A me a-analysis on he e ec o echnology on he achie emen o
less ad an aged s uden s
☆,☆☆
Gio gio Di Pie o
a,b,1,*
, Jona an Cas a˜
no Mu˜
noz
c
a
Eu opean Commission- Join Resea ch Cen e, Edi icio Expo, Calle Inca Ga cilaso, 3, 41092, Se ille, Spain
b
Ins i u e o Labou Economics (IZA), Schaumbu g-Lippe-S aße 5-9, 53113, Bonn, Ge many
c
Uni e si y o Se ille, Depa men o Teaching and Educa ional O ganiza ion, Facul y o Educa ion, C/ Pi o ecnia, S/N, 41013, Se ille, Spain
ARTICLE INFO
Keywo ds:
Less ad an aged s uden s
S uden achie emen
Educa ional echnology
Me a-analysis
ABSTRACT
This pape p esen s a me a-analysis ha in es iga es he impac ha he educa ional use o digi al
echnologies has on less ad an aged s uden s’ achie emen . We use a comp ehensi e de ini ion
o his g oup o s uden s ha includes all s uden s in less de eloped coun ies as well as mo e
disad an aged s uden s in mo e de eloped coun ies. 740 es ima es om 72 s udies employing
expe imen al and quasi-expe imen al esea ch designs a e collec ed. O e all, educa ional ech-
nology ini ia i es a e ound o ha e a small, posi i e, s a is ically signi ican e ec ha emains
e en a e co ec ing o publica ion bias. Addi ionally, ou esul s indica e ha compu e -
assis ed lea ning and beha iou al in e en ions a e mo e e ec i e in aising he achie emen
o less ad an aged s uden s han simple access o echnology. In e es ingly, he e ec o hese wo
in e en ions appea s o be o a simila magni ude. Finally, he use o digi al echnologies is
associa ed wi h sligh ly g ea e achie emen s in ma h and science han humani ies.
1. In oduc ion
The expansion o echnology has a ec ed many a eas o ou li e, including educa ion. Digi al lea ning ools such as able s,
sma boa ds and online applica ions ha e become inc easingly impo an elemen s o eaching and cou se deli e y. The e is a lo o
e idence showing ha he in oduc ion o hese ools can imp o e child en’s eaching and lea ning expe iences. McEwan (2015)
a gues ha echnology-based in e en ions may be as e ec i e in aising s uden achie emen as well-known and popula policies such
as smalle class size, eache aining and pe o mance incen i es.
Discussions exis in he li e a u e conce ning he ela ionship be ween digi al echnologies and equi y in educa ional ou comes
(Wa schaue & Xu, 2018). On he one hand, he e a e conce ns ha mo e ulne able s uden s can miss ou on he bene i s ha hese
echnologies b ing. The use o digi al echnologies may be less e ec i e o low socio-economic s a us child en as hey end o ha e
☆
The au ho s would like o hank h ee anonymous e e ees o hei help ul and cons uc i e commen s. The usual disclaime
applies.
☆☆
Jona an Cas a˜
no Mu˜
noz acknowledges he suppo o he ‘Ram´
on y Cajal’ g an RYC 2020-030157 unded by MCIN (Minis e io de
Ciencia, Inno aci´
on y Uni esidades)/AEI (Minis e io de Ciencia, Inno aci´
on y Uni esidades)/10.13039/501100011033 and by “ESF In es ing in
you u u e”, and he Uni e si y o Se ille "VI Uni e si y esea ch plan" (VI plan p opio de in es igaci´
on) o his wo k.
* Co esponding au ho . Eu opean Commission- Join Resea ch Cen e, Edi icio Expo, Calle Inca Ga cilaso, 3, 41092, Se ille, Spain.
E-mail add esses: [email p o ec ed] (G. Di Pie o), [email p o ec ed] (J. Cas a˜
no Mu˜
noz).
1
The iews exp essed a e pu ely hose o he au ho and may no in any ci cums ances be ega ded as s a ing an o icial posi ion o he Eu opean
Commission.
Con en s lis s a ailable a ScienceDi ec
Compu e s & Educa ion
jou nal homepage: www.else ie .com/loca e/compedu
h ps://doi.o g/10.1016/j.compedu.2024.105197
Recei ed 1 May 2024; Recei ed in e ised o m 2 No embe 2024; Accep ed 11 No embe 2024
Compu e s & Educa ion 226 (2025) 105197
A ailable online 13 No embe 2024
0360-1315/© 2024 The Au ho s. Published by Else ie L d. This is an open access a icle unde he CC BY license
( h p://c ea i ecommons.o g/licenses/by/4.0/ ).
limi ed access o echnological in as uc u e (Hohl eld, Ri zhaup , Ba on, & Kemke hon, 2008), exhibi mode a e le els o digi al
compe ence (F aillon, Ainley, Schulz, F iedman, & Duckwo h, 2019), can coun on li le ex e nal suppo o he use o hese ech-
nologies (Reich, 2020), and a e less likely o adop a sel - egula ed lea ning s a egy (Yang, Cheng, & Chen, 2018).
On he o he hand, howe e , o he a gumen s sugges ha he applica ion o echnology o he educa ion sec o has he po en ial o
be e y bene icial o mo e disad an aged s uden s. Fi s , echnology may enable pe sonalised lea ning by ailo ing educa ional ex-
pe iences o indi idual s uden s’ needs and abili ies. This may be ad an ageous especially o s uden s who need ex a help (e.g.,
s uden s wi h lea ning di icul ies), ensu ing ha hey ecei e he suppo ha is igh o hem and empowe ing hem o choose when
and how hey lea n (Ogan e al., 2012; Wagne , 2016). Second, he inco po a ion o echnologies in o he lea ning p ocess may in-
c ease s uden engagemen . In e ac i e and mul imedia- ich digi al esou ces ( ha include, o ins ance, games, simula ion, quizzes)
may make lea ning enjoyable and exci ing, doing a be e job in cap u ing s uden s’ a en ion han adi ional class oom se ings.
S uden s lacking mo i a ions, who a e mo e likely o come om less ad an aged backg ounds (Fejes, 2012), may pa icula ly bene i
om his lea ning app oach. Thi d, echnology has b oken down physical ba ie s, enabling lea ne s o ha e access o high-quali y
educa ional esou ces i espec i e o hei geog aphical loca ion (e.g., h ough online lea ning pla o ms, ideo con e encing ools).
This is especially ele an o indi iduals om emo e and u al a eas who end o ace signi ican mo e challenges in ob aining
lea ning ma e ials han hose om u ban a eas.
Gi en he g owing impo ance o echnologies in educa ion, one needs o gain a be e unde s anding o hei ole in enhancing he
achie emen o all s uden s. I is impo an o know ha he de elopmen and di usion o digi al echnologies is no lea ing behind
hose om mo e ulne able g oups. The pu pose o his e iew is o analyse e idence om igo ous e alua ions on he e ec o
educa ional echnology (ed- ech) in e en ions on he academic pe o mance o less ad an aged s uden s.
1.1. P io e iews on echnologies and he achie emen o less ad an aged s uden s
Ea lie e iews a emp ing o summa ise exis ing e idence on he impac o digi al echnologies on he achie emen o less
ad an aged s uden s ha e ollowed wo di e en app oaches.
Fi s , a ew me a-analyses and sys ema ic e iews ha e in es iga ed how ed- ech in e en ions a ec he lea ning ou comes o
s uden s in disad an aged con ex s such as less de eloped coun ies. Some o hese s udies ha e ocused hei a en ion on he impac
o a speci ic se o echnologies. Fo ins ance, Majo , F ancis, and Tsapali (2021) ha e looked a he e ec o echnology-suppo ed
pe sonalised lea ning on academic ou comes o school-aged lea ne s in low- and middle-income coun ies. They ound ha hese
in e en ions ha e a s a is ically signi ican posi i e e ec on s uden s’ lea ning, epo ing an o e all e ec size o 0.18. O he s udies
ha e adop ed a b oade app oach, conside ing all ypes o echnologies. Fo example, Rod iguez-Segu a (2022) has a emp ed o
syn he ise he esul s o s udies analysing he e ec o any ype o ed- ech in e en ion on s uden pe o mance in less de eloped
coun ies. He obse es ha , while access o echnology in e en ions alone a e no su icien o enhance lea ning, in e en ions
cen ed a ound sel -led lea ning and imp o emen s in ins uc ion a e he mos p omising ones.
The second line o esea ch consis s in me a-analyses on he associa ion be ween ed- ech and s uden achie emen in which socio-
economic s a us is used as a mode a o a iable. Fo ins ance, Cheung and Sla in (2012 & 2013) conside K-12 s uden s in he US and
examine whe he he e any di e ences in he impac o echnology on s uden pe o mance in eading and ma h be ween s uden s
om high and low socio-economic s a us. No s a is ically signi ican di e ences ac oss socio-economic s a us we e ound.
1.2. The cu en s udy
The p esen me a-analysis in es iga es he impac o ed- ech in e en ions on academic ou comes among less ad an aged s uden s.
We ex end p e ious ele an wo k by employing a comp ehensi e de ini ion o his g oup o s uden s ha comp ises all s uden s in
less de eloped coun ies as well as mo e disad an aged s uden s in mo e de eloped coun ies.
To he bes o ou knowledge, he e is no me a-analysis examining he impac o digi al echnologies on he achie emen o mo e
disad an aged s uden s in mo e de eloped coun ies. This is impo an because, in con as o ea lie me a-analyses including socio-
economic s a us as a mode a o a iable and he eby employing mo e p i ileged s uden s as a con ol g oup, we use simila s uden s (i.
e., s uden s om mo e disad an aged backg ounds no being exposed o he ed- ech in e en ion) as a compa ison g oup. In addi ion
o s udies whe e he majo i y o he sample o he whole sample consis o mo e disad an aged s uden s in mo e de eloped coun ies,
we also conside s udies analysing s uden s in less de eloped coun ies. This allows us o ake a holis ic app oach o he mo e
disad an aged s uden s. Analysing and compa ing esul s om he a o emen ioned wo g oups o s udies is indeed an impo an alue
added o ou esea ch as his may p o ide in o ma ion on di e en ial e u ns o in es men in ed- ech in di e en pa s o he wo ld. In
less de eloped coun ies ed- ech may help o add ess issues such as low supply o quali ied eache s, eache s’ absen eeism,
2
sca ce
quali y lea ning ma e ials, and la ge s uden - o- eache a ios,
3
bu i may also play an impo an ole in mo e de eloped coun ies, o
ins ance by enhancing he quali y o educa ion o s uden s om u al a eas, boos ing s uden s’ mo i a ion, and pe sonalising eaching
p ac ices (Escue a, Nickow, O eopoulos, & Quan, 2020).
2
Chaudhu y, Hamme , K eme , Mu alidha an, and Roge s (2006) obse e ha in less de eloped coun ies he p opo ion o absen eache s
du ing unannounced isi s is 19%.
3
Fo ins ance, acco ding o he UNESCO da abase, while in 2018 he pupil- eache a io in p ima y educa ion was 15.3 in OECD coun ies, he
simila igu e in he leas de eloped coun ies ( ollowing he Uni ed Na ions’ de ini ion) was 37.2.
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
2
Since he e ms “mo e de eloped” and “less de eloped” coun ies ha e been used loosely in he li e a u e, i is impo an o p o ide
a wo king de ini ion o hese e ms. In his e iew, “mo e de eloped” coun ies e e o high-income o uppe middle-income coun ies
as de ined by he Wo ld Bank (WB)’s classi ica ion o coun ies by income le els. “Less de eloped” coun ies e e o low-income o
lowe middle-income coun ies, again as de ined by he WB’s classi ica ion o coun ies by income le els.
Simila ly, i is also impo an o p o ide a wo king de ini ion o he exp ession “educa ional echnology”. This e e s o hose digi al
ools and esou ces designed o deli e lea ning ma e ials, suppo o enhance s uden achie emen ha complemen , and no eplace,
in-pe son eaching (e.g., compu e games, lea ning so wa e, apps, ex messages). Fully online cou ses as well as eache - ocused ools
(e.g., lea ning analy ics, AI o esou ce gene a ion) a e excluded.
This me a-analysis seeks o add ess he ollowing h ee esea ch ques ions (RQ).
RQ1. Wha is he o e all impac o ed- ech in e en ions on he academic pe o mance o his b oade g oup o less ad an aged
s uden s?
RQ2. Is he impac o ed- ech in e en ions on achie emen di e en be ween all s uden s in less de eloped coun ies and mo e
disad an aged s uden s in mo e de eloped coun ies?
RQ3. Wha ype/s o ed- ech in e en ions is/a e mo e success ul in aising he achie emen o his b oade g oup o less ad an aged
s uden s?
2. Da a and me hods
2.1. Inclusion and exclusion c i e ia
Table 1 shows he en p ede ined inclusion and exclusion c i e ia de eloped and applied in he sc eening p ocess. We chose o
conside he pe iod om 2000 onwa ds because digi al educa ion gained momen um a he s a o he millennium. Acco ding o a
KPMG epo (Wildi-Yune & Co de o, 2015), he global e-lea ning ma ke has massi ely g own since 2000. Following he ecom-
menda ions by he Coch ane S a is ical Me hods G oup,
4
s udies ha e no been excluded pu ely based on he sample size
5
(G ainge,
2015). In addi ion o pee - e iewed jou nal a icles and schola ly book chap e s, we decided o conside also con e ence pape s,
epo s and wo king pape s. The a ionale behind his is o ha e a balanced pic u e o a ailable e idence gi en ha he g ey li e a u e
ep esen s an impo an ehicle o dissemina ing s udies wi h null o nega i e esul s ha migh no o he wise be dissemina ed (Paez,
2017). We only included s udies p esen ing e idence om expe imen al o quasi-expe imen al esea ch designs. These echniques,
which es causal hypo heses (Whi e & Saba wal, 2014), p o ide he mo e igo ous e idence o he e alua ion o ed- ech on s uden
academic ou comes. We es ic ed ou a en ion o s udies ocusing on p ima y (including kinde ga en), lowe and uppe seconda y
educa ion and using objec i e indica o s o measu e s uden academic ou comes. Achie emen in all subjec s (excep o digi al li -
e acy
6
) is conside ed.
