Mas e ’s Deg ee P og am in
In o ma ion Managemen
E ec i eness o Mobile Apps in Language Lea ning: A Compa a i e
S udy o Duolingo and T adi ional Class oom Me hods
Guilhe me Oli ei a da Di ei a Calhei os
Mas e Thesis
p esen ed as pa ial equi emen o ob aining he Mas e Deg ee P og am in In o ma ion Managemen
NOVA In o ma ion Managemen School
Ins i u o Supe io de Es a ís ica e Ges ão de In o mação
Uni e sidade No a de Lisboa
MGI
NOVA In o ma ion Managemen School
Ins i u o Supe io de Es a ís ica e Ges ão de In o mação
Uni e sidade No a de Lisboa
EFFECTIVENESS OF MOBILE APPS IN LANGUAGE LEARNING: A
COMPARATIVE STUDY OF DUOLINGO AND TRADITIONAL
CLASSROOM METHODS
By
Guilhe me Calhei os
Mas e Thesis p esen ed as pa ial equi emen o ob aining he mas e ’s deg ee in in o ma ion
managemen , wi h a specializa ion in In o ma ion Sys ems and Technologies Managemen
Supe ised by
Ví o Manuel Pe ei a Dua e dos San os, PhD, NOVA In o ma ion Managemen School
July 2024
ii
STATEMENT OF INTEGRITY
I he eby decla e ha ing conduc ed his academic wo k wi h in eg i y. I con i m ha I ha e no used
plagia ism o any o m o undue use o in o ma ion o alsi ica ion o esul s along he p ocess leading
o i s elabo a ion. I u he decla e ha I ha e ully acknowledge he Rules o Conduc and Code o
Hono om he NOVA In o ma ion Managemen School.
Lisbon, 15/07/2024
iii
DEDICATION
To my mum, o he elen less suppo and unwa e ing belie , whe e lo e is a symbol o esis ance.
To my dad, o he calm suppo and quie con idence, whe e aun ing mee s s eng h.
To my b o He , o he o e e p esence and my con idence, whe e legacy mee s he i age.
To my b o heR, o he men al boos and comple e unde s anding, whe e design mee s pa ience.
To my sIs e , o he ceaseless companionship and obnoxious us , whe e nu i ion mee s c eed.
To my iends, ha me me in hose la e nigh s, wi h las -minu e in o ma ion, o wo ied abou how
his wo k was p oceeding, hank you, his would no be possible alone.
To hose who no longe a e wi h us, o he p o ec ion and igilance, whe e loneliness was a mi age.
To he playlis s, I ha e c ea ed along he way, I will che ish e e y bea wi h lo e.
i
ACKNOWLEDGEMENTS
I would like o exp ess my g a i ude o my ad iso P o esso Ví o San os o his dedica ed
assis ance in guiding me h oughou he Thesis and he s udy ha I ealized o subs an ia e
my hesis.
I would, also, like o hank all he pa icipan s ha we e pa o my “Duolingo e sus
T adi ional Chinese classes” s udy, who ha e helped me each he esul s p esen ed in his
hesis and dedica ed hei ime and e o o lea ning a new and ha d language, such as
Chinese, aiding me p o ing he main p emise o my hesis and o he eache as well as o 李
老师. Thank you all.

ABSTRACT
Educa ion is su e ing a massi e mode nisa ion, and he ield o Language Lea ning is no excep ion.
T adi ional me hods used o each a language om sc a ch a e being challenged by con empo a y
In o ma ion Sys ems, especially by Mobile Language Lea ning Apps (LLAs). The popula i y, con enience
and gami ied expe ience o he mos popula LLA - Duolingo – has a ac ed millions o use s o lea n
a language e e y day. This s udy de elops bo h app oaches and compa es hem, es ing i s
e ec i eness in Language eaching, in a complex Language se ing: Chinese Manda in. Da a associa ed
wi h he pe o mance in language acquisi ion (LA), was mainly collec ed wice: be o e he s udy, in a
Diagnos ic es and a e wa ds, in a Final Exam ha comp ehended speci ic opics like Cha ac e s
ecogni ion, Lis ening comp ehension and P onuncia ion. In a s udy o 31 inal pa icipan s, T adi ional
app oach s uden s, sco ed on a e age, g ea e han Duolingo s uden s, in e e y aspec o e alua ion,
including he Final Sco e which is di ided in a W i ing Pa and an O al one. Al hough, he e iciency o
he Duolingo g oup was p o ed o be highe , on a e age, han he ones on T adi ional side. Bo h
g oups, o e all sco ed poo ly in ce ain a eas, such as Cha ac e ecogni ion, exposing he complexi y
o he Chinese Language om sc a ch, al hough he di e ence be ween bo h is obse able. The lack
o dialogue capabili ies is also unimp essi e om he Duolingo app oach, whe e s uden s a e no able
o answe a simple ques ion, hinde ing luency. Duolingo, as i s ands, is no able o ully eplace a
p o esso , by no p o iding op imal esul s, despi e i s po en ial as a complemen a y ool, due o i s
cons an access abili y and o he in oduc ion o new ocabula y.
KEYWORDS
m-Lea ning; Language Lea ning App; Duolingo; Chinese Language; T adi ional Educa ion; Lea ning
Sus ainable De elopmen Goals (SGD):
i
INDEX
1. In oduc ion ......................................................................................................... 1
1.1. Backg ound and p oblem iden i ica ion ................................................................ 1
1.2. Objec i es .............................................................................................................. 2
1.3. Impo ance and Rele ance .................................................................................... 2
2. Li e a u e Re iew ................................................................................................. 4
2.1. O e iew o he Language eaching p oblema ic ................................................. 4
2.1.1. Communica ion .............................................................................................. 4
2.1.2. Fi s Language & Second Language Acquisi ion ............................................. 4
2.1.3. Mo i a ion ...................................................................................................... 4
2.1.4. Age .................................................................................................................. 5
2.2. Pedagogical app oaches o Language eaching ................................................... 6
2.2.1. L2 S a egy ...................................................................................................... 6
2.2.2. Sel -Regula ion ............................................................................................... 6
2.2.3. Language Lea ning Theo ies ........................................................................... 7
2.3. The ole o echnology in Language eaching ....................................................... 8
2.3.1. O e iew ......................................................................................................... 8
2.3.2. Educa ion Roles .............................................................................................. 8
2.3.3. Technology in Language Lea ning .................................................................. 8
2.3.4. Dis ance Lea ning ........................................................................................... 9
2.3.5. Types o Technology in FL Lea ning ................................................................ 9
2.3.6. Mobile Lea ning ............................................................................................ 10
2.3.7. Language Lea ning Apps (LLAs) .................................................................... 11
2.4. Rela ed Wo k on he use o LLA o Language eaching ..................................... 12
3. Me hodology...................................................................................................... 14
4. Empi ical S udy .................................................................................................. 16
4.1. Expe imen al Design ............................................................................................ 16
4.1.1. Pa icipan s ................................................................................................... 16
4.1.2. G oup Assignmen ........................................................................................ 16
4.1.3. Lea ning Me hods ........................................................................................ 16
4.1.4. S udy Du a ion and Daily Commi men ....................................................... 16
4.1.5. Con ol and Moni o ing ................................................................................ 16
4.1.6. Rele an Fac o s Assessmen & In e iews ................................................. 17
4.1.7. Da a Collec ion ............................................................................................. 17
ii
4.2. Execu ion ............................................................................................................. 18
4.2.1. P e s udy ....................................................................................................... 18
4.2.2. S udy ............................................................................................................. 19
5. Resul s and discussion ........................................................................................ 25
5.1. Time Spen Lea ning ............................................................................................ 25
5.2. Pa icipa ion ........................................................................................................ 26
5.3. Final Exam Resul s ............................................................................................... 26
5.4. E ec i eness ........................................................................................................ 27
5.5. Subskills E alua ion ............................................................................................. 28
5.5.1. Chinese Cha ac e s ....................................................................................... 28
5.5.2. Lis ening........................................................................................................ 29
5.5.3. Pinyin ............................................................................................................ 29
5.5.4. P onuncia ion skills....................................................................................... 30
5.5.5. Dialogue ........................................................................................................ 32
5.6. Quali a i e analysis .............................................................................................. 32
6. Conclusions and u u e wo ks ............................................................................. 33
6.1. Syn hesis o he de eloped wo k ........................................................................ 33
6.2. Limi a ions ........................................................................................................... 34
6.2.1. S udy ............................................................................................................. 34
6.2.2. Ma e ial ........................................................................................................ 34
6.2.3. In luences ..................................................................................................... 34
6.3. Fu u e wo k ......................................................................................................... 34
Bibliog aphical REFERENCES ...................................................................................... 2
Appendix A. Duolingo S uden s Sc eensho s ........................................................... 11
Appendix B. Ques ions om he W i en exam ....................................................... 13
Appendix C. O al Exam CRITERIA ............................................................................. 14
Appendix D. Duolingo egis a ion able .................................................................. 15
Appendix E. S uden s gene al sco es ....................................................................... 16
Appendix F. comp ehensi e O al sco e .................................................................... 17
iii
LIST OF FIGURES
Figu e 1 - Example o a Diagnos ic Exam’s ques ion ................................................................ 19
Figu e 2 - One o he ques ions asked when con igu ing Duolingo's Chinese class……………... 20
Figu e 3 - Sc eensho om Duolingo's exe cise o sen ence o ganisa ion .............................. 21
Figu e 4 - Slideshow p esen ed du ing a T adi ional App oach Class…………………………………… 21
Figu e 5 - Full Hea s in Duolingo……………………………………………………………………………………..… 22
Figu e 6 - Losing all hea s in he middle o a lesson…………………………………………………………….22
Figu e 7 - Example o ques ions ep esen ing sounds ............................................................. 23
Figu e 8 - Example o ques ions ep esen ing Cha ac e s and Meaning…………………………….. 23
Figu e 9 - Final G ade A e age o S uden s o each App oach, in which g een ep esen s
Duolingo, and blue ep esen s T adi ional Classes. Maximum 29 poin s. Minimum
Posi i e Line in 14.5…………………………………………………………………………………………………… 27
Figu e 10 - Pe cen age o co ec answe s, be ween Duolingo ( ep esen ed by he g een
colou ) and T adi ional S uden s ( ep esen ed by he blue colou ), by componen s o he
W i ing Exam .................................................................................................................... 30
Figu e 11 - Box Plo o P onuncia ion, Duolingo & T adi ional……………………………………………. 31
5
The enowned Linguis Resea che Zol án Dö nyei (2009) exp esses ha e en a lea ne which has
excellen ins uc ion and cu icula, wi h an ou s anding capaci y will no ha e a gua an eed success in
i s long- e m objec i es wi hou mo i a ion and ha , con e sely, a s ong d i e and ambi ion can easily
compensa e o he “de iciencies bo h in one’s language ap i ude and lea ning condi ions.” (Dö nyei,
1998).
