa d/.
Figu e 1: Repo ed Du a ion o School Closu es, % Households
No es: The abb e ia ions (iso3) o he coun ies a e as ollows: A ghanis an (AFG), Aze baijan (AZE), Geo gia (GEO),
Kazakhs an (KAZ), Ky gyz Republic (KGZ), Mongolia (MNG), Pakis an (PAK), Tajikis an (TAJ), and Uzbekis an (UZB).
Sou ce: Au ho s’ es ima es om he su ey.
ADBI Wo king Pape 1450 W. Kodama e al.
11
Figu e 4: P obabili y o “Se e e Lea ning Loss” by Remo e Lea ning Modes
No es: This igu e epo s he p edic ed p obabili y o e y li le o no p og ess du ing school closu e. The es ima ion is
based on Column 1 in Table 4.
4.3 E ec i eness o Mul iple Dis ance Lea ning Modes
E en hough we ound ha only online classes we e e ec i e in mi iga ing se e e
lea ning loss due o longe school closu es, he ole o using mul iple emo e lea ning
modes is no conside ed in ou p e ious eg ession. Speci ically, pape lessons may
enhance he e ec i eness o online classes o TV p og ams. To elici he e ec i eness
o using mul iple emo e lea ning modes, we es ima e he same equa ion as in Table 4
bu wi h (i) a bina y a iable ha ep esen s longe school closu e (= 1 i longe han
six mon hs and = 0 i o he wise) o simpli y he econome ic model, and (ii) in e ac ion
e ms be ween mul iple lea ning modes (online classes and pape lessons; TV
p og ams and pape lessons) and he du a ion o school closu e.
Table 5 epo s he esul s. As in Table 4, a longe school closu e du a ion inc eases
he p obabili y o se e e lea ning loss (Column 1) o highe se e i y o lea ning loss
(Column 2). The use o online classes ela es o a lowe p obabili y o se e e lea ning
loss, bu i s combina ion wi h pape lessons does no b ing abou any u he bene i s.
In con as , e en hough he use o TV p og ams alone is no ela ed o a lowe
p obabili y, i s combined usage wi h pape lessons mi iga es he nega i e e ec s o
longe school closu es.
To demons a e he ole o single and mul iple emo e lea ning modes, Table 6 shows
he es ima ed p obabili y o se e e lea ning loss by du a ion o school closu e (less
han s. mo e han six mon hs) and ypes o lea ning modes, based on he esul s om
he eg ession epo ed in Column 1 o Table 5. Wi hou aking any emo e lea ning
mode, on a e age he p obabili y inc eases by 33% when he school closu e du a ion is
longe han six mon hs. This p obabili y inc ease is s ill 26% and 42% wi h online
classes alone and TV p og ams alone, espec i ely. The usage o only pape lessons
ela es o he lowes p obabili y o se e e lea ning loss unde sho e school closu es
ADBI Wo king Pape 1450 W. Kodama e al.
12
(< 6 mon hs), bu he p obabili y inc eases sha ply by 90% when he du a ion is longe
(> 6 mon hs). Unde longe school closu es, online classes, online and pape lessons,
and TV and pape lessons ela e o he lowes p obabili ies o se e e lea ning loss
(0.430, 0.429, and 0.426, espec i ely). The bene i s o using mul iple lea ning modes
a e seen o TV p og ams unde longe school closu e pe iods. Compa ed o online
classes, TV p og ams a e no in e ac i e, and s uden s end o be less ac i e. Pape
lessons (e.g., homewo k) may supplemen his limi a ion o TV p og ams by enhancing
s uden s’ e o s wi h lea ning.
