Proceedings of the 8th International Conference on English Pronunciation: Issues and Practices

Held a he Uni e si y o Can ab ia, Spain May 2024
Published 2025
Uni e si é G enoble-Alpes
h ps://doi.o g/10.5281/zenodo.17319582
2025
Table o Con en s
Acknowledgemen s
i
In oduc ion: Looking back and mo ing o wa d o online EPIP p oceedings
Ped o Humánez-Be al, Alice Hende son and Anas azija Ki ko a-Nasko a
i
The ole o p onuncia ion in he assessmen o lis ening skills in he Polish Ma u a
exam
Agnieszka B yła-C uz
1
EFL lea ne s’ pe cep ions o na u alness, di icul y, and speech a e in na i e
connec ed speech
I ana Duckinoska-Mihajlo ska
15
Hunga ian seconda y school eache s’ p onuncia ion eaching p ac ices: A
ques ionnai e s udy
Noémi Gyu ka
28
Fluency in spon aneous desc ip ion e sus s o y- eading: Lea ne pe o mance and
iews
Xa ie Ma in-Rubió
43
A c i ical examina ion o esea ch on he eaching o p onuncia ion in a second
language
Ma ha C. Penning on
55
Li e o lea e! Vowel quali y and owel leng h in he in elligibili y o ad anced
Spanish-accen ed English
Ma eusz Pie azek
71
Wo d-ini ial oicing in L2 English: E alua ions o /z/ p oduc ions be o e and a e
pe cep ual aining
Sidsel H. Rasmussen
84
P epa ed o escue Cinde ella? Explo ing Aus ian EFL s uden eache s’ belie s
abou p onuncia ion eaching and lea ning
Ka in Rich e
97
P onuncia ion ea u es no iced by L1 Vie namese MOOC use s: In sea ch o
sociolinguis ic salience
Lau a Rupp and Alice Hende son
109
The e ec s o explici p onuncia ion ins uc ion: A s udy o ad anced Aus ian
EFL s uden s
Vale ie Whi e
-
Hau zinge , Ma ija Djenadic, Ca olin Rumple ,
and
Mi iam Fiala
118
I
ndex
129
i
Acknowledgemen s
Wo king on he EPIP8 P oceedings has been a joy ul expe ience o us as edi o s. Wi h e e y
published p oceedings, we eel like we ha e achie ed ano he miles one – his ime adding
ano he book o he EPIP online collec ion hos ed on he HAL pla o m. We uly hope ha he
open access publica ion app oach will make he au ho s mo e isible and hei esea ch mo e
a ailable.
Exchanging ideas wi h o he colleagues in he ield, in such a iendly and suppo i e
a mosphe e, has been en iching and ewa ding in many ways. We a e e e so g a e ul o he
b illian e iewe s we we e o una e o ge on boa d. They gene ously dedica ed hei ime and
expe ise o p o ide cons uc i e eedback by gi ing insigh ul commen s and sugges ions,
which ha e signi ican ly enhanced he quali y o he chap e s. We a e indeb ed o he ollowing
schola s (in alphabe ical o de ):
Gemma A che Uni e si y o S a hclyde, Sco land
Alex Ba a a Uni e si y o Manches e , England
Małgo za a Ba an-Łuca z Uni e si y o W ocław, Poland
Vincen Chane hom P ince on Uni e si y, USA
T acey De wing Uni e si y o Albe a, Canada
Jonás Fouz-González Uni e si y o Mu cia, Spain
Joshua Go don Uni e si y o No he n Iowa, USA
Nadine He y-Béni Pa is Nan e e Uni e si y, F ance
Sophie He men Aix-Ma seille Uni e si y, F ance
Pekka Lin unen Uni e si y o Tu ku, Finland
John Le is Iowa S a e Uni e si y, USA
José Mompéan-Gonzalez Uni e si y o Mu cia, Spain
Wayne Rimme Uni e si y o Manches e , England
A kadiusz Royczyk Uni e si y o Silesia in Ka owice, Poland
Radek Ska ni zl Cha les Uni e si y in P ague, Czech Republic
Sinem Sonsaa -Hegelheime Iowa S a e Uni e si y, USA
Nick T a e s Camosun College, Canada
Gabo Tu scan Aix-Ma seille Uni e si y, F ance
Elina Vasu Tampe e Uni e si y, Finland
Jan Volin Cha les Uni e si y in P ague, Czech Republic
We also ex end ou hea el app ecia ion o he inc edible au ho s who con ibu ed hei
esea ch o publica ion. Thei p o essionalism and b oad-mindedness made ou coope a ion a
shee deligh .
ii
We a e hank ul o he eam a he Uni e si y o G enoble-Alpes o lending suppo wi h
he echnical aspec o he publica ion p ocess, in pa icula Sand ine Co ey-Bi on and
Anne-Ch is ine Jacob a he uni e si y lib a ies. Thei gen le guidance has been ins umen al
in o e coming a ious challenges and ensu ing he accu acy o he ex s when published on he
pla o m. Special hanks a e due o Aynu Kaso om Ss. Cy il and Me hodius Uni e si y in
Skopje o helping us wi h he design o he book co e .
Finally, we wish o exp ess ou deepes and since es g a i ude o ou lo ed ones, ou pilla s
o pa ience, encou agemen , and unwa e ing suppo . You ha e kep us ocused and mo i a ed
du ing his c ea i e jou ney.
Alice Hende son, Anas azija Ki ko a-Nasko a and Ped o Humánez-Be al
F ance, No h Macedonia, and Spain
Oc obe 2025

iii
This chap e is he in oduc ion o he p oceedings o he 8 h In e na ional
Con e ence English P onuncia ion: Issues and P ac ices (EPIP 8) held May
8–10, 2024 a he Uni e si y o Can ab ia in San ande , Spain. I is licensed
unde he C ea i e Commons A ibu ion 4.0 In e na ional License, a copy o
which can be ound a
h p://c ea
i ecommons.o g/licenses/by/4.0/
.
Humánez-Be al, P., Hende son, A., & Ki ko a-Nasko a, A. (2025). In oduc ion: Looking back and mo ing
o wa d o online EPIP p oceedings. In A. Ki ko a-Nasko a, P. Humanez-Be al, & A. Hende son (Eds.),
P oceedings o he 8 h In e na ional Con e ence on English P onuncia ion: Issues and P ac ices (pp. i –xii).
Uni e si é G enoble-Alpes. h ps://doi.o g/10.5281/zenodo.17273047
In oduc ion: Looking back and mo ing o wa d o online EPIP
p oceedings
Ped o Humánez-Be al
Uni e si y o Can ab ia
Alice Hende son
G enoble-Alpes Uni e si y
Anas azija Ki ko a-Nasko a
Ss. Cy il and Me hodius Uni e si y, Skopje
These p oceedings encompass ex ended accoun s o selec ed o al p esen a ions and
plena ies om he con e ence English P onuncia ion: Issues and P ac ices – EPIP8, held in
2024 a he Uni e si y o Can ab ia in San ande , Spain. EPIP is an in e na ional, bi-annual
con e ence de o ed o how English p onuncia ion is augh and lea n . I s main objec i e is o
b ing oge he and inspi e connec ions be ween eache s and esea che s om all o e he
wo ld, be hey p e- o in-se ice eache s, unde g adua e o g adua e s uden s, ea ly ca ee o
in e na ionally enowned esea che s. People om all ca ee s ages a end EPIP con e ences,
whe e hey can ne wo k in pe son bu also explo e scien i ic, social and pedagogical issues
ela ed o English p onuncia ion. Some con e ence pa icipan s ha e he luxu y o jus
a ending, o he s in es a g ea deal o ime and ene gy in o p epa ing (and deli e ing) a alk –
and hen some o he la e submi p oposals o he p oceedings.
A labou -in ensi e, h ee-s age p ocess lies behind hese compiled p oceedings: i s ,
con e ence p esen e s commi o shaping hei wo k in o a w i en ex , which hen unde goes
a me iculous double-blind pee - e iew p ocess, and inally he e ised ex is ca e ully edi ed
by he co-edi o s – o en in ol ing many back-and- o h exchanges. Some con ibu ions a e
ejec ed bu , as EPIP p ides i sel on welcoming esea che s in o he ield, e iewe s a e
encou aged o p o ide cons uc i e c i icism and edi o s wo k ha d o hone as many
con ibu ions as possible, espec ing he p inciples o anspa ency and scien i ic igou .
To each as many eade s as possible and as quickly as possible, as edi o s we chose o make
EPIP P oceedings a ailable eely online s a ing in 2023 a e he EPIP7 con e ence. Simila
Humánez-Be al, Hende son & Ki ko a-Nasko a
Looking back, mo ing o wa d
o he PSLLT a chi es1 a Iowa S a e Uni e si y, he online EPIP p oceedings2 a e a eliable
place o ead inno a i e wo k being done now in ou ield, helping us all o upda e ou
knowledge e en when we canno a end con e ences. We decided o use an Open Access digi al
a chi e and o p ocu e Digi al Objec Iden i ie s (DOI) o each ex . DOIs a e impo an in he
long e m o he ield, because documen s emain accessible a a s able link. They a e also
c ucial in he sho e m o many o ou colleagues who li e and wo k in coun ies whe e DOIs
a e (o a e becoming) s a egically impo an o hei na ional ins i u ions. Mo eo e , we sough
o ob ain he DOIs om a Eu opean ins i u ion (Zenodo3), because we suppo Eu opean Union
aims o eedom, peace, social jus ice, scien i ic p og ess, and di e si y in e ms o cul u e and
languages.4
Pa icipa ing in he in e na ional open access and open da a mo emen equi es mo e han
jus good in en ions. We also need an online space o ‘house’ he p oceedings, and o his we
chose HAL, he open science a chi e c ea ed by F ance’s minis y o esea ch and highe
educa ion. Each ime a eade opens o downloads a ex ia HAL, hei ac ions a e c oss-
e e enced wi h sea ch engines and in e connec ed wi h o he se ices, such as ORCID. This
is a win-win si ua ion: he on-line p o iles o au ho s a e boos ed, and so is ou ield’s
ep esen a ion in ci a ion and abs ac da abases. HAL p o ides basic s a is ics on how eade s
use he a chi e, as Figu e 1 shows.
Figu e 1
Downloads o 25 Tex s in EPIP7 P oceedings5: Augus 1 and Sep embe 21, 2025
In Augus 2023 he EPI7 P oceedings wen li e online, ye he peak o in e es came in 2024
– pe haps as colleagues began o p epa e o EPIP8 which was held in May 2024. In he un-
up o EPIP96 in Ap il 2026, a he Uni e si y o Manches e , downloads ha e been inc easing
1 h ps://apling.engl.ias a e.edu/con e ences/p onuncia ion-in-second-language-lea ning-and- eaching-
con e ence/psll -a chi e/
2 h ps://hal.uni -g enoble-alpes. /EPIP/
3 h ps://abou .zenodo.o g/:“Zenodo is de i ed om Zenodo us, he i s lib a ian o he Ancien Lib a y
o Alexand ia and a he o he i s eco ded use o me ada a, a landma k in lib a y his o y.”
4h ps://eu opean-union.eu opa.eu/p inciples-coun ies-his o y/p inciples-and- alues/aims-and-
alues_en
5 h ps://hal.science/EPIP7/s a /dashboa d. The numbe 25 includes he edi o s’ In oduc ion as well as
he ile compiling he In oduc ion and 23 con ibu ions by au ho s.
6 h ps://epip9manches e .weebly.com/
Humánez-Be al, Hende son & Ki ko a-Nasko a
Looking back, mo ing o wa d
i
in au umn 2025 (e.g., up om app oxima ely 1200 o 1800) and i will be in e es ing o con inue
acking his ac i i y ia HAL.
The EPIP7 p oceedings include 23 inalised con ibu ions, whe eas EPIP8’s ha e only 10.
To compa e he esea ch hemes add essed a he wo con e ences, we hus u ned o he books
o abs ac s: EPIP7 ecei ed 37 abs ac s (wo d okens 13,904: 2,737 wo d ypes) and EPIP8
46 abs ac s (17,982 wo d okens: 3,714 wo d ypes). As such, he analysis can e eal opics
which we e discussed a each con e ence, ega dless o whe he o no he p esen e hen
submi ed a ex o publica ion in he P oceedings. Appendix 1 displays all he esul s o
compa ing he keywo ds7 as gene a ed by An Conc.8
The keywo ds we e gene a ed by he An Conc so wa e (An hony, 2022), d awing on he
con e ences’ abs ac bookle s as co po a. One co pus was c ea ed o each con e ence, in
which each abs ac was a sepa a e ex , wi h e e ences and he lis o au ho -p o ided
keywo ds emo ed. This helps o show how opics change o e ime, e.g., ges u es and poin ing
appea ed in he EPIP7 abs ac s bu do no appea o EPIP8. Howe e , he numbe o
occu ences o a e m mus be combined wi h i s ange is, i.e., in how many di e en abs ac s
i appea s. A la ge numbe o occu ences may no necessa ily be meaning ul, e.g., he wo d
<English> (EPIP7: 124 occu ences) and (EPIP8: 198 occu ences) occu s espec i ely ac oss
86% and 82% o abs ac s, me ely con i ming ha i is a key opic o he con e ence.
Fo each con e ence, he ollowing pa ag aphs will d aw on he selec ion o esul s p esen ed
in Table 1, in o de o highligh which e ms appea ed equen ly and in se e al abs ac s, as
well as wha is missing o a e, wha has inc eased o dec eased, appea ed o disappea ed. A
inal pa ag aph ocuses on which speci ic languages, language a ie ies o con ex s we e
men ioned, e e ing o Table 2.
Rega ding e ms used he leas equen ly, <CLIL>, <Fo eign Language Anxie y> and
<mul ilingual> do no appea in abs ac s un il EPIP8, whe e < o eign> (+ <accen > o +
<accen edness>) is used in wo abs ac s, while <ges u e> and <poin ing> a e los . In bo h
co po a, <shadowing>, <s a egy> and <syllable> hold s eady, bu a e men ioned in only one
abs ac .