2.2. Li e a u e sea ch
S udies included in ou me a-analysis a e iden i ied h ough h ee main s eps: 1) elec onic da abase sea ch, 2) ances y sea ch
ac oss he s udies selec ed a he end o he i s s ep, and 3) ances y sea ch ac oss p e ious ele an sys ema ic e iews and me a-
analyses.
In he i s s ep, ollowing he ecommenda ion ha me a-analyses and sys ema ic e iews should employ mul iple bibliog aphic
da abases o sea ch o ele an li e a u e (Ha a i, Pa ola, Ha well, & Riegelman, 2020), we used ou o hem (i.e., Web o Science,
Scopus, Educa ion In o ma ion Resea ch Cen e (ERIC) and Google Schola ). Ewald e al. (2022) ind ha sea ching wo o mo e
da abases educes he isk o missing eligible s udies and He nandez, Ma i, and Roman (2020) sugges ha a leas h ee sea ch
engines should be u ilised. While ERIC is an educa ion- ocused da abase, Web o Science and Scopus a e comp ehensi e gene al
da abases co e ing high-quali y publica ions (Fan & Beh, 2023). Bo ego and F oyd (2014) s ess he impo ance o using gene al
da abases in addi ion o subjec -speci ic ones when ca ying ou sys ema ic e iews. The a ionale o his is ha esea ch on a gi en
opic o in e es is o en no only published in specialis jou nals o epo se ies, bu also in gene alis jou nals o epo se ies.
Addi ionally, i is impo an o use Google Schola as his da abase can be use ul o inding g ey li e a u e (Haddaway, Collins,
Coughlin, & Ki k, 2015).
Appendix A se s ou he clus e s o keywo ds used o iden i y he s udies o be included in his me a-analysis. These s udies we e
ound using keywo ds co e ing i e di e en concep s: (1) he se ing o in e es (kinde ga en, p ima y and seconda y educa ion), (2)
4
I s membe s we e polled abou whe he i is app op ia e o a Coch ane sys ema ic e iew o exclude small s udies. 26 ou o 26 ep esen a i es
o ed agains his p oposal.
5
While some p e iously published me a-analyses (e.g., Zhang e al., 2020) do no conside s udies wi h a sample size o less han 5, his exclusion
c i e ion is also me in ou case (in ou me a-analysis he s udy by Aunio and Mononen (2018) has he lowes sample size, i.e., 22).
6
This is because digi al compe ences a e no always included among he basic skills expec ed o be lea ned by s uden s. Fo ins ance, in PISA
(P og amme o In e na ional S uden Assessmen ) basic skills a e eading, ma hema ics and science. While eading in p ima y g ades e e s
especially o ocabula y acquisi ion and ex comp ehension, in he in e media e and uppe g ade le els ac ual knowledge and unde s anding a e
p og essi ely expanded and inc easingly applied o ope a ional use.
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
3
he me hodology (expe imen al and quasi-expe imen al esea ch designs), (3) he exposu e (ed- ech in e en ions), (4) he ou come
(s uden achie emen ), and (5) he socio-economic condi ion (mo e disad an aged s a us). We employed Boolean e ms ‘OR’ and
‘AND’ o combine sea ches wi hin and be ween concep s, espec i ely. In ERIC, he sea ch e ms we e sea ched in he ‘abs ac ’ and
‘desc ip o ’. As ega ds Google Schola , we sea ched ‘wi h all o he keywo ds, anywhe e in he a icle’ and ocused ou a en ion on
he i s 100 esul s o he sea ch (Romanelli e al., 2021). In Scopus, he sea ch e ms we e sough in he ‘ i le’, ‘abs ac ’ and
‘keywo ds’. In Web o Science, he sea ch e ms we e sough only in he ‘abs ac ’. Ou sea ch, which ended in Decembe 2023,
deli e ed 141 hi s. A e emo ing duplica es (i.e., 5), he wo au ho s, wo king independen ly, sc eened he i les and he abs ac s o
he s udies. Following his, 105 i ems we e excluded. Nex , he ull ex o he emaining 31 s udies was e ie ed and ca e ully
examined. This exe cise was again conduc ed by bo h au ho s who independen ly classi ied he s udies as ele an and i ele an based
on he p ede ined inclusion and exclusion c i e ia. While in e nal consis ency was qui e s ong (Cohen’s Kappa was 0.88), s udies on
which he e was disag eemen we e discussed in dep h un il consensus was eached. 11 s udies we e ound as a esul o his i s s ep.
In he second s ep, we sc eened he e e ences sec ions o he 11 s udies selec ed in he i s s ep wi h he pu pose o inding
addi ional s udies (By on & Pos , 2016). Employing he ances y app oach, 10 ele an s udies we e added.
Finally, in he hi d s ep we expanded esul s om i s wo sea ches by e iewing he bibliog aphies o p e ious ele an sys-
ema ic e iews and me a-analyses.
7
51 ele an s udies we e iden i ied h ough his addi ional ances y sea ch.
Since he numbe o ele an s udies ound h ough ances y sea ches is signi ican ly highe compa ed o he se o pape s e ie ed
by he sea ch in elec onic da abases, ollowing he app oach o Wilke and Pyka (2024), we pe o med some amendmen s o he lis o
keywo ds displayed in Appendix A. Howe e , his did no lead o he inclusion o any addi ional ele an s udy, bu only inc eased he
numbe o i ele an pape s. The impo ance o using o he sou ces o in o ma ion in addi ion o elec onic da abases when conduc ing
e iews on he opic o educa ion and echnology has been highligh ed by Escue a e al. (2020) and Rod iguez-Segu a (2022).
Fu he mo e, in ou case an addi ional di icul y lies in he iden i ica ion o s udies in mo e de eloped coun ies ocusing on s uden s
wi h disad an aged backg ounds. This is pa icula ly challenging gi en ha in o ma ion abou s uden s’ socio-economic condi ions is
o en no included in he s udies’ i les, keywo ds and abs ac s.
Table 1
Inclusion and exclusion c i e ia.
C i e ion Included Excluded
Language English O he languages, e.g. Spanish, Ge man, F ench, Po uguese
Da a on an e ec size o
su icien in o ma ion o
calcula e i
Da a on an e ec size (o su icien in o ma ion o compu e i )
and i s s anda d e o (o -s a is ic, o p- alue, o su icien
in o ma ion o calcula e i )
Lack o da a on an e ec size (o insu icien in o ma ion o
compu e i ) o i s s anda d e o (o -s a is ic, o p- alue, o
insu icien in o ma ion o calcula e i )
Publica ion ype Pee - e iewed jou nals, schola ly book chap e s, con e ence
pape s, epo s o wo king pape s published be ween 2000
and 2023
Mas e ’s and PhD disse a ions as well as pee - e iewed
jou nals, schola ly book chap e s, con e ence pape s, epo s
o wo king pape s published ea lie han 2000 and la e han
2023
Measu emen o s uden
achie emen
Objec i e indica o s including s anda dized es sco es as well
as sco es om es s de eloped by eache s o esea che s
Achie emen s in digi al li e acy, s uden ’s sel -assessed
g ades, non-cogni i e skills, and o he ou comes (e.g., school
a endance)
Educa ion con ex P ima y (including kinde ga en), lowe and uppe seconda y
educa ion
Highe educa ion
Resea ch design Expe imen al (i.e., andomized con olled ials (RCTs)) o
quasi-expe imen al (i.e., p e- es pos - es s udy, eg ession
discon inui y, ins umen al a iable, p opensi y sco e,
di e ence in di e ences) esea ch designs
Non-expe imen al esea ch designs
Resea ch se ing Clea dis inc ion be ween a ea ed g oup (exposu e o ed- ech
in e en ion) and a con ol g oup (no exposu e o ed- ech
in e en ion)
Compa ison o wo al e na i e ea men s wi hin he ea ed
g oup
Disad an aged con ex E idence om:
coun ies classi ied by he WB as high- o uppe -middle income
coun ies a he ime o he ed- ech in e en ion and whe e,
ollowing he app oach o Die ichson, Klin Jø gensen, and
Filges (2017), a leas 50% o he sample pa icipan s a e om
disad an aged backg ounds de ined in e ms o pa en al
occupa ion, educa ion o income, access o ee o educed
school meals, a ea o esidence (e.g., u al o emo e a eas,
low-income egions), mig an o mino i y s a us
o
coun ies classi ied by he WB as low- o lowe -middle income
coun ies a he ime o he ed- ech in e en ion
E idence om coun ies classi ied by he WB as high- o uppe -
middle income coun ies a he ime o he ed- ech
in e en ion no ocusing on mo e disad an aged s uden s o
ocusing on low-pe o ming s uden s o s uden s wi h lea ning
p oblems bu wi h no in o ma ion on hei socio-economic
backg ound
De ini ion o educa ional
echnology in e en ion
In e en ions explici ly aiming a imp o ing s uden
achie emen h ough he use o echnologies. They should
complemen in-pe son lea ning and no ully eplacing i
In e en ions consis ing o cou ses en i ely deli e ed online
Sample size Any sample size
7
These s udies we e selec ed among hose excluded in he i s and second s eps.
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
4
A o al o 72 s udies was included in his me a-analysis.
8
The li e a u e sea ch and he sc eening p ocedu e a e summa ised in
Fig. 1
9
2.3. S udy coding
Be o e coding, he wo au ho s c ea ed a coding manual ha p o ides guidelines on how o consis en ly ex ac in o ma ion abou
e ec sizes and mode a o a iables om he s udies included in he sample. Commen s and sugges ions on his coding manual we e
gi en by a esea che in he a ea o educa ion and echnology. The e ised e sion o he coding manual was hen es ed by compa ing
he codes gene a ed by he wo au ho s o 5 andomly selec ed s udies. Following some mino modi ica ions, he wo au ho s, wo king
independen ly, pe o med he coding on he emaining 67 s udies. The Kappa alue be ween hem was 0.83, indica ing ela i ely good
consis ency. Howe e , when disag eemen a ose, he s udies in ques ion we e e-examined by bo h au ho s oge he un il a inal
ag eemen was eached.
2.3.1. E ec size calcula ions
In an a emp o compa e he es ima es o a ious ed- ech in e en ions on di e en academic ou comes, ollowing he app oach o
simila ea lie sys ema ic e iews (e.g., Escue a e al., 2020) and me a-analyses (e.g., Ni, Cheung, & Shi, 2022), we used Cohen’s d.
Al hough esul s om di e en s udies a e ne e ully compa able, Cohen’s d does o e aluable insigh s in o he o e all magni ude o
impac ac oss di e se p og amme con ex s. No only is Cohen’s d he mos widely employed e ec size o measu e he magni ude o
g oup di e ences (McCoach & Siegle, 2009), bu i is also he mos used among he s udies in ou sample ha epo e ec sizes. In
hose s udies whe e Cohen’s d alues a e no p o ided, i was possible o compu e hese using in o ma ion he ein con ained. Cohen’s
d was calcula ed by di iding he mean di e ence in pe o mance be ween ea men (exposu e o ed- ech in e en ion) and con ol
condi ions (no exposu e o ed- ech in e en ion) by he pooled s anda d de ia ion (d =M1−M2
Sp; Sp=
(n1−1)S2
1+(n2−1)S2
2
(n1−1)+(n2−1)
√, wi h M
1
, M
2
, S
p
,
n
1
, n
2
, S
1
, S
2
, deno ing he means o he ea men g oup and con ol g oup, pooled s anda d de ia ion o bo h g oups, he sample sizes
o he ea men g oup and con ol g oup, and he s anda d de ia ion o he ea men g oup and con ol g oup). Addi ionally, i e ec
sizes a e epo ed using Hedge’s g, hese we e con e ed in o Cohen’s d s a is ics using he o mula in Ha e , Cuijpe s, Fu ukawa, and
Ebe (2021) (d =g
1−(3
4(n1+n2−2)−1)⎞
⎟
⎟
⎠. Finally, Cohen’s d was calcula ed om p e- es pos - es designs employing he o mula in Mo is
(2008).
Cohen’s d s anda d e o is also missing in a numbe o s udies. A ew s a egies ha e been employed o add ess his si ua ion. Fo
example, i in o ma ion on sample sizes is a ailable, Cohen’s d s anda d e o was calcula ed h ough he o mula epo ed in Coope
and Hedges (1994). Whe e in o ma ion on sample sizes is no included in he s udies bu exac p- alues a e ins ead epo ed, he
o mula p o ided by Higgins and G een (2011) was employed o calcula e s anda d e o s.