Table 5: Single and Mul iple Remo e Lea ning Modes
and Pe cei ed Lea ning P og ess
(1) Logi Model
(2) O de ed Logi Model
(= 1 i Ve y Li le o No P og ess;
= 0 O he wise)
(= 1 i Same, = 2 i Slowe , = 3 i Ve y Li le,
= 4 i No P og ess)
Coe icien
Ma ginal E ec
Coe icien
Ma ginal E ec
Du a ion (> 6 mon hs)
0.686***
(0.200)
0.160***
(0.047)
0.582***
(0.173)
0.064***
(0.019)
Mode o emo e lea ning
Online classes
–0.355**
(0.180)
–0.083**
(0.042)
–0.317**
(0.156)
–0.035**
(0.017)
TV p og ams
–0.099
(0.196)
–0.023
(0.046)
0.017
(0.147)
0.002
(0.016)
Pape lessons
–0.927***
(0.167)
–0.217***
(0.039)
–1.023***
(0.157)
–0.112***
(0.018)
Online # Pape
0.646**
(0.292)
0.151**
(0.068)
0.799***
(0.250)
0.088***
(0.028)
TV # Pape
1.210***
(0.321)
0.283***
(0.075)
1.109***
(0.267)
0.121***
(0.029)
Du a ion # Mode
Du a ion # Online
–0.355
(0.230)
–0.083
(0.054)
–0.349*
(0.194)
–0.038*
(0.021)
Du a ion # TV
0.156
(0.242)
0.037
(0.057)
–0.053
(0.185)
–0.006
(0.020)
Du a ion # Pape
0.430
(0.271)
0.100
(0.063)
0.418*
(0.235)
0.046*
(0.026)
Du a ion # Online # Pape
0.028
(0.399)
0.007
(0.093)
–0.223
(0.340)
–0.024
(0.037)
Du a ion # TV # Pape
–1.533***
(0.422)
–0.358***
(0.098)
–1.273***
(0.354)
–0.139***
(0.039)
Coun y ixed e ec s
✔
✔
O he con ol
✔
✔
Obse a ions
3,553
3,553
No es: As e isks (*, **, and ***) deno e signi icance a 10, 5, and 1 % le els, espec i ely. S anda d e o s a e epo ed
in pa en hesis. The ma ginal e ec o he o de ed logi model shows he e ec on he p obabili y o “no p og ess.”
Sou ce: Au ho s’ es ima es om he su ey.
Table 6: P obabili y o Se e e Lea ning Loss by Single
and Mul iple Remo e Lea ning Modes
Du a ion o School Closu e
Di e ences
Less han 6 Mon hs
Mo e han 6 Mon hs
% Change
Chi-squa ed
None
0.424
(0.027)
0.564
(0.033)
33.02
11.97***
Online classes only
0.341
(0.021)
0.430
(0.022)
26.10
9.62***
TV p og ams only
0.394
(0.042)
0.561
(0.039)
42.39
8.76***
Pape lessons only
0.239
(0.217)
0.455
(0.037)
90.38
29.41***
Online and Pape
0.324
(0.043)
0.429
(0.040)
32.41
3.37*
TV and Pape
0.434
(0.057)
0.426
(0.053)
–1.84
0.01
Obse a ions
3,553
No es: The es ima es a e based on he eg ession epo ed in Column 1 o Table 5. As e isks (*, **, and ***) deno e
signi icance a 10, 5, and 1 % le els, espec i ely. S anda d e o s a e epo ed in pa en hesis.
Sou ce: Au ho s’ es ima es om he su ey.
ADBI Wo king Pape 1450 W. Kodama e al.
13
4.4 He e ogenei y
We also in es iga e he he e ogenei y ac oss g ade le els in 2019 (be o e he school
closu es) using he same speci ica ion as in Table 5 o dis inguish he ole o single and
mul iple emo e lea ning modes. We conside he g ade le els o : (i) G ades 1 o 3 in
elemen a y school; (ii) G ades 4 o 6 in elemen a y school; (iii) middle school ( ypically
G ades 6 o 9); and (i ) high and oca ional school ( ypically G ades 10 o 12).