The c ea ion o amewo ks ha ha e been accoun ing mo i a ion o a mo e success ul Second
Language Lea ning ha e been a ound o se e al decades, being he mos p eeminen wo ks
“Ga dne ’s (1985) o iginal socio-educa ional model” (Anczewski, 1987) , “Deci and Ryan’s (1985) sel -
de e mina ion heo y (SDT)” (Deci & Ryan, 1985) and “Dö nyei’s L2 Mo i a ional Sel Sys em”, he
la e became ully cons uc ed in 2009, con ibu ing o he c ea ion and de elopmen o mo e
empi ical esea ches in o de o analyse i s use in p ac ical and eal en i onmen s (Dö nyei, 2009).
F om he e, Teache s may be amilia wi h hese amewo ks and heo ies, al hough he u ilisa ion o
hese “ heo e ical p inciples” in o he lea ne en i onmen is no di ec and simple (Ushioda, 2020).
2.1.4. Age
Ano he de e minan ac o ha may a ec he abili y o lea n second Language is Age (Hu, 2016).
Resea che s ha e s udied he younge age lea ne ’s abili y o be mo e p o icien , in e ms o acqui ing
a second language (Lennebe g, 1967), as i s a ed o de elop as a common belie e, ha in adul hood
a pe son can no longe de elop o al p o iciency in L2, whe eas be o e i , a young lea ne can
comple ely e ol e in o a na i elike L2 speake due o he occu ence o a de elopmen al pe iod
(Bi dsong, 2005). Suppo ing Lennebe g, Child en’s b ains ha e a s ong plas ici y which would allow
a g a e second Language acquisi ion, in a mo e na u al and e o less way, belie ing in an op imal age
o lea n a L2 (Xu, 2023). Howe e , a my iad o s udies we e analysed, which ega ded he age o
lea ne s and he du a ion o he o eign language lea ning, did no concluded a clea and explici
ad an age o younge lea ne s (Huang, 2016). Mo eo e , esea che s ha e concluded in empi ical
s udies ha olde lea ne s a e mo e likely o ob ain g ea e abili y in sho - e m analysis (Huang, 2016;
K ashen e al., 1979). And i we ex end his pe iod, o a long- e m analysis we may obse e a ca ch-up
om he younge lea ne s, as we can obse e in a la ge -scale expe imen s, such as in he Ba celona
Age Fac o , despi e he o e all p o iciency did no su pass i s pee s (Muñoz, 2006).

6
2.2. PEDAGOGICAL APPROACHES FOR LANGUAGE TEACHING
Th oughou he XX cen u y, conce ns a ose ega ding he e ec i eness o ce ain lea ne s in acqui ing
a second language, p omp ing s udies o e alua e a ious lea ning s a egies, especially he ones
success ul a i . Se e al s udies we e ini ially unde aken o unde s and he di e ence be ween
lea ne s o de elop a second language p o iciency, which poin ed 3 inhe en s a egies as he ones
used by success ul second language lea ne s: ac i e guesse , meaning an e ec i e s o age o
in o ma ion, which can be e ie ed e o lessly, ea less ega ding his appea ance o ain i s
communica ion and new-sen ence-make , by mixing acqui ed con en (Rubin, 1975).
2.2.1. L2 S a egy
Since hen, pedagogical s a egies ha e e ol ed as well as i s de ini ion, so o dis inguish be ween
“ heo y and p inciples and a se o de i ed p ocedu es” in he Language Teaching a ea, i was a
de eloped a de ini ion o each concep (Richa ds & Rodge s, 2014), sepa a ed by a hie a chy, ini ia ing
in he app oach - “axioma ic” - desc ibed as a collec ion o p esump ions and belie s ega ding bo h
language eaching and lea ning, explaining he essence o wha is going o be augh , whe eas he
me hod – “p ocedu al” – is connec ed o he s uc u ed gene alized s a egy employed including he
skills used, he p o ided con en and i s p esen a ion o de o Language Lea ning, always ega ding
i s p e ious selec ed app oach, inishing in he undamen al laye whe e we ind he echnique –
“implemen a ional” - oo ed o i s p ac ice in a class oom, using a speci ic “ ick, s a agem o
con i ance” applied o achie e a compelling goal (An hony, 1963).
Mo eo e , he e o o execu e a axonomy di iding all exis en s a egies ha exis ed a a ime was
in o me acogni i e ( ocusing on he lea ning p ocess by paying a en ion, men al links, c ea ing goals,
plans and oppo uni ies, and sel -e alua ion o he p ocess), a ec i e (by con olling he anxie y, by
elaxa ion echniques and laugh e , posi i e ein o cemen , and lis ening o he own body, ou inely),
social (by bonding wi h mul i-cul u al comp ehension and awa eness o o he s eelings, asking
ques ions and h ough colleagues coope a ion) comp ised he indi ec me hods g oup, opposi e o he
di ec s a egies g oup which include memo y ones (using associa ions, con ex ualiza ion o new
wo ds, image y, key wo ds and sounds use, o ganized e iew and physical ac ion employmen ),
gene al cogni i e (by p ac icing a language, wi h aid o eple ion, also messaging, deeply mind
easoning, and summa izing and no e w i ing) and compensa ion used o o e come es ain s (using
guessing, and he esou ce o help by ge ing i o using mo he ongue and miming) (Ox o d, 1990,
2002); which is ela ed o he axonomy accomplished by O’Malley and Chamo (1990) ha di ides he
s a egy in o 3 di e en ypes: cogni i e, me acogni i e and social/a ec i e s a egies whose au ho s
ecognized he la e as a gene alized concep , o include all he o he s a egies ha would no i in o
he i s wo ca ego ies (O’Malley & Chamo , 1990).
2.2.2. Sel -Regula ion
Howe e , scep icism a ound he e m s a egy s a ed o g ow, when Dö nyei (2005) exposed he
eplacemen made by he scien i ic communi y in o a mo e lexible domina ion o sel - egula ion, due
o he change in he second Language acquisi ion a ea o s udy in o a “mo e applied ealm o language
eaching me hodology” om a b oade and mo e simple esea ch subjec (Dö nyei, 2005). This shi
occu ed because i was unde s ood ha lea ning s a egies we e demons a ed as a a mo e complex
opic han p e iously desc ibed, wi h an ambiguous de ini ion, concen a ing solely on i s “su ace
7
mani es a ions” - which is he ac ics, echniques and s a egies used by lea ne s - will no p ope ly
add ess he ield, so he in oduc ion o he e m sel - egula ion, o e en “sel - egula ed Lea ning” was
undamen al o swi ch he esea ch ocus o he examina ion o mo e dynamic and p ocess-o ien ed
a iables om he p e ious lea ning/ cogni i e echniques, as well as, he ex ension o o he academic
ields o psychological and cogni i e sciences as a majo opic.
2.2.3. Language Lea ning Theo ies
The de elopmen o Language Lea ning heo ies ha e pe mi ed an ad ancemen on he s udy o he
language acquisi ion a ea, which in i s co e mus be able o conside he quan i y o elemen s inhe en
and he numbe o si ua ions and con ex s o he lea ne ’s language acquisi ion (Ligh bown & Spada,
2021), as well as he “cogni i e, pe sonal, in e pe sonal, and social” cha ac e is ics o he lea ne when
lea ning a second language (Richa ds & Rodge s, 2014).
Beha iou ism has eme ged, h ough a se ies o expe imen s, ha essen ially a pe son may lea n
h ough he esponse o i s en i onmen o hei ac ions, in which h ough “ope an condi ioning” is
epo ed ha a posi i e o ewa ding o a ce ain beha iou , i “inc eases he p obabili y o u he
esponding” (SKINNER, 1957). Howe e , humans a e capable o “ a ional hough ”, wi h an inhe en
capaci y o p oblem-sol ing, in e p e ing, memo ising, and hinking, and a e no only a esul o he
ex e io , which is desc ibed by he Lea ning Theo y Cogni i ism (Cla k, 2018). Cons uc i ism is a
heo y, gaining popula i y among lea ning li e a u e despi e i s di icul y o be de ined (Null, 2004),
based on he na u e o knowledge, indi iduals de elop hei knowledge om hei own expe iences,
al hough i may no be always indica i e o " he eal wo ld" and i s inhe en expe iences (K e chma ,
2021).
In o de o be a p o icien lea ne , one mus espec h ee undamen al p inciples: ac i i y h ough
ac ion, concep o be lea ned, and he con ex o he lea ning p ocess, ha mus be e ol ed and used
o esul in knowledge (B own e al., 1989), whe e bo h he lea ne and i s en i onmen a e “c i ical
o he cons uc i is ”: Knowledge is c ea ed by he in e ac ion o bo h ac o s (E me & Newby, 2013).