2.3.2. Mode a o a iables
Fo each e ec size, we coded se e al mode a o a iables
10
, ha is, ac o s po en ially in luencing he size o he impac o ed- ech
in e en ions on s uden achie emen .
a) Type o publica ion
We dis inguished be ween pee - e iewed jou nal a icles and o he s udies. Publica ion ype is a common mode a o in me a-
analysis. I is expec ed ha pee - e iewed jou nal a icles a e o highe scien i ic igou and less likely o include ypos o e o s in
he epo ed esul s.
b) Publica ion yea
Publica ion yea is ye ano he ypical mode a o a iable in me a-analysis. In ou s udy, his ac o may p o ide an indica ion
abou how he o e all e ec i eness o ed- ech applica ions has changed o e ime
11
(Cheung & Sla in, 2013). This is ele an because
8
In o ma ion on he iming o he ed- ech in e en ion is missing in some o he selec ed s udies (e.g., Ve hallen & Bus, 2010). Howe e , hese
s udies ake place in coun ies whose income classi ica ion by he WB has no changed o e ime (e.g., Ne he lands has always been a high-income
coun y). This means ha he exp ession “a he ime o he ed- ech in e en ion” in he eigh inclusion and exclusion c i e ion (see Table 1) is
edundan in hese cases. Please no e also ha , al hough he i le o he s udies by Be linski e al. (2021) and Be linski e al. (2022) is he same, he
la e is a e ised e sion o he o me and includes di e en es ima es.
9
As illus a ed in Fig. 1, we we e able o ha e access o he ull ex o all he s udies iden i ied in each s ep o he li e a u e sea ch.
10
In o ma ion on all mode a o a iables was a ailable o all he e ec sizes included in ou sample.
11
Al hough he iming o he ed- ech in e en ions is he app op ia e a iable o look a in assessing how he o e all e ec i eness o hese in-
e en ions has changed o e ime, as s a ed ea lie , his in o ma ion is missing in some o he s udies included in ou sample. The a ionale o using
yea o publica ion is ha i may be co ela ed wi h he iming o he ed- ech in e en ions.
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
5
one migh expec ha ed- ech in e en ions will become mo e e ec i e as ime goes on because o he con inued and signi ican
ad ances in echnology.
c) Type o ed- ech in e en ion
Following Escue a e al. (2020), ed- ech in e en ions a e classi ied in o h ee ca ego ies: access o echnology, compu e -assis ed
lea ning (CAL), and beha iou al in e en ions.
12
The i s ca ego y comp ises measu es p o iding o acili a ing he p o ision o
compu e s/ able s/so wa e and he in e ne o s uden s. This ype o in e en ion is common in u al a eas and in less de eloped
coun ies whe e many s uden s lack he echnology o in e ne access equi ed o academic lea ning. Fo example, C is ia, Iba a an,
Cue o, San iago, and Se e in (2017) looked a he impac o a p og amme p o iding lap ops o child en in Pe u, whe eas Leu en,
Lindahl, Oos e beek, and Webbink (2007) analysed he e ec s o a compu e subsidy in he Ne he lands, designed o schools wi h a
Fig. 1. Flowcha illus a ing he e iew selec ion p ocess.
12
Fou ca ego ies a e included in he o iginal classi ica ion de eloped by Escue a e al. (2020). Howe e , he ca ego y o online cou ses canno be
conside ed in his wo k because s udies e alua ing in e en ions consis ing o cou ses en i ely deli e ed h ough he in e ne a e excluded om ou
analysis (see inclusion and exclusion c i e ia).
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
6
la ge p opo ion o disad an aged s uden s.
The second ca ego y (CAL) comp ises in e en ions u ilizing echnology o supplemen adi ional class oom ins uc ion. Common
examples o his ca ego y a e educa ional so wa e and applica ions o enhance ma h (e.g., Ru he o d e al., 2014) and language (e.g.,
Maca uso, Hook, & McCabe, 2006) skills. Howe e , CAL can also include mo e sophis ica ed ools like in elligen u o ing sys ems,
educa ional games, i ual eali y en i onmen s, o e en so wa e acili a ing online u o ing p og ammes (e.g., Go aza , Hupkau, &
Rold´
an, 2022). Finally, one should no e ha CAL includes also Compu e -Aided Ins uc ion (CAI).
Beha iou al in e en ions a e de ined as echnology-media ed in e en ions designed o o e come o compensa e o non-
cogni i e skill de ici s ha lead o nega i e s uden academic ou comes. They a e o en a ge ed a inc easing pa en al in ol emen
in hei child en’s lea ning ac i i ies (e.g., p og ammes h ough which pa en s ecei e egula ex messages con aining ips on how o
engage hei child en in eading ac i i ies).
13
P e ious e iew s udies (e.g., Escue a e al., 2020) ques ion he ole o access o echnology in posi i ely a ec ing s uden academic
achie emen . On he o he hand, hey highligh he po en ial bene i s associa ed wi h he o he wo ypes o in e en ions, especially
CAL.
d) Le el o educa ion
In ou analysis, we dis inguished be ween p ima y (including kinde ga ens) and seconda y educa ion.
14
Ea lie me a-analyses
show mixed esul s abou he ole o educa ional le el in explaining a ia ions in he e ec s o ed- ech in e en ions on s uden
pe o mance. In hei me a-analysis, Kazu and Ku o˘
glu Yalçın (2022) show ha he e is no s a is ically signi ican di e ence in he
impac o lipped class oom lea ning on s uden pe o mance among s udies ocusing on elemen a y, seconda y and pos seconda y
educa ion. On he o he hand, Ran, Kasli, and Secada (2021) obse e ha compu e echnology in e en ions ha e a la ge e ec o
kinde ga en and p ima y school s uden s han o high school s uden s. Digi al ools can play a key ole in enhancing he lea ning
ou comes o s uden s in he i s s ages o hei educa ion. They enable s uden s o enjoy and ha e a posi i e a i ude owa ds lea ning,
p omo e engagemen and con ibu e o he o ma ion o p oblem-sol ing skills (Sun, Chen, & Ruokamo, 2021).
e) Subjec a ea
Following he app oach o se e al p e ious me a-analyses (e.g., Di Pie o, 2023), we g ouped subjec s in o h ee di e en b oad
ca ego ies: ma h/science, humani ies and a mix ca ego y. In addi ion o gene al ma h and science, he i s ca ego y co e s biology as
well as a ious a eas o ma h (e.g., algeb a) and speci ic ma h unc ions (e.g., sub ac ion). The e is also one s udy (Beg, Lucas, Halim,
& Sai , 2022) using combined ma h and science es sco es. As o he ca ego y o humani ies, his comp ises he subjec s o language,
o eign language, and social s udies. Di e en aspec s o language lea ning a e conside ed (e.g., eading, wo d ecogni ion). Finally,
he mix ca ego y e e s o es s combining di e en subjec s belonging o he i s wo ca ego ies oge he (e.g., ma h +Chinese in Mo,
Huang, e al., 2015), GPA o o e all academic achie emen .
The hypo hesis o he e ogeneous e ec s o ed- ech in e en ions by subjec is suppo ed by he indings o se e al s udies. Bulman
and Fai lie (2016) a gue ha i is easie o de elop e ec i e so wa e packages o ma h han o language. In he me a-analysis by
Chauhan (2017), echnological applica ions a e ound o ha e a smalle impac on s uden lea ning in social s udies compa ed o
language, ma h, and science and echnology. Zheng, Wa schaue , Lin, and Chang (2016) conclude ha , while he impac o one- o-one
lap op p og ammes on academic achie emen is posi i e ac oss i e di e en subjec a eas, he la ges e ec is ound o science.
On he o he hand, in hei me a-analysis Majo e al. (2021) ind ha in low- and middle-income coun ies echnology-suppo ed
lea ning ini ia i es a e equally impo an in enhancing s uden pe o mance in ma h and li e acy. Simila e ec s in ma h and li e acy
a e also ound by Kim, Gilbe , Yu, and Gale (2021).
) Geog aphical loca ion
As ou lined in he inclusion and exclusion c i e ia, we selec ed s udies examining s uden s in less de eloped coun ies as well as
s udies on mo e de eloped coun ies whe e mo e disad an aged s uden s make up he whole sample o i s majo i y. Based on he
li e a u e, i is unclea whe he ed- ech in e en ions can be mo e e ec i e o he o me o la e g oup o s uden s. On he one hand,
one may a gue ha echnology is especially bene icial in less de eloped coun ies whe e eache s a e, on a e age, less p epa ed and
mo e absen han hose in mo e de eloped coun ies. The e o e, in he o me coun ies, as s a ed by Bane jee, Cole, Du lo, and Linden
(2007), compu e s can eplace eache s wi h less mo i a ion and aining. A simila bu mo e gene al a gumen is ad anced by Bulman
13
In s udies examining he e ec o access o echnology p og ammes he un ea ed g oup consis s o s uden s who did no ecei e any compu e /
able /so wa e, we e no connec ed o he in e ne o we e no inancially incen i ised o pu chase educa ional echnology (Leu en e al., 2007). In
s udies looking a he impac o echnology-enabled beha iou al in e en ions he con ol g oup is composed by indi iduals (e.g., pa en s, eache s)
ha we e no exposed o he ea men . Finally, in s udies assessing he e ec i eness o CAL he con ol g oup is made up by s uden s (classes/-
schools) ha did no use echnology o supplemen in- and ou -o -class eaching.
14
Al hough his is a b oad ca ego iza ion, one needs o conside ha i is e y di icul o dis inguish be ween lowe p ima y and uppe p ima y
educa ion and be ween lowe seconda y and uppe seconda y educa ion in a wide c oss-na ional con ex . Fo example, in Tanzania junio seconda y
educa ion ends a g ade 11 (Seo, 2017), whe eas in he US senio high school begins a g ade 9.
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
7
and Fai lie (2016). They claim ha in less de eloped coun ies ed- ech may make up o he lowe quali y o educa ion. On he o he
hand, howe e , as sugges ed by DeWi and Alias (2019), se e al issues may make i mo e di icul o implemen ed- ech p og ammes
in less de eloped coun ies, which in u n may unde mine hei e ec i eness. Fi s , hese coun ies a e mo e likely o expe ience
i egula elec ical supply and slow in e ne speeds. This, o ins ance, could educe he po en ial bene i s associa ed wi h he use o a
lea ning so wa e ha can only be accessed online. Second, he e is also he possibili y ha speci ic ins uc ional ma e ials deli e ed
h ough echnology o e en some lea ning ools may u n ou no o be sui able o he cul u e, cus oms and mo ale o a less de eloped
coun y. Fo example, be ween 2005 and 2017 in Malaysia eache s s ongly objec ed o he in oduc ion o mobile phones o lea ning
and eaching as hey a gued ha hei use would cause a lo o disciplina y p oblems. Thi d, in he li e a u e i is o en sugges ed ha o
be success ul in enhancing s uden achie emen , echnology needs o be p ope ly in eg a ed in o he eache s’ ins uc ion and cu -
iculum (Rod iguez-Segu a, 2022). Howe e , his is less likely o occu in less de eloped coun ies whe e many eache s ha e no
ecei ed aining in he pedagogies o compu e s in educa ion. Addi ionally, in hese coun ies eache s may be especially eluc an o
acqui e hese compe encies as hey pe cei e his change o be a h ea o adi ional eaching p ac ices (Hinos oza, 2018).
g) Con ol/s
Finally, we coded a a iable which equals one i he model om which he e ec size is ex ac ed includes one o mo e con ol
a iables, and ze o o he wise. One would expec he p esence o con ol a iables o educe he magni ude o he e ec o ed- ech
in e en ions on s uden achie emen .
2.4. Sample cha ac e is ics
Table 2 p esen s he s udies included in he da ase . Fo each s udy, we epo in o ma ion on he au ho (s), yea o publica ion,
coun y examined, numbe o he e ec sizes collec ed and hei mean alue.
The da ase used o he me a-analysis includes 740 e ec sizes om 72 s udies published be ween 2004 and 2023. Each s udy
included in ou da ase con ains a numbe o e ec sizes ha a y om 1 o 48. The s udies co e a o al o 26 coun ies. The la ges
sou ce coun ies a e India (168 e ec sizes) and he US (144 e ec sizes).
Appendix B epo s he de ini ion and he desc ip i e s a is ics o he a iables used in ou analysis. I also indica es he numbe o
e ec sizes o each mode a o a iable and he numbe o s udies ha include each mode a o a iable.
2.5. Risk o bias assessmen
The isk o bias in each o he s udies included in ou sample was independen ly assessed by he wo au ho s. In e - a e eliabili y
epo ed by Cohen’s Kappa was a he high (0.78), bu any disag eemen was esol ed h ough discussions. While he Risk O Bias in
Non- andomised S udies o In e en ions (ROBINS-I) (S e ne e al., 2016) was used o e alua e s udies wi h a quasi-expe imen al
design, Ve sion 2 o he Coch ane Risk o Bias ool o andomised ials (RoB 2) (S e ne e al., 2019) was employed o assess he
o he s udies. As shown in Appendix C, a a he common issue wi h he la e g oup o s udies lies in limi ed in o ma ion on, o
p oblems wi h, he andomisa ion p ocess. An addi ional issue is he use o an un eliable measu e o s uden achie emen . As ega d
s udies using quasi-expe imen al app oaches, as epo ed in Appendix D, sample selec ion issues and missing da a p oblems (e.g., no
ou come da a o some membe s o he con ol ( ea ed) g oup) a e he mos common sou ces o po en ial bias.