Table 7 shows he esul s. Fo all g ade g oups excep o high school, a longe school
closu e du a ion is ela ed o a highe likelihood o “se e e lea ning loss.” When he
school closu e du a ion is sho e (less han six mon hs), pape lessons a e ela ed o a
lowe p obabili y o lea ning losses o all g ade g oups excep o middle school.
Online classes also ela e o lowe p obabili ies o middle and high school s uden s.
Howe e , TV p og ams ha e no posi i e e ec s on lea ning p og ess.
Table 7: Remo e Lea ning Modes and Pe cei ed Lea ning Loss
by G ade Le el in 2019, O e all Ma ginal E ec s o Logi Model
(1)
(2)
(3)
(4)
Elemen a y 1–3
Elemen a y 4–6
Middle School
High School
Du a ion (> 6 mon hs)
0.220***
(0.072)
0.179*
(0.093)
0.195**
(0.090)
–0.119
(0.142)
Mode o emo e lea ning
Online classes
0.058
(0.066)
–0.055
(0.096)
–0.142*
(0.082)
–0.382***
(0.131)
TV p og ams
–0.019
(0.068)
0.163*
(0.088)
–0.152
(0.096)
–0.119
(0.233)
Pape lessons
–0.259***
(0.059)
–0.053
(0.085)
–0.297***
(0.095)
–0.269**
(0.118)
Online # Pape
0.099
(0.110)
0.263**
(0.130)
0.295**
(0.147)
0.560***
(0.209)
TV # Pape
0.408***
(0.113)
–0.183
(0.149)
0.156
(0.188)
0.128
(0.350)
Du a ion # Mode
Du a ion # Online
–0.161**
(0.082)
–0.110
(0.111)
–0.008
(0.109)
0.278*
(0.166)
Du a ion # TV
0.083
(0.087)
–0.171
(0.105)
0.069
(0.112)
0.095
(0.266)
Du a ion # Pape
0.046
(0.096)
0.001
(0.128)
0.225
(0.151)
0.702**
(0.304)
Du a ion # Online # Pape
–0.038
(0.148)
–0.080
(0.175)
–0.193
(0.202)
–1.031***
(0.373)
Du a ion # TV # Pape
–0.330**
(0.152)
0.084
(0.185)
–0.244
(0.230)
–0.117
(0.457)
Coun y ixed e ec s
✔
✔
✔
✔
O he con ol
✔
✔
✔
✔
Obse a ions
1,263
876
830
324
No es: As e isks (*, **, and ***) deno e signi icance a 10, 5, and 1 % le els, espec i ely. S anda d e o s a e epo ed
in pa en hesis.
Sou ce: Au ho s’ es ima es om he su ey.
When he school closu e du a ion is longe (longe han six mon hs), o elemen a y
G ade 1–3 s uden s, online classes appea o be e ec i e in mi iga ing se e e lea ning
loss. Combining online classes and pape lessons is also e ec i e o high school
s uden s. E en hough combined TV p og ams and pape lessons ela e o a lowe
p obabili y o elemen a y G ade 1–3 s uden s, he e ec s a e canceled ou by hei
e ec wi hou he in e ac ion e m. O e all, hese he e ogenei ies con i m he
e ec i eness o online classes and sugges he impo ance o p o iding a mode o
emo e lea ning ha is app op ia e o a child’s g ade le el.
ADBI Wo king Pape 1450 W. Kodama e al.
14
5. CONCLUSION AND POLICY RECOMMENDATIONS
School closu es nega i ely a ec child en’s educa ion ou comes. Du ing school
closu es caused by he COVID-19 pandemic in Cen al Asia and he Caucasus, online
classes and TV p og ams we e in oduced as emo e lea ning ools o educe lea ning
loss, along wi h adi ional pape lessons. Based on a household su ey co e ing nine
coun ies, his s udy documen s pa en s’ pe cei ed lea ning losses o child en and
he e ec i eness o emo e lea ning modes du ing school closu es. Ou su ey da a
sugges ha 65% and 30% o child en used online classes and TV p og ams,
espec i ely, and 36% elied on con en ional pape lessons. Some used mul iple
modes, while as many as 24% o child en did no use any emo e lea ning ools.