8
2.3. THE ROLE OF TECHNOLOGY IN LANGUAGE TEACHING
2.3.1. O e iew
The in eg a ion o communica ion and in o ma ion echnology in o Language educa ion has boos ed
he in e es and sc u iny among esea che s and specialis s, ini ially in class ooms en i onmen s, due
o i s capaci y o p o ide “lea ning ma e ials and inpu s” o s uden s, which may be p oduced di ec ly
o language lea ne s (Skehan, 2003), pe o mance enhancemen o s uden s in he majo ca ego ies
o a L2 and de elopmen o in e es and exci emen (F igaa d, 2002) and a isible luidi y, whe e i was
ecognized by FL lea ne s he g ea e independence gi en, whe e i was possible o con ol hei own
lea ning p ocess (Lam & Law ence, 2002). The educa o s, on he o he hand, ha e a c ucial ole on he
implemen a ion o echnologies, which i can be a new sou ce o ideas and ad ancemen s in language
eaching (Timucin, 2006) and unde s and how undamen al was “indi idualizing his eaching p ac ice”
(Lam & Law ence, 2002).
2.3.2. Educa ion Roles
The implemen a ion o echnology has ma ked a u n on he ela ionship’s dynamic, u ning he
Language Lea ning class oom in o a mo e “lea ne -cen ed” one, whe e he eache plays a ole o he
class di ec assis an o he s uden s’ guide ha is eadily a ailable o he s uden and i s lea ning
p ocess (Lam & Law ence, 2002; Lee & VanPa en, 2003), ins ead o only g ading a w i en pape and
make s uden s comple e “unmanageable e ision asks”, he eache will be ocusing on g oup p ojec s
and hei p og ess epo , changing i s ole o a suppo e and no only ad ise bu guide as men ioned
be o e. Hence, he class oom will eplace he use o a mo e o mal “ ansmission o in o ma ion”,
p ocess also known as A las Complex, a e m ha desc ibes he o al esponsibili y and au ho i y o
he eache on he lea ning p ocess, whe e s uden s con ibu e no hing o he class, being me e
ecei e s o in o ma ion, esul ing on he lack o powe “ o make hings happen o hei s uden s” in o
a “lea ning ac i i y” p ocess ha u ns s uden s in o hei own lea ning manage s, by he eache being
capable o moni o each s uden indi idually (Bes , 1997; Finkel & Monk, 1983; Lee & VanPa en,
2003).
I is also impo an o no e he impo ance o p o esso s o he success o Language Lea ning
Technologies, as Timuçin (2006) desc ibes he implemen a ion o echnologies in any eaching o
lea ning en i onmen is deeply dependen on how engaged and commi ed he ela ed pa ies a e,
whe e epo s show how eache s a e no longe “ he physical o in ellec ual on o he class” due o
he decen aliza ion p oduced by compu e echnology (Ci ello, 1999; Timucin, 2006).
No wi hs anding, eache s may ea he “loss o con ol” in he class oom, due o he b oad use o
compu e s, which may esul in eaching s ic e linguis ic ules and de ini ions due o he lack o
knowledge on how hey unc ion, di ec ly hi by a nega i e iew o new echnologies (Picka d, 1994)
o , i may be pe cei ed in a bene icial way, as s uden s will assis he p o esso in any oubles ha
may a ise in he use o echnologies, allowing he la e o ocus on i s pedagogic unc ions,
con ibu ing o a heal hy implemen a ion o echnology in he pedagogical cu iculum (Nellen, 1999).
2.3.3. Technology in Language Lea ning
The use o echnology, and i s subsequen de elopmen in he p ac ice o Language eaching and
Lea ning, may be coined as CALL (Compu e Assis ed Language Lea ning), al hough he de ini ion and
9
meaning o his umb ella e m is highly deba able (Le y & Hubba d, 2005). The Longman Dic iona y o
Language Teaching and Applied Linguis ics, desc ibes CALL as no only “ he use o a compu e in he
eaching and lea ning o a second o o eign language” bu also he lea ning h ough ac i i ies ha
includes he use o o he ypes o media (al hough a compu e is needed o o he uses), class oom
ac i i ies ha uses compu e p og ams ha helps s uden s in di e en ca ego ies, such as w i ing by
helping s uden s choosing a heme and “by checking a composi ion o ocabula y, g amma and opic
de elopmen ” and also, CALL exclusi e ac i i ies (Richa ds & Schmid , 2013).
Essen ially, he main hemes ha ha e eme ged om CALL s udy and p ac ice e ol e a ound he
educa ional oppo uni ies ha come wi h using echnology (Chapelle, 2005).
2.3.4. Dis ance Lea ning
The expansion o In e ne use, he no able g ow h o use in social compu ing, he con inuous esea ch
o CALL, and he inhe en in e es in eaching languages, allowed he c ea ion o ano he e m ha is
subs an ially impo an in hese “new lea ning en i onmen s”: Dis ance Lea ning (DL) (Blake, 2009).
Resea ch in ol ing his opic desc ibes DL as a lea ning ype ha occu s any ime he e is a empo al o
spa ial sepa a ion be ween he lea ne and i s eache , ha may be subdi ided in o di e en gen es,
such as “online lea ning, e-lea ning, open lea ning, dis ibu ed lea ning, elecon e ence o ideo-
con e ence lea ning, blended o hyb id lea ning, li e-long lea ning, o independen s udy” (Olson,
2015) al hough i may be also di e en ia ed by he use o hei own media (i.e. p in , audio, compu e -
based, In e ne -media sou ce) as well as, he exis ence o epo ed di e ences in whe he he class
has li e al echnological imp o ed media o i i is jus a mix u e be ween scheduled p esen ial classes
in he class oom and online lessons ( his may be used as synonym, by some esea che s) (Blake, 2009).
2.3.5. Types o Technology in FL Lea ning
Resea che s ha e been measu ing he ela i e impac and e ec i eness o di e en na u es o
echnology in L2, whe he hey a e p ac ically es ablished in ou socie y like PC and sma phones, o
a e s ill in i s ma u ing phase, such as na u al language p ocessing (NLP) (Golonka e al., 2014).
Technology’s p og ession has na u ally in luenced he Language Teaching ield by i s di e en ypes,
shown by se e al s udies, whe e he mos p ominen and mode n a e included in di e en ca ego ies,
a ising Ne wo k-Based Language Teaching (NBLT), which is he p ocess o eaching a language ha
uses “local o global ne wo ks”, ia compu e s connec ed o hemsel es opening he access o lea ning
o he wo ld, wi h a mul i ude o language lea ning ools (Ke n & Wa schaue , 2000). The s udies
associa ed wi h NBLT in ol e he unde s anding o lea ne s and hei ela ionship wi h speake s o he
language and media, h ough “compu e -media ed con ex s o in e ac ion” (Ke n e al., 2017) .
This e m may be di ided in o synch onous compu e -media ed communica ion (CMC), whe e he e
is in e ac ion wi h ano he pe son in eal ime (Blake, 2009) ha includes online cha s – bene icial as
i was obse ed ha lea ne s would be mo e eage o co ec hemsel es when p oducing a message
han in ace- o- ace dialogues, as an example (Lai & Zhao, 2006; Yamada, 2009) and ideo con e encing,
conside ed he mos widesp ead known ype o dis ance lea ning is he connec ion be ween
compu e s, equi ing a sc een, a cama a and loud speake s, wi h some ypes a ached: Desk op
ideocon e encing, whe e he communica ion is one- o-one, al hough i also con ains an in e ac ion
“Teache - o-Class”, wi h a pe sonal compu e connec ed (O’Dowd, 2006). Synch onous CMC will
10
enhance collabo a i e lea ning en i onmen s - eaching echniques ha encou age g oup
collabo a ion among s uden s o p oduce he bes possible lea ning esul (Sun & Yuan, 2018) – and
o ced s uden s o communica e o hei pee s (wi h di e en na i e languages), in he esea ch
de eloped wi h child en o di e en coun ies whe e English was a L2, lea ne s would in oduce
hemsel es in English o would pa icipa e in a coope a i e way o a common objec i e, in o de o
ensu e “ eal pu pose o he s uden s o use he a ge language (communica ing eal ideas)”(Dooly &
Sadle , 2016); and in o asynch onous CMC, wi h a long esea ch wi h i s in ol emen in DL, eeing
e e yone in ol ed om space/ ime cons ain s (Allen & Seaman, 2006; Blake, 2009), as he mos
p ominen cu en ool is social media: Ins an messaging applica ion s udies, such as Wha sApp which
p o ed his me hod e icacy in ime sa ing among s uden s as well as pe mi ing bo h asynch onous
and synch onous communica ion and ease o p ocess and deli e in o ma ion, in a s udy o g oup
magazine co-c ea ion (A ci & Adiguzel, 2017), pe sonal Blogs whe e s uden s we e engaged o “build
on hei w i ing skills” as obse ed in a s udy in Spain, conduc ed by Ruby Vu dien whe e hey ha e
de eloped awa eness and p ecau ion in ask planning and in e o co ec ion (Vu dien, 2013) , and
Facebook as i is capable o in e ac ion wi h use s om o he s coun ies and und hei in e es in he
language o mula ion (Lan z-Ande sson e al., 2013) al hough long ime use was o epo ed o a ec
s uden s posi i e pe cep ion on “ease o access” and he sea ch o “school- ela ed in o ma ion”
(Gamble & Wilkins, 2014).
2.3.6. Mobile Lea ning
Addi ionally, he signi ican g ow h o “po able and wi eless de ices and echnologies” (examples a e
mobile phones, lap ops, Wi-Fi, Blue oo h, 3G o 5G) o e he yea s ha e pe mi ed indi iduals o access
pedagogical con en o “any ime, anywhe e lea ning” (A ani is & K ys alli, 2021; Golonka e al.,
2014). Al hough Mobile Lea ning (m-Lea ning) is a he di icul o desc ibe due o i s e olu ion o e
ime, which leads o he uppe -men ioned Longman Dic iona y o Language Teaching and Applied
Linguis ics poin ing i as he lea ning ha akes ad an age o po able echnologies, ha may occu
ac oss loca ions, whe he wi h he use o po able de ices o Language Lea ning o g oup pa icipa ion
in he class oom h ough mobile de ices (Richa ds & Schmid , 2013), al hough 5 cha ac e is ics we e
comp essed and p esen ed in he ea ly yea s o Mobile Lea ning Resea ch: Pe manency, Accessibili y,
Immediacy, In e ac i i y and Si ua ing o Ins uc ional ac i i ies (Oga a & Yano, 2004). Pe manency
ela es o he lea ne s’ pe pe ual wo k eco d, i he e is no pu pose ul dele ion o he con en ; access
o hei educa ion is p o ided wi h hei eques ; wi h access o hei educa ion anywhe e means he
Accessibili y, whe eas ge ing in o ma ion any ime i is eques ed, is in he Immediacy domain,
In e ac i i y desc ibes he capaci y o in e ac wi h pee s o eache s, whe he in a synch onous o
asynch onous o m, and he Si ua ing o ins uc ional ac i i ies ansla es in he imme sion o he
lea ning p ocess in he daily li e, in an au hen ic o m (Chen e al., 2002; Oga a & Yano, 2004).