Once a e m appea s in a leas 10 abs ac s, i begins o cons i u e oughly a qua e o he
o al numbe o a con e ence’s abs ac s, wi h a ew de elopmen s wo h highligh ing. Fi s , in
he upwa d sense, < o eign language> has climbed om appea ing in only 3 abs ac s o 14
(30%) and <lea ne s> bounds om 54% o 72%. The numbe o abs ac s wi h he e m <EFL>
leaps om 27% o 39%, <ins uc ion> ises om 32% o 43%, < eedback> ises om 13% o
22%, <acquisi ion> appea s in 37% o EPIP8 abs ac s (up om 19%). Mino inc eases include
<in elligibili y>, which mo es up o 28% om 22%, while <comp ehensibili y> eaches 26%
om 24%9. Re lec ing a key objec i e o he con e ence – o b ing oge he esea che s and
eache s – he ollowing co e opics a e holding s eady in EPIP abs ac s, espec i ely he 7 h
and 8 h con e ences: < eaching> (54% and 48%), <lea ning> (43% and 48%), and < eache s>
(32% and 37%). The sole no able in e sion conce ns <accen >, which has declined om 32%
o 24% o abs ac s, p obably e lec ing he no able sociolinguis ic p esence a he G enoble
EPIP.
In ela ion o he ongoing e minological deba e (Dewaele, 2018; Thomas & Osmen , 2020),
no occu ences o <LX> o <addi ional language> we e ound; 51% and 48% o abs ac s
con inue o use he e m <na i e>, espec i ely o EPIP7 and EPIP8, and he e m <second
language> appea s in oughly 24% abs ac s o bo h.
7 In con as , o he In oduc ion o he EPIP7 P oceedings (Hende son & Ki ko a-Nasko a, 2023), we
analysed he equency o occu ence o au ho -p o ided keywo ds.
8 An Conc so wa e de eloped by An hony (2022).
9 See Appendix 1 o some e ms occu ing in be ween 3 and 10 abs ac s.
This chap e is based on he o al p esen a ion gi en by he au ho a he 8 h
In e na ional Con e ence English P onuncia ion: Issues and P ac ices (EPIP
8) held May 8–10, 2024 a he Uni e si y o Can ab ia in San ande , Spain.
I is licensed unde he C ea i e Commons A ibu ion 4.0 In e na ional
License. To iew a copy o he license, please go o:
h p://c ea i ecommons.o g/licenses/by/4.0/
.
B yła-C uz, A. (2025). The ole o p onuncia ion in he assessmen o lis ening skills in he Polish Ma u a exam.
In A. Ki ko a-Nasko a, P. Humánez-Be al, & A. Hende son (Eds.), P oceedings o he 8 h In e na ional
Con e ence on English P onuncia ion: Issues and P ac ices (pp. 1–14). Uni e si é G enoble-Alpes.
h ps://doi.o g/10.5281/zenodo.16020252
The ole o p onuncia ion in he assessmen o lis ening skills in he Polish
Ma u a exam
Agnieszka B yła-C uz
Ma ia Cu ie-Skłodowska Uni e si y
Abs ac
In he second language (L2) class oom, p onuncia ion has been pe cei ed as a na u al
componen o speaking and has adi ionally been assessed based on he lea ne ’s speech, and
ye p onuncia ion also plays a ole in he assessmen o L2 lis ening skills. While phonological
compe ence is c ucial o he ac i a ion o bo om-up p ocesses in lis ening, which in u n a e
indispensable o he au oma isa ion o wo d ecogni ion and op-down p ocessing (Buck,
2001; Field, 2008), he ole o phone ics in ecep i e skills ends o be ma ginalised.
The p esen s udy in es iga es he ex en o which he ole o p onuncia ion is ecognised in
lis ening comp ehension es s in he Polish Ma u a, based on he exam pape s om 2023. The
s udy has shown ha while he spoken ex s used in he es s a he basic and he ad anced
le els a e di e si ied in e ms o leng h and gen e, hey ha dly di e wi h espec o phone ic
ai s. The main di e ence be ween he wo le els is he leng h o ex s and he a e o speech,
which is a bi as e o he ad anced le el, bu s ill slowe han in na u al speech.
Keywo ds: L2 p onuncia ion, p onuncia ion assessmen , lis ening comp ehension es s,
sc ip ed ex s, spoken ex s

B yła-C uz
P onuncia ion & assessmen o lis ening skills
2
1. In oduc ion
In he second language (L2) class oom, p onuncia ion has been ea ed as a na u al componen
o speaking skills and has ypically been assessed in lea ne s’ o al p oduc ion. Howe e ,
p onuncia ion also plays a ole in he assessmen o L2 lis ening abili y. The ole o phone ics
in lis ening comp ehension has been con i med in empi ical s udies, as in Wagne and To h
(2017), whe e ex s iden ical in lexis and g amma we e unde s ood di e en ly by lis ene s,
when he speake ’s p onuncia ion was modi ied. While phonological compe ence is c ucial o
he ac i a ion o bo om-up p ocesses in lis ening, which in u n a e indispensable o he
au oma isa ion o wo d ecogni ion and op-down p ocessing (Buck, 2001; Field, 2008), he
ole o phone ics in ecep i e skills ends o be ma ginalised.
This s udy in es iga es he ex en o which he ole o p onuncia ion is ecognised in
lis ening comp ehension es s in he Polish Ma u a, based on he English exam pape s om
2023. The phone ic ea u es in he audio ma e ials used in he basic and ad anced e sion o
he exam we e analysed in o de o es ablish whe he he phone ic di icul y o he spoken ex s
co esponds o he le el o lea ne s’ linguis ic compe ence.
2. Theo e ical amewo k
The aim o a es is o assess lea ne s’ language skills beyond he es con ex . Thus, L2 lis ening
es s should mi o genuine spoken ex s in e ms o phonology, lexis, g amma , o ganisa ion
and p agma ics (Buck, 2001; Wagne & To h, 2017). Ye , many o he spoken ex s used in L2
lis ening es s a e no ep esen a i e o na u al speech. Ra he han e lec ing con e sa ional,
spoken language in he eal-wo ld, audio ma e ials in lis ening comp ehension es s a e usually
sc ip ed, i.e., w i en down, edi ed, ehea sed and ead aloud wi h an excessi ely clea
enuncia ion (Wagne , 2013). This con adic s wha is pos ula ed by many esea che s, namely
ha “gi en he inhe en di e ences be ween o al and w i en discou se, i is impe a i e o
in es iga e comp ehension using lis ening passages ha app oxima e na u al speech”
(Schmid -Rineha , 1994, p. 180).
The di e ences be ween sc ip ed and unsc ip ed ex s a e mani old and ha e been
ex ensi ely discussed in he li e a u e (e.g., Cauldwell, 2014; Cha e, 1982; Gilmo e, 2004;
Hende son & Cauldwell, 2020; Ockey & Wagne , 2018). Fo ins ance, speci ic speech ea u es
impac he o e all di icul y which a gi en ex poses o L2 lis ene s, including: 1) hesi a ion
phenomena (HP) p esen in unplanned speech; 2) speech a e; 3) phonological modi ica ions
cha ac e is ic o unplanned spoken ex s; and 4) o ganisa ional and lexico-g amma ical
cha ac e is ics. The es o his sec ion discusses he i s h ee ca ego ies o ea u es, as he las
lies beyond he scope o he p esen pape .
Hesi a i e phenomena abound in spon aneous discou se and include illed and un illed
pauses, epe i ions, alse s a s, edundancies, and hesi a ions. These e lec he ac ha he
spoken ex is simul aneously composed and u e ed. S udies in o HP in L2 pe cep ion ha e
ende ed somewha inconsis en esul s. On he one hand, HP may cons i u e a majo pe cep ual
p oblem o non-na i e speake s (Voss, 1979), because hey may be con used wi h uns essed
o ms o pa s o wo ds and may be assigned seman ic meaning (G i i hs, 1991; Voss, 1979).
On he o he hand, HP can help lis ene s, e.g., p o iding hem ex a ime o p ocess he message
(Vande g i & Goh, 2012). I has been sugges ed ha such ea u es make he ex easie o
lis en o, by e oking na u al o al language o a g ea e ex en han planned speech (Shohamy
& Inba , 1991)
I is widely ecognised ha he a e o speech in luences L2 comp ehension. The gene al
conclusion om ample empi ical esea ch is ha spoken ex s deli e ed a a as e speech a e
a e mo e di icul o language lea ne s o unde s and han ex s spoken mo e slowly (Chiu &
B yła-C uz
P onuncia ion & assessmen o lis ening skills
3
Chen, 2023; Fuji a, 2017; G i i hs, 1990). Howe e , he eali y is mo e nuanced. Acco ding
o Wang and Na ayanan (2007), “speech a e is p ima ily dependen on wo ac o s: speaking
s yle and he ype o speech p oduc ion (e.g., sc ip ed o spon aneous)” (p. 2190). Typically,
sc ip ed ex s a e enuncia ed mo e slowly and mo e clea ly. Bloom ield e al. (2011) no e ha
“ as e speech is usually less clea han slowe speech al hough speech a e and audi o y cla i y
a e dis inc p ope ies” (p. 67). Mo eo e , while L2 lis ene s o en epo di icul y in
comp ehension due o as e deli e y (Goh, 2000), hey may in ac con la e p oblems wi h
a ec i e ac o s (anxie y). Fo example, he in insic na u e o lis ening in many L2 language
class ooms – namely, ime p essu e and he lack o con ol o e he acous ic signal (Bloom ield
e al., 2011) – is cause o anxie y in many lea ne s.
Fu he mo e, a ia ions in speech a e a e ypical o na u al, spon aneous speech; speake s
ha e been ound o speed up when deli e ing less ele an pa s o u e ances, such as
pa en hesis, dig essions o epai sequences (Fon, 1999; Uhmann, 1992). The a e o speech
also inc eases in chunks ypical o spon aneous discou se, p e- ab ica ed exp essions o
o mulaic sequences (McCa hy, 2010; Wood, 2006). Such clus e s o wo ds cause decoding
di icul y due o hei educed p ominence and hey may a y in leng h om he wo-wo d
chunks (e.g., “I mean”) o as many as six (e.g., “do you know wha I mean”) (Hende son &
Cauldwell, 2020).
Mos eal-wo ld speaking e en s a e cha ac e ised by connec ed speech p ocesses (CSP).
Alameen and Le is (2015), p opose ha CSPs can be di ided in o six main ca ego ies: linking,
dele ion, inse ion, modi ica ion (pala alisa ion/yod-coalescence, assimila ion, glo alisa ion,
lapping), educ ion, and o he p ocesses (lexical combina ions, con ac ed o ms). B own and
Kondo-B own (2006) poin ou ha al hough connec ed speech is mo e common in casual and
in o mal con ex s, i is a no mal ea u e o all egis e s and speaking s yles and should no be
conside ed a ma ke o sloppy o ca eless speech. Bowen (1975) asse s ha in in o mal speech
he use o con ac ions is almos uni e sal and ha educ ion is an e e -p esen ea u e o o al
English. Ye , a speake may wish o educe CSPs i hey a end consciously o hei speech in
pu sui o clea e enuncia ion and inc eased in elligibili y (I o, 2001). This is especially he
case when he speake eads a ex aloud (Cha e, 1982).
Agains his backd op i is he e o e alid o check whe he he wo le els o he Ma u a
exam acknowledge he ole o phone ics in lis ening comp ehension. This s udy ocuses on es
assessmen a he han o he , less o mal ypes o e alua ion, because “ es s will o en ha e an
e ec on class oom eaching” (Buck, 2001, p. 196). In o he wo ds, analysing he es p o ides
insigh in o he eaching p ocess wi h espec o i s aims and esou ces. In he case o he high-
s akes Ma u a exam, he analysis o he lis ening componen is expec ed o p o ide some
indica ions as o whe he he washback e ec is bene icial o no .
3. Resea ch me hodology
The ins umen used in he p esen s udy consis s o 18 eco dings and hei ansc ip s om
he basic and he ad anced le el o he Ma u a exam om 2023 (see Table 1 o a de ailed
desc ip ion o asks). Fi s , he ex ac s we e analysed wi h e e ence o he main di e ences
be ween sc ip ed and unsc ip ed ex s, ocusing on hose ela ed o p onuncia ion (HP,
phonological modi ica ions, and speech a e). The numbe o wo ds and syllables in each o
he eco dings was coun ed wi h he use o Tex Analyse .1 Speech a e was calcula ed as
syllables pe second (s/s), ollowing common p ac ice (Sh ibe g e al., 2000). Pauses and non-
linguis ic u e ances a e ea ed as indi idual uni s, because “ hey migh be indica o s o
concep ual planning and hei exis ence migh also con ibu e o a e pe cep ion” (Fon, 1999,
1 Online U ili y – F ee Online So wa e U ili ies h ps://www.online-u ili y.o g/
B yła-C uz
P onuncia ion & assessmen o lis ening skills
4
p. 663). Phonological modi ica ions (assimila ions, elisions, linking de ices) and HP
(hesi a i e pauses, alse s a s, epe i ions, edundancies) we e aced audi o ily and coun ed
manually using bo h he ansc ip and he eco ding.
The s udy explo es he ollowing esea ch ques ions:
RQ1: Is he phone ic di icul y o he spoken ex s modi ied o co espond o speci ic
le els o linguis ic p o iciency?
RQ2: Is p onuncia ion aligned wi h he ype o discou se?