Despi e he go e nmen s’ e o s o minimize lea ning losses wi h emo e lea ning ools,
79% o households wi h child en en olled in school el ha hei child en’s lea ning
p og ess was slowe han in in-pe son schooling, while 39% belie ed ha hei child en
had shown “ e y li le p og ess” o “no p og ess.”
Ou econome ic analysis e ealed some impo an indings. Fi s , he usage o online
classes is s ongly co ela ed wi h ad an ageous household cha ac e is ics such as a
good in e ne connec ion, u ban esidence, and a highe household income. TV
p og ams we e mo e commonly used by u al households. Second, online classes a e
ound o educe lea ning loss du ing school closu es, bu his does no hold o TV
p og ams. Online classes also educe he nega i e e ec s o a longe school closu e
du a ion on pe cei ed lea ning p og ess. Thi d, on a e age, pape lessons a e also
ound o be an e ec i e mode o emo e lea ning, bu hei e ec i eness is limi ed
o sho e school closu e du a ions. Finally, combined usage o TV p og ams and
pape lessons seems o be e ec i e in mi iga ing he nega i e e ec s o longe school
closu e du a ions.
The abo e indings ha e he ollowing policy implica ions. Du ing school closu es,
which may also be due o na u al disas e s, ai pollu ion, ex eme wea he , con lic s,
e c., o mo e gene ally when a child is unable o go o school, online classes appea o
be he bes op ion o minimizing lea ning loss (as his ool p o ides in e ac ion).
Howe e , in places wi h limi ed access o he in e ne o sma de ices hey a e unlikely
o be used by all child en. In such places, pape lessons could be used as a way o
educe lea ning loss o a sho school closu e du a ion. O e all, building he capaci y
o p o ide child en wi h disad an aged backg ounds wi h e ec i e access o online
classes is c i ically impo an o educing lea ning losses and o p e en ing widening
gaps in e ms o lea ning losses du ing u u e school closu es. In addi ion, in he
absence o access o online lessons du ing long- e m school closu es, he usage o TV
p og ams oge he wi h pape lessons could be an e ec i e al e na i e ool, which
equi es u he in es iga ion in o i s e ec i eness.
ADBI Wo king Pape 1450 W. Kodama e al.
15
REFERENCES
ADB. 2022. Lea ning and Ea ning Losses om COVID-19 School Closu es in
De eloping Asia. In Asian De elopmen Ou look 2022: Mobilizing Taxes o
De elopmen . Manila: Asian De elopmen Bank.
Agos inelli, F., M. Doepke, G. So en i, and F. Zilibo i. 2022. When he G ea Equalize
Shu s Down: Schools, Pee s, and Pa en s in Pandemic Times. Jou nal o Public
Economics 206: 104574. h ps://doi.o g/10.1016/j.jpubeco.2021.104574.
Alasino, E., M. J. Ramí ez, M. Rome o, N. Schady, and D. U ibe. 2024. Lea ning
Losses Du ing he COVID-19 Pandemic: E idence om Mexico.
Economics o Educa ion Re iew 98: 102492. h ps://doi.o g/10.1016/
j.econedu e .2023.102492
Awal, R. 2023. Measu ing he E ec i eness o Online Classes Du ing he COVID-19
Pandemic: Case S udy o a Go e nmen College in Bangladesh. Jou nal o
Managemen and Business Educa ion 6(1): 43–57. h ps://doi.o g/10.35564/
jmbe.2023.0003.
Azhgaliye a, D., R. Mish a, T. Long, P. Mo gan, and W. Kodama. 2022. Impac s o
COVID-19 on Households’ Business, Employmen and School Educa ion:
E idence om Household Su ey in CAREC Coun ies. ADBI Wo king Pape ,
1335. h ps://doi.o g/10.56506/spwh1535.