O e he yea s, mobile echnology esea ch has been gaining ele ance, s anding as he “ he mos
equen ly in es iga ed”, due o he ad an ages ha a e associa ed o i , including he ema kable
accessibili y, he lexibili y in lea ning s yle and he ex a con en wi h ad an ageous cha ac e is ics
(Zhang & Zou, 2022).
The e olu ion o mobile de ices is e lec ed on he s udies p oceeded on Mobile Lea ning, whe e
Chinne y (2006) in oduced he exp ession MALL (Mobile-Assis ed Language Lea ning), i s domina ed
by he Pe sonal Digi al Assis an s (PDAs) which we e p ima ily used as a ansla o , bu also due o

11
access o In e ne , sha ing documen s be ween he s uden s/ lea ne en i onmen was a ailable
(Chinne y, 2006). A e iew o se e al s udies in mobile lea ning segmen , epo ed ha his ool was
only use ul in “ eading and w i ing ac i i ies”(Golonka e al., 2014)
Po able media playe s, such as iPods, could ansmi high-quali y sound, highly a ed by s uden s and
“downloadable so wa e” including Language Lea ning ones, whe e i s use was documen ed in class
and uni e si y en i onmen s, bu wi h no “empi ical e idence” o he e ec i eness o his sys em
(Chinne y, 2006; Golonka e al., 2014). Howe e , iPod popula ized Podcas s, whe e s uden s eel p one
o use his ype o media as a pedagogical ool and, consequen ly hey assess his use as a bene i ,
when aced by i , o a my iad o easons o de ed by he p o esso (Abdous e al., 2009).
2.3.7. Language Lea ning Apps (LLAs)
The s a e-o - he-a , on he mobile lea ning en i onmen , begins wi h he p incipal app oach o mobile
applica ions o Language Lea ning, which is anked as he mos s udied ype o echnology o
Language Lea ning, h oughou ecen yea s, whose majo pu poses o he esea ch a e, in a
descending o de : P omo ing P ac ises, Deli e ing Ins uc ional Con en , Facili a ing In e ac ions and
Res uc u ing Teaching App oaches (Zhang & Zou, 2022).
Mobile Assis ed Language Lea ning (MALL) bene i s s uden s by enabling lea ning anywhe e and
any ime wi h hei mobile de ices. In addi ion, i is possible o, om p e ious esea ch, ex apola e
ad an ages in mobile echnology applied o Lea ning: adap abili y, consis ency, immedia e eedback,
socialisa ion, pe sonalisa ion, ac i e engagemen , pee coaching, sel -e alua ion, ou doo s inspi a ion
sou ces, and cul u al au hen ici y (Kukulska‐Hulme & Vibe g, 2018). P e ious esea ches ha e been
ocusing p ima ily on he English Language as a L2, and in he acquisi ion o ocabula y (Hwang & Fu,
2019), showing ha “62.8% o he measu emen s showed posi i e esul s”, on mobile language
lea ning.
Fu he mo e, Resea ch ques ions ound emphasises in which ype o games exis ed, which mobile
de ices do people use and he use o esea ch o me hodologies and wha we e he employed lea ning
s a egies (Cheng, 2023).
I was also possible o obse e ha indi idualized eedback on pe o mance is some hing many
lea ne s a ely ge ou side o mal ui ion, ge ing answe s co ec in quizzes, o using apps o memo ize
e b o ms and ocabula y, a e ewa ding ac i i ies, and use s a e sa is ied o knowing hey go
some hing igh . While some apps con inue o o e d illing wi h li le eaching and lack o meaning ul
eedback o suppo , some examples o good p ac ice a e now a ailable, pa icula ly among apps ha
o e a ull language-lea ning expe ience (e.g. Duolingo, Busuu) (Rosell-Aguila , 2017).
Applica ions may also be used o online dic iona ies ha allow lea ne s o imp o e hei p onuncia ion
and expand hei ocabula y (Na iya i e al., 2020). I is usual o apps o combine bo h unc ions, as
ansla o s and dic iona ies including Google T ansla e, DeepL, Wo dRe e ence, he Camb idge, Ox o d
and Collins dic iona ies, and applica ions like Fo o, which ac s as eposi o y o a ious p onuncia ions,
including a ie ies and dialec s.
The u ili y o LLA also include he mo i a ion o lea ning using echnology amilia o s uden s such as
sma phones and able s, gi ing he oppo uni y o hem o de elop all six communica ion skills:
“comp ehension and p oduc ion o w i en speech, comp ehension and p oduc ion o spoken
12
language, au al in e ac ion and au al and w i en media ion” and he encou agemen and suppo o
use he a ge language as he unique mean o communica ion (A ani is & K ys alli, 2021), al hough
a lea ning p og amme exclusi ely sma phone o ien ed as i may exclude many lea ne s (Godwin-
Jones, 2011).
I was ound ha LLAs each ocabula y in a he “isola ed uni s” han in no mal con ex s, a e
unadap able o indi idual needs and in e es s, and seldom exhibi explana o y co ec i e eedback o
lea ns (Heil e al., 2016), which may be disas ous since lea ning “becomes mo e eal” i he e a e
e e ences o he lea ne i sel , whe e mobile phone pe o ms a c ucial ole, due o i s abili y o going
beyond he academic en i onmen (Godwin-Jones, 2011)
2.4. RELATED WORK ON THE USE OF LLA FOR LANGUAGE TEACHING
Resea ch on he e ec i eness o MALL in SLA is “limi ed”, exis ing only a ew wo ks s udying i h ough
“quan i a i e da a ” o “expe imen al design”, bu when i is possible o measu e i , MALL e eals
“lea ning ad an ages” o a my iad o a eas like eading and lis ening (Loewen e al., 2019; Shadie e
al., 2017).
Howe e , he e ec i eness o comme cial p og ams o online language lea ning is unce ain due o
he exis ence o s udies es ing se e al echnologies including Duolingo whose “imp o emen was
s a is ically signi ican ” (Vesselino & G ego, 2012), al hough some ese a ions a e made ega ding
hese esul s, and he absence o da a a ailable , “ o comme cial p oduc s”, which u n he esul s
dubious (Loewen e al., 2019; Van Deusen–Scholl, 2015).
Loewen e al. (2019) de eloped a simila s udy, wi h 12 s uden s on he e ec i eness o Duolingo, bu
o s udy Tu kish which esul s ha e come “somewha mixed”, whe e s uden s knew Tu kish on he end
o he cou se bu no enough o pass he es p esen ed (Loewen e al., 2019).
In a quali a i e analysis on he e ec i eness o Duolingo, i was claimed ha DL is an “easy, ee and
p ac ical lea ning p ocess” whe e he esul s con idence was high, al hough i was epo ed ha
despi e some s uden s liked lea ning ocabula y and p onuncia ion, o he s complain abou he
G amma and how needs imp o emen (Psychogyiou & Ka asimos, 2019)
The compa ison be ween Asynch onous CMC and Synch onous CMC wi h adi ional cu iculum
(con ol g oup) was esea ched by Ab ams (2003) who examined and ied o con i m he e ec i eness
o CMC in “ ace- o- ace in e ac ions” and he di e ences be ween bo h CMC in o al p oduc ion. The
esea che ound ha he g oup in ol ing Synch onous CMC p oduced be e han he o he wo
g oups, and ha su p isingly he con ol g oup wen be e han he asynch onous CMC (Ab ams,
2003).
I was ound in a su ey, ha al hough mos s uden s enjoyed using a LLA (in his case Busuu), 64%
e ealed ha acqui ing a o eign language only by LLA is impossible, ecognizing i as mo e o a “help ul
ool”(Ké yi, 2013). In e ms o mo i a ion, a esea che ound he e was an in insic mo i a ion wi h
he s uden s in ol ed, when using a speci ic LLA (Gamlo, 2019).
13
When es ing he app ecia ion, Resea che s ound LLA o ha e a c ucial ac o : i s “ubiqui y”, so
s uden s will ha e a mo e pe manen con ac wi h he language, due o he mobili y o a sma phone,
which ha e also p o ed he s imula ion due o “sel -lea ning” and “gaming aspec s” cha ac e is ics
(Ga ni e al., 2017). Al hough, in he same s udy, In e ne dependence was also s a ed as a
disad an age.
14
3. METHODOLOGY
To achie e he p oposed esea ch objec i es, as shown in Table 1, he S udy will be composed o h ee
majo phases: explo a ion, analy ical, conclusi e. These phases a e essen ial o an o e a ching
de elopmen on he esea ch ocused on he Duolingo e ec i eness, on s uden s lea ning a language
om he absolu e beginning, assis ing he p ocess o decision-making in he conclusion o he hesis.
In he i s phase, he heme will be explo ed by e iewing exis en pape s, on LL me hodologies,
o e iewing he heo ies a ound he p ocesses o eaching a new language and how echnology is
in luencing SLA. How lea ning can be a ec ed and wha in luences i s success will be esea ched, by
compa ing ea ly s udies o unde s and he cha ac e is ics o each app oach and hei in e ac ion.