Wi h e e ence o RQ1, i is expec ed ha es s used o assessing lis ening comp ehension
skills should a y wi h espec o p onuncia ion depending on he es ake ’s le el. Rega ding
RQ2, he alignmen o p onuncia ion wi h he ype o discou se means ha ex s used in L2
lis ening comp ehension es s should mi o spoken ex s om he a ge con ex o language
use. In eal li e, spoken ex s di e in o ali y, meaning ha some o hem con ain mo e ea u es
o spoken language as opposed o w i en language (Tannen, 1982), e.g., adio o TV news a e
sc ip ed and hus less o al han spon aneous in e ac ions o unplanned monologues. The
passages used o L2 lis ening comp ehension should be low o high in o ali y depending on
hei eal-li e coun e pa s.
4. Da a analysis
Table 1 p esen s he s uc u e o he lis ening comp ehension es s a bo h he basic and he
ad anced le el. I shows ha bo h e sions o he exam consis o h ee asks, which include
Table 1
The S uc u e o he Lis ening Comp ehension Tes s
Tes
ype
Lis ening
sec ion
Tex
ype Ques ion ype
No o imes
he ex is
lis ened o
No o
ques ions
Basic
Task 1 1 dialogue ue o alse 2 5
Task 2 5 monologues ma ching a
s a emen o
he speake
2 5
Task 3 2 monologues
1 dialogue
mul iple-choice 2 6
Ad anced
Task 1 2 monologues
1 mini in e iew
(a ques ion and an
ex ended answe )
mul iple-choice 2 6
Task 2 5 monologues ma ching a
s a emen o
he speake
2 5
Task 3 1 dialogue cloze es 2 4
B yła-C uz
P onuncia ion & assessmen o lis ening skills
5
dialogues and monologues. Comp ehension is checked by means o mul iple-choice ques ions
and ma ching a s a emen o he speake . Addi ionally, ue o alse ques ions appea in he basic
le el and a cloze es in he ad anced one. All ex s a e lis ened o wice.
5. Resul s
The esul s o he analysis o he spoken ex s used in he lis ening comp ehension es s a e
p esen ed in his sec ion. The in o ma ion pe aining o HP and he a e o speech is included
in Tables 2 and 3, whe eas CSPs a e discussed sepa a ely in he ex . In Tables 2 and 3 he las
wo columns jux apose he deg ee o o ali y in he audio ma e ial wi h he deg ee o o ali y ha
i s coun e pa in eal li e is supposed o ha e.
5.1 Basic le el
The nine ex s used in he basic-le el exam a y wi h espec o leng h, s yle, and gen e (see
Table 2). Th ee ex s could be classi ied as a he o mal and he es as semi- o mal o
in o mal. I espec i e o he ype and leng h, howe e , all ex s a e sc ip ed, which is jus i ied
only in ou cases: wo pa s o a adio b oadcas , a TV announcemen , and pa ly in he ou is
guide announcemen whe e some p e-planning/ ehea sing would be expec ed. The emaining
ex s a e ei he pe sonal na a i es o con e sa ions, which would be non-sc ip ed ye speci ic
o and/o ypical o he a ge ed si ua ion o language use. Only one eco ding (Tex 3 in Task
3) con ains some ea u es o unplanned ex s, such as one hesi a i e pause (Hmmm… his ace
is de ini ely iangula ) and wo silen pauses (I only need one pai …bu hold on,
Well…ob iously you wan he bes pai o you ace).
All speake s enuncia e clea ly, wi h a subsequen es ic ion on he numbe o educ ions
which would no mally occu in hei speech. The mos common ea u e o connec ed speech is
linking / / (n = 11), ound in examples (1)–(5), and in usi e / / (n = 2) in examples (6) and (7).
(1) we e a eenage /wə ə ' i:neɪʤə/
(2) o only / ə 'əʊnlɪ/
(3) o as li le as / ə əz 'lɪ l əz/
(4) bea a he same shop /'beə ə ðə 'seɪm 'ʃɒp/
(5) ea o people /' ɪə ə 'pi:pəl/
(6) I saw a ew people / 'sɔ: ə ' ju: 'pi:pəl/
(7) d awing /'d ɔ: ɪŋ/
Elisions and assimila ion a e sca ce. Fo ins ance, he e is yod-coalescence in example (8),
bu i is no consis en , i.e., i does no appea in example (9) when he same speake says a
di e en ph ase ha mee s he phonological condi ions o assimila ion. In Task 2, Tex 2, he e
a e some dele ions and con ac ions, illus a ed in (10) and (11) bu no in all ins ances, as in
sen ence (12). The e is no h-d opping, e.g., (13) and (14), o / /-dele ion as shown in (15)–(17).
Plosi e dele ions we e obse ed in (18)–(20), bu no in (21). Simila ly, ins ances o
assimila ion can be obse ed in (22) and (23), bu no in (24) and (25).
(8) would you /wʊʤʊ/
(9) wha you need /wɒ jə 'ni:d/
B yła-C uz
P onuncia ion & assessmen o lis ening skills
6
Table 2
Tex Types ( e Leng h, S yle, Gen e) Used a he Basic Le el
Lis ening
ask
(pa s)
Tex ype Leng h HP Speech
a e O ali y O ali y in
eal li e
Task 1 dialogue
(an in e iew)
2 min.
330 wo ds
none 4.2 s/s low
(sc ip ed)
high
(unsc ip ed)
Task 2
Tex
1
monologue
(a pe sonal ilm
ecommenda ion)
32 s
92 wo ds
none 3.1 s/s low
(sc ip ed)
high
(unsc ip ed)
Tex
2
monologue
(a ou is guide
explana ion)
31 s
93 wo ds
none 3.9 s/s low
(sc ip ed)
a he low
(p e-
planned)
Tex
3
monologue
(a TV b oadcas )
39 s
107 wo ds
none 3.7 s/s low
(sc ip ed)
low (p e-
planned)
Tex
4
monologue
(a adio
p og amme)
38 s
102 wo ds
none 3.8 s/s low
(sc ip ed)
a he low
(semi-
sc ip ed)
Tex
5
monologue
(a pe sonal
na a i e abou
a pas e en )
42 s
119 wo ds
none 3.7 s/s low
(sc ip ed)
high
(unsc ip ed)
Task 3 Tex
1
monologue
(a pe sonal
na a i e abou
a pas e en )
46 s
139 wo ds
none 4 s/s low
(sc ip ed)
high
(unsc ip ed)
Tex
2
monologue
(a pa o a TV
o adio
p og amme)
50 s
125 wo ds
none 3.5 s/s low
(sc ip ed)
low
(sc ip ed)
Tex
3
dialogue
(a con e sa ion
be ween a
cus ome and a
shop assis an )
1 min 19 s
214 wo ds
1 hesi
a i e
pause,
2 silen
pauses
3.4 s/s low
(sc ip ed)
high
(unsc ip ed)
(10) why don’ you /waɪ dəʊn jə/
(11) you won’ eg e i /jʊ wəʊn ɪ'g e ɪ /
(12) I is pe ec o up o ou gues s /ɪ ɪz 'pɜ: ɪk ə ʌp ə ɔ: 'ges s/
(13) wi h him /wɪð hɪm/
(14) om him / ɒm hɪm/
(15) nex week /neks 'wi:k/

B yła-C uz
P onuncia ion & assessmen o lis ening skills
7
(16) jus i e days /ʤʌs ' aɪ 'deɪz/
(17) jus like Lego /ʤʌs 'laɪk 'legəʊ/
(18) jus lo ed /ʤʌs 'lʌ d/
(19) second yea /'sekən 'jɪə/
(20) i s job /' ɜ:s 'ʤɒb/
(21) mos o my ime /məʊs ə maɪ ' aɪm/
(22) did you wo k /dɪʤʊ 'wɜ:k/
(23) did you e e ind /dɪʤʊ e ə ' aɪnd/
(24) abou you wo k /ə'baʊ jɔ: 'wɜ:k/
(25) wha a e you doing /'wɒ jə 'du:ɪŋ/
Resul s show ha he e a e a ew con ac ions, such as I’d like o ask you, i ’s g ea , I’m
wo king, i ’s in ac , i ’s beau i ul, bu hey a e no always used consis en ly, o ins ance,
because i is, which had jus been hi , and i had been buil .
The speech a e anges om 3.1–4.2 s/s, wi h he slowes empo in he na a i e and he
as es in he dialogue. While a as e a e o in e ac ion is in line wi h wha akes place in eal
li e, i is s ill conside ably slowe han in casual con e sa ions: 5–6 s/s (Deese, 1984) o 6.19
s/s (Pelleg ino e al., 2011), and e en slowe han in adio b oadcas s whe e he a e o speech
is on a e age a ound 5.43 s/s (Kendall, 2009). No ably, due o he p e- ab ica ed cha ac e o
he passages, he e a e no dig essions, a e hough s, epai s, o e o mula ions, which would
in oduce a a ia ion in he speech a e.
5.2 Ad anced le el
The ex s used in he ad anced e sion o he exam a e di e si ied wi h espec o leng h and
gen e. The egis e o he ex s is semi- o mal o in o mal and all nine ex s a e sc ip ed. This
seems comple ely unjus i ied, because in eal li e none o hese ex s would be w i en.
Wi h espec o p onuncia ion, i is ca e ul and clea wi h a ai ly limi ed numbe o CSPs.
Needless o say, some educ ions a e used as in English, i espec i e o whe he a ex is
sc ip ed o non-sc ip ed. These include con ac ions and weak o ms o unc ion wo ds, bu he
la e ha e no consonan d opped, as shown in examples (26)–(29). Ins ances o assimila ion
a e ew bu p esen , as in examples (30) and (31). The mos equen phenomenon is linking
/ /, e iden in he examples (32)–(36), and he e is one case o in usi e / / in example (37).
(26) mus be whi e /mʌs 'bi: 'waɪ /
(27) li e has e ol ed /'laɪ həz ɪ' ɒl d/
(28) whe e I can pu chase /'weə aɪ kən 'pɜ:ʧəs/
(29) I say ha my iend /aɪ 'seɪ ðə maɪ ' end/
(30) wha could go w ong /'wɒ kəg 'gəʊ ' ɒŋ/
(31) does you /dʌʒjɔ:/
(32) p oduce o /p ə'dʒu:sə ə /
(33) he e any /ðeə 'enɪ/
(34) a e a pe ec /ɑ: ə 'pɜ: ɪk /
(35) o a u key / ə ə ' ɜ:kɪ/
(36) o inside /ɔ: ɪn'saɪd/
(37) pizza in he idge /'pi: sə ɪn ðə ' ɪdʒ/
B yła-C uz
P onuncia ion & assessmen o lis ening skills
8
Table 3
Tex Types ( e Leng h, S yle, Gen e) Used a he Ad anced Le el
Lis ening
ask
(pa s)
Tex
ype Leng h HP Speech
a e O ali y O ali y in
eal li e
Task 1 Tex
1
monologue
(an anecdo e)
57 s
173 wo ds
none 4.2 s/s low
(sc ip ed)
high
(unsc ip ed)
Tex
2
dialogue
(a adio in e iew
- a ques ion and
an ex ended
answe )
44 s
129 wo ds
none 4.4 s/s low
(sc ip ed)
high
(unsc ip ed)
Tex
3
monologue
(a pe sonal
na a i e o a pas
e en )
2 min. 2 s
385 wo ds
none 4.3 s/s low
(sc ip ed)
high
(unsc ip ed)
Task 2 Tex
1
monologue
(pe sonal
na a i e ela ing
o he speake ’s
wo k expe ience)
43 s
141 wo ds
none 4.6 s/s low
(sc ip ed)
high
(unsc ip ed)
Tex
2
monologue
(pe sonal
na a i e ela ing
o he speake ’s
wo k expe ience)
37 s
108 wo ds
none 4.5 s/s low
(sc ip ed)
high
(unsc ip ed)
Tex
3
monologue
(pe sonal
na a i e ela ing
o he speake ’s
wo k expe ience)
31 s
106 wo ds
none 5.4 s/s low
(sc ip ed)
high
(unsc ip ed)
Tex
4
monologue
(pe sonal
na a i e ela ing
o he speake ’s
wo k expe ience)
38 s
118 wo ds
none 4.3 s/s low
(sc ip ed)
high
(unsc ip ed)
Tex
5
monologue
(pe sonal
na a i e ela ing
o he speake ’s
wo k expe ience)
34 s
95 wo ds
none 4.2 s/s low
(sc ip ed)
high
(unsc ip ed)
Task 3 dialogue
(a TV in e iew wi h a
ood p oduce )
2 min. 10 s
365 wo ds
none 4.3 s/s low
(sc ip ed)
a he high
(semi-
sc ip ed)
B yła-C uz
P onuncia ion & assessmen o lis ening skills
9
Con ac ions a e used, bu inconsis en ly wi hin he same speake s. Fo ins ance, hey a e
p esen in he examples we’ e all, you wouldn’ , I’m alking, i ’s usually shown, can’ wai ,
hey’ e always, ha ’s ano he , a en’ , wouldn’ las , I won’ wa ch, I’ e been wo king, I wasn’
su e, and I’ll, bu absen in hey a e a pe ec subs i u e, you would ge , po a oes ha e been
eshly baked, and as i is abou .
The a e o speech anges be ween 4.2–4.6 s/s. The slowes a e o speech appea s in a
na a i e (anecdo e; Task 2, Tex 1) and he as es also in a na a i e (Task 1). The dialogues
a e no cha ac e ised by an inc eased a e o speech compa ed o he monologues, which is no
ypical in e e yday speech. The o e all a e o speech in all ex s is a bi slowe han in
co esponding spon aneous ex s in eal li e whe e i would be a ound 5.04 s/s (G osz &
Hi schbe g, 1992). Because he e a e no phenomena e lec ing concep ual planning, he a e o
speech is de oid o any a ia ion.