2.6. Models and es ima o s
2.6.1. Model o compu e summa y e ec size es ima es
The ixed e ec s (FE) and andom e ec s (RE) models a e wo app oaches equen ly employed in me a-analysis. They a e based on
di e en assump ions. The FE model assumes ha he e is one ue e ec size common o all s udies and ha all di e ences in he
obse ed e ec s can be asc ibed o wi hin-s udy sampling e o . In con as o his, he RE model assumes ha he e ec size may a y
be ween s udies no only as a esul o he wi hin-s udy sampling e o , bu also because he e is he e ogenei y in ue e ec s be ween
s udies. This addi ional a iabili y is ypically modelled using a be ween-s udy a iance pa ame e (o en called
τ
2). Conside ing he
di e en cha ac e is ics o he s udies included in ou sample, i is di icul o assume ha he e is a common ue e ec sha ed by all
s udies. Thus, i is an icipa ed ha he RE model would be mo e app op ia e, i.e., es ima ing he mean o he dis ibu ion o ue e ec s.
Speci ically, in line wi h he app oach o Kaise and Menkho (2020), we es ima e he mean o he dis ibu ion o ue e ec s using a
RE me a-analysis based on a Robus Va iance Es ima ion (RVE) de eloped by Tanne -Smi h and Tip on (2014). The RVE app oach
enables o accoun o he possibili y ha mul iple e ec sizes om he same s udy a e co ela ed be ween each o he . The ad an ages
o his me hod a e ha he e is no need o elimina e any e ec size ( o ensu e hei s a is ical independency) and no in o ma ion is
needed on he in e co ela ion be ween e ec sizes wi hin s udies.
2.6.2. Me hods o es and co ec o publica ion bias
The e a e wo ac o s sugges ing a po en ial bias owa ds posi i e esul s. Fi s , manusc ip s inding s a is ically signi ican posi i e
indings a e mo e a ac i e o esea che s, e e ees and edi o s (Begg & Belin, 1988). Second, companies signi ican ly in es ing in
ed- ech p oduc s and se ices may also welcome esea ch esul s sugges ing a posi i e e ec o echnology on s uden lea ning (Escue a
& Holloway, 2019).
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
8
Table 2
Sou ces o me a-analysis.
Au ho s Yea o publica ion Coun y Numbe o e ec sizes collec ed Mean e ec size
Ab ami e al. 2016 Kenya 3 0.38
Amendum e al. 2011 US 7 0.49
Aunio & Mononen 2018 Finland 9 −0.17
Bai e al. 2016 China 6 0.05
Bai e al. 2023 China 4 0.18
Bake e al. 2017 US 6 −0.01
Bando e al. 2017 Hondu as 6 −0.06
Bane jee e al. 2007 India 13 0.21
Ba ow e al. 2009 US 8 0.2
Beg e al. 2022 Pakis an 18 0.06
Be gman 2021 US 12 0.12
Be gman & Chan 2021 US 14 0.03
Be gman & Roge s 2016 US 4 0.06
Be linski e al. 2022 Chile 5 0.09
Be linski e al. 2021 Chile 1 0.08
Beue mann e al. 2015 Pe u 2 0.07
Bianchi e al. 2022 China 4 0.19
Blimpo e al. 2020 Gambia 6 0.51
Bo zekowski 2018 Tanzania 7 0.13
Bo zekowski e al. 2019 India 48 0.16
B own e al. 2020 Sudan 2 0.95
Büchel e al. 2022 El Sal ado 6 0.28
Ca dim e al. 2023 Angola 3 0.02
Ca illo e al. 2010 Ecuado 30 0.12
Chambe s e al. 2006 US 10 0.14
Chambe s e al. 2008 US 10 0.44
Cillie s e al. 2022 Sou h A ica 9 0.05
C is ia e al. 2017 Pe u 24 0.02
de Hoop e al. 2023 Zambia 21 0.25
De ksen e al. 2020 Malawi 6 0.02
Du lo e al. 2012 India 12 0.16
Go aza e al. 2022 Spain 10 0.38
Ibe & Abamuche 2019 Nige ia 1 0.97
I o e al. 2021 Cambodia 6 0.69
Johns on & Ksoll 2022 Ghana 29 0.16
K a & Mon i-Nussbaum 2017 US 20 0.11
Kuma & Meh a 2018 India 1 0.16
Lai e al. 2015 China 4 0.1
Lai e al. 2016 China 8 0.14
Lai e al. 2013 China 6 0.12
Leh e e al. 2019 Senegal 21 0.24
Leu en e al. 2007 Ne he lands 18 −0.04
Linden 2008 India 26 −0.1
Lineba ge e al. 2010 US 8 0.29
Lysenko e al. 2019 Kenya 15 0.58
Maca uso e al. 2006 US 4 0.96
Malamud e al. 2019 Pe u 6 0.01
McManis & McManis 2016 US 2 0.38
Mille & Robe son 2011 Sco land 1 0.09
Mo e al. 2020 China 4 0.06
Mo e al. 2015(a) China 24 0.12
Mo e al. 2013 China 4 0.06
Mo e al. 2014 China 6 0.18
Mo e al. 2015(b) China 8 0.19
Mu alidha an e al. 2019 India 29 0.26
Naik e al. 2020 India 36 0.13
N aila & Mba aka 2023 Malawi 2 2.18
Pipe e al. 2016 Kenya 6 0.26
Pi ch o d 2015 Malawi 13 0.41
Riley 2018 Uganda 16 0.05
Rouse & K uege 2004 US 13 0.15
Ru he o d e al. 2014 US 17 0.08
San ana e al. 2019 Chile 7 0.42
Schac e & Jo 2016 US 1 1.09
Schac e e al. 2016 US 1 0.57
Seo 2017 Tanzania 24 0.02
Se en 2023 US 4 0.12
Silande e al. 2016 US 3 0.12
(con inued on nex page)
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
9
his way, one can maximise he bene i s, po en ially leading o syne gis ic ou comes. Fu he mo e, gi en he low cos o imple-
men a ion o beha iou al in e en ions, such a combina ion may u n ou o be a p omising cos -e ec i e app oach o using ech-
nologies o educe educa ional inequali ies.
5. Limi a ions and u u e di ec ions
This a icle has h ee limi a ions ha may ha e implica ions o u u e esea ch. Fi s , we we e unable o di e en ia e be ween
lowe and uppe p ima y o seconda y educa ion because o he challenges o conduc ing his analysis in a wide ansna ional con ex .
Fu u e esea ch could look a hese dis inc ions ocusing on a g oup o mo e homogeneous coun ies (e.g., he EU). Second, while we
ha e no conside ed he impac o echnology on digi al li e acy, he e is e idence sugges ing a po en ial link be ween he wo, e en
when he in e en ion is limi ed o echnology p o ision (e.g., Beue mann, C is ia, Cue o, Malamud, & C uz-Aguayo, 2015; Malamud,
Cue o, C is ia, & Beue mann, 2019). Thi d, we acknowledge ha educa ional echnology is a as -e ol ing a ea and some o ou
indings may possibly change in he u u e due o an inc ease in he numbe o s udies examining he e ec o eme ging echnologies on
he achie emen o less ad an aged s uden s. Fo ins ance, he s udy by Ga cía-Vandewalle Ga cía, Ga cía-Ca mona, T ujillo To es,
and Moya-Fe n´
andez (2022), which elies on he iews o eigh in e na ional expe s in educa ion, concludes ha new echnologies
can play an impo an ole in imp o ing he achie emen o s uden s in disad an aged con ex s.
CRediT au ho ship con ibu ion s a emen
Gio gio Di Pie o: W i ing – e iew & edi ing, W i ing – o iginal d a , Me hodology, In es iga ion, Fo mal analysis, Da a cu a ion,
Concep ualiza ion. Jona an Cas a˜
no Mu˜
noz: W i ing – e iew & edi ing, Me hodology, In es iga ion, Da a cu a ion,
Concep ualiza ion.
Decla a ions o compe ing in e es
None.
Appendix A. Sea ch e ms
Ca ego y Keywo ds
Se ing o in e es p ima y (OR educa ion OR s uden OR school) OR elemen a y (OR educa ion OR s uden OR school) OR seconda y (OR educa ion OR
s uden OR school) OR p eschool OR kinde ga en OR middle (school OR s uden )
AND
Exposu e compu e OR mobile OR lap op OR able OR so wa e OR in e ne OR apps (OR applica ions) OR digi al OR i ual OR echnolog* OR
ex messages OR SMS
AND
Me hodology expe imen al OR quasi-expe imen al OR ins umen al OR eg ession discon inui y OR andomized con ol ial OR andomised con ol
ial OR RCT OR p opensi y sco e OR di e ence-in-di e ence*
AND
Ou come s uden (OR academic OR scholas ic) pe o mance (OR achie emen OR lea ning OR ou come) OR (pupil (OR academic OR scholas ic)
pe o mance (OR achie emen OR lea ning OR ou come) OR es sco e OR lea ning abili y
AND
Socio-economic
condi ion
less de eloped coun * OR de eloping coun * OR unde de eloped coun * OR unde -de eloped coun * OR low* GDP coun * OR low*
socio-economic (OR socioeconomic) OR disad an aged OR less ad an aged OR less p i ileged OR unp i ileged OR ulne able OR less
a luen OR less weal hy OR mino i y OR low*-income OR poo * income OR mig an OR u al OR emo e
Appendix B. De ini ion and desc ip i e s a is ics
Va iable name Va iable desc ip ion Unweigh ed
Mean (S anda d
de ia ion)
(1)
Weigh ed (by he in e se o
he numbe o es ima es
epo ed in each s udy)
Mean (S anda d de ia ion)
(2)
Numbe o e ec sizes/
numbe o s udies including
each mode a o a iable
(3)
E ec size Es ima ed e ec size (Cohen’s d) 0.177 0.260 740/72
(0.288) (0.394)
E ec size’s s anda d e o Es ima ed s anda d e o o Cohen’s d0.114 0.125 740/72
(0.099) (0.102)
Yea o publica ion Yea o publica ion o he s udy whe e he
e ec size is ex ac ed
2016.197 2016.386 740/72
(4.916) (4.755)
(con inued on nex page)
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
16
(con inued)
Va iable name Va iable desc ip ion Unweigh ed
Mean (S anda d
de ia ion)
(1)
Weigh ed (by he in e se o
he numbe o es ima es
epo ed in each s udy)
Mean (S anda d de ia ion)
(2)
Numbe o e ec sizes/
numbe o s udies including
each mode a o a iable
(3)
Con ol/s Dummy, 1 i he model om which he e ec
size is ex ac ed includes one o mo e con ol
a iables, 0 o he wise
0.696 0.592 515/52
(0.460) (0.492)
Subjec a ea
Ma h/Science Dummy, 1 i he subjec a ea is ma h o science,
0 o he wise
0.451 0.501 334/56
(0.498) (0.500)
Humani ies (base
ca ego y)
Dummy, 1 i he subjec a ea is humani ies,
0 o he wise
0.423 0.406 313/51
(0.494) (0.491)
Mix Dummy, 1 i s uden achie emen in di e en
subjec a eas is conside ed, 0 o he wise
0.126 0.093 93/16
(0.332) (0.290)
Le el o educa ion
P ima y (base ca ego y) Dummy, 1 i he educa ion le el is p ima y
school (o kinde ga en), 0 o he wise
0.727 0.752 538/54
(0.446) (0.432)
Seconda y Dummy, 1 i he educa ion le el is seconda y
school, 0 o he wise
0.250 0.221 185/16
(0.433) (0.415)
P ima y and Seconda y Dummy, 1 i he educa ion le el is bo h
p ima y and seconda y school, 0 o he wise
0.023 0.028 17/2
(0.150) (0.164)
Type o ed- ech in e en ion
CAL Dummy, 1 i he ed- ech in e en ion is a
compu e -assis ed lea ning p og amme,
0 o he wise
0.757 0.742 560/54
(0.429) (0.438)
Beha iou al in e en ions Dummy, 1 i he ed- ech in e en ion is
beha iou al in na u e, 0 o he wise
0.138 0.152 102/11
(0.345) (0.359)
Access o echnology (base
ca ego y)
Dummy, 1 i he ed- ech in e en ion p o ides
s uden s wi h access o echnology and/o he
in e ne , 0 o he wise
0.105 0.107 78/10
(0.307) (0.309)
Geog aphical loca ion
S uden s in less de eloped
coun ies
(base ca ego y)
Dummy, 1 i he e ec size is ex ac ed om a
s udy examining s uden s in less de eloped
coun ies, 0 o he wise
0.604 0.476 447/34
(0.489) (0.500)
Mo e disad an aged
s uden s in mo e
de eloped coun ies
Dummy, 1 i he e ec size is ex ac ed om a
s udy examining mo e disad an aged s uden s
in mo e de eloped coun ies, 0 o he wise
0.396 0.524 293/38
(0.489) (0.500)
Type o publica ion
Pee - e iewed jou nal
a icles
Dummy, 1 i he e ec size is ex ac ed om a
pee - e iewed jou nal a icle, 0 o he wise
0.788 0.807 583/58
(0.409) (0.395)
O he publica ions
(base ca ego y)
Dummy, 1 i he e ec size is ex ac ed om a
publica ion ha is no a pee - e iewed jou nal
a icle, 0 o he wise
0.212 0.193 157/14
(0.409) (0.396)
Appendix C. Quali y assessmen o RCTs (RoB 2)
S udy Risk o bias a ising
om he
andomisa ion p ocess
Risk o bias due o
de ia ions om he
in ended in e en ion
Missing
ou come
da a
Risk o bias in he
measu emen o he
ou come
Risk o bias in he
selec ion o he
epo ed esul s
O e all
isk o bias
Ab ami, Wade, Lysenko,
Ma sh, and Gioko
(2016)
High Low Some