The li e a u e e iew will in luence how he me hodology will occu , by assessing he expe imen al
amewo k used in analogous s udies, he e o e implemen ing mo e app op ia e p ocesses, measu es
and p ocedu es on he p esen s udy.
The analy ical phase cons i u es he cons uc ion o he Expe imen al F amewo k which englobes he
P e-S udy Phase whe e he e is Selec ion o he ma e ial ha will be aim o s udy and he Pa icipan ’s
Rec ui men . Du ing his phase, he p io knowledge o he s uden s will be assessed and consequen ly
assigned in o he s udy in a andom app oach.
The esea ch execu ion, s ep 4 in he Analy ical phase, will con ain he pedagogical pe iod, whe e he
s uden s will lea n a language uniquely by one o he app oaches. Las ly, he s uden s will comple e
he cou se h oughou a mon h, which is he ime spen on he lea ning endea ou ha will be
moni o ed and con olled. On he LLAs, compliance will be ollowed by he opening and closu e o he
app, while on he class oom he p esence will be eco ded.
Explo a ion
Phase S ep 1 - Li e a u e e iew
S ep 2 - Me hodology e iew and imp o emen
Analy ical
Phase S ep 3 - Build a expe imen al amewo k
S ep 4 - Execu e he expe ience
S ep 5 - Re ei e expe imen al esul s
Conclusi e
Phase S ep 6 - Resul s e alua ion and discussion
S ep 7 - Conclusions and esul s ap esen a ion
S ep 8 - Final e isions
Table 1 - Th ee-S ep Me hodology
Figu e 1 - Example o ques ions ep esen ing sounds
Figu e 2 - Example o ques ions ep esen ing sounds
Figu e 3 - Example o ques ions ep esen ing sounds
21
Th oughou he cou se, Duolingo uni s a e di ided in o: Basic Food and D inks, Na ionali ies,
P o essions, P e e ences, Family, and ad e bs o equency deli e ing i s con en by epea ing a my iad
o exe cises, om ma ching pai s (Cha ac e s & Pinyin o i s ansla ion), mul iple choice ques ions,
and w i ing he sen ence lis ened o he mos equen ype o ques ion which is o ganising he
ansla ion o a sen ence ei he o Chinese o o English, in small lesson.
Figu e 3 - Sc eensho om Duolingo's exe cise o sen ence o ganisa ion
P o esso s usually ollow a p e-chosen s uden wo kbook, which i is used wi h eache s own ma e ial,
exhibi ing a slide show, o example. The P o esso will in oduce he Manda in language, by explaining
and co ec ing s uden s’ speaking (Pinyin and ones p onuncia ion), which was ollowed by: G ee ings,
In oduc ion o Syn ax, Wo ds ( e bs, nouns, ad e bs…), In e oga i e s uc u e, Numbe s, Dialogue
aining, In oduc ion o Cha ac e ’s W i ing, Food, Shopping and Cu ency. Uni s we e accompanied
by Chinese Cul u al e e ences and explana ions, o a u he unde s anding o he language
con ex ualising li e in China, and he in e ac ion by eache ’s asking s uden s o pa icipa e by
epea ing new wo ds and sen ences, un il he co ec p onuncia ion was spoken as well as, he
dialogues be ween s uden s, encou aged o gi e elemen s a ce ain deg ee o eedom, educing he
online ba ie , keeping s uden s engaged and ocused on he class which also simula es a eal-wo ld
en i onmen .
Figu e 4 - Slideshow p esen ed du ing a T adi ional App oach Class

22
4.2.2.4. S uden ’s Lea ning P ocess
Lea ne s on he Duolingo comple ed se e al lessons so hey could pass om uni o uni , ini ially
a e aging 12 minu es on he app o un il hey ha e los all hea s - Duolingo as a eemium pla o m
only allows 5 o al e o s in he lessons, disallowing he s uden o con inue lea ning in he ee e sion,
e illing each one e e y 5 hou s - which ep esen ed, on he ini ial days 12.5% o he o al o he
pa icipan s. This p oblem al hough b u ally sup essed, o e ime, ha e exis ed h oughou he whole
expe imen , whe e i i happened wice in a ow, s uden s’ mo i a ion would clea ly d op.
T adi ional app oach’s s uden s we e di ec ly accompanied by he P o esso and he esea che du ing
classes, whe e se e al di icul ies we e obse able du ing he i s hou s o he classes, mainly he
p onuncia ion o pinyin and he unde s anding o ones.
Figu e 5 - Full Hea s in Duolingo Figu e 6 - Losing all hea s in he middle o a lesson
4.2.2.5. Final Assessmen
The inal exam was sepa a ed in a w i ing pa and an o al pa . The w i ing pa was di ided in o 2
main pa s, o e alua e bo h app oaches ai ly, s a ing wi h 9 ques ions wi h con en ha ma ched
bo h lea ning app oaches, ollowing he ypes o exe cise p oposed by Xiwen e al. (2019), whe e i
was s udied in an online en i onmen , how he lea ne ’s s udy be ween each o “ he componen s o
Chinese wo d acquisi ion (sounds, meanings and cha ac e s)” could in luence s uden s’ inal esul s.
Besides, exe cises ocused on sounds, meanings and cha ac e s, as o me ly men ioned, i was also
included Pinyin ocused exe cises. The second pa o he W i ing Exam included 4 ques ions exclusi e
o he app oach o he s uden (Duolingo o Classes wi h Teache ), o e alua e he p og ess in he
espec i e app oaches.
All ques ions we e o mul iple choice, always wi h an op ion o “I don’ know”, coun ing 0 poin s. I he
lea ne , chooses he co ec answe will be awa ded wi h a poin (1), bu i i chooses he inco ec
one, a poin will be deduc ed om he inal g ade (-1). The maximum g ade is 13.
The ollowing images (Images 7 and 8) exempli y he ques ions ha we e used in he Final Assessmen ,
in i s Po uguese Ve sion, showing he di e si y o he componen s e alua ed, bu also he quiz o e all
simplici y. (see Appendix B o see o he examples o ques ions on he Exam)
23
Figu e 8 - Example o ques ions ep esen ing Cha ac e s and Meaning
To ensu e an accu a e esul om he es ake s, e lec ing he impo ance o a oiding e o s,
s uden s will ha e i s poin s deduc ed i hey choose an inco ec answe , which may help de e mine
p o iciency be ween s uden s, ensu ing a ai assessmen , educing he impac o andom decisions.
The exam had a du a ion o 7 minu es, plus ex a ime, i necessa y, whose only access was a p e-
scheduled Zoom session, wi h he esea che , o moni o he exam, and ensu e esul s’ eliabili y and
us wo hiness o he inal g ades. The ques ionnai e also included a sec ion o s uden s o pu hei
own names.
The o al pa is done a e he end o he w i ing pa . I is composed, ini ially, by 5 main ques ions,
which we e di ided, in ol ing he p onuncia ion o wo ds, he eading o sen ences in bo h cha ac e s
and Pinyin, and he ansla ion o spoken sen ences. A e wa ds, he s uden s a e asked, acco ding o
hei app oach, ques ions o de elop a dialogue, which we e 3 o he adi ional app oach (T1, T2 and
T3), and only one o he Duolingo class (D1), due o he lack o con en in his domain. (See Appendix
C, o he c i e ia p oposed on he O al Exam)
The O al Exam was di ec ed as ollowed:
1) How do you p onounce wa e in Chinese?
2) A e you able o ead "你是中国人"? (You a e Chinese)
2.1) Do you know wha i means?
2.2) I unable o comple e 2), may you ead i s pinyin "nǐ shì zhōngguó
én"?
3) Can you ansla e he sen ence "我是老师 - wǒ shì lǎoshī." when spoken?
(I am a eache )
4) Can you say "Nice o mee you" in Manda in?
Figu e 7 - Example o ques ions ep esen ing sounds
24
4.1) Read he ollowing pinyin "Wǒ hěn gāoxìng ènshí nǐ".
5) I i is said “这是米饭”, can you ansla e i ? (This is ice)
T1) 你叫什么名字? (Wha is you name?)
T2) 你爱吃什么? (Wha do you lo e ea ing?)
T3) 咖啡多少钱? (How much does i cos he co ee)
D1) 你喜欢茶吗? (Do you like ea?)
The sco ing o he w i en exam was au oma ic, h ough a Qual ics su ey, which was ansposed in o
a sp eadshee , wi h he g ade in each ques ion, and he espec i e name o he s uden , whe eas he
ques ions in he o al e alua ion, we e g aded om 0 o 5, conduc ed in wo ways: a di ec e alua ion,
in ol ing di ec ques ions like ansla ions, cha ac e eading and ques ion answe ing and; an
in e p e a ion assessmen , whe e he p onuncia ion was g aded by mo e han one na i e Manda in
speake who ha e g aded owa ds an ID numbe ha was p o ided o ensu e sec ecy and anonymi y
o he s uden .
The inal g ade o each s uden was 50% o he w i en pa and 50% o he o al pa .
To inish he e alua ion, s uden s we e andomly asked some ques ions in he end, whe he i hey
we e sa is ied and mo i a ed and i hey ha e any ecommenda ion o u u e language lea ning apps.
25
5. RESULTS AND DISCUSSION
The analysis o he collec ed da a om he s udy and i s ollowing esul s, compa ing how ab ini io
s uden s who exclusi ely lea n he Chinese Manda in language h ough he Duolingo app wi h
s uden s who lea n om adi ional app oach wi h a p o esso on a i ual en i onmen , will ocus
no only on speci ic exe cises sco e, inal g ades, he ime spen on lea ning, and e ec i eness o he
lea ning me hods, bu also on a quali a i e analysis wi h he eedback p o ided by he pa icipan s.