6. Discussion and key indings
A summa y o he indings is p esen ed in Table 4. In bo h he basic and he ad anced e sion
o he exam, he same pai o speake s is employed. The ex s a e di e si ied wi h espec o
gen e, ye none o he exam e sions exploi s a casual con e sa ion. A mo e a ied selec ion o
monologues is ound in he basic e sion – TV and adio announcemen s, pe sonal na a i es
and an ex ac deli e ed by a ou is guide. In he ad anced e sion he monologues a e o wo
ypes – a pe sonal na a i e and an anecdo e. The basic e sion includes wo ex s which would
also be sc ip ed in eal li e, whe eas hey a e no p esen in he ad anced e sion. This may
ha e been a delibe a e choice on he pa o he es designe s, bu i has no p ac ical
consequence, as e en in he ad anced e sion all ex s a e sc ip ed and ead ou . Pa adoxically,
he e o e, he ex s in he basic e sion, no in he ad anced one, be e ep esen hei eal-li e
coun e pa s.
Rega ding RQ1, he phone ic di icul y o he spoken ex s is no modi ied o co espond o
le els o linguis ic p o iciency. This is pa ly due o he ac ha bo h he basic and ad anced
e sion o he exam include sc ip ed ex s, which la gely de e mines hei p onuncia ion
ea u es. The main di e ence be ween he wo e sions o he exam is du a ion and he a e o
speech. While he mean leng h o sho e passages is oughly he same o he wo le els (39
sec. s. 40 sec.), he longe ex ac s ange om 1:19–2:00 minu es in he basic le el and 2:02–
2:10 in he ad anced le el. Since a longe ex usually means mo e in o ma ion o p ocess, he
es designe s may ha e delibe a ely in oduced his di e ence o inc ease he di icul y o he
ad anced e sion.
The numbe o phonological modi ica ions is simila in bo h e sions. The e a e
con ac ions, weak o ms and ce ain p ocesses which appea in English i espec i e o he ype
o ex (w i en o spoken), bu hey a e p onounced mo e clea ly han in na u al speech. In he
basic e sion he e a e h ee ins ances o assimila ion, speci ically yod-coalescence, and i e
ins ances o elision; 11 cases o linking / / and wo cases o in usi e / /. The ad anced e sion
includes wo ins ances o assimila ion and wo cases o linking / /.
The a e o speech in he wo e sions di e s as well. O e all, he a e o speech in he exam
ex s is slowe han na u ally occu ing monologues and dialogues. Bo h he ange o speech
and he mean alues a e lowe o he basic e sion han he ad anced e sion, espec i ely:
3.1–4 s/s (m = 3.6 s/s) and 4.2.–4.6 s/s (m = 4.3 s/s). As a as e speech a e usually equi es
mo e au oma ic p ocessing, inc easing i o he ad anced le el inc eases ask di icul y.
Fu he mo e, na u al unplanned discou se is cha ac e ised by speech a e a ia ion, ye
a ia ion is absen om he eco dings; hey a e deli e ed a a s eady a e o speech, wi hou
he no mal dig essions, a e hough s, alse s a s, and edundancies which would occu a an
accele a ed empo in eal li e.
B yła-C uz
P onuncia ion & assessmen o lis ening skills
10
Table 4
Compa ison o he Main Fea u es o Audio Ma e ials Used in he Basic and Ad anced Ve sion
o he Ma u a Exam
Fea u e Basic Ad anced
Numbe o speake s 2 (male & emale) 2 (male & emale)
Accen SSBE SSBE
Type o ex s sc ip ed sc ip ed
Type o gen e a ied
(monologues & dialogues)
a ied
(monologues & dialogues)
Ra e o
speech
Range 3.1–4 s/s 4.2–4.6 s/s
Mean 3.6 s/s 4.3 s/s
Leng h in
sho ex ac s
Range 31–50 s 31–57 s
Mean 39 s 40 s
Leng h in
long ex ac s
Range 1min. 19 s–2:00 2 min. 2 s–2:10
Mean 1min. 39 s 2 min. 6 s
Hesi a i e pauses 1 illed pause (hmmm)
2 silen pauses
none
Repe i ions
False s a s
Re o mula ions
Un inished sen ences
none none
Tu n-opene s 2 ins ances o well 1 ins ance o well
Connec ed speech phenomena
o he han weak o ms
3 ins ances o assimila ion
(yod-coalescence)
5 ins ances o elision
11 ins ances o linking / /
2 ins ances o in usi e / /
2 ins ances o assimila ion 24
ins ances o linking / /
Apa om du a ion and he a e o speech, he wo e sions do no di e conside ably wi h
espec o p onuncia ion. In o he wo ds, he phone ic compe ence equi ed o pe o m he
lis ening comp ehension asks appea s no o ha e been aken in o conside a ion. The es
designe s a e no p ima ily ocused on es ing phone ic compe ence e en hough lis ening
compe ence consis s o , abo e all, dealing wi h he speech signal. The ad anced e sion o he
Ma u a exam is cha ac e ised by mo e ad anced lexis and mo e complex g amma ical
s uc u es, bu in e ms o p onuncia ion, i emains la gely a he same le el as he basic one.
This is cong uen wi h Buck’s (2001) obse a ion ha “un o una ely, many s uden s p og ess
Duckinoska-Mihajlo ska
Pe cep ions o na u alness, di icul y & speech a e
17
impo ance o exposing lea ne s o di e se English a ie ies. Fo example, Miao e al. (2024)
in es iga ed he e ec s o in eg a ing Global English (GE) a ie ies in o class oom ins uc ion
and he impac on EFL lea ne s’ lis ening comp ehension and p onuncia ion. Thei s udy
compa ed an in e en ion g oup exposed o di e en accen ed English a ie ies wi h a con ol
g oup ha hea d only Ame ican English. Da a was collec ed h ough TOEFL-like lis ening
exam ques ions and a speaking ask, using a p e/pos - es design. The esul s showed ha
lis ening comp ehension o GE a ie ies o he in e en ion g oup imp o ed, bu only sho
e m, and hei p onuncia ion emained una ec ed. Despi e he lack o sus ained long- e m
e ec s, he indings sugges ha exposu e o di e en English a ie ies does no nega i ely
a ec lea ne s’ comp ehension.
Fu he esea ch is needed on EFL lea ne s’ pe spec i es on na i e connec ed speech o
be e in o m bo h esea ch and eaching p ac ice. To ha end, he cu en s udy examines
S anda d Sco ish English (SSE). This a ie y was chosen because i emains unde ep esen ed
in cou se ma e ials, whe e S anda d Sou he n B i ish English (SSBE) and Gene al Ame ican
(GA) con inue o domina e as eaching models. P e ious esea ch has explo ed a i udes owa d
one a ie y o SSE, Glasgow S anda d English, in e ms o speake compe ence and social
a ac i eness (McKenzie, 2008). The cu en s udy, howe e , ocuses on he pe cei ed
na u alness o SSE, wi h pa icula a en ion o connec ed speech p ocesses.
2.1 Resea ch ques ions
The li e a u e e iewed abo e highligh s he impo ance o exposing L2 lea ne s o a ious
language a ie ies, bo h na i e and non-na i e. Achie ing in elligibili y always in ol es a wo-
way p ocess. Ne e heless, in NNS–NS in e ac ions, CSPs place an addi ional cogni i e
demand on NNS speake s, as au hen ic speech can de ia e om hei expec a ions o spoken
English (Alameen & Le is, 2015). Wi h his in mind, he cu en s udy in es iga es he
exposu e o a less commonly ep esen ed na i e a ie y, Sco ish English. I examines
Macedonian EFL lea ne s’ pe cep ions o he p esence o absence o CSPs in na i e speech,
by asking hem which o wo eco dings sounds mo e na u al, and hen o explain hei choice.
Na u alness is no de ined o pa icipan s bu a he i is hoped ha hei explana ions will
men ion CSPs. Thus, he p esen s udy a emp s o answe he ollowing esea ch ques ions:
RQ1: How do Macedonian EFL lea ne s e alua e he na u alness o na i e speech in
ela ion o: a) speech a e, and b) connec ed speech p ocesses?
RQ2: How do Macedonian EFL lea ne s ecognise and iden i y speci ic connec ed
speech p ocesses ( owel educ ion, elision, in usi e sounds) in na i e speech?
While Macedonian EFL lea ne s may ini ially ind speech wi h a less amilia accen mo e
challenging, we assume ha hey will gene ally pe cei e bo h eco dings as being easy o
unde s and. Howe e , we hypo hesise ha hey will expe ience g ea e di icul y iden i ying
speci ic connec ed speech p ocesses, pa icula ly in as e -paced speech.
3. Resea ch me hodology
3.1 Pa icipan s
The pa icipan s in his s udy we e 50 Macedonian EFL lea ne s (Mage = 20.8, age ange = 20–
25). The e we e 9 male and 40 emale pa icipan s, wi h one pa icipan iden i ying as non-
bina y. All pa icipan s we e second-yea English majo s a he Uni e si y o Ss. Cy il and
Me hodius in Skopje. They we e en olled in he English Phone ics and Phonology cou se and

Duckinoska-Mihajlo ska
Pe cep ions o na u alness, di icul y & speech a e
18
had comple ed one semes e o ins uc ion bu had no ye s udied connec ed speech p ocesses.
All had lea ned English h ough o mal educa ion, and 50% epo ed ha ing basic knowledge
o an addi ional o eign language. The class is obliga o y bu pa icipa ion in he s udy was
olun a y.
A na i e emale speake o Sco ish S anda d English se ed as a speech model. She holds
a Mas e o Resea ch in Linguis ics and wo ks as a uni e si y u o , eaching English
p onuncia ion classes o in e na ional s uden s.
3.2 P ocedu e
The s udy consis ed o wo s ages. Fi s , he na i e speake eco ded a 295-wo d passage
(adap ed om The Gua dian) wice. Fo he i s eco ding (R1), he speake was ins uc ed o
ead he passage a a slowe , mo e delibe a e pace, aiming o p oduce mo e wo ds in ull o ms.
Fo he second eco ding (R2), he speake ead he passage a a as e pace, esul ing in speech
wi h mo e educed o ms. Bo h eco dings, along wi h selec ed exce p s, we e embedded in a
ques ionnai e. In he second s age, he pa icipan s comple ed he ques ionnai e in a con olled
en i onmen using desk op compu e s and headphones and we e supe ised by a esea che .1
The ask ook app oxima ely 40 minu es.
3.3 Da a collec ion and analysis
The main da a collec ion ins umen was a ques ionnai e comp ising wo pa s: Pa 1 ga he ed
demog aphic in o ma ion, whe eas Pa 2 ocused on esponses o he eco dings (see
Appendix). I comp ised 14 ques ions: 8 open-ended, 2 Like scale, and 4 mul iple-choice
ques ions, along wi h wo eco dings and i e exce p s. The pa icipan s lis ened o Reco dings
1 and 2 wice be o e answe ing he ques ions, while he exce p s could be played up o i e
imes. These exce p s a ge ed h ee connec ed speech ea u es: owel educ ion, elision, and
in usi e sounds. Table 1 p esen s he selec ed ea u es and ansc ip ions o hei occu ence
in a sen ence.
These i e exce p s we e chosen due o hei ep esen a i eness and he p ominence o he
es ed ea u e, hus acili a ing ecogni ion and dis inc ion o hei p esence o absence in
speech. In Exce p s 3, 4, and 5, he a ge ed ea u e was p esen in bo h eco dings, u he
emphasising he ac ha connec ed speech occu s na u ally, i espec i e o speech a e.
Responses om he comple ed ques ionnai es we e manually en e ed in o Excel. The open-
ended ques ions we e analysed o he p esence o ecu ing hemes, while quan i a i e da a
was p ocessed in SPSS 25 (IBM Co p., 2017). A Wilcoxon signed- ank es was used,
app op ia e o anked da a such as Like -scale i ems.
4. Resul s
The esul s, based on he analysis o bo h he quali a i e and quan i a i e da a om he
ques ionnai e, a e p esen ed in wo subsec ions. Quo es om indi idual pa icipan s a e
e e enced wi h he pa icipan s’ codes P01–P50, whe e P s ands o pa icipan ollowed by
he pa icipan numbe .
1 The esea che is he au ho o his pape .
Duckinoska-Mihajlo ska
Pe cep ions o na u alness, di icul y & speech a e
19
Table 1
Ta ge ed Connec ed Speech Fea u es in Exce p s
Exce p
Ta ge ea u e Ph ase T ansc ip ion
1
Vowel educ ion in
g amma ical wo ds
This was a p oblem.
R1 /ðɪs wɒz ə ˈp ɒbləm/
R2 /ðɪs wəz ə ˈp ɒbləm/
2
Vowel educ ion in
g amma ical wo ds
…including poli icians,
p esen ed as spo s
ans.
R1 /ɪnˈkluːdɪŋ ˌpɒlɪˈ ɪʃənz
p ɪˈzen ɪd æz ˈspɔː s ˈ ænz/
R2 /ɪnˈkluːdɪŋ ˌpɒlɪˈ ɪʃənz
p ɪˈzen ɪd əz ˈspɔː s ˈ ænz/
3
Elision
Bu i can’ ha e
emained mo e elusi e
in eal li e.
R1 /kæn æ ɪˈmeɪnd/
R2 /kæn ə ɪˈmeɪnd/
4
In usi e /w/
… o i s apid laye ed
hy hms…
R1 & R2
/ u (w) ɪ s ˈ æpɪd ˈleɪəd ˈ ɪðəmz/
5
In usi e /j/
… why all public
igu es, including
poli icians…
R1 & R2
/waɪ(j) ɔːl ˈpʌblɪk ˈ ɪɡəz ɪnˈkluːd
ɪŋ ˌpɒlɪˈ ɪʃənz/
4.1 Reco dings 1 and 2
The i s se o ques ions in Pa 2 o he ques ionnai e elici ed pa icipan s’ iews ega ding
hei pe cei ed ease o unde s anding o he speech in R1 and R2, he challenges aced while
lis ening, and he di e ences be ween R1 and R2.