conce ns
Low Low High
Amendum,
Ve non-Feagans, and
Ginsbe g (2011)
Some conce ns Low Low Some conce ns Low Some
conce ns
Aunio and Mononen
(2018)
High Low Low Low Low High
Bai, Mo, Zhang, Boswell,
and Rozelle (2016)
Low Low Low Low Low Low
Bai e al. (2023) Some conce ns Low Low Low Low Some
conce ns
Bake e al. (2017) Some conce ns Low Some
conce ns
Low Low Some
conce ns
Bando, Gallego, Ge le ,
and Rome o Fonseca
(2017)
Some conce ns Low Low Low Low Some
conce ns
(con inued on nex page)
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
17
(con inued)
S udy Risk o bias a ising
om he
andomisa ion p ocess
Risk o bias due o
de ia ions om he
in ended in e en ion
Missing
ou come
da a
Risk o bias in he
measu emen o he
ou come
Risk o bias in he
selec ion o he
epo ed esul s
O e all
isk o bias
Bane jee e al. (2007) Low Low Low Low Low Low
Ba ow e al. (2009) Low Some conce ns Low Low Low Some
conce ns
Beg e al. (2022) Some conce ns Low Low Low Low Some
conce ns
Be gman (2021) Low Low Low Low Low Low
Be gman and Chan (2021) Low Low Low Low Low Low
Be gman and Roge s
(2016)
Some conce ns Some conce ns Low Low Low Some
conce ns
Be linski, Busso,
Dinkelman, and
Ma inez (2022)
Low Low Some
conce ns
Low Low Some
conce ns
Be linski, Busso,
Dinkelman, and
Ma inez (2021)
Low Low Some
conce ns
Low Low Some
conce ns
Beue mann e al. (2015) Low Low Low Low Low Low
Bo zekowski (2018) Some conce ns High Low Low Low High
Bo zekowski,
Singpu walla,
Meh o a, and Howa d
(2019)
Some conce ns Low Low Some conce ns Low Some
conce ns
Büchel, Jakob, Kuhnhanss,
S e en, & B une i
(2022)
Some conce ns Low Low Low Low Some
conce ns
Ca dim, Molina-Mill´
an,
and Vicen e (2023)
Some conce ns Low Low Some conce ns Low Some
conce ns
Ca illo, Ono a, and Ponce
(2010)
Some conce ns Low Low Low Low Some
conce ns
Chambe s, Cheung,
Gi o d, Madden, and
Sla in (2006)
Low Low Low Low Low Low
Chambe s e al. (2008) Some conce ns Low Low Low Low Some
conce ns
Cillie s e al. (2022) Low Low Low Low Low Low
C is ia e al. (2017) Some conce ns Low Low Low Low Some
conce ns
de Hoop e al. (2023) Some conce ns Low Low Low Low Some
conce ns
De ksen, Lecle c, and
Souza (2020)
Low High Some
conce ns
Low Low High
Du lo, Hanna, and Ryan
(2012)
Low Low Low Low Low Low
Go aza e al. (2022) Low Low Low Some conce ns Low Some
conce ns
Ibe and Abamuche (2019) High Low Low Low Low High
I o, Kasai, and Nakamu o
(2021)
Low Some conce ns Low Low Low Some
conce ns
Johns on and Ksoll (2022) Some conce ns Low Low Some conce ns Low Some
conce ns
K a and Mon i-Nussbaum
(2017)
Low Low Some
conce ns
Low Low Some
conce ns
Kuma and Meh a (2018) Low Some conce ns Low Low Low Some
conce ns
Lai, Luo, Zhang, Huang,
and Rozelle (2015)
Low Low Low Low Low Low
Lai e al. (2016) Some conce ns Low Low Low Low Some
conce ns
Lai e al. (2013) Low Low Low Low Low Low
Linden (2008) Low Low Low Low Low Low
Lineba ge , Pio owski,
and G eenwood
(2010)
Low Low Low Some conce ns Low Some
conce ns
Malamud e al. (2019) Some conce ns Low Some
conce ns
Low Low Some
conce ns
McManis and McManis
(2016)
High Low Some
conce ns
Low Low High
Mille and Robe son
(2011)
Some conce ns Low Some
conce ns
Low Low Some
conce ns
(con inued on nex page)
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
18
(con inued)
S udy Risk o bias a ising
om he
andomisa ion p ocess
Risk o bias due o
de ia ions om he
in ended in e en ion
Missing
ou come
da a
Risk o bias in he
measu emen o he
ou come
Risk o bias in he
selec ion o he
epo ed esul s
O e all
isk o bias
Mo e al. (2020) Low Low Low Low Low Low
Mo, Huang, e al. (2015) Low Low Low Low Low Low
Mo e al. (2013) Low Low Low Low Low Low
Mo e al. (2014) Some conce ns Low Low Low Low Some
conce ns
Mo, Zhang, e al. (2015) Low Low Low Low Low Low
Mu alidha an, Singh, and
Ganimian (2019)
Low Low Low Low Low Low
Naik, Chi e, Bhalla, and
Rajan (2020)
Some conce ns Low Low Low Low Some
conce ns
Pipe , Simmons
Zuilkowski,
Kwayumba, and
S igel (2016)
Some conce ns Low Low Some conce ns Low Some
conce ns
Pi ch o d (2015) High Low Low Some conce ns Low High
Riley (2018) Some conce ns Low Low Low Some conce ns Some
conce ns
Rouse and K uege (2004) Low Low Some
conce ns
Low Low Some
conce ns
Ru he o d e al. (2014) Low Low Low Low Low Low
San ana e al. (2019) Low Low Some
conce ns
Low Low Some
conce ns
Schac e e al. (2016) High Low High Low Low High
Seo (2017) Low Low Some
conce ns
Low Low Some
conce ns
Se en (2023) Low Some conce ns Some
conce ns
Low Low Some
conce ns
Silande e al. (2016) High Low Some
conce ns
Some conce ns Low High
Ve hallen and Bus (2010) Some conce ns Low Low Low Low Some
conce ns
Wol , Abe , Beh man, and
Tsinigo (2019)
Low Low Low Some conce ns Low Some
conce ns
Yang e al. (2013) Some conce ns Low Low Low Low Some
conce ns
Appendix D. Quali y assessmen o quasi-expe imen al s udies (ROBINS-I)
S udy Bias due o
con ounding
Bias in selec ion
o pa icipan s
in o he s udy
Bias in
classi ica ion o
in e en ions
Bias due o
de ia ions om
in ended
in e en ions
Bias due o
missing
da a
Bias in
measu emen o
ou comes
Bias in
selec ion o
epo ed
esul s
O e all
bias
Bianchi, Lu, and
Song (2022)
Low Low Low Low Low Low Low Low
Blimpo, Gajigo,
Owusu,
Tomi a, and
Xu (2020)
Low Mode a e Low Low Low Low Mode a e Mode a e
B own e al.
(2020)
Low Mode a e Low Low Mode a e Low Low Mode a e
Leh e ,
Mawoyo,
and Mbaye
(2019)
Low Mode a e Low Low Mode a e Low Low Mode a e
Leu en e al.
(2007)
Low Low Low Low Low Low Low Low
Lysenko e al.
(2019)
Low Mode a e Low Low Mode a e Low Low Mode a e
Maca uso e al.
(2006)
Low Mode a e Low Low Mode a e Low Low Mode a e
N aila and
Mba aka
(2023)
High Mode a e Low Low Mode a e Low Low High
(con inued on nex page)
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
19
(con inued)
S udy Bias due o
con ounding
Bias in selec ion
o pa icipan s
in o he s udy
Bias in
classi ica ion o
in e en ions
Bias due o
de ia ions om
in ended
in e en ions
Bias due o
missing
da a
Bias in
measu emen o
ou comes
Bias in
selec ion o
epo ed
esul s
O e all
bias
Schac e and Jo
(2016)
Low Low Low Low Low Low Low Low
Wenne s en,
Qu aishy,
and
Velamu i
(2015)
Low Low Low Low Low Low Low Low
Da a a ailabili y
Da a will be made a ailable on eques .
Re e ences
*A icles included in he me a-analysis
* Ab ami, P. C., Wade, C. A., Lysenko, L., Ma sh, J., & Gioko, A. (2016). Using educa ional echnology o de elop ea ly li e acy skills in Sub-Saha an A ica. Educa ion
and In o ma ion Technologies, 21(4), 945–964. h ps://doi.o g/10.1007/s10639-014-9362-4.
* Amendum, S. J., Ve non-Feagans, L., & Ginsbe g, M. C. (2011). The e ec i eness o a echnologically acili a ed class oom-based ea ly eading in e en ion. The
Elemen a y School Jou nal, 112(1), 107–131. h ps://doi.o g/10.1086/660684.
Ande son, E., Jalles D’O ey, M. A., Du endack, M., & Esposi o, L. (2017). Does go e nmen spending a ec income inequali y? A me a- eg ession analysis. Jou nal o
Economic Su eys, 31(4), 961–987. h ps://doi.o g/10.1111/joes.12173
Aung, T. N., & Khaing, S. S. (2015). Challenges o implemen ing e-lea ning in de eloping coun ies: A e iew. In T. T. Zin, J. W. Lin, J. S. Pan, P. Tin, & M. Yoko a
(Eds.), Gene ic and e olu iona y compu ing: Ad ances in in elligen sys ems and compu ing (pp. 405–411). Sp inge . h ps://doi.o g/10.1007/978-3-319-23207-2_41.
* Aunio, P., & Mononen, R. (2018). The e ec s o educa ional compu e game on low-pe o ming child en’s ea ly nume acy skills – an in e en ion s udy in a
p eschool se ing. Eu opean Jou nal o Special Needs Educa ion, 33(5), 677–691. h ps://doi.o g/10.1080/08856257.2017.1412640.
* Bai, Y., Mo, D., Zhang, L., Boswell, M., & Rozelle, S. (2016). The impac o in eg a ing ICT wi h eaching: E idence om a andomized con olled ial in u al
schools in China. Compu e s & Educa ion, 96(1), 1–14. h ps://doi.o g/10.1016/j.compedu.2016.02.005.
* Bai, Y., Tang, B., Wang, B., Mo, D., Zhang, L., Rozelle, S., e al. (2023). Impac o online compu e assis ed lea ning on educa ion: Expe imen al e idence om
economically ulne able a eas o China. Economics o Educa ion Re iew, 94, A icle 102385. h ps://doi.o g/10.1016/j.econedu e .2023.102385.
* Bake , D. L., Basa aba, D. L., Smolkowski, K., Con y, J., Hau ala, J., Richa dson, U., e al. (2017). Explo ing he c oss-linguis ic ans e o eading skills in Spanish o
English in he con ex o a compu e adap i e eading in e en ion. Bilingual Resea ch Jou nal, 40(2), 222–239. h ps://doi.o g/10.1080/
15235882.2017.1309719.
* Bando, R., Gallego, F., Ge le , P., & Rome o Fonseca, D. (2017). Books o lap ops? The e ec o shi ing om p in ed o digi al deli e y o educa ional con en on
lea ning. Economics o Educa ion Re iew, 61, 162–173. h ps://doi.o g/10.1016/j.econedu e .2017.07.005.
* Bane jee, A. V., Cole, S., Du lo, E., & Linden, L. (2007). Remedying educa ion: E idence om wo andomized expe imen s in India. Qua e ly Jou nal o Economics,
122(3), 1235–1264. h ps://doi.o g/10.1162/qjec.122.3.1235.
* Ba ow, L., Ma kman, L., & Rouse, C. E. (2009). Technology’s edge: The Educa ional bene i s o compu e -aided ins uc ion. Ame ican Economic Jou nal: Economic
Policy, 1(1), 52–74. h ps://doi.o g/10.1257/pol.1.1.52.
Becke , H. J. (2000). Who’s wi ed and who’s no : Child en’s access o and use o compu e echnology. The Fu u e o Child en, 10(2), 44–75. h ps://doi.o g/10.2307/
1602689
* Beg, S. A., Lucas, A. M., Halim, W., & Sai , U. (2022). Engaging eache s wi h echnology inc eased achie emen , bypassing eache s did no . Ame ican Economic
Jou nal: Economic Policy, 14(2), 61–90. h ps://doi.o g/10.1257/pol.20200713.
Begg, C. B., & Belin, J. A. (1988). Publica ion bias: A p oblem in in e p e ing medical da a. Jou nal o he Royal S a is ical Socie y, 151(3), 419–445. h ps://doi.o g/
10.2307/2982993
* Be gman, P. (2021). Pa en -child in o ma ion ic ions and human capi al in es men : E idence om a ield expe imen . Jou nal o Poli ical Economy, 129(1),
286–322. h ps://doi.o g/10.1086/711410.
* Be gman, P., & Chan, E. W. (2021). Le e aging pa en s h ough low-cos echnology: The impac o high- equency in o ma ion on s uden achie emen . Jou nal o
Human Resou ces, 56(1), 125–158. h ps://doi.o g/10.3368/jh .56.1.1118-9837R1.
* Be gman, P., & Roge s, T. (2016). Pa en adop ion o school communica ions echnology: A 12-school expe imen o de aul en ollmen policies. Pape p esen ed a
he socie y o esea ch on educa ional e ec i eness sp ing 2016 con e ence. Washing on, DC, Ma ch 4.
* Be linski, S., Busso, M., Dinkelman, T., & Ma inez, C. (2021). Reducing pa en -school in o ma ion gaps and imp o ing educa ion ou comes: E idence om high
equency ex messaging in Chile. IDB-WP-1221. h ps://doi.o g/10.18235/0003257.
* Be linski, S., Busso, M., Dinkelman, T., & Ma inez, C. (2022). Reducing pa en -school in o ma ion gaps and imp o ing educa ion ou comes: E idence om high equency
ex messaging in Chile. NBER Wo king Pape 28581.