5.1. TIME SPENT LEARNING
O iginally, i was eques ed ha s uden s in he class oom, should a end all classes, o alling 6 hou s,
whe eas Duolingo s uden s we e eques ed o pe o m a o al o 4 hou s and 40 minu es o 7 hou s
wi hin 28 days. (To see he cou se comple ion, please see he ollowing image and Appendix D)
Howe e , he e we e ins ances o class absences and skipped days on Duolingo, which may accu a ely
e lec eal-li e lea ning en i onmen s. Due o he need o sel -managemen , ha has come ac oss as
a he di icul o main ain, he Duolingo’s s uden s eco ded, on a e age, a lowe ime spen lea ning,
o 03:56 hou s (3.94), wi h Classes’ s uden s a e aging, an hou mo e: 4:57 hou s (4.96).
To be accep ed in he esul s, s uden s mus ha e comple ed, a leas a ull hou .
In he Duolingo domain, due o i s eedom, compa a i ely wi h a scheduled class wi h a p o esso , i
was possible o obse e he s iking di e ences o ime be ween lea ne s, anging om 01h18m o a
s iking 9h30m, o eal ime spen on he app.
Table 2 - A endance egis a ion o he T adi ional App oach s uden s
26
5.2. PARTICIPATION
The pa icipa ion in he inal exam was sligh ly highe om membe in he Duolingo side (n=17), han
he ones who ha e had classes wi h he Chinese P o esso (n=14).
5.3. FINAL EXAM RESULTS
The inal g ade o each s uden is he alue ha sums, he g ade in bo h w i ing and o al exam, whe e
each skill ep esen s 50% o he g ade. The maximum g ade achie able is 29, wi hs anding he
maximum ma k o 13 in he W i ing exam and he 45 poin s in he o al pa .
The w i ing exam has exposed signi ican di e ences be ween candida es, in ei he app oach.
Duolingo’s s uden s ha e, a e aged 6.82 poin s, sligh ly highe han he minimum posi i e g ade,
whe e 47% o he s uden s ha e ob ained a nega i e g ade (6 poin s o less). In he o he hand, he
adi ional app oach has a e aged, 8.93 poin s be ween i s s uden s, whe e wo ha e eached he
maximum punc ua ion. Howe e , i is he la e who ha e egis e ed he highes di e ence be ween
he maximum alue and he minimum alue (2 and 13).
In he o al pa , Teache ’s me hod has su passed wi h s uden s, a e aging a punc ua ion o 25.02,
opposing he 20.24 poin s ha he Duolingo s uden a e aged. The di e ence accen ua es be ween
he op s uden s om each app oach, wi h a di e ence o 11 poin s, o he adi ional side. On
a e age, bo h app oaches, ha e classi ied poo ly, anking nex o he posi i e h eshold.
The inal g ade ollows he pa e n o he p e ious esul s, whe e he e is a lack o consis ency on he
poin s sco ed, wi h a signi ican di e ence be ween g ades, whe e i is ound ha he T adi ional
app oach s uden s, on a e age, (16.86 poin s) su passes he Duolingo ones (13.53 poin s) in 3.33
poin s.
Fo a comple e and exhaus i e sco e in bo h pa s, o each s uden , please see Appendix E.

27
Figu e 9 - Final G ade A e age o S uden s o each App oach, in which g een ep esen s Duolingo, and blue ep esen s
T adi ional Classes. Maximum 29 poin s. Minimum Posi i e Line in 14.5.
5.4. EFFECTIVENESS
Based on he Duolingo e ec i eness o mula, i s measu e accoun s, no only he esul s, bu also he
ime spen dedica ed on he lea ning p ocess, which will be a “ ai measu e” o he e ec i eness o
bo h Duolingo and Class ooms, measu ing he a ion be ween e ec and e o s (Vesselino & G ego,
2012):
𝐸𝑓𝑓𝑒𝑐𝑡𝑖𝑣𝑒𝑛𝑒𝑠𝑠 = 𝐸𝑓𝑓𝑒𝑐𝑡
𝐸𝑓𝑓𝑜𝑟𝑡𝑠 = 𝐼𝑚𝑝𝑟𝑜𝑣𝑒𝑚𝑒𝑛𝑡 𝑜𝑓 𝐿𝑎𝑛𝑔𝑢𝑎𝑔𝑒 𝑠𝑘𝑖𝑙𝑙𝑠
𝑆𝑡𝑢𝑑𝑦 𝑇𝑖𝑚𝑒 = 𝐹𝑖𝑛𝑎𝑙 − 𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑡𝑒𝑠𝑡 𝑠𝑐𝑜𝑟𝑒
𝐻𝑜𝑢𝑟𝑠 𝑜𝑓 𝑠𝑡𝑢𝑑𝑦
Equa ion 1 – E ec i eness Fo mula
The esul s o he equa ion ha e pu he e ec i eness o Duolingo - 4.21 - abo e he e ec i eness o
T adi ional me hods - 3.36.
Al hough, he esul s on he exam ha e sugges ed a be e a e age on he inal exams, o he
adi ional app oach, Duolingo has shown a be e deli e o lea ning me hods pe hou o s udy.
13,53
16,86
0,00
14,50
29,00
Duolingo T adi ional
Final G ade by Ap oach, On A e age
28
Table 3 - E ec i eness Calcula ion o each S uden , and espec i e esul s in W i ing and O al pa
5.5. SUBSKILLS EVALUATION
Th oughou he inal exam, as a mean o u e ly e alua e he language acquisi ion, di e en ques ions
we e composed o iden i y 5 key componen s, essen ial o a success ul p o iciency on he Chinese
Language: eading and comp ehension o Chinese Cha ac e s, lis ening and ansla ion o sen ences,
p onuncia ion skills, unde s anding o Pinyin, and capaci y o p oduce dialogues and speech.
5.5.1. Chinese Cha ac e s
Bo h w i ing and o al pa s, assessed s uden s’ knowledge on he Cha ac e s o he Chinese Language,
also called, Hanzi, by asking 3 ques ions in he o m o mul iple choice, on he w i ing pa whe eas on
he o al examina ion, s uden s we e asked o ansla e one sen ence.
When he ques ions we e app oached in he w i ing exam, s uden s on he T adi ional app oach ha e
had a g ea e success, on answe ing co ec ly, wi h a success a e o 52.38% pa icipan s, agains he
21.57% o Duolingo pa icipan s who ha e go he co ec answe s. Consequen ly, in he ollowing
pa , i was asked, he ansla ion o he sen ence “你是中国人。”, meaning “You a e Chinese”, augh
in bo h app oaches, which mos o he s uden s ha e comp ehended he na u e o he sen ence:
na ionali y, al hough no s uden could ha e de eloped which one, o s uc u ing a comple e co ec
answe ou o i (5 poin s). The sco e 0 was g an ed o 74% o all pa icipan s, whose a e age g ade
was 0.35 and 0.64, o he Duolingo app oach and adi ional one, espec i ely.
29
Bo h eams ha e achie ed ex emely low esul s, especially when he e a e no op ions a ailable,
e ealing he impac o he Language di e ence be ween a La in language wi h he Chinese Manda in
language, especially in he w i ing pa , which is comple ely di e en , u ning, o beginne s, Hanzi
ecogni ion ex emely ha d. Despi e low g ades, he P o esso ’s s uden s we e able o achie e a be e
g ade, on cha ac e s ecogni ion, ha could be associa ed wi h he emphasis placed by he eache on
his a ea, and wi hou an ex e nal guidance, Duolingo lea ne s did no unde s ood i s impo ance, and
neglec ed he s udy o Hanzi.
5.5.2. Lis ening
On he o he hand, lis ening o e all esul s ha e been abo e he a e age, whe e s uden s ha e also
been ques ioned in bo h pa s.
In he W i ing pa , s uden s lis ened o wo di e en audios and had o co espond o he co ec
pinyin and he co ec ansla ion. The success a e o each app oach su passed he 70% ma k, on bo h
ques ions. S uden s o he LLA had a success a e o 71%, while he s uden ’s p o esso had a a e o
co ec answe s o 79%.
Howe e , in he second pa o he exam, in a scale o 0 o 5, Duolingo s uden s ha e sco ed be e
esul s ha ing on a e age, a sco e o 3.91 poin s, while Class oom’s s uden s had on a e age, a sco e
o 3 poin s, on bo h ques ions gi en.
5.5.3. Pinyin
Simila o he lis ening e alua ion, he comp ehension o Pinyin was a conside able success. I was only
p esen in he w i ing pa , composed wi h 3 ques ions, ha we e basically Pinyin o i s meaning, o
meaning o i s Pinyin.
On he o al o he 3 ques ions, he success a e be ween each app oach was minimal, whe e Duolingo
app oach ha e sco ed less han he adi ional app oach, wi h 88,2% and 92,9%, espec i ely. This
success is also suppo ed by he absolu e numbe o people ha ha e answe ed co ec ly o all h ee
ques ions, which ou o 17 people, 12 Duolingo lea ne s we e co ec on Pinyin ocused ques ion,
whe e ou o he 14 s uden s, ep esen ing he class oom s uden s on he inal exam, 11 ha e s ood
co ec in all h ee ques ions.
These esul s ha e e ealed how Pinyin is undamen al o Chinese language lea ning beginne s, wi h a
simple and mo e amilia look, whe e a lea ne om he La in language quickly associa es wi h i s
own alphabe and sounds, o e coming he exigency and he di icul y o lea ning Hanzi om he s a ,
allowing a s eady lea ning.
30
Figu e 10 - Pe cen age o co ec answe s, be ween Duolingo ( ep esen ed by he g een colou ) and T adi ional S uden s
( ep esen ed by he blue colou ), by componen s o he W i ing Exam
5.5.4. P onuncia ion skills
The assessmen o o al p oduc ion and p onuncia ion o Chinese Manda in is a he complex, due o
he na u e o i s in e p e a ion and subjec i i y o he e alua o , in luencing he g ading p ocess,
despi e he exis ence o guidelines o punc ua ion, and he simila i y o di e en sounds. To o e come
con ex ual di e ences and cons ain s, s uden s we e asked i s ly o ead a wo d gi en, hen a simple
sen ence and in he end a complex sen ence, all in pinyin o main ain ai ness be ween con es an s.