Fo sel -pe cei ed ease o unde s anding, he pa icipan s a ed each eco ding on a scale o
1–5 (1 = e y di icul , 5 = e y easy). The a e age sco e o R1 was M = 4.34, compa ed o M
= 4.50 o R2, indica ing ha he pa icipan s did no ind ei he eco ding pa icula ly di icul
o unde s and. To es ablish whe he he di e ence in unde s anding was s a is ically signi ican ,
a Wilcoxon es was used. The esul s showed he di e ence be ween R1 and R2 was no
s a is ically signi ican (p = .106). Rega ding he challenges aced, he pa icipan s men ioned
di icul ies ela ed o accen o R1 (n = 24, 48%), as illus a ed in he ollowing esponses:
“some wo ds we e di icul o unde s and because she was p onouncing hem in an odd way”
(P06), “i ook me some ime o adjus , bu i wasn' di icul ” (P12), o ”incomp ehensible due
o un amilia p onuncia ion” (P43). The majo i y ound R2 less challenging han R1, al hough
Duckinoska-Mihajlo ska
Pe cep ions o na u alness, di icul y & speech a e
20
a ew (n = 10, 20%) epo ed di icul ies wi h unde s anding R2 due o he hea y accen , as
pace o speech, p onuncia ion o some wo ds, o he need o mo e p ocessing ime.
When asked which eco ding sounded mo e na u al, some pa icipan s (n = 8, 16%) selec ed
R1, he majo i y (n = 33, 66%) selec ed R2, while he emaining (n = 9, 18%) said ha bo h
sounded na u al o hem. The main easons o R1 no sounding na u al included – again, in
pa icipan s’ own wo ds – i s slow o obo ic speech, and he pe cep ion ha he speake was
non-na i e. In con as , R2 was pe cei ed as mo e na u al due o i s na u al low o speech,
lack o pauses, he speake sounding na i e despi e he accen , and he o e all one being mo e
con e sa ional.
The pa icipan s we e also asked o iden i y any ob ious di e ences be ween R1 and R2,
choosing om a lis o op ions; one se ed as a dis ac o (op ion d). Table 2 lis s he main
di e ences and he numbe o esponses o each op ion.
Table 2
Di e ences be ween Reco ding 1 and Reco ding 2 and he Numbe o Selec ed Answe s
Possible di e ences be ween
Reco ding 1 and Reco ding 2 No o esponses
a) The speake speaks as e . 38
b) Some sounds ha e been le ou . 17
c) Some ex a sounds ha e been added. 15
d) Some wo ds ha e been le ou . 5
e) Some wo ds a e p onounced sho e . 36
) Some imes he e a e no pauses be ween wo wo ds. 36
The esul s show ha mos pa icipan s no iced ha he speake in R2 spoke as e , esul ing
in some wo ds being p onounced sho e o wi hou pauses. Fewe pa icipan s no iced he
omission o addi ion o sounds, while only i e pa icipan s selec ed he dis ac o op ion ha
some wo ds had been le ou .
4.2 Exce p s
The second se o ques ions elici ed pa icipan s’ iews ega ding speci ic connec ed speech
ea u es p esen o absen in i e di e en exce p s. Exce p s 1 and 2 a ge ed owel educ ions
in g amma ical wo ds. In bo h exce p s, R1 con ained a s ong o m o he g amma ical wo ds
(i.e., was /wɒz/ and as /æz/), while R2 con ained a weak o m (i.e., was /wəz/ and as /əz/).
When asked whe he he g amma ical wo d in Exce p 1 was he same o di e en in bo h
eco dings, ew pa icipan s (n = 8, 16%) said i was he same, while he majo i y (n = 42, 84%)
no ed ha i was di e en . The pa icipan s who iden i ied a di e ence ci ed ac o s such as
Duckinoska-Mihajlo ska
Pe cep ions o na u alness, di icul y & speech a e
21
emphasis on he wo d, he p esence o a s ong e sus a weak owel, and he p esence o
absence o pauses be o e and a e he wo d, as e iden in he ollowing ema ks: “In R2 he
one is mo e con e sa ional and he ocus is no on <was>” (P04); “In eco ding one <was> is
p onounced ha de and i sound[s] mo e impo an , while in eco ding wo <was> is
p onounced mo e so ly” (P14). 2 Howe e , some esponses e e ed o dialec al di e ences o
p o ided easoning no di ec ly ela ed o he a ge ea u e, such as: “The i s one has an
accen o i making i sound like <wos> and he second one is clea e and sounds like he
[A]me ican <was>” (P06), o “In he i s eco ding, he <s> is p onounced as /z/, while in he
second i ’s /s/” (P42).
The esponses o Exce p 2, a ge ing he wo d as, we e mo e a ied. Hal o he
pa icipan s (n = 25, 50%) epo ed ha he p onuncia ion was he same in bo h eco dings,
many (n = 22, 44%) indica ed ha i was di e en , and ew (n = 3, 6%) p o ided no answe .
Those who pe cei ed a di e ence a ibu ed i o he p esence o a schwa sound and hea ing
only he /s/ sound in R2, o no icing clea e pauses be o e and a e he wo d in R1 compa ed
o R2.
Exce p 3 in ol ed ins ances o elision, speci ically he omission o /h/, in bo h eco dings.
The only di e ence be ween R1 and R2 was he absence and p esence o a weak o m in he
g amma ical wo d ha e, espec i ely. When asked i he wo ds in he e b ph ase we e ully
p onounced, esponses a ied signi ican ly. Twen y pa icipan s (40%) indica ed ha he wo ds
we e ei he ully p onounced in bo h eco dings, o no ully p onounced bu hei answe s did
no men ion omission o /h/. Examples included commen s such as he wo ds sounding like
one wo d in R1 bu like wo sepa a e wo ds in R2, o ha wo ds we e ully p onounced only
in R2. The emaining pa icipan s (n = 30, 60%) co ec ly no iced he elision o /h/ co ec ly,
in a leas one o he eco dings, o iden i ied he con ac ed o m o ha e in he e b ph ase,
as illus a ed in he ollowing commen s: “In Reco ding 2, he wo ds <can’ > and <ha e> a e
connec ed wi h he neu al owel in hei p onuncia ion, esul ing in he sho ening o he wo d
<ha e>, which is no he case in Reco ding 1” (P03) and “All o he wo ds a e p onounced in
Reco ding 1[;] howe e , in Reco ding 2, he /h/ in <ha e> is le ou and he wo d is connec ed
o he p e ious one, <can’ > ” (P38).
Conce ning sel -pe cei ed ease o unde s anding o he ph ase in Exce p 3, he majo i y
(n = 28, 56%) epo ed no di icul y. A smalle p opo ion ound he speech easie in R1 han
in R2 (n = 5, 10%), while o he s ound i easie in R2 han in R1 (n = 6, 12%). The emaining
pa icipan s (n = 10, 20%) said ha i was no e y easy, and hey ei he had o ely on he
con ex o lis en o he ph ase mul iple imes. One pa icipan men ioned ha amilia i y wi h
he accen helped. One pa icipan did no answe he ques ion.
The las wo exce p s, 4 and 5, ocused on he in usi e sounds /w/ and /j/, espec i ely. Fo
Exce p 4, only one pa icipan co ec ly iden i ied he in usi e sound, s a ing: “I sounds like
he e is a /w/ sound in be ween hem” (P11). The es ei he did no no ice any in usi e sound,
o p o ided i ele an in o ma ion, such as hea ing an ex a / / o a p olonged /s/ sound. No
pa icipan iden i ied he in usi e /j/ sound in Exce p 5. Howe e , one s uden epo ed
di icul y unde s anding he ph ase, s a ing: “I was pa ially unde s andable, since he i s
wo d o he ph ase was incomp ehensible.” (P34), sugges ing ha he p esence o he in usi e
sound may ha e a ec ed comp ehension. Rega ding hei sel -pe cei ed ease o unde s anding
he ph ase in Exce p 4, he majo i y did no expe ience any di icul y (n = 29, 58%), while he
es (n = 20, 40%) no ed ha hey had o lis en o i mul iple imes, o ha pauses be ween
wo ds aided comp ehension. One pa icipan did no answe he ques ion.
2 In all o iginal commen s by pa icipan s, hei inconsis en punc ua ion has been eplaced: < > is
used o g aphemes and / / o phonemes, o main ain a ex ha is easy o ead.
Duckinoska-Mihajlo ska
Pe cep ions o na u alness, di icul y & speech a e
22
5. Discussion
The i s esea ch ques ion in ended o in es iga e he pa icipan s’ pe cep ions o he
na u alness o na i e speech and he di icul ies hey encoun e ed unde s anding i . The as -
paced speech, which ea u ed mo e ins ances o connec ed speech ea u es, was gene ally
pe cei ed as mo e na u al due o i s absence o pauses, he apid speech a e, and he speake
sounding na i e. These esul s sugges ha as speech is associa ed wi h luency, despi e he
pa icipan s epo ing di icul ies in pe cei ing speci ic CSPs. This suppo s he iew ha CSPs
con ibu e o he pe cei ed egula i y o English hy hm (Alameen & Le is, 2015).
In e ms o sel -pe cei ed ease o unde s anding, nei he eco ding was pe cei ed as
especially di icul , and he di e ence was no s a is ically signi ican . This may be due o he
pa icipan s’ o e all amilia i y wi h English h ough o mal ins uc ion and exposu e o
a ious na i e accen s in he media. Howe e , u he esea ch is needed o sys ema ically
in es iga e how such p io exposu e a ec s he abili y o p ocess na i e speech and whe he
sel - epo ed ease o unde s anding co ela es wi h ac ual unde s anding, as au hen ic speech
o en emains challenging (Wong e al., 2021).
When compa ing he wo eco dings, he pa icipan s we e able o iden i y mo e ob ious
dis inc ions, such as a ia ions in speech pace and he absence o pauses be ween wo ds.
Ne e heless, mo e sub le CSPs, such as elision and in usi e sounds, p o ed mo e di icul .
This was e lec ed in he esponses o he speech exce p s explo ed in he second esea ch
ques ion. Fo example, while mo e han hal o he pa icipan s iden i ied /h/ elision in Exce p
3, his ecogni ion was inconsis en ac oss bo h eco dings and mainly obse ed in R2. In
con as , he in usi e sounds /w/ and /j/ in Exce p s 4 and 5, espec i ely, we e sca cely
no iced. Only one pa icipan iden i ied he in usi e /w/, and ano he appea ed o misin e p e
he ini ial pa o he ph ase in Exce p 5, po en ially due o he p esence o in usi e /j/. These
indings align wi h p e ious esea ch, ein o cing he no ion ha CSPs a e di icul o EFL
lea ne s o pe cei e and decode accu a ely.
Pa icipan s’ pe cep ion o owel educ ion in Exce p s 1 and 2 showed ha hey ound i
easie o de ec he educed sound in /wəz/ in R2 o Exce p 1 han in /əz/ in Exce p 2. The
majo i y o pa icipan s (84%) iden i ied he schwa sound in R2 o Exce p 1, whe eas nea ly
hal as many (44%) no iced i o epo ed hea ing only a /s/ sound in R2 o Exce p 2. While
hese indings a e based on a limi ed numbe o examples and should no be gene alised, hey
a e consis en wi h he esul s o a p oduc ion s udy o owel educ ion among Macedonian
EFL lea ne s (Duckinoska, 2021). This s udy ound ha he L2 lea ne s mo e eadily p oduce
he weak o m o g amma ical wo ds con aining he s ong owel /ɒ/ (e.g., was) han he weak
o m o wo ds wi h he s ong owel /æ/ (e.g., am, can, shall, has). Ne e heless, u he
esea ch is needed o es ablish a clea e link be ween lea ne s’ pe cep ion and p oduc ion o
weak o ms.
In e es ingly, he pa icipan s equen ly e e enced e ms such as accen , na i e speake , o
speci ic dialec . Accen was o en ci ed as a sou ce o di icul y, while he deli e y o he na i e
speake in R2 was pe cei ed as mo e na u al. These indings sugges ha he pa icipan s’
pe cep ion o speech na u alness was p ima ily in luenced by wo ac o s: 1) he speake being
pe cei ed as na i e, and 2) he absence o a no iceable accen . This insigh ca ies signi ican
pedagogical implica ions: lea ne s should be made awa e o he di e si y o English accen s –
bo h na i e and non-na i e – and lis en o a wide ange o speech models. As exposu e o Global
Englishes does no hinde comp ehension (Miao e al., 2024), such p ac ice can b oaden
lea ne s’ awa eness o wha cons i u es in elligible and na u al English.
The esul s also sugges ha ce ain CSPs a e ha de o de ec han o he s. Thus, explici
ins uc ion on how CSPs aid luency and hy hm could be bene icial, as inc eased awa eness
imp o es decoding o apid speech (Kennedy & Blanche , 2014). Fo ins ance, eache s could

Duckinoska-Mihajlo ska
Pe cep ions o na u alness, di icul y & speech a e
23
con as sen ences wi h and wi hou in usi e sounds, highligh ing how hei p esence helps
main ain luency. They could also design ac i i ies whe e s uden s lis en speci ically o
in usi e sounds in au hen ic speech and ma k hem in w i en ansc ip s. Mo eo e , since
owel educ ion was easie o de ec in was han in as, EFL ins uc ion should explici ly
highligh how owel educ ion a ies ac oss di e en g amma ical wo ds. Finally, he ac ha
only 60% o pa icipan s co ec ly no iced /h/ elision in ha e sugges s ha lea ne s may no
always be awa e o common elisions in connec ed speech. Class oom ac i i ies should include
a ge ed lis ening asks wi h common examples o elision, ollowed by p onuncia ion p ac ice
o ein o ce ecogni ion.