* Beue mann, D. W., C is ia, J., Cue o, S., Malamud, O., & C uz-Aguayo, Y. (2015). One lap op pe child a home: Sho - e m impac s om a andomized expe imen
in Pe u. Ame ican Economic Jou nal: Applied Economics, 7(2), 53–80. h ps://doi.o g/10.1257/app.20130267.
* Bianchi, N., Lu, Y., & Song, H. (2022). The e ec o compu e -assis ed lea ning on s uden s’ long- e m de elopmen . Jou nal o De elopmen Economics, 158, A icle
102919. h ps://doi.o g/10.1016/j.jde eco.2022.102919.
* Blimpo, M. P., Gajigo, O., Owusu, S., Tomi a, R., & Xu, Y. (2020). Technology in he class oom and lea ning in seconda y schools. Wo ld Bank G oup. h ps://doi.o g/
10.1596/1813-9450-9288. Policy Resea ch Wo king Pape 9288.
Bo ego, M., Fos e , M. J., & F oyd, J. E. (2014). Sys ema ic li e a u e e iews in enginee ing educa ion and o he de eloping in e disciplina y ields. Jou nal o
Enginee ing Educa ion, 103(1), 45–76. h ps://doi.o g/10.1002/jee.20038
* Bo zekowski, D. (2018). A quasi-expe imen examining he impac o educa ional ca oons on Tanzanian child en. Jou nal o Applied De elopmen al Psychology, 54
(1), 53–59. h ps://doi.o g/10.1016/j.appde .2017.11.007.
* Bo zekowski, D., Singpu walla, D., Meh o a, D., & Howa d, D. (2019). The impac o galli galli sim sim on Indian p eschoole s. Jou nal o Applied De elopmen al
Psychology, 64, A icle 101054. h ps://doi.o g/10.1016/j.appde .2019.101054.
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
20
* B own, F., Fa ag, A., Hussein, F., Mille , L., Rad o d, K., Abbadi, A. A., e al. (2020). Can’ wai o lea n: A quasi-expe imen al e alua ion o a digi al game-based
lea ning p og amme o ou o school child en in Sudan. Jou nal o De elopmen E ec i eness, 15(3), 320–341. h ps://doi.o g/10.1080/19439342.2020.1829000.
* Büchel, K., Jakob, M., Kuhnhanss, C., S e en, D., & B une i, A. (2022). The ela i e e ec i eness o eache s and lea ning so wa e: E idence om a ield
expe imen in El Sal ado . Jou nal o Labo Economics, 40(3), 737–777. h ps://doi.o g/10.1086/717727.
Bulman, G., & Fai lie, R. W. (2016). Technology and educa ion: Compu e s, so wa e, and he in e ne . In E. A. Hanushek, S. Machin, & L. Woessmann (Eds.),
Handbook o he economics o educa ion (Vol. 5, pp. 239–280). Else ie .
By on, K., & Pos , C. (2016). Women on boa ds o di ec o s and co po a e social pe o mance: A me a-analysis. Co po a e Go e nance: An In e na ional Re iew, 24(4),
428–442. h ps://doi.o g/10.1111/co g.12165
* Ca dim, J., Molina-Mill´
an, T., & Vicen e, P. C. (2023). Technology-aided ins uc ion in p ima y educa ion: Expe imen al e idence om Angola. Jou nal o
De elopmen Economics, 164, A icle A icle103145. h ps://doi.o g/10.1016/j.jde eco.2023.103145.
* Ca illo, P. E., Ono a, M., & Ponce, J. (2010). In o ma ion echnology and s uden achie emen : E idence om a andomized expe imen in Ecuado . In e -Ame ican
De elopmen Bank. IDB Wo king Pape No. IDB-WP-223.
* Chambe s, B., Cheung, A., Gi o d, R., Madden, N. A., & Sla in, R. E. (2006). Achie emen e ec s o embedded mul imedia in a Success o All eading p og am.
Jou nal o Educa ional Psychology, 98(1), 232–237. h ps://doi.o g/10.1037/0022-0663.98.1.232.
* Chambe s, B., Sla in, R. E., Madden, N. A., Ab ami, P. C., Tucke , B. J., Cheung, A., e al. (2008). Technology in usion in Success o All: Reading ou comes o i s
g ade s. The Elemen a y School Jou nal, 109(1), 1–15. h ps://doi.o g/10.1086/592364.
Chang, H. Y., Binali, T., Liang, J. C., Chiou, G. L., Cheng, K. H., Lee, S. W. Y., e al. (2022). Ten yea s o augmen ed eali y in educa ion: A me a-analysis o (quasi-)
expe imen al s udies o in es iga e he impac . Compu e s & Educa ion, 191, A icle 104641. h ps://doi.o g/10.1016/j.compedu.2022.104641
Chaudhu y, N., Hamme , J., K eme , M., Mu alidha an, K., & Roge s, F. H. (2006). Missing in ac ion: Teache and heal h wo ke absence in de eloping coun ies. The
Jou nal o Economic Pe spec i es, 20(1), 91–116. h ps://doi.o g/10.1257/089533006776526058
Chauhan, S. (2017). A me a-analysis o he impac o echnology on lea ning e ec i eness o elemen a y s uden s. Compu e s & Educa ion, 105, 14–30. h ps://doi.o g/
10.1016/j.compedu.2016.11.005
Che y, R., F iedman, J. N., & Rocko , J. E. (2014). Measu ing he impac s o eache s II: Teache alue-added and s uden ou comes in adul hood. The Ame ican
Economic Re iew, 104, 2633–2679. h ps://doi.o g/10.1257/ae .104.9.2633
Cheung, A. C. K., & Sla in, R. E. (2012). How ea u es o educa ional echnology applica ions a ec s uden eading ou comes: A me a-analysis. Educa ional Resea ch
Re iew, 7, 198–215. h ps://doi.o g/10.1016/j.edu e .2012.05.002
Cheung, A. C. K., & Sla in, R. E. (2013). The e ec i eness o educa ional echnology applica ions o enhancing ma hema ics achie emen in K-12 class ooms: A me a-
analysis. Educa ional Resea ch Re iew, 9, 88–113. h ps://doi.o g/10.1016/j.edu e .2013.01.001
* Cillie s, J., Fleisch, B., Ko ze, J., Mohohlwane, N., Thula e, T., & Taylo , S. (2022). Can i ual eplace in-pe son coaching? Expe imen al e idence on eache
p o essional de elopmen and s uden lea ning in Sou h A ica. Jou nal o De elopmen Economics, 155, A icle 102815. h ps://doi.o g/10.1016/j.
jde eco.2021.102815.
Coope , H., & Hedges, L. V. (1994). The handbook o esea ch syn hesis. Russell Sage Founda ion.
* C is ia, J., Iba a an, P., Cue o, S., San iago, A., & Se e in, E. (2017). Technology and child de elopmen : E idence om he one lap op pe child p og am. Ame ican
Economic Jou nal: Applied Economics, 9(3), 295–320. h ps://doi.o g/10.1257/app.20150385.
* de Hoop, T., Ring, H., Siwach, G., Dias, P., Tembo, G., Ro hba d, V., e al. (2023). Impac o echnology-aided ac i i y-based lea ning app oaches on lea ning
ou comes: Expe imen al e idence om communi y schools in u al Zambia. Jou nal o Resea ch on Educa ional E ec i eness. h ps://doi.o g/10.1080/
19345747.2023.2268072 (in p ess).
de Linde Leona d, M., S anley, T. D., & Doucouliagos, H. (2014). Does he UK minimum wage educe employmen ? A me a- eg ession analysis. B i ish Jou nal o
Indus ial Rela ions, 52(3), 499–520. h ps://doi.o g/10.1111/bji .12031
De Luca, G., & Magnus, J. R. (2011). Bayesian model a e aging and weigh ed-a e age leas squa es: Equi a iance, s abili y, and nume ical issues. STATA Jou nal, 11
(4), 518–544. h ps://doi.o g/10.1177/1536867X12011004
* De ksen, L., Lecle c, C., & Souza, P. C. (2020). Sea ching o answe s: The impac o s uden access o Wikipedia. Uni e si y o Wa wick. Wo king Pape No. 450, Cen e
o Compe i i e Ad an age in he Global Economy.
DeWi , D., & Alias, N. (2019). Compu e s in educa ion in de eloping coun ies: Manage ial issues. In A. Ta nall (Ed.), Encyclopedia o educa ion and in o ma ion
echnologies (pp. 1–11). Cham: Sp inge . h ps://doi.o g/10.1007/978-3-319-60013-0_125-1.
Di Pie o, G. (2022). S udying ab oad and ea nings: A me a-analysis. Jou nal o Economic Su eys, 36, 1096–1129. h ps://doi.o g/10.1111/joes.12472
Di Pie o, G. (2023). The impac o Co id-19 on s uden achie emen : E idence om a ecen me a-analysis. Educa ional Resea ch Re iew, 39, A icle 100530. h ps://
doi.o g/10.1016/j.edu e .2023.100530
Dicke son, A., McIn osh, S., & Valen e, C. (2015). Do he ma hs: An analysis o he gende gap in ma hema ics in A ica. Economics o Educa ion Re iew, 46, 1–22.
h ps://doi.o g/10.1016/j.econedu e .2015.02.005
Die ichson, J., Klin Jø gensen, A.-M., & Filges, T. (2017). Academic in e en ions o elemen a y and middle school s uden s wi h low socioeconomic s a us: A
sys ema ic e iew and me a-analysis. Re iew o Educa ional Resea ch, 87(2), 243–282. h ps://doi.o g/10.3102/0034654316687036
* Du lo, E., Hanna, R., & Ryan, S. P. (2012). Incen i es wo k: Ge ing eache s o come o school. The Ame ican Economic Re iew, 102(4), 1241–1278. h ps://doi.o g/
10.1257/ae .102.4.1241.
Du al, S., & Tweedie, R. (2000). T im and ill: A simple unnel-plo –based me hod o es ing and adjus ing o publica ion bias in me a-analysis. Biome ics, 56(2),
455–463. h ps://doi.o g/10.1111/j.0006-341x.2000.00455.x
Egge , M., Da ey Smi h, G., Schneide , M., & Minde , C. (1997). Bias in me a-analysis de ec ed by a simple, g aphical es . B i ish Medical Jou nal, 315(7109),
629–634. h ps://doi.o g/10.1136/bmj.315.7109.629
El ahi , M. E. (2019). E-Lea ning in de eloping coun ies: Is i a panacea? A case s udy o Sudan. IEEE Access, 7, 97784–97792. h ps://doi.o g/10.1109/
ACCESS.2019.2930411
Escue a, M., & Holloway, S. (2019). In es men in educa ion echnology ac oss he globe: Whe e p o i mee s pu pose. In Columbia SIPA en ep eneu ship & policy
wo king pape se ies.
Escue a, M., Nickow, A. J., O eopoulos, P., & Quan, V. (2020). Upg ading educa ion wi h echnology: Insigh s om expe imen al esea ch. Jou nal o Economic
Li e a u e, 58(4), 897–996. h ps://doi.o g/10.1257/jel.20191507
Ewald, H., Kle ings, I., Wagne , G., Heise, T. L., e al. (2022). Sea ching wo o mo e da abases dec eased he isk o missing ele an s udies: A me a esea ch s udy.
Jou nal o Clinical Epidemiology, 149, 154–164. h ps://doi.o g/10.1016/j.jclinepi.2022.05.022
Fan, Z., & Beh, L. S. (2023). Knowledge sha ing among academics in highe educa ion: A sys ema ic li e a u e e iew and u u e agenda. Educa ional Resea ch Re iew,
42, A icle 100573. h ps://doi.o g/10.1016/j.edu e .2023.100573
Fejes, J. B. (2012). Lea ning mo i a ion o disad an aged s uden s. In N. M. Seel (Ed.), Encyclopedia o he sciences o lea ning (pp. 1935–1937). New Yo k: Sp inge .
h ps://doi.o g/10.1007/978-1-4419-1428-6_680.
Fodo , L. A., Co eț, C. D., Cuijpe s, P., Szamoskozi, S., Da id, D., & C is ea, I. A. (2018). The e ec i eness o i ual eali y based in e en ions o symp oms o anxie y
and dep ession: A me a-analysis. Scien i ic Repo , 8, A icle 10323. h ps://doi.o g/10.1038/s41598-018-28113-6
F aillon, J., Ainley, J., Schulz, W., F iedman, T., & Duckwo h, D. (2019). P epa ing o li e in a digi al Wo ld: IEA in e na ional compu e and in o ma ion li e acy s udy
2018 in e na ional epo . Ams e dam. In e na ional Associa ion o he E alua ion o Educa ional Achie emen .
F iese, M., F ankenbach, J., Job, V., & Loschelde , D. D. (2017). Does sel -con ol aining imp o e sel -con ol? A me a-analysis. Pe spec i es on Psychological Science,
12(6), 1077–1099. h ps://doi.o g/10.1177/1745691617697076
Fu uya-Kanamo i, L., Ba end eg , J. J., & Doi, S. A. R. (2018). A new imp o ed g aphical and quan i a i e me hod o de ec ing bias in me a-analysis. In e na ional
Jou nal o E idence-Based Heal hca e, 16, 195–203. h ps://doi.o g/10.1097/XEB.0000000000000141
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
21
Ga cía-Vandewalle Ga cía, J. M., Ga cía-Ca mona, M., T ujillo To es, J. M., & Moya-Fe n´
andez, P. (2022). The in eg a ion o eme ging echnologies in
socioeconomically disad an aged educa ional con ex s. The iew o in e na ional expe s. Jou nal o Compu e Assis ed Lea ning, 38(4), 1185–1197. h ps://doi.
o g/10.1111/jcal.12677
Goodall, J. S. (2016). Technology and school–home communica ion. In e na ional Jou nal o pedagogies and lea ning, 11(2), 118–131. h ps://doi.o g/10.1080/
22040552.2016.1227252
* Go aza , L., Hupkau, C., & Rold´
an, A. (2022). Online u o ing wo ks: Expe imen al e idence om a p og am wi h ulne able child en. Wo king Pape No. 2, esade.