Ac oss he h ee ques ions, which we e all g aded om 0 o 5, he adi ional app oach had on a e age,
3.40 poin s, a minimal di e ence compa a i ely wi h Duolingo’s s uden s – 3.25 poin s. O e all, he
g oups a e age was posi i e by a whole poin (3.32 poin s agains he 2.5 poin s o minimal posi i e
g ade), ansmi ing how bo h app oaches could be e ec i e in his ma e , al hough i may exhibi a
g ea e in e es and engagemen in his a ea o knowledge.
The numbe s also show a dis inc ion on he ecu ence o a g ade, be ween bo h me hods o lea ning:
he mode sco e on he Duolingo app oach was g ea e han he one in he T adi ional segmen , 4 and
3, espec i ely. Al hough, he s a is ical median was 3 on bo h.
The sco e 5 a e is he numbe o ques ions ha we e awa ded wi h 5 poin s, by app oach. 26,2% o
he ques ions answe ed by s uden s o he T adi ional side, had he pe ec a ing o 5, in
p onuncia ion skills, whe eas Duolingo candida es ha e answe ed 13.7% o ques ions wi h an excellen
ma k.
P onuncia ion skills, in he i ual class oom, a e es ed equen ly, whe he by di ec asking o by he
p omo ion o speaking ac i i ies be ween s uden s whe e eache may poin mis akes s aigh away,
deli e ing indi idual eedback abou he accen , ones, and syllables. This allows signi ican
imp o emen s in engaged s uden s and an o e all consis en imp o emen o he class, which
ansla es on he be e a e age and in he quali y o some answe s by some s uden s di ec ly ela ed
o a highe sco e 5 a e. The highe mode on he Duolingo s uden s, may be due o he epe i i e
lis ening o sounds on he exe cises and he comple ion o speaking exe cises, whe e al hough hey
0
10
20
30
40
50
60
70
80
90
100
Ra e o Co ec Answe s in %
Cha ac e s Lis ening Pinyin
Success Ra e on he W i ing Pa , by App oach
Blake, R. J. (2009). The Use o Technology o Second Language Dis ance Lea ning. The Mode n
Language Jou nal, 93(s1), 822–835. h ps://doi.o g/10.1111/j.1540-4781.2009.00975.x
Blake, R. J. (2013). B a e new digi al class oom: Technology and o eign language lea ning (Second
edi ion). Geo ge own Uni e si y P ess.
B own, J. S., Collins, A., & Duguid, P. (1989). Si ua ed Cogni ion and he Cul u e o Lea ning.
Educa ional Resea che , 18(1), 32–42. h ps://doi.o g/10.3102/0013189X018001032
Cenoz, J. (2013). De ining Mul ilingualism. Annual Re iew o Applied Linguis ics, 33, 3–18.
h ps://doi.o g/10.1017/S026719051300007X
Chapelle, C. A. (2005). Compu e -assis ed language lea ning. In Handbook o esea ch in second
language eaching and lea ning (pp. 743–755). Rou ledge.
Chen, Y.-S., Kao, T.-C., Sheu, J.-P., & Chiang, C.-Y. (2002). A mobile sca olding-aid-based bi d-
wa ching lea ning sys em. 15–22.
Cheng, F. (2023). The Role o Fi s Language in Second Language Acquisi ion. In B. Majoul, D. Pandya,
& L. Wang (Eds.), P oceedings o he 2022 4 h In e na ional Con e ence on Li e a u e, A and
Human De elopmen (ICLAHD 2022) (pp. 1236–1243). A lan is P ess SARL.
h ps://doi.o g/10.2991/978-2-494069-97-8_158
Chinne y, G. M. (2006). Going o he MALL: Mobile Assis ed Language Lea ning. Eme ging
Technologies, 10(1), 9–16.
Ci ello, C. A. (1999). ‘Mo e o e , Please’: The Decen aliza ion o he Teache in he Compu e -Based
Class oom. The English Jou nal, 88(4), 89. h ps://doi.o g/10.2307/822426
Cla k, K. R. (2018). Lea ning Theo ies: Cogni i ism. Radiologic Technology, 90(2), 176–179.
C ys al, D. (2012). English as a global language (Can o classics edi ion). Camb idge Uni e si y P ess.
Deci, E. L., & Ryan, R. M. (1985). In insic Mo i a ion and Sel -De e mina ion in Human Beha io .
Sp inge US. h ps://doi.o g/10.1007/978-1-4899-2271-7
Dooly, M., & Sadle , R. (2016). Becoming Li le Scien is s: Technologically-Enhanced P ojec -Based
Language Lea ning. Language Lea ning, 20(1), 54–78.

Dö nyei, Z. (1998). Mo i a ion in second and o eign language lea ning. Language Teaching, 31(3),
117–135. h ps://doi.o g/10.1017/S026144480001315X
Dö nyei, Z. (2005). The psychology o he language lea ne : Indi idual di e ences in second language
acquisi ion (digi al p in ing 2010). Rou ledge.
Dö nyei, Z. (2009). 2. The L2 Mo i a ional Sel Sys em. In Z. Dö nyei & E. Ushioda (Eds.), Mo i a ion,
Language Iden i y and he L2 Sel (pp. 9–42). Mul ilingual Ma e s.
h ps://doi.o g/10.21832/9781847691293-003
E me , P. A., & Newby, T. J. (2013). Beha io ism, Cogni i ism, Cons uc i ism: Compa ing C i ical
Fea u es F om an Ins uc ional Design Pe spec i e. Pe o mance Imp o emen Qua e ly,
26(2), 43–71. h ps://doi.o g/10.1002/piq.21143
Finkel, D. L., & Monk, G. S. (1983). Teache s and lea ning g oups: Dissolu ion o he a las complex.
New Di ec ions o Teaching and Lea ning, 1983(14), 83–97.
h ps://doi.o g/10.1002/ l.37219831411
Fishe , S. E. (2005). On Genes, Speech, and Language. New England Jou nal o Medicine, 353(16),
1655–1657. h ps://doi.o g/10.1056/NEJMp058207
F iedmann, N., & Rusou, D. (2015). C i ical pe iod o i s language: The c ucial ole o language inpu
du ing he i s yea o li e. Cu en Opinion in Neu obiology, 35, 27–34.
h ps://doi.o g/10.1016/j.conb.2015.06.003
F igaa d, A. (2002). Does he Compu e Lab Imp o e S uden Pe o mance on Vocabula y, G amma ,
and Lis ening Comp ehension?
Ga ni, R., Achi u , D. B., & Rachmani, G. J. (2017). Lea ning Fo eign Languages Using Mobile
Applica ions. Jou nal o In o ma ion Technology Educa ion: Resea ch, 16, 301–317.
Gamble, C., & Wilkins, M. (2014). S uden A i udes and Pe cep ions o Using Facebook o Language
Lea ning. Dimension, 49, 72.
Gamlo, N. (2019). The Impac o Mobile Game-Based Language Lea ning Apps on EFL Lea ne s’
Mo i a ion. English Language Teaching, 12(4), 49. h ps://doi.o g/10.5539/el . 12n4p49
Godwin-Jones, R. (2011). MOBILE APPS FOR LANGUAGE LEARNING. Language Lea ning &
Technology, 2, 2–11.
Golonka, E. M., Bowles, A. R., F ank, V. M., Richa dson, D. L., & F eynik, S. (2014). Technologies o
o eign language lea ning: A e iew o echnology ypes and hei e ec i eness. Compu e
Assis ed Language Lea ning, 27(1), 70–105. h ps://doi.o g/10.1080/09588221.2012.700315
Hu, R. (2016). The Age Fac o in Second Language Lea ning. Theo y and P ac ice in Language S udies,
6(11), 2164. h ps://doi.o g/10.17507/ pls.0611.13
Huang, B. H. (2016). A syn hesis o empi ical esea ch on he linguis ic ou comes o ea ly o eign
language ins uc ion. In e na ional Jou nal o Mul ilingualism, 13(3), 257–273.
h ps://doi.o g/10.1080/14790718.2015.1066792
Hummel, K. M. (2020). In oducing second language acquisi ion: Pe spec i es and p ac ices (2nd
edi ion). Wiley-Blackwell.
Hu o d, J. R. (2014). O igins o Language: A Slim Guide. OUP Ox o d.
Hwang, G.-J., & Fu, Q.-K. (2019). T ends in he esea ch design and applica ion o mobile language
lea ning: A e iew o 2007–2016 publica ions in selec ed SSCI jou nals. In e ac i e Lea ning
En i onmen s, 27(4), 567–581. h ps://doi.o g/10.1080/10494820.2018.1486861
Jodai, H., Za a ghandi, A. M. V., & Tous, M. D. (2013). Mo i a ion, In eg a i eness, O ganiza ional
In luence, Anxie y, and English Achie emen . Glo o heo y, 4(2).
h ps://doi.o g/10.1524/glo .2013.0012
Ke n, R., Wa e, P., & Wa schaue , M. (2017). Ne wo k-Based Language Teaching. In N. Van Deusen-
Scholl & S. May (Eds.), Second and Fo eign Language Educa ion (pp. 197–209). Sp inge
In e na ional Publishing. h ps://doi.o g/10.1007/978-3-319-02246-8_30
Ke n, R., & Wa schaue , M. (2000). In oduc ion: Theo y and p ac ice o ne wo k-based language
eaching. In M. Wa schaue & R. Ke n (Eds.), Ne wo k-based Language Teaching (pp. 1–19).
Camb idge Uni e si y P ess. h ps://doi.o g/10.1017/CBO9781139524735.003
Ké yi, A. (2013). Using Sma Phones in Language Lea ning – A Pilo S udy o Tu n CALL in o MALL. 20
Yea s o EUROCALL: Lea ning om he Pas , Looking o he Fu u e, 129–134.
h ps://doi.o g/10.14705/ pne .2013.000150
K ashen, S. D., Long, M. A., & Sca cella, R. C. (1979). Age, Ra e and E en ual A ainmen in Second
Language Acquisi ion. TESOL Qua e ly, 13(4), 573. h ps://doi.o g/10.2307/3586451
K e chma , J. (2021). Cons uc i ism. In Salem P ess Encyclopedia.
h ps://sea ch.ebscohos .com/login.aspx?di ec = ue&Au hType=ip,shib&db=e s&AN=89164
134&lang=p -p &si e=eds-li e&scope=si e
Kukulska‐Hulme, A., & Vibe g, O. (2 18). Mobile collabo a i e language lea ning: S a e o he a .