6. Conclusion and limi a ions
This s udy in es iga ed Macedonian EFL lea ne s’ pe cep ions o na i e speech wi h he
p esence and absence o connec ed speech ea u es. Al hough he pa icipan s epo ed
ela i ely high ease o unde s anding, many no ed he need o lis en o he exce p s mul iple
imes, o ely on con ex , unde sco ing he ac ha au hen ic speech emains challenging.
Simul aneously, he pe cep ion o Reco ding 2 as mo e na u al sugges s ha connec ed speech
ea u es con ibu e o he na u al hy hm and pe cei ed luency o English.
The cu en s udy has se e al limi a ions. The esul s a e based on speci ic samples aken
om a speci ic con ex ; di e en exce p s om he eco dings, o di e en con ex s may ha e
elici ed di e en esponses. In addi ion, using di e en speech models (e.g., Aus alian
English) may in luence pa icipan s’ esponses. Fu he esea ch could explo e hese a iables
and measu e he e ec s o exposu e o connec ed speech on pe cep ion and p oduc ion. Despi e
hese limi a ions, he indings con ibu e o a be e unde s anding o how EFL lea ne s
pe cei e au hen ic spoken English and emphasise he impo ance o d awing lea ne s’ a en ion
o connec ed speech ea u es as a s ep owa ds imp o ed lis ening comp ehension.
Re e ences
Ahmadian, M., & Ma ou , R. (2014). The e ec o explici ins uc ion o connec ed speech
ea u es on I anian EFL lea ne s’ lis ening comp ehension skills. In e na ional Jou nal o
Applied Linguis ics and English Li e a u e, 3(2), 227–236.
h p://dx.doi.o g/10.7575/aiac.ijalel. .3n.2p.227
Alameen, G., and Le is, J. M. (2015). Connec ed speech. In M. Reed & J. M. Le is (Eds.), The
handbook o English p onuncia ion (pp. 159–174). Wiley Blackwell.
h p://doi.o g/10.1002/9781118346952.ch9
B own, J.D., & Hil e y, A. (2006). The e ec i eness o eaching educed o ms o lis ening
comp ehension. In J. D. B own & K. Kondo-B own, (Eds.), Pe spec i es on eaching
connec ed speech o second language speake s (pp. 51–58). Uni e si y o Hawai’i, Na ional
Fo eign Language Resou ce Cen e .  h ps://n l c.hawaii.edu/PDFs/MG01TOC.pd
De wing, T. M., & Mun o, M. J. (1997). Accen , in elligibili y, and comp ehensibili y:
E idence om ou L1s. S udies in Second Language Acquisi ion, 19, 1–16.
h ps://doi.o g/10.1017/S0272263197001010
Duckinoska, I. (2021). Vowel educ ion o English g amma ical wo ds by Macedonian EFL
lea ne s. In A. Ki ko a-Nasko a, A. Hende son, & J., Fouz-González (Eds.), English
p onuncia ion ins uc ion: Resea ch-based insigh s (pp. 279–302). John Benjamins
Publishing Company. h ps://doi.o g/10.1075/aals.19.12duc
IBM Co p. (2017). IBM SPSS S a is ics o Windows, Ve sion 25.0 [Compu e so wa e]. IBM
Co p.
Duckinoska-Mihajlo ska
Pe cep ions o na u alness, di icul y & speech a e
24
h ps://www.ibm.com/suppo /pages/downloading-ibm-spss-s a is ics-25
I o, Y. (2006). The signi icance o educed o ms in L2 pedagogy. In J. D. B own & K. Kondo-
B own, (Eds.), Pe spec i es on eaching connec ed speech o second language speake s (pp.
17–25). Uni e si y o Hawai‘i, Na ional Fo eign Language Resou ce Cen e .
h ps://n l c.hawaii.edu/PDFs/MG01TOC.pd
Kang, O., Thomson, R., & Mo an, M. (2020). Which ea u es o accen a ec unde s anding?
Explo ing he in elligibili y h eshold o di e se accen a ie ies. Applied Linguis ics, 41(4),
453–480. h ps://doi.o g/10.1093/applin/amy053
Kennedy, S., & Blanche , J. (2014). Language awa eness and pe cep ion o connec ed speech
in second language. Language Awa eness, 23, 92–106.
h ps://doi.o g/10.1080/09658416.2013.863904
McKenzie, R. M. (2008). Social ac o s and non‐na i e a i udes owa ds a ie ies o spoken
English: A Japanese case s udy. In e na ional Jou nal o Applied Linguis ics, 18(1), 63–88.
h ps://doi.o g/10.1111/j.1473-4192.2008.00179.x
Miao, Y., Kang, O., & Meng, X. (2024). Inco po a ing global Englishes a ie ies in o EFL
class ooms: De elopmen o lis ening comp ehension and p onuncia ion. TESOL
Qua e ly. h ps://doi.o g/10.1002/ esq.3328
Poeso á, K, & Weinga o á, L. (2018). Cha ac e o owel educ ion in Czech English. In J.
Volín & R. Ska ni zl (Eds.), The p onuncia ion o English by speake s o o he languages
(pp. 96–116). Camb idge Schola s.
Rahimi, M. & Chalak, A. (2017). The e ec o connec ed speech eaching on lis ening
comp ehension o I anian EFL lea ne s. Jou nal o Applied Linguis ics and Language
Resea ch, 4(8), 280–291. h ps://www.jall .com/index.php/JALLR/a icle/ iew/722
Suzukida, Y., & Sai o, K. (2019). Which segmen al ea u es ma e o success ul L2
comp ehensibili y? Re isi ing and gene alizing he pedagogical alue o he unc ional load
p inciple. Language Teaching Resea ch, 25, 431–450.
h ps://doi.o g/10.1177%2F1362168819858246
Wong, S. W., Leung, V. W., Tsui, J. K., Dealey, J., & Cheung, A. (2021). Chinese ESL lea ne s'
pe cep ual e o s o English connec ed speech: Insigh s in o lis ening comp ehension.
Sys em, 98, 1–10. h ps://doi.o g/10.1016/j.sys em.2021.102480
Appendix
Pa icipan s’ ques ionnai e
Pa 1
Please comple e he ques ionnai e below. The in o ma ion will only be accessible o and used by he
esea che o esea ch pu poses and all he da a will be anonymised.
Name and su name
Age
Gende
Male
Female
O he :
Wha na ionali y a e you?
Wha is you i s language/mo he ongue?
Duckinoska-Mihajlo ska
Pe cep ions o na u alness, di icul y & speech a e
25
P io o s a ing uni e si y, did you s udy English ou side
school?
a) p i a e language cou ses
Time pe iod: ______
b) p i a e lessons
Time pe iod: ______
Ha e you s ayed in an English-speaking coun y? a) Yes
Which one:
Time pe iod:
b) No
How would you a e you sa is ac ion wi h you English
p onuncia ion on a scale om 1 o 5?
1 = no sa is ied a all
5 = ex emely sa is ied
1 2 3 4 5
Do you hink ha o mal aining in English phone ics and
phonology helps you imp o e you p onuncia ion?
a) I is de ini ely help ul.
b) I is help ul in some ways.
c) I is no help ul a all.
Do you speak any o he o eign language(s)?
I yes, please s a e which one(s).
Pa 2
In his pa , you will lis en o se e al eco dings. We a e in e es ed in you hough s abou he
eco dings. You will no be es ed on any phone ic/phonological aspec . Please ollow he ins uc ions
as o wha eco ding you need o lis en o and how many imes. You need o double click on he icon
o he audio o s a .
S ep 1: Lis en o Reco ding 1 only once and hen answe he ques ions below.
Reco ding 1.mp3
How easy was i o unde s and eco ding 1?
1 = e y di icul
5 = e y easy 1 2 3 4 5
Did you ind any p onuncia ion aspec pa icula ly
challenging o unde s and?
Please desc ibe i in you own wo ds.
S ep 2: Lis en o Reco ding 2 only once and hen answe he ques ions below.
Reco ding 2.mp3
Duckinoska-Mihajlo ska
Pe cep ions o na u alness, di icul y & speech a e
26
How easy was i o unde s and eco ding 2?
1 = e y di icul
5 = e y easy 1 2 3 4 5
Did you ind any p onuncia ion aspec pa icula ly
challenging o unde s and?
Please desc ibe i in you own wo ds.
S ep 3: Lis en o Reco ding 1 and Reco ding 2 one mo e ime, i necessa y, and hen compa e
Reco ding 1 and Reco ding 2. Please ead he ques ions i s be o e lis ening.
Which o he wo eco dings
sounds mo e na u al o you?
a) Reco ding 1
Please explain why:
b) Reco ding 2
Please explain why:
c) They bo h sound he same o me.
Did you no ice any ob ious
di e ences be ween he wo
eco dings? Choose as many as
applicable.
a) The speake speaks as e .
b) Some sounds ha e been le ou .
c) Some ex a sounds ha e been added.
d) Some wo ds ha e been le ou .
e) Some wo ds a e p onounced sho e .
) Some imes he e a e no pauses be ween wo wo ds.
S ep 4: You will lis en o exce p s om bo h eco dings and answe he ques ions below. You can
lis en o each exce p up o i e imes.
Lis en o Exce p 1. Pay a en ion o he way ‘was’ is p onounced in bo h eco dings.
Exce p 1.wa
Is ‘was’ p onounced he same in bo h
eco dings?
a) Yes
b) No
I no , wha is di e en ?
Lis en o Exce p 2. Pay a en ion o he way ‘as’ is p onounced in bo h eco dings.
Exce p 2.wa
Is ‘as’ p onounced he same in bo h eco dings? a) Yes
b) No
I no , wha is di e en ?
Lis en o Exce p 3. Pay a en ion o he e b ph ase in he clause in bo h eco dings.
Gyu ka
P onuncia ion eaching p ac ices in Hunga ian schools
33
p onuncia ion speci ically. These open-ended esul s a e seemingly in line wi h he mos
commonly used p onuncia ion eaching echnique he eache s epo edly used (see Table 2).
Table 2
Desc ip i e S a is ics o P onuncia ion Teaching Techniques
P onuncia ion eaching echnique M SD
using lis en-and- epea 3.77 1.40
eaching an exe cise om he cou sebook 3.39 1.34
using pai wo k 3.06 1.50
ma king p onuncia ion in d awing
(e.g., unde lining he s essed syllable) 2.91 1.47
using he minimal pai echnique 2.89 1.40
eaching how o ead he IPA symbols 2.39 1.33
using a game o p ac ice p onuncia ion 2.30 1.23
gi ing me alinguis ic explana ions; poin ing ou he
di e ences be ween English and Hunga ian 2.20 1.12
eaching p onuncia ion by elying on ouch
(e.g., ouching he neck o eel oiced consonan s) 2.13 1.23
using ideos 2.06 1.18
using online games o p ac ice 2.03 1.25
eaching how o ecognise IPA symbols 1.97 1.20
using Jazz Chan s 1.89 1.10
using mo emen (e.g., clapping) 1.77 1.18
ha ing he s uden s eco d hemsel es 1.69 0.87
The pa icipan s s a ed ha hey mos equen ly employed he “lis en and epea ” echnique
o p onuncia ion eaching, once e e y mon h. This sugges s ha hey only used ‘lis en and
epea ’ when co ec ing e o s, because ocabula y i ems a e in oduced mo e o en han jus
e e y mon h. The second mos commonly used echnique was using exe cises om a
cou sebook e e y wo o h ee mon hs. In e es ingly, esponden s epo ed e y limi ed use o
IPA symbols in he class oom. The esul s showed ha IPA symbols we e used o
demons a ion pu poses less equen ly han e e y wo o h ee mon hs, whe eas eaching how
o w i e and use IPA symbols was e en a e .
In e ms o p onuncia ion ea u es (see Table 3), silen le e s and he den al ica i es we e
augh he mos o en; bo h ea u es we e epo ed o come up a ound e e y mon h. Among he
leas equen ly augh ea u es we e he s ess- imed hy hm, he weak p onuncia ion o
g amma wo ds, and aspi a ion. These ea u es we e a ely add essed, less equen ly han wo
o h ee mon hs.

Gyu ka
P onuncia ion eaching p ac ices in Hunga ian schools
34
Table 3
Desc ip i e S a is ics o Teaching P onuncia ion Fea u es
P onuncia ion ea u e M SD
silen le e s 4.05 1.03
he den al ica i es 3.87 1.24
wo d s ess 3.84 1.35
he di e ences be ween SSBE (S anda d
Sou he n B i ish English) and GA (Gene al
Ame ican)
3.55 1.25
he -ed su ix 3.53 1.28
he -s su ix 3.52 1.17
he di e ence be ween /w/ and / / 3.50 1.40
he /ŋ/ sound 3.25 1.32
owel educ ion in uns essed syllables 3.23 1.49
syllabic consonan s 3.06 1.44
he di e ence be ween /əʊ/ and /ɔː/ 3.03 1.28
aspi a ion 2.83 1.28
he di e ences be ween na i e accen s o English
(o he han SSBE and GA) 2.50 1.14
weak o ms o g amma wo ds 2.39 1.39
s essed- ime hy hm 2.36 1.24
The o e all esul s o hese wo scales sugges ha Hunga ian seconda y school eache s
include p onuncia ion ea u es e e y wo o h ee mon hs (M=3.23; SD=0.91), and ha hey
use some eaching echniques only occasionally (M=2.30; SD=0.71).
4.3 Va iables in luencing Hunga ian eache s’ p onuncia ion eaching p ac ices (RQ3)
Fi s , o de e mine which a iables migh impac on eache s’ p onuncia ion eaching p ac ices,
a se ies o eg ession analyses we e un using he equency o p onuncia ion eaching
echniques, and hen he equency o eaching p onuncia ion ea u es as a dependen a iable.
The 16 i ems measu ing belie s abou p onuncia ion eaching we e ini ially en e ed in o he
model as independen a iables. I ems ha did no exe a s a is ically signi ican in luence on
he dependen a iables we e subsequen ly emo ed one a a ime, wi h he analyses being
e un a e each emo al.