G ainge, M. (2015). Excluding small s udies om a sys ema ic e iew o me a-analysis. Pape p esen ed a CSG annual mee ing. Ge many: D esden. Ma ch 12-18.
Haddaway, N. R., Collins, A. M., Coughlin, D., & Ki k, S. (2015). The ole o Google schola in e idence e iews and i s applicabili y o g ey li e a u e sea ching. PLoS
One, 10(9), A icle e0138237. h ps://doi.o g/10.1371/jou nal.pone.0138237
Ha a i, M. B., Pa ola, H. R., Ha well, C. J., & Riegelman, A. (2020). Li e a u e sea ches in sys ema ic e iews and me a-analyses: A e iew, e alua ion, and
ecommenda ions. Jou nal o Voca ional Beha io , 118, A icle 103377. h ps://doi.o g/10.1016/j.j b.2020.103377
Ha e , M., Cuijpe s, P., Fu ukawa, T., & Ebe , D. (2021). Doing me a-analysis wi h R. A hands-on guide. CRC P ess.
Hee s, M., Van Kla e en, C., G oo , W., & Maassen an den B ink, H. (2016). Communi y schools: Wha we know and wha we need o know. Re iew o Educa ional
Resea ch, 86(4), 1016–1051. h ps://doi.o g/10.3102/0034654315627365
Heppen, J. B., Ku ki, A., & B own, S. (2020). Can ex ing pa en s imp o e a endance in elemen a y school? A es o an adap i e messaging s a egy (ncee 2020–006a).
Washing on, DC: U.S. Depa men o Educa ion, Ins i u e o Educa ion Sciences, Na ional Cen e o Educa ion E alua ion and Regional Assis ance.
He nandez, A. V., Ma i, K. M., & Roman, Y. M. (2020). Me a-analysis. Ches , 158(1), S97–S102. h ps://doi.o g/10.1016/j.ches .2020.03.003
Higgins, J. P. T., & G een, S. (2011). Coch ane handbook o sys ema ic e iews o in e en ions. The Coch ane Collabo a ion [upda ed Ma ch 2011(]) Ve sion 5.1.0.
Hinos oza, J. E. (2018). New challenges o ICT in educa ion policies in de eloping coun ies: The need o accoun o he widesp ead use o ICT o eaching and
lea ning ou side he school. In I. A. Lubin (Ed.), ICT-Suppo ed inno a ions in small coun ies and de eloping egions: Pe spec i es and ecommenda ions o in e na ional
educa ion (pp. 99–119). Cham: Sp inge In e na ional. h ps://doi.o g/10.1007/978-3-319-67657-9_5.
Hohl eld, T. N., Ri zhaup , A. D., Ba on, A., & Kemke hon, K. (2008). Examining he digi al di ide in K-12 public schools: Fou -yea ends o suppo ing ICT li e acy
in Flo ida. Compu e s & Educa ion, 51(4), 1648–1663. h ps://doi.o g/10.1016/j.compedu.2008.04.002
* Ibe, E., & Abamuche, J. (2019). E ec s o audio isual echnological aids on s uden s’ achie emen and in e es in seconda y school biology in Nige ia. Heliyon, 5,
A icle e01812. h ps://doi.o g/10.1016/j.heliyon.2019.e01812.
Ioannidis, J. P., S anley, T. D., & Doucouliagos, H. (2017). The powe o bias in economics esea ch. Economic Jou nal, 127(605), F236–F265. h ps://doi.o g/
10.1111/ecoj.12461
* I o, H., Kasai, K., & Nakamu o, M. (2021). Does compu e -aided ins uc ion imp o e child en’s cogni i e and non-cogni i e skills? E idence om Cambodia. Asian
De elopmen Re iew, 38(1), 98–118. h ps://doi.o g/10.1162/ade _a_00159.
* Johns on, J., & Ksoll, C. (2022). E ec i eness o in e ac i e sa elli e- ansmi ed ins uc ion: Expe imen al e idence om Ghanaian p ima y schools. Economics o
Educa ion Re iew, 91, A icle 102315. h ps://doi.o g/10.1016/j.econedu e .2022.102315.
Kaise , T., & Menkho , L. (2020). Financial educa ion in schools: A me a-analysis o expe imen al s udies. Economics o Educa ion Re iew, 78, A icle 101930. h ps://
doi.o g/10.1016/j.econedu e .2019.101930
Kazu, I. Y., & Ku o˘
glu Yalçın, C. (2022). A me a-analysis s udy on he e ec i eness o lipped class oom lea ning on s uden s’ academic achie emen . E-In e na ional
Jou nal o Educa ional Resea ch, 13(1), 85–102. h ps://doi.o g/10.19160/e-ije .1033589
* K a , M. A., & Mon i-Nussbaum, M. (2017). Can schools enable pa en s o p e en summe lea ning loss? A ex messaging ield expe imen o p omo e li e acy
skills. The Annals o he Ame ican Academy o Poli ical and Social Science, 674(1), 85–112. h ps://doi.o g/10.1177/0002716217732009.
K oupo a, K., Ha anek, T., & I so a, Z. (2024). S uden employmen and educa ion: A me a-analysis. Economics o Educa ion Re iew, 100, A icle 102539. h ps://doi.
o g/10.1016/j.econedu e .2024.102539
* Kuma , A., & Meh a, A. (2018). Remedying educa ion wi h pe sonalized homewo k: E idence om a andomized ield expe imen in India. Social Science Resea ch
Ne wo k. h ps://doi.o g/10.2139/ss n.2756059 (SSRN Schola ly Pape ID 2756059; Issue ID 2756059).
Ku e, A. E., B e ik, L. M., & Bliks ad-Balas, M. (2023). Digi al skills c i ical o educa ion: Video analysis o s uden s’ echnology use in No wegian seconda y English
class ooms. Jou nal o Compu e Assis ed Lea ning, 39(1), 269–285. h ps://doi.o g/10.1111/jcal.12745
* Lai, F., Luo, R., Zhang, L., Huang, X., & Rozelle, S. (2015). Does compu e -assis ed lea ning imp o e lea ning ou comes? E idence om a andomized expe imen in
mig an schools in beijing. Economics o Educa ion Re iew, 47, 34–48. h ps://doi.o g/10.1016/j.econedu e .2015.03.005.
* Lai, F., Zhang, L., Bai, Y., Liu, C., Shi, Y., Chang, F., e al. (2016). Mo e is no always be e : E idence om a andomised expe imen o compu e -assis ed lea ning in
u al mino i y schools in qinghai. Jou nal o De elopmen E ec i eness, 8(4), 449–472. h ps://doi.o g/10.1080/19439342.2016.1220412.
* Lai, F., Zhang, L., Hu, X., Qu, Q., Shi, Y., Qiao, Y., e al. (2013). Compu e assis ed lea ning as ex acu icula u o ? E idence om a andomised expe imen in u al
boa ding schools in shaanxi. Jou nal o De elopmen E ec i eness, 5(2), 208–231. h ps://doi.o g/10.1080/19439342.2013.780089.
* Leh e , K., Mawoyo, M., & Mbaye, S. (2019). The impac s o in e ac i e sma boa ds on lea ning achie emen in Senegalese p ima y schools (3ie G an ee Final Repo ).
* Leu en, E., Lindahl, M., Oos e beek, H., & Webbink, D. (2007). The e ec o ex a unding o disad an aged pupils on achie emen . The Re iew o Economics and
S a is ics, 89(4), 721–736. h ps://doi.o g/10.1162/ es .89.4.721.
* Linden, L. (2008). Complemen o subs i u e? The e ec o echnology on s uden achie emen in India. in oDe Wo king Pape No. 17.
* Lineba ge , D., Pio owski, J. T., & G eenwood, C. R. (2010). On-sc een p in : The ole o cap ions as a supplemen al li e acy ool. Jou nal o Resea ch in Reading, 33
(2), 148–167. h ps://doi.o g/10.1111/j.1467-9817.2009.01407.x.
* Lysenko, L., Ab ami, P. C., Wade, C. A., Ma sh, J. P., WaGioko, M., & Ki o o, E. (2019). P omo ing young Kenyans’ g ow h in li e acy wi h educa ional echnology: A
ale o wo yea s o implemen a ion. In e na ional Jou nal o Educa ional Resea ch, 95, 176–189. h ps://doi.o g/10.1016/j.ije .2019.02.013.
* Maca uso, P., Hook, P. E., & McCabe, R. (2006). The e icacy o compu e -based supplemen a y phonics p og ams o ad ancing eading skills in a - isk elemen a y
s uden s. Jou nal o Resea ch in Reading, 29(2), 162–172. h ps://doi.o g/10.1111/j.1467-9817.2006.00282.x.
Magnus, J. R., Powell, O., & P u e , P. (2010). A compa ison o wo model a e aging echniques wi h an applica ion o g ow h empi ics. Jou nal o Econome ics, 154
(2), 139–153. h ps://doi.o g/10.1016/j.jeconom.2009.07.004
Majo , L., F ancis, G. A., & Tsapali, M. (2021). The e ec i eness o echnology-suppo ed pe sonalised lea ning in low-and middle-income coun ies: A me a-analysis.
B i ish Jou nal o Educa ional Technology, 52, 1935–1964. h ps://doi.o g/10.1111/bje .13116
* Malamud, O., Cue o, S., C is ia, J., & Beue mann, D. W. (2019). Do child en bene i om in e ne access? Expe imen al e idence om Pe u. Jou nal o De elopmen
Economics, 138, 41–56. h ps://doi.o g/10.1016/j.jde eco.2018.11.005.
McCoach, D. B., & Siegle, D. (2009). The i s wo d: A le e om he co-edi o s: E ec sizes—an explana ion o jaa edi o ial policy. Jou nal o Ad anced Academics, 20
(2), 209–212. h ps://doi.o g/10.1177/1932202X0902000201
McEwan, P. J. (2015). Imp o ing lea ning in p ima y schools o de eloping coun ies: A me a-analysis o andomized expe imen s. Re iew o Educa ional Resea ch, 85
(3), 353–394. h ps://doi.o g/10.3102/003465431455312
* McManis, M. H., & McManis, L. D. (2016). Using a ouch-based, compu e -assis ed lea ning sys em o p omo e li e acy and ma h skills o low-income p eschoole s.
Jou nal o In o ma ion Technology Educa ion: Resea ch, 15, 409–429. h ps://doi.o g/10.28945/3550.
* Mille , D. J., & Robe son, D. P. (2011). Educa ional bene i s o using game consoles in a p ima y class oom: A andomised con olled ial. B i ish Jou nal o
Educa ional Technology, 42(5), 850–864. h ps://doi.o g/10.1111/j.1467-8535.2010.01114.x.
* Mo, D., Bai, Y., Shi, Y., Abbey, C., Zhang, L., Rozelle, S., e al. (2020). Ins i u ions, implemen a ion, and p og am e ec i eness: E idence om a andomized
e alua ion o compu e -assis ed lea ning in u al China. Jou nal o De elopmen Economics, 146, A icle 102487. h ps://doi.o g/10.1016/j.jde eco.2020.102487.
* Mo, D., Huang, W., Shi, Y., Zhang, L., Boswell, M., & Rozelle, S. (2015). Compu e echnology in educa ion: E idence om a pooled s udy o compu e -assis ed
lea ning p og ams among u al s uden s in China. China Economic Re iew, 36, 131–145. h ps://doi.o g/10.1016/j.chieco.2015.09.001.
* Mo, D., Swinnen, J., Zhang, L., Yi, H., Qu, Q., Boswell, M., e al. (2013). Can one- o-one compu ing na ow he digi al di ide and he educa ional gap in China? The
case o Beijing mig an schools. Wo ld De elopmen , 46, 14–29. h ps://doi.o g/10.1016/j.wo ldde .2012.12.019.
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
22
* Mo, D., Zhang, L., Luo, R., Qu, Q., Huang, W., Wang, J., e al. (2014). In eg a ing compu e -assis ed lea ning in o a egula cu iculum: E idence om a andomised
expe imen in u al schools in Shaanxi. Jou nal o De elopmen E ec i eness, 6(3), 300–323. h ps://doi.o g/10.1080/19439342.2014.911770.
* Mo, D., Zhang, L., Wang, J., Huang, W., Shi, Y., Boswell, M., e al. (2015b). Pe sis ence o lea ning gains om compu e assis ed lea ning: Expe imen al e idence
om China. Jou nal o Compu e Assis ed Lea ning, 31, 562–581. h ps://doi.o g/10.1111/jcal.12106.
Mo is, S. B. (2008). Es ima ing e ec sizes om p e es -pos es -con ol g oup designs. O ganiza ional Resea ch Me hods, 11(2), 364–386. h ps://doi.o g/10.1177/
109442810629105
* Mu alidha an, K., Singh, A., & Ganimian, A. J. (2019). Dis up ing educa ion? Expe imen al e idence on echnology-aided ins uc ion in India. The Ame ican
Economic Re iew, 109(4), 1426–1460. h ps://doi.o g/10.1257/ae .20171112.