Engzell, P., A. F ey, and M. D. Ve hagen. 2021. Lea ning Loss Due o School Closu es
Du ing he COVID-19 Pandemic. P oceedings o he Na ional Academy o
Sciences 118(17). h ps://doi.o g/10.1073/pnas.2022376118.
Haele mans, C. e al. 2022. Sha p Inc ease in Inequali y in Educa ion in Times o he
COVID-19 Pandemic. PLoS ONE 17: e0261114. h ps://doi.o g/10.1080/
00313831.2021.2006307.
Hanushek, E., and L. Woessmann. 2020. The Economic Impac s o Lea ning Losses.
OECD Educa ion Wo king Pape s, No. 225. OECD Publishing, Pa is,
h ps://doi.o g/10.1787/21908d74-en.
Iqbal, S. A., and H. A. Pa inos. 2023. Lea ning Du ing he Pandemic: E idence om
Uzbekis an. Policy Resea ch Wo king Pape s. h ps://doi.o g/10.1596/1813-
9450–10474.
Kodama, W., P. Mo gan, D. Azhgaliye a, T. Long, and K. Kim. Fo hcoming. Family
Business du ing he COVID-19 Pandemic in Asia: Role o Go e nmen
Financial Aid and Coping S a egies. Wo ld De elopmen .
Lichand, G., C. A. Do ia, O. Leal-Ne o, and J. P. Fe nandes. 2022. The Impac s o
Remo e Lea ning in Seconda y Educa ion Du ing he Pandemic in B azil.
Na u e Human Beha iou 6(8): 1180–1180. h ps://doi.o g/10.1038/s41562-022-
01420-9.
Maddawin, A., P. Mo gan, A. Pa k, D. Su yada ma, T. Q. Long, and P. Vandenbe g.
2024. Lea ning Dis up ions du ing he COVID-19 Pandemic: E idence om
Household Su eys in Sou heas Asia. In e na ional Jou nal o Educa ional
De elopmen , 107: 103053. h ps://doi.o g/10.1016/j.ijedude .2024.103053.
ADBI Wo king Pape 1450 W. Kodama e al.
16
Ma eau, J. F. 2020. Pos -COVID Educa ion in Kazakhs an: Hea y Losses and
Deepening Inequali y. Wo ld Bank Blogs. h ps://blogs.wo ldbank.o g/
eu opeandcen alasia/pos -co id-educa ion-kazakhs an-hea y-losses-and-
deepening-inequali y.
Meliboe a, N., H. Pa inos, and J. Teixei a. 2020. Uzbekis an: A Timely Response
on Lea ning Du ing School Closu es. h ps://blogs.wo ldbank.o g/
eu opeandcen alasia/uzbekis an- imely- esponse-lea ning-du ing-school-
closu es (accessed 22 Decembe 2023).
Minis y o Educa ion, Science, Cul u e, and Spo s o Geo gia. 2020. Pandemic and
Gene al Educa ion in Geo gia. h ps://mes.go .ge/mesgi s/1609844890_2020-
Annual- epo -ENG.pd (accessed 22 Decembe 2023).
Minis y o Science and Educa ion o Aze baijan. 2020. Aze baijan: How One Minis y
Found he Righ S a egy, Resou ces, and Technology o Quickly C ea e Online
Class ooms. h ps://www.edu.go .az/en/news-and-upda es/18629 (assessed
22 Decembe 2023).
Mosco iz, L., and D. E ans. 2022. Lea ning Loss and S uden D opou s du ing he
COVID-19 Pandemic: A Re iew o he E idence Two Yea s a e Schools Shu
Down. Washing on, DC: Cen e o Global De elopmen . h ps://www.cgde .o g/
si es/de aul / iles/lea ning-loss-and-s uden -d opou s-du ing-co id-19-pandemic-
e iew-e idence- wo-yea s.pd .