The esul s (see Table 4) show ha he explana o y powe o he i s model is somewha
s onge as i accoun ed o a g ea e pe cen age o a iance in he case o he p onuncia ion
ea u es (R2 = 0.54) No wi hs anding, he second model also accoun s o a conside able
p opo ion o he a iance (R2 = 0.49). The belie ha p onuncia ion eaching is impo an a a
seconda y school le el had a posi i e, albei weak impac on he equency o p onuncia ion
ea u es add essed in he lessons. In e es ingly, he belie ha p onuncia ion eaching can
imp o e he lea ne s’ con idence had a mode a ely posi i e e ec on he a ious echniques
employed du ing he lessons, bu his did no ha e an impac on he p onuncia ion ea u es
Gyu ka
P onuncia ion eaching p ac ices in Hunga ian schools
35
augh . The simila i y be ween he wo models is ha he belie ha p onuncia ion can imp o e
on i s own had a weak, bu nega i e in luence on how equen ly eache s augh p onuncia ion
ea u es and used a ious p onuncia ion eaching echniques.
Table 4
Linea Reg ession P edic ing F equency o P onuncia ion Fea u es and Teaching
Techniques
R2 Impo ance Con idence Imp o e
P onuncia ion ea u es 0.54 0.36 n.s. -0.29
P onuncia ion eaching echniques 0.49 n.s. 0.40 -0.23
No es. (p<0.05). Impo ance = belie ha eaching p onuncia ion is impo an a a seconda y school
le el; Con idence = belie ha p onuncia ion eaching imp o es he lea ne s’ con idence; Imp o e =
belie ha he lea ne s’ p onuncia ion can imp o e on i s own.
4.4 The e ec o aining on p onuncia ion eaching p ac ices (RQ4)
To de e mine whe he he aining eache s ecei ed ela ed o p onuncia ion eaching
me hodology played a signi ican ole in e ms o hei p ac ices, independen -samples - es s
we e un on he cons uc o p onuncia ion eaching echniques and p onuncia ion ea u es. As
he e was no signi ican di e ence be ween eache s wi h and wi hou p io educa ion in
ela ion o ei he he equency o eaching p onuncia ion ea u es o he equency o using
p onuncia ion eaching echniques, he la e cons uc was examined o possible la en
dimensions o see whe he he di e en dimensions would show a signi ican di e ence. The
15 i ems o he scale we e subjec ed o ac o analysis.
Table 5
Resul o he Fac o Analysis
P onuncia ion eaching echniques Fac o loadings
1 2
eading IPA symbols .837
w i ing IPA symbols .888
eaching ules (me alinguis ic explana ion)
.480
he minimal pai echnique
.692
lis en and epea
.467
d awing
.841
pai wo k
.700
No es. Ex ac ion me hod: Maximum likelihood; Ro a ion me hod: Oblimin. Loadings smalle han
0.45 we e emo ed.
Gyu ka
P onuncia ion eaching p ac ices in Hunga ian schools
36
P io o his, he sui abili y o he da a o ac o analysis was e i ied.1 The esul s yielded
a wo- ac o solu ion, which can be seen in Table 5. Fac o 1 p esen s a dimension ha
subsumes eaching he IPA symbols, while ac o 2 con ains eaching echniques o he han he
use o IPA in he class oom.
The independen -samples - es s we e conduc ed o compa e eache s wi h p io educa ion
in p onuncia ion eaching me hodology (n = 28) and eache s wi hou p io educa ion (n = 36).
As he esul s in Table 6 indica e, he e was a signi ican di e ence be ween he g oups in e ms
o he equency o p onuncia ion eaching when i comes o eaching IPA in he class oom,
bu no in he case o he o he p onuncia ion eaching echniques.
Table 6
The Resul s o he Independen -Samples - es s
M SD p
equency o using o he echniques wi h educa ion 3.20 0.99 1.69
0.10
wi hou educa ion 2.78 0.97
equency o eaching IPA wi h educa ion 2.57 1.11 2.30
0.03
wi hou educa ion 1.90 0.84
5. Discussion
The esul s e ealed ha Hunga ian seconda y school eache s indica e ha hey did no add ess
p onuncia ion o en du ing lessons, and whene e hey did de o e ime o p onuncia ion
eaching, i was o each ocabula y i ems. Dealing wi h p onuncia ion only when eaching a
new ocabula y i em is no su icien on i s own; p onuncia ion should be conside ed as wo hy
o being augh , jus like o he language aspec s, e.g., g amma .
In e ms o eaching echniques, “lis en and epea ” and cou sebook exe cises we e selec ed
as among he mos equen ly used echniques, showing he same endency as in o he
geog aphical con ex s (see Ja a i e al., 2021; Szyszka 2016). Fu he mo e, he ex ensi e use
o lis en-and- epea ” is no su p ising, as a conside able numbe o he pa icipan s conside ed
ocabula y eaching o me e co ec i e eedback o cons i u e explici p onuncia ion eaching.
Responden s selec ed silen le e s as he mos commonly add essed p onuncia ion ea u e.
This may be linked o p onuncia ion being included in he lessons whene e he p onuncia ion
o a ocabula y i em seemed p oblema ic. While i is possible o each he appea ance o silen
le e s as a sys em, i is highly unlikely ha eache s de o ed ime o demons a e he pa e n o
hei s uden s. Based on he quali a i e answe s, eache s p obably add essed his ea u e
whene e i caused an issue ela ed o a single ocabula y i em. The o he p onuncia ion ea u e
wi h he highes equency o eaching was he den al ica i es. This inding leads o an
in e es ing disc epancy be ween he heo e ical unde pinnings o p onuncia ion eaching and
1 The Kaise -Meye -Olkin alue was 0.73, which exceeded he 0.60 ecommended minimal alue
(Geo ge & Malle y, 2019); he Ba le ’s Tes o Sphe ici y p o ed o be s a is ically signi ican (Ⲭ2 =
138.70, p < 0.001), which u he demons a ed ha he da a was su icien o he analysis. Then,
ac o analysis was pe o med using maximal likelihood and Di ec oblimin o a ion as i was
assumed ha he esul s would co ela e.
Gyu ka
P onuncia ion eaching p ac ices in Hunga ian schools
37
sac ual p ac ices, which indica es ha he pa icipan s’ eaching p ac ices migh no ha e been
in o med by p onuncia ion eaching heo y. As i has been es ablished ha p onouncing he
den al ica i es does no in luence in elligibili y (Jenkins, 2000), i migh be unnecessa y o
ocus on eaching hem. I may ne e heless be aluable o iden i y and a ge speci ic
p onuncia ion ea u es ha p esen pe sis en challenges o Hunga ian lea ne s (e.g., wo d
s ess) and whose imp o emen could u he enhance o e all in elligibili y.
In ela ion o he di e en a iables which migh in luence eache s’ p onuncia ion eaching
p ac ices, i was ound ha he belie ha p onuncia ion imp o es on i s own nega i ely impac s
he equency o eaching. Along hese lines, he belie ha p onuncia ion should be augh in
seconda y educa ion and ha i imp o es he s uden s’ con idence had a posi i e impac on he
equency o inclusion. These indings undeniably highligh he need o well-designed
p onuncia ion eaching me hodology cou ses in eache educa ion, since only aining can
p o ide insigh o eache s ha can equip hem wi h he necessa y knowledge o in o m hei
p ac ices (Bo g, 2011).
The impo an ole o eache educa ion in in luencing eache s’ p ac ices was also e iden
as he e was a di e ence be ween eache s who had a cou se on p onuncia ion eaching and
eache s who did no . As could be seen abo e (Table 6), he e was a signi ican di e ence in
e ms o eaching IPA symbols, which was expec ed, since using he IPA symbols o adequa ely
p o ide me alinguis ic explana ions migh be di icul o un ained eache s. All o hese
esul s poin o he same endency obse ed in con ex s ou side Hunga y: eache educa ion
plays a pi o al ole in la e p onuncia ion inclusion (see Hende son e al., 2012; Shah e al.,
2017).
6. Implica ions
The indings o his s udy unde sco e se e al impo an pedagogical implica ions o
p onuncia ion ins uc ion in Hunga ian seconda y educa ion. Fi s , he limi ed na u e o
p onuncia ion eaching ( ypically ied o ocabula y ins uc ion) sugges s a need o mo e
beyond his app oach. P onuncia ion should be augh sys ema ically in he same manne as
o he language skills, and no me ely add essed when indi idual lexical i ems a e p oblema ic
o when in oducing new ocabula y i ems. Second, he esul s o his s udy poin o he
impo ance o eache educa ion in p onuncia ion eaching me hodology. Mo e han hal o he
eache s indica ed ha hey had no ecei ed o mal ins uc ion in his a ea. The e o e, he e is
a p essing need o p onuncia ion eaching me hodology cou ses in p e-se ice and in-se ice
eache educa ion in Hunga y. I his came o pass, and eache s had adequa e heo e ical
knowledge o why and how o in eg a e p onuncia ion e ec i ely in o hei class oom p ac ice,
hei belie s migh change and hey migh p omo e p onuncia ion in hei class ooms jus like
any o he language skill.
Acknowledgemen s
The esea ch has been suppo ed by he EKÖP-24 Uni e si y Excellence Schola ship P og am
o he Minis y o Cul u e and Inno a ion om he Sou ce o he Na ional Resea ch,
De elopmen and Inno a ion Fund. This s udy was also conduc ed in he ame o p ojec no.
PPKE-BTK-KUT-23-2, suppo ed by he Facul y o Humani ies and Social Sciences o
Pázmány Pé e Ca holic Uni e si y.
Gyu ka
P onuncia ion eaching p ac ices in Hunga ian schools
38
Re e ences
Bai, B., & Yuan, R. (2019). EFL eache s’ belie s and p ac ices abou p onuncia ion eaching.
ELT Jou nal, 73(2), 134–143. h ps://doi.o g/10.1093/el /ccy040
Bake , A. (2011). ESL eache s and p onuncia ion pedagogy: Explo ing he de elopmen o
eache s’ cogni ions and class oom p ac ices. In J. Le is & K. LeVelle (Eds.), P oceedings
o he 2nd p onuncia ion in second language lea ning and eaching con e ence (pp. 82–94).
Iowa S a e Uni e si y. h ps://apling.engl.ias a e.edu/wp-
con en /uploads/si es/221/2015/05/PSLLT_2nd_P oceedings_2010.pd
Bake , A. (2014). Explo ing eache s’ knowledge o second language p onuncia ion
echniques: Teache cogni ions, obse ed class oom p ac ices, and s uden pe cep ions.
TESOL Qua e ly, 48(1), 136–163. h ps://doi.o g/10.1002/ esq.99
Ba an-Łuca z, M. (2017). FL p onuncia ion anxie y and mo i a ion: Resul s o a mixed-
me hod s udy. In E. Piechu ska-Kuciel, E. Szymańska-Czaplak & M. Szyszka (Eds.), A he
C oss oads: Challenges o Fo eign Language Lea ning (pp. 107–133). Sp inge Cham.
h ps://doi.o g/10.1007/978-3-319-55155-5_7
Bo g, S. (2003). Teache cogni ion in language eaching: A e iew o esea ch on wha
language eache s hink, know, belie e, and do. Language Teaching, 36(2), 81–109.
h ps://doi.o g/10.1017/S0261444803001903
Bo g, S. (2005). Teache cogni ion on language eaching. In K. Johnson (Ed.), Expe ise in
second language lea ning and eaching (pp. 190–209). Palg a e Macmillan.
h ps://doi.o g/10.1057/9780230523470_10
Bo g, S. (2011). The impac o in-se ice eache educa ion on language eache s’ belie s.
Sys em, 39(3), 370–380. h ps://doi.o g/10.1016/j.sys em.2011.07.009
Buss, L. (2016). Belie s and p ac ices o B azilian EFL eache s ega ding p onuncia ion.
Language Teaching Resea ch, 20(5), 619–637. h ps://doi.o g/10.1177/1362168815574145
Dö nyei Z. (2007). Resea ch me hods in applied linguis ics. Ox o d Uni e si y P ess.
Dö nyei, Z., & Csizé , K. (2012). How o design and analyze su eys in SLA esea ch? In A.
Mackey & S. Gass (Eds.), Resea ch me hods in second language acquisi ion: A p ac ical
guide (pp. 74–94). Wiley-Blackwell. h ps://doi.o g/10.1002/9781444347340.ch5
Geo giou, G. P. (2019). EFL eache s’ cogni ions abou p onuncia ion in Cyp us. Jou nal o
Mul ilingual and Mul icul u al De elopmen , 40(6), 538–550.
h ps://doi.o g/10.1080/01434632.2018.1539090
Gilbe , J. (1995). P onuncia ion p ac ice as an aid o lis ening comp ehension. In D. J.
Mendelsohn & J. Rubin (Eds.), A guide o he eaching o second language lis ening (pp.
97–102). Dominie P ess.
Gyu ka, N. (2022). P onuncia ion ac i i ies in he EFL class oom: Impo an o useless? In K.