* Naik, G., Chi e, C., Bhalla, M., & Rajan, J. (2020). Impac o use o echnology on s uden lea ning ou comes: E idence om a la ge-scale expe imen in India. Wo ld
De elopmen , 127, A icle 104736. h ps://doi.o g/10.1016/j.wo ldde .2019.104736.
Ni, A., Cheung, A. C. K., & Shi, J. (2022). E ec s o educa ional echnology on eading achie emen o Chinese K-12 English second language lea ne s: A me a-
analysis. F on ie s in Psychology, 13, A icle 1025761. h ps://doi.o g/10.3389/ psyg.2022.1025761
* N aila, Y. W., & Mba aka, S. R. (2023). Examining he impac o in e ac i e mul imedia ins uc ion on he pe o mance o seconda y school s uden s in biology in
Dedza dis ic , Malawi. Eu opean Jou nal o Educa ional Resea ch, 12(4), 1697–1708. h ps://doi.o g/10.12973/eu-je .12.4.1697.
OECD. (2016). A e he e di e ences in how ad an aged and disad an aged s uden s use he In e ne ? PISA in Focus, 64, A icle Pa is.
Ogan, A., Walke , E., Bake , R. S. J. D., Rebolledo Mendez, G., Jimenez Cas o, M., Lau en ino, T., e al. (2012). Collabo a ion in cogni i e u o use in La in Ame ica:
Field s udy and design ecommenda ions. In P oceedings o he SIGCHI con e ence on human ac o s in compu ing sys ems (pp. 1381–1390). h ps://doi.o g/10.1145/
2207676.2208597
Paez, A. (2017). G ay li e a u e: An impo an esou ce in sys ema ic e iews. Jou nal o E idence-Based Medicine, 10(3), 233–240. h ps://doi.o g/10.1111/
jebm.12266
* Pipe , B., Simmons Zuilkowski, S., Kwayumba, D., & S igel, C. (2016). Does echnology imp o e eading ou comes? Compa ing he e ec i eness and cos -
e ec i eness o ICT in e en ions o ea ly g ade eading in Kenya. In e na ional Jou nal o Educa ional De elopmen , 49, 204–214. h ps://doi.o g/10.1016/j.
ijedude .2016.03.006.
* Pi ch o d, N. J. (2015). De elopmen o ea ly ma hema ical skills wi h a able in e en ion: A andomized con ol ial in Malawi. F on ie s in Psychology, 6. h ps://
doi.o g/10.3389/ psyg.2015.00485.
Popp, M. (2023). How elas ic is labo demand? A me a-analysis o he Ge man labo ma ke . Jou nal o Labou Ma ke Resea ch, 57(1), 1–21. h ps://doi.o g/
10.1186/s12651-023-00337-8
Ran, H., Kasli, M., & Secada, W. G. (2021). A me a-analysis on compu e echnology in e en ion e ec s on ma hema ics achie emen o low-pe o ming s uden s in
K-12 class ooms. Jou nal o Educa ional Compu ing Resea ch, 59(1), 119–153. h ps://doi.o g/10.1177/0735633120952063
Reich, J. (2020). Failu e o dis up : Why echnology alone can’ ans o m educa ion. Camb idge, MA: Ha a d Uni e si y P ess. h ps://doi.o g/10.2307/j.c 322 4cp
* Riley, E. (2018). Role models in mo ies: The impac o Queen o Ka we on s uden s’ educa ional a ainmen . Cen e o he S udy o A ican Economies, Uni e si y o
Ox o d. CSAE Wo king Pape WPS/2017-13.
Rod iguez-Segu a, S. (2022). EdTech in de eloping coun ies: A e iew o he e idence. The Wo ld Bank Resea ch Obse e , 37(2), 171–203. h ps://doi.o g/10.1093/
wb o/lkab011
Romanelli, R. J., Dixon, W. G., Rod iguez-Wa son, C., Sau e , B. C., Alb igh , D., & Ma cum, Z. A. (2021). The use o na a i e elec onic p esc ibing ins uc ions in
pha macoepidemiology: A scoping e iew o he in e na ional socie y o pha macoepidemiology. PDS Pha macoepidemiology & D ug Sa e y, 30(10), 1281–1292.
h ps://doi.o g/10.1002/pds.5331
* Rouse, C. E., & K uege , A. B. (2004). Pu ing compu e ized ins uc ion o he es : A andomized e alua ion o a ‘‘scien i ically based’’ eading p og am. Economics
o Educa ion Re iew, 23, 323–338. h ps://doi.o g/10.1016/j.econedu e .2003.10.005.
* Ru he o d, T., Fa kas, G., Duncan, G., Bu chinal, M., Kib ick, M., G aham, J., e al. (2014). A andomized ial o an elemen a y school ma hema ics so wa e
in e en ion: Spa ial-Tempo al Ma h. Jou nal o Resea ch on Educa ional E ec i eness, 7(4), 358–383. h ps://doi.o g/10.1080/19345747.2013.856978.
* San ana, M., Mon i-Nussbaum, M., Ca mona, R., & Cla o, S. (2019). Ha ing un doing ma h: Tex messages p omo ing pa en in ol emen inc eased s uden
lea ning. Jou nal o Resea ch on Educa ional E ec i eness, 12(2), 251–273. h ps://doi.o g/10.1080/19345747.2018.1543374.
* Schac e , J., & Jo, B. (2016). Imp o ing low-income p eschoole s ma hema ics achie emen wi h Ma h Shel , a p eschool able compu e cu iculum. Compu e s in
Human Beha iou , 55, 223–229. h ps://doi.o g/10.1016/j.chb.2015.09.013.
* Schac e , J., Shih, J., Allen, C. M., DeVaul, L., Adkins, A. B., I o, T., e al. (2016). Ma h shel : A andomized ial o a p ekinde ga en able numbe sense
cu iculum. Ea ly Educa ion & De elopmen , 27(1), 74–88. h ps://doi.o g/10.1080/10409289.2015.1057462.
* Seo, H. K. (2017). Do school elec i ica ion and p o ision o digi al media deli e educa ional bene i s? Fi s -yea e idence om 164 Tanzanian seconda y schools. London:
In e na ional G ow h Cen e. Wo king Pape E-40308-TZA-2.
* Se en, E. (2023). Race o he able ? The impac o a pe sonalized able educa ional p og am. Educa ion, Finance and Policy, 18(2), 213–231. h ps://doi.o g/
10.1162/ed p_a_00359.
* Silande , M., Moo hy, S., Dominguez, X., Hupe , N., Pasnik, S., & Llo en e, C. (2016). Using digi al media a home o p omo e young child en’s ma hema ics lea ning:
Resul s o a andomized con olled ial. Socie y o Resea ch on Educa ional E ec i eness.
S anley, T. D., & Doucouliagos, H. (2012). Me a-Reg ession analysis in economics and business. London: Rou ledge.
S anley, T. D., & Doucouliagos, H. (2014). Me a- eg ession app oxima ions o educe publica ion selec ion bias. Resea ch Syn hesis Me hods, 5(1), 60–78. h ps://doi.
o g/10.1002/j sm.1095
S anley, T. D., & Doucouliagos, H. (2015). Nei he ixed no andom: Weigh ed leas squa es me a-analysis. S a is ics in Medicine, 34(13), 2116–2127. h ps://doi.o g/
10.1002/sim.6481
S anley, T. D., Ja ell, S. B., & Doucouliagos, H. (2010). Could i be be e o disca d 90% o he da a? A s a is ical pa adox. The Ame ican S a is ician, 64(1), 70–77.
h ps://doi.o g/10.1198/ as .2009.08205
S e ne, J. A. C., He n´
an, M. A., Ree es, B. C., Sa o i´
c, J., Be kman, N. D., Viswana han, M., e al. (2016). ROBINS-I: A ool o assessing isk o bias in non- andomised
s udies o in e en ions. B i ish Medical Jou nal, 355. h ps://doi.o g/10.1136/bmj.i4919
S e ne, J. A. C., Sa o i´
c, J., Page, M. J., e al. (2019). RoB 2: A e ised ool o assessing isk o bias in andomised ials. B i ish Medical Jou nal, 366, A icle
A icle14898. h ps://doi.o g/10.1136/bmj.l4898
Sun, L., Chen, X., & Ruokamo, H. (2021). Digi al game-based pedagogical ac i i ies in p ima y educa ion: A e iew o en yea s’ s udies. In e na ional Jou nal o
Technology in Teaching and Lea ning, 16(2), 78–92. h ps://doi.o g/10.37120/ij l.2020.16.2.02
Tanne -Smi h, E. E., & Tip on, E. (2014). Robus a iance es ima ion wi h dependen e ec sizes: P ac ical conside a ions and a so wa e u o ial in S a a and SPSS.
Resea ch Syn hesis Me hods, 5(1), 13–30. h ps://doi.o g/10.1002/j sm.1091
Ugu , M., Chu chill, S. A., & Luong, H. M. (2020). Wha do we know abou R&D spillo e s and p oduc i i y? Me a-Analysis e idence on he e ogenei y and s a is ical
powe . Resea ch Policy, 49, A icle 103866. h ps://doi.o g/10.1016/j. espol.2019.103866
an de We ho s , H. G., Kessenich, E., & Ge en, S. (2022). The digi al di ide in online educa ion: Inequali y in digi al eadiness o s uden s and schools. Compu e s and
Educa ion Open, 3, A icle 100100. h ps://doi.o g/10.1016/j.caeo.2022.100100
* Ve hallen, M. J. A. J., & Bus, A. G. (2010). Low-income immig an pupils lea ning ocabula y h ough digi al pic u e s o ybooks. Jou nal o Educa ional Psychology,
102(1), 54–61. h ps://doi.o g/10.1037/a0017133.
Wagne , T. (2016). Technology o educa ion in low-income coun ies: Suppo ing he UN sus ainable de elopmen goals. In I. A. Lubin (Ed.), ICT-suppo ed
inno a ions in small coun ies and de eloping egions. Educa ional communica ions and echnology: Issues and inno a ions (pp. 51–74). Cham: Sp inge . h ps://doi.o g/
10.1007/978-3-319-67657-9_3.
Wa schaue , M., & Xu, Y. (2018). Technology and equi y in educa ion. In J. Voog , G. Knezek, R. Ch is ensen, & K. W. Lai (Eds.), Second Handbook o in o ma ion
Technology in P ima y and seconda y educa ion. Sp inge in e na ional handbooks o educa ion. Cham: Sp inge . h ps://doi.o g/10.1007/978-3-319-71054-9_76.
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
23
* Wenne s en, M., Qu aishy, Z. B., & Velamu i, M. (2015). Imp o ing s uden lea ning ia mobile phone ideo con en : E idence om he B idgeIT India p ojec .
In e na ional Re iew o Educa ion, 61(4), 503–528. h ps://doi.o g/10.1007/s11159-015-9504-y.
Whi e, H., & Saba wal, S. (2014). Quasi-expe imen al design and me hods. In Me hodological b ie s: Impac e alua ion (Vol. 8). UNICEF.
Wildi-Yune, J., & Co de o, C. (2015). Co po a e digi al lea ning – how o ge i igh . KPMG AG, Posi ion Pape .
Wilke, U., & Pyka, A. (2024). Sus ainable inno a ions, knowledge and he ole o p oximi y: A sys ema ic li e a u e e iew. Jou nal o Economic Su eys. h ps://doi.
o g/10.1111/joes.12617 (in p ess).
* Wol , S., Abe , J. L., Beh man, J. R., & Tsinigo, E. (2019). Expe imen al impac s o he “Quali y P eschool o Ghana” in e en ions on eache p o essional well-
being, class oom quali y, and child en’s school eadiness. Jou nal o Resea ch on Educa ional E ec i eness, 12(1), 10–37. h ps://doi.o g/10.1080/
19345747.2018.1517199.
Yang, T. C., Cheng, M. C., & Chen, Y. (2018). The in luences o sel - egula ed lea ning suppo and p io knowledge on imp o ing lea ning pe o mance. Compu e s &
Educa ion, 126, 37–52. h ps://doi.o g/10.1016/j.compedu.2018.06.025
* Yang, Y., Zhang, L., Zeng, J., Pang, X., Lai, F., & Rozelle, S. (2013). Compu e s and he academic pe o mance o elemen a y school-aged gi ls in China’s poo
communi ies. Compu e s & Educa ion, 60, 335–346. h ps://doi.o g/10.1016/j.compedu.2012.08.011.
Zhang, J. J. Y., Lee, K. S., Voisin, M. R., He ey-Jumpe , S. L., Be ge , M. S., & Zadeh, G. (2020). Awake c anio omy o esec ion o sup a en o ial glioblas oma: A
sys ema ic e iew and me a-analysis. Neu o-Oncology Ad ances, 2(1), 1–12. h ps://doi.o g/10.1093/noajnl/ daa111
Zheng, B., Wa schaue , M., Lin, C.-H., & Chang, C. (2016). Lea ning in one- o-one lap op en i onmen s: A me a-analysis and esea ch syn hesis. Re iew o Educa ional
Resea ch, 86(4), 1052–1084. h ps://doi.o g/10.3102/0034654316628645
Zig aio a, D., Ha anek, T., I so a, Z., & No ak, J. (2021). How puzzling is he o wa d p emium puzzle? A me a-analysis. Eu opean Economic Re iew, 134, A icle
103714. h ps://doi.o g/10.1016/j.eu oeco e .2021.103714
G. Di Pie o and J. Cas a˜
no Mu˜
noz
Compu e s & Educa ion 226 (2025) 105197
24