Pa inos, H. A. 2023. The Longe S uden s We e Ou o School, he Less They
Lea ned. Jou nal o School Choice 17(2): 161–175. h ps://doi.o g/10.1080/
15582159.2023.2210941.
Pa inos, H., Vegas, E., and Ca e -Rau, R. 2023. An Analysis o COVID-19 S uden
Lea ning Loss. In The Ox o d Resea ch Encyclopedia o Economics and
Finance, edi ed by Jona han Hamil on. Ox o d: Ox o d Uni e si y P ess.
Roman, M., and A. Plopeanu. 2021. The E ec i eness o he Eme gency eLea ning
du ing COVID-19 Pandemic. The Case o Highe Educa ion in Economics
in Romania. In e na ional Re iew o Economics Educa ion 37: 100218.
h ps://doi.o g/10.1016/j.i ee.2021.100218.
Swan, K. 2003 . Lea ning E ec i eness: Wha he Resea ch Tells Us. In Elemen s o
Quali y Online Educa ion, P ac ice and Di ec ion, edi ed by J. Bou ne and
J. C. Moo e. Needham, MA: Sloan Cen e o Online Educa ion, pp. 13–45.
Uni ed Na ions. 2020. Dis ance Lea ning in Kazakhs an unde Qua an ine.
h ps://www.as anaci ilse icehub.o g/uploads/co id/ iles/Dis ance%20lea ning_
Kazakhs an%20case.pd (accessed 22 Decembe 2023).
UNESCO. 2021. School Closu es and Regional Policies o Mi iga e Lea ning Loss Due
o COVID-19: A Focus on he Asia-Paci ic. Pa is: UNESCO Publishing.
UNESCO, UNICEF, and Wo ld Bank. 2021. The S a e o he Global Educa ion C isis: A
Pa hway o Reco e y. Pa is: UNESCO Publishing.
UNICEF. 2020, May. UNICEF Aze baijan Coun y O ice COVID-19 Si ua ion Repo
No. 5. h ps:// elie web.in / epo /aze baijan/unice -aze baijan-coun y-o ice-
co id-19-si ua ion- epo -no-5-1-may-2020.
ADBI Wo king Pape 1450 W. Kodama e al.
17
APPENDIX
Table A1: Income Le els in Local Cu ency o Each Coun y
Uzbekis an,
sum
≤ 1,200,000
1,200,001–
2,000,000
2,000,001–
3,200,000
< 3,200,000
No e: Since di e en coun ies ha e di e en numbe s o income classes, in ou empi ical analyses we eca ego ized he income classes in o ou g oups o make hem consis en ac oss
coun ies. See Azhgaliye a e al. (2022) o mo e de ails.
Tajikis an,
somoni
≤ 800
801–
1,400
1,401–
2,400
<
2,400
Pakis an,
upee
≤ 5,000
5,001–
15,000
15,001–
30,000
30,001–
60,000
> 60,000
Mongolia,
ug ug
≤ 500,000
500,001–
900,000
900,001–
1,100,000
1,100,001–
2,100,000
> 2,100,001
Ky gyz Republic,
som
≤ 6,000
6,001–
12,000
12,001–
20,000
> 20,000
Kazakhs an,
enge
≤ 60,000
60,001–
100,000
100,001–
150,000
150,001–
250,000
250,001–
300,000
> 300,000
Geo gia,
gel
≤ 299
300–599
600–999
1,000–
1,499
> 1,500
Aze baijan,
mana
≤ 820
820,1–
1,
035,0
1,
035,1
–
1,
240,0
1,
240,1
–
1,
650,0
> 1,
650,1
A ghanis an,
a n
≤ 5,
000
5,001–
15,000
15,001–
30,000
30,001–
60,000
> 60,000
Re ised Income
Class
1
2
3
4
O iginal Income
Class
1
2
3
4
5
6