B i ish Jou nal o Educa ional Technology, 49(2), 207–218.
h ps://doi.o g/10.1111/bje .12580
Lai, C., & Zhao, Y. (2006). No icing and Tex -Based Cha . Language Lea ning & Technology, 10(3),
102–120.
Lam, Y., & Law ence, G. (2002). Teache -S uden Role Rede ini ion Du ing a Compu e -Based Second
Language P ojec : A e Compu e s Ca alys s o Empowe ing Change? Compu e Assis ed
Language Lea ning, 15(3), 295–315. h ps://doi.o g/10.1076/call.15.3.295.8185
Lan z-Ande sson, A., Vigmo, S., & Bowen, R. (2 1 ). C ossing bounda ies in Facebook: S uden s’
aming o language lea ning ac i i ies as ex ended spaces. In e na ional Jou nal o
Compu e -Suppo ed Collabo a i e Lea ning, 8(3), 293–312. h ps://doi.o g/10.1007/s11412-
013-9177-0
Lee, J. F., & VanPa en, B. (2003). Making communica i e language eaching happen (Second
edi ion). McG aw-Hill; Wo ldCa . h ps://www.js o .o g/s able/328644?o igin=c oss e
Lennebe g, E. H. (1967). The Biological Founda ions o Language. Hospi al P ac ice, 2(12), 59–67.
h ps://doi.o g/10.1080/21548331.1967.11707799
Le y, M., & Hubba d, P. (2 ). Why call CALL “CALL”? Compu e Assis ed Language Lea ning, 18(3),
143–149. h ps://doi.o g/10.1080/09588220500208884
Ligh bown, P., & Spada, N. (2021). How languages a e lea ned (Fi h edi ion). Ox o d Uni e si y P ess.
Loewen, S., C ow he , D., Isbell, D. R., Kim, K. M., Maloney, J., Mille , Z. F., & Rawal, H. (2019).
Mobile-assis ed language lea ning: A Duolingo case s udy. ReCALL, 31(3), 293–311.
h ps://doi.o g/10.1017/S0958344019000065
Mihalje ic Djiguno ic, J. (2012). A i udes and mo i a ion in ea ly o eign language lea ning. CEPS
Jou nal, 2(3), 55–74. h ps://doi.o g/10.25656/01:6787
Mille , G. A., & Gildea, P. M. (1987). How Child en Lea n Wo ds. Scien i ic Ame ican, 257(3), 94–99.
JSTOR.
Muñoz, C. (2006). Age and he a e o o eign language lea ning (Vol. 19). Mul ilingual Ma e s.
Na iya i, N. P. L., Sudi man, S., & P a iwi, N. P. A. (2020). EFL P e-Se ice Teache sâ€TM Pe cep ion
owa d he Use o Mobile Assis ed Language Lea ning in Teaching English. In e na ional
Jou nal o Language Educa ion, 38–47. h ps://doi.o g/10.26858/ijole. 4i2.10052
Nellen, T. (1999). Mo phing om Teache o Cyb a ian. Mul imedia Schools, 6(1), 20–25.
Null, J. W. (2004). Is Cons uc i ism T adi ional? His o ical and P ac ical Pe spec i es on a Popula
Ad ocacy. The Educa ional Fo um, 68(2), 180–188.
h ps://doi.o g/10.1080/00131720408984625
O’Dowd, R. (2 6, Janua y 1). Telecollabo a ion and he de elopmen o in e cul u al communica i e
compe ence. Langenscheid .
h ps://www.academia.edu/362172/Telecollabo a ion_and_ he_de elopmen _o _in e cul u
al_communica i e_compe ence
Oga a, H., & Yano, Y. (2004). Knowledge awa eness o a compu e -assis ed language lea ning using
handhelds. In e na ional Jou nal o Con inuing Enginee ing Educa ion and Li elong Lea ning,
14(4/5), 435. h ps://doi.o g/10.1504/IJCEELL.2004.005731
Olson, H. (2015). Key elemen s o e ec i e hyb id English as a second language p og ams o adul s.
O’Malley, J. M., & Chamo , A. U. (199 ). Lea ning S a egies in Second Language Acquisi ion (1s ed.).
Camb idge Uni e si y P ess. h ps://doi.o g/10.1017/CBO9781139524490
Ox o d, R. L. (1990). Language lea ning s a egies: Wha e e y eache should know. Newbu y House
Publ.
Ox o d, R. L. (2002). Language Lea ning S a egies in a Nu shell: Upda e and ESL Sugges ions. In J. C.
Richa ds & W. A. Renandya (Eds.), Me hodology in Language Teaching (1s ed., pp. 124–132).
Camb idge Uni e si y P ess. h ps://doi.o g/10.1017/CBO9780511667190.018
Picka d, V. (1994). Conco dancing o Schools: P oblems and Po en ial.
Psychogyiou, A., & Ka asimos, A. (2019). The e ec i eness o lea ning a o eign language ia a
dis ance lea ning ool: Τes ing he Duolingo applica ion. Selec ed Pape s on Theo e ical and
Applied Linguis ics, Vol 23, 364-380 Pages. h ps://doi.o g/10.26262/ISTAL.V23I0.7354
Richa ds, J. C., & Rodge s, T. S. (2014). App oaches and me hods in language eaching. Camb idge
uni e si y p ess.
Richa ds, J. C., & Schmid , R. W. (2013). Longman Dic iona y o Language Teaching and Applied
Linguis ics (0 ed.). Rou ledge. h ps://doi.o g/10.4324/9781315833835
Robins, R. H., & C ys al, D. (2024). Language. In Encyclopedia B i annica.
h ps://www.b i annica.com/ opic/language
Rosell-Aguila , F. (2017). S a e o he App: A Taxonomy and F amewo k o E alua ing Language
Lea ning Mobile Applica ions. CALICO Jou nal, 34(2). h ps://doi.o g/10.1558/cj.27623
Rubin, J. (197 ). Wha he ‘Good Language Lea ne ’ Can Teach Us. TESOL Qua e ly, 9(1), 41.
h ps://doi.o g/10.2307/3586011
Shadie , R., Hwang, W.-Y., & Huang, Y.-M. (2017). Re iew o esea ch on mobile language lea ning in
au hen ic en i onmen s. Compu e Assis ed Language Lea ning, 30(3–4), 284–303.
h ps://doi.o g/10.1080/09588221.2017.1308383
Sil e i, M. M. (2012). Adolescen B ain De elopmen and Unde age D inking in he Uni ed S a es:
Iden i ying Risks o Alcohol Use in College Popula ions. Ha a d Re iew o Psychia y, 20(4),
189–200. h ps://doi.o g/10.3109/10673229.2012.714642

Skehan, P. (2003). Focus on Fo m, Tasks, and Technology. Compu e Assis ed Language Lea ning,
16(5), 391–411. h ps://doi.o g/10.1076/call.16.5.391.29489
SKINNER, B. F. (1957). THE EXPERIMENTAL ANALYSIS OF BEHAVIOR. Ame ican Scien is , 45(4), 343–
371. JSTOR.
Sun, P. (Paul), & Yuan, R. (E ic). (2018). Unde s anding collabo a i e language lea ning in no ice-le el
o eign language class ooms: Pe cep ions o eache s and s uden s. In e ac i e Lea ning
En i onmen s, 26(2), 189–205. h ps://doi.o g/10.1080/10494820.2017.1285790
Timucin, M. (2006). Implemen ing CALL in an EFL con ex . ELT Jou nal, 60(3), 262–271.
h ps://doi.o g/10.1093/el /ccl006
Uni ed S a es Go e nmen Accoun abili y O ice. (2017). DEPARTMENT OF STATE Fo eign Language
P o iciency Has Imp o ed, bu E o s o Reduce Gaps Need E alua ion.
Ushioda, E. (2020). Language lea ning mo i a ion. Ox o d Uni e si y P ess.
Van Deusen–Scholl, N. (2015). Assessing Ou comes in Online Fo eign Language Educa ion: Wha A e
Key Measu es o Success? The Mode n Language Jou nal, 99(2), 398–400.
h ps://doi.o g/10.1111/modl.12234_2
Vesselino , R., & G ego, J. (2012). Duolingo e ec i eness s udy. Ci y Uni e si y o New Yo k, USA,
28(1–25).
Vu dien, R. (2013). Enhancing w i ing skills h ough blogging in an ad anced English as a Fo eign
Language class in Spain. Compu e Assis ed Language Lea ning, 26(2), 126–143.
h ps://doi.o g/10.1080/09588221.2011.639784
Xu, L. (2023). The Age E ec in Second Language Acquisi ion and I s S udy Design Me hod. Jou nal o
Educa ion, Humani ies and Social Sciences, 8, 1215–1222.
h ps://doi.o g/10.54097/ehss. 8i.4454
Yamada, M. (2009). The ole o social p esence in lea ne -cen e ed communica i e language lea ning
using synch onous compu e -media ed communica ion: Expe imen al s udy. Compu e s &
Educa ion, 52(4), 820–833. h ps://doi.o g/10.1016/j.compedu.2008.12.007
Zhang, R., & Zou, D. (2022). Types, pu poses, and e ec i eness o s a e-o - he-a echnologies o
second and o eign language lea ning. Compu e Assis ed Language Lea ning, 35(4), 696–
742. h ps://doi.o g/10.1080/09588221.2020.1744666
APPENDIX A. DUOLINGO STUDENTS SCREENSHOTS
14:02 – 14:12 -- 10 minu es
2 days in a ow
Uni 1
19:52 – 20:02 -- 10 minu es
17 days in a ow
Uni 2