Helle & I. S eine (Eds.), Unde he umb ella o applied linguis ics (pp. 200–214). ELTE
BTK Alkalmazo Nyel észe i és Fone ikai Tanszék. [o iginal i le: Kiej és elada ok a
középiskolai angol nyel ó án: on os agy elesleges? In K. Helle & I. S eine (Eds.), Az
alkalmazo nyel észe ese nyője ala (pp. 200–214). ELTE BTK Alkalmazo Nyel észe i
és Fone ikai Tanszék.] h ps://www.el e eade .hu/media/2022/10/alknyel -2021-
22_ja .pd
Gyu ka, N., & Piuko ics, Á. (2023). Tailo ing in e na ional ac i i ies o Hunga ian lea ne s
o English. Resea ch in Language, 21(3), 313–332. h ps://doi.o g/10.18778/1731-
7533.21.3.06
Hende son, A., F os , D., Te guje , E., Kau zsch, A., Mu phy, D., Ki ako a-Nasko a, A.,
Waniek-Klimczak, E., Le ey, D., Cunningham, U., & Cu nick, L. (2012). The English
p onuncia ion eaching in Eu ope su ey: Selec ed esul s. Resea ch in Language, 10(1), 5–
27. h ps://doi.o g/10.2478/ 10015-011-0047-4

Gyu ka
P onuncia ion eaching p ac ices in Hunga ian schools
39
Ja a i, S., Ka imi, M. R., & Ja a i, S. (2021). Belie s and p ac ices o EFL ins uc o s in
eaching p onuncia ion. Jou nal o Language and Fo eign Language Lea ning, 10(2), 147–
166. h ps://doi.o g/10.21580/ j 11i110812
Jenkins, J. (2000). The phonology o English as an in e na ional language: New models, new
no ms, new goals. Ox o d Uni e si y P ess.
Kelly, L. G. (1969). 25 cen u ies o language eaching. Newbu y House Publishe s.
Le is, J. M. (2005). Changing con ex s and shi ing pa adigms in p onuncia ion eaching.
TESOL Qua e ly, 39(3), 369–377. h ps://doi.o g/10.2307/3588485
Le is, J. M. (2018). In elligibili y, o al communica ion, and he eaching o p onuncia ion.
Camb idge Uni e si y P ess. h ps://doi.o g/10.1017/9781108241564
Mu phy, J. M., & Bake , A. A. (2015). His o y o ESL p onuncia ion eaching. In M. Reed &
J. M. Le is (Eds.), The handbook o English p onuncia ion (pp. 36–65). John Wiley & Sons.
h ps://doi.o g/10.1002/9781118346952.ch3
Nagle, C., Sachs, R., & Zá a e-Sández, G. (2018). Explo ing he in e sec ion be ween eache s’
belie s and esea ch indings in p onuncia ion ins uc ion. The Mode n Language Jou nal,
102(3), 512–532. h ps://doi.o g/10.1111/modl.12493
Nunan, D., & Mille , L. (Eds.). (1995). New ways in eaching lis ening (2nd ed.). TESOL.
Penning on, M. C. (2021). Teaching p onuncia ion: The s a e o he a 2021. RELC Jou nal,
52(1), 3–21. h ps://doi.o g/10.1177/00336882211002283
Piuko ics, Á. (2016). The changing ole o he IPA in he EFL class oom. In A. Á, Reményi,
Cs. Sá di, & Zs. Tó h (Eds.), Pe spec i es in con empo a y Hunga ian applied linguis ics
(pp. 155–164). Tin a Köny kiadó. [o iginal i le: The changing ole o he IPA in he EFL
class oom. In A. Á, Reményi, Cs. Sá di, & Zs. Tó h (Eds.), Tá la ok a mai magya
alkalmazo nyel észe ben (pp. 155–164). Tin a Köny kiadó.]
Quoc, T. X., Thanh, V. Q., Dang, T. D. M., Mai, N. D. N., & Nguyen, P. N. K. (2021).
Teache s’ pe spec i es and p ac ices in eaching English p onuncia ion a Menglish Cen e .
In e na ional Jou nal o TESOL & Educa ion, 1(2), 158–175. h ps://i-
j e.o g/index.php/jou nal/a icle/ iew/62
Shah, S. S. A., O hman, J., & Senom, F. (2017). The p onuncia ion componen in ESL lessons:
Teache s’ belie s and p ac ices. Indonesian Jou nal o Applied Linguis ics, 6(2), 193–203.
h ps://doi.o g/10.17509/ijal. 6i2.4844
Si akis, N. C., & Souga i, A. (2005). P onuncia ion issues and EIL pedagogy in he pe iphe y:
A su ey o G eek s a e school eache s’ belie s. TESOL Qua e ly, 39(3), 467–488.
h ps://doi.o g/10.2307/3588490
Szyszka, M. (2016). English p onuncia ion eaching a di e en educa ional le els: Insigh s
in o eache s’ pe cep ions and ac ions. Resea ch in Language, 14(2), 165–180.
h ps://doi.o g/10.1515/ ela-2016-0007
Zheng, H. (2015). Teache belie s as a complex sys em: English language eache s in China.
Sp inge . h ps://doi.o g/10.1007/978-3-319-23009-2
Gyu ka
P onuncia ion eaching p ac ices in Hunga ian schools
40
Appendix
S udy ques ionnai e
Pa 1. The equency o p onuncia ion eaching in gene al
How o en do you each p onuncia ion in you lessons?
1 – ne e
2 – less equen ly han e e y 2-3 mon hs
3 – e e y 2-3 mon hs
4 – weekly
5 – almos e e y single lesson
Can you explain in a ew sen ences why?
________________________________________________________________________
Pa 2. Belie s abou p onuncia ion eaching
Below you will ind s a emen s abou p onuncia ion eaching. Please indica e you le el o
ag eemen on a scale o 1 o 5.
1
–
comple ely disag ee; 5
–
comple ely ag ee
1. Teaching p onuncia ion is impo an a seconda y school le el.
2. P onuncia ion should be augh o beginne s.
3. The language lea ne ’s age in luences he e ec i eness o p onuncia ion eaching.
4. English spoken wi h a s ong Hunga ian accen leads o disc imina ion ab oad.
5. The aim o p onuncia ion eaching a seconda y school le el is o elimina e he
Hunga ian accen .
6. The aim o p onuncia ion eaching a seconda y school le el is o make he lea ne ’s
accen in elligible.
7. The bes eache o eaching p onuncia ion is a na i e speake .
8. When eaching p onuncia ion, eache s should poin ou he di e ences be ween he
sound sys ems o English and Hunga ian.
9. P onuncia ion is bes augh by explaining ules.
10. S uden s’ p onuncia ion imp o es spon aneously wi h language use.
11. Spending a leas six mon hs in he a ge language coun y helps imp o e he
lea ne ’s p onuncia ion.
12. Explici p onuncia ion ins uc ion helps imp o e p onuncia ion.
13. My s uden s like i when I co ec hei p onuncia ion.
14. I hink my s uden s would like o change hei p onuncia ion.
15. Teaching p onuncia ion gi es s uden s con idence when using he language.
16. Knowing he p onuncia ion o wo ds imp o es he lea ne s’ lis ening comp ehension.
Gyu ka
P onuncia ion eaching p ac ices in Hunga ian schools
41
Pa 3. A i udes ela ed o p onuncia ion eaching
The ollowing a e s a emen s ela ed o you expe ience when eaching p onuncia ion. Please
also indica e he e how much you ag ee wi h hem on a scale o 1 o 5.
1
–
comple ely disag ee; 5
–
comple ely ag ee
1. I enjoy eaching p onuncia ion.
2. I am con iden when eaching p onuncia ion.
3. Teaching p onuncia ion is di icul .
4. I is exhaus ing o each p onuncia ion.
Pa 4. P onuncia ion ea u es
How o en do you each he ollowing ea u es? Conside all you g oups ega dless o le el
and age.
1 – ne e , 2 – less equen ly han e e y 2-3 mon hs, 3 – e e y 2-3 mon hs, 4 –
mon hly, 5
–
weekly
1. “ h” sounds, e.g., hanks, mo he
2. he di e ence be ween we and e
3. he di e ence be ween boa and bough
4. in wo ds ending in -ng, he e is no p onounced ‘ɡ’, e.g., sing
5. he h ee-way p onuncia ion o he -ed su ix, e.g., loo‘k ’ – clea‘nd’ – nee‘id’
6. he h ee-way p onuncia ion o he -s su ix, e.g., ca‘ sz’ – do‘ɡz’ – bus‘iz’2
7. wo d s ess, e.g., poLICE, pa TIcula
8. silen le e s, e.g., deb , pneumonia, gnome, w i e
9. p onouncing a sho ‘h’ a e ‘p’, ‘ ’, ‘k’, e.g., ‘ph’ie, ‘ h’ap, ‘kh’i e
10. he p onuncia ion o -ism a he end o wo ds, e.g., ou ism p onounced ‘ u izöm’ and
no ‘ u izmö’3
11. he p onuncia ion o g amma wo ds wi h an ö-like sound, e.g., he sen en ence ‘The e
was a boy’ sounds like ‘dö öz ö’
12. uns essed syllables, e.g., pencil, lemon sound like ‘penszöl’, ‘lemön’, espec i ely
13. hy hm, e.g., p onouncing he sen ence ‘Bob died’ akes he same amoun o ime as
‘Bob could ha e died’
14. he di e ence be ween B i ish and Ame ican accen s
15. di e ences be ween o he accen s, e.g. Canadian, Aus alian, No h o England, e c.
Pa 5. P onuncia ion eaching echniques
How o en do you use he ollowing echniques? Conside all you g oups ega dless o le el
and age.
1 – ne e , 2 – less equen ly han e e y 2-3 mon hs, 3 – e e y 2-3 mon hs, 4 –
mon hly, 5
–
weekly
2 Hunga ian o hog aphy was used o ep esen he endings (/s/, /z/, and /ɪz/) o easie unde s anding.
3 Hunga ians o en equa e he sound /ø/ ( ep esen ed wi h <ö> in spelling) wi h he schwa, which is
why his i em was o mula ed as is.
Gyu ka
P onuncia ion eaching p ac ices in Hunga ian schools
42
1. Comple ing he p onuncia ion asks in he cou sebook.
2. Teaching how o ecognise phone ic symbols.
3. Lea ning how o w i e phone ic symbols.
4. S uden s lis en o he eache o a eco ding and hen epea he u e ance/wo d, ying
o imi a e i as accu a ely as possible.
5. P ac ising he di e ence be ween wo ds which only di e in one sound, e.g., bed –
bad, bough – boa , we – e , e c.
6. Ma king p onuncia ion in a ex , e.g., ma king s ess, d awing in ona ion a ows, e c.
7. Teaching p onuncia ion by eeling, e.g., ouching he neck wi h a hand while
p onouncing oiced and oiceless sounds
8. Teaching p onuncia ion h ough mo emen , e.g., clapping, head bobbing, e c.
9. Using p onuncia ion eaching ideos in class.
10. Using class oom games ocusing on p onuncia ion.
11. Teaching p onuncia ion using online games.
12. P ac ising in pai s o g oups wi h dialogue o ole-play, wi h pa icula a en ion o
p onuncia ion.
13. Explaining English p onuncia ion ules and compa ing hem wi h Hunga ian.
14. Using Jazz Chan s o hy hmic p ac ice.
15. Lea ne s eco d hei p onuncia ion and lis en o i .
Pa 6. The equency o p onuncia ion eaching in gene al (again)
I would like o ask you he e y i s ques ion again. Please answe how o en you usually
each p onuncia ion and i you hink you opinion has changed, please desc ibe why.
1 – ne e
2 – less equen ly han e e y 2-3 mon hs
3 – e e y 2-3 mon hs
4 – weekly
5 – almos e e y single lesson
Can you explain in a ew sen ences why?
________________________________________________________________________
Abou he au ho
Noémi Gyu ka is Assis an Lec u e a Pázmány Pé e Ca holic Uni e si y (PPCU). She holds
an MA in Teaching English as a Fo eign Language and Teaching Hunga ian Language and
Li e a u e as a Fi s Language (PPCU, 2022). She is doing he PhD in Language Pedagogy and
English Applied Linguis ics a Eö ös Lo ánd Uni e si y. He esea ch in e es s a e
p onuncia ion eaching and p onuncia ion in eg a ion in he Hunga ian English as a o eign
language class oom. She has been wo king a PPCU since 2022, as a ull- ime lec u e since
Sep embe 2024.
Email: [email p o ec ed]
Ma in-Rubió
Fluency in 2 asks: Lea ne pe o mance & iews
49
he s o y: he la e lines ep esen hose who paused mo e o en, and he s eepe lines ep esen
hose who ead en i e lines wi hou pausing.
4.2 Fluency and p osody o s o y- eading
The impac o luency and p osody in s o y- eading can be illus a ed by compa ing wo
s uden s wi h opposi e app oaches o he ask. S12 and S22 had e y simila MSRs o T1
(11.18 and 11.22 espec i ely), bu hei app oach o T2 was di e en in many espec s. In
e ms o he numbe o syllables, S22 p oduces he canonical 116 syllables, whe eas S12
p oduced 119. S12 mis akenly said be awa e (which adds one syllable) a he han bewa e,
adds ah (one mo e syllable) be o e he second I’ll d op i ha is no in he ex , and does no
use he con ac ion I’m he hi d ime S ick Man ells us who he is, i.e., I AM STICK MAN,
which has one mo e syllable han I’M STICK MAN.
Mo e de ail o his compa ison is p o ided in Table 2, whe e numbe s in b acke s indica e
he ounded leng hs (in milliseconds) o he pauses. S12 pauses 13 mo e imes han S22, wi h
a o al leng h o pausing o 12.5 seconds o S12 and jus 4.2 seconds o S22. S22 eads he
i s wo lines wi hou pausing, whe eas S12 pauses h ee imes. The same happens when he
dog s a s alking; S22 deli e s he i s wo lines by he dog wi hou pauses, whe eas S12
pauses h ee imes.
Table 2
The Two Rende ings o he Fi s Fou Pages by S12 and S22
S12
S22
S ick Man (0.5) li es in he amily ee (0.8)
wi h his S ick Lady Lo e (0.5) and hei s ick
child en h ee (1.2) one day (0.4) he wakes ea ly
and goes o a jog (0.8) S ick Man (0.4) oh S ick
Man (1.0) be awa e o he dog (1.0) a s ick (0.9)
ba ks he dog an excellen s ick (0.6) he igh
kind o s ick (0.3) o my a ou i e ick (0.7)
I’ll e ch i (0.5) and d op i (0.4) and e ch i
(0.3) and hen ah I’ll d op i and e ch i and d op
i again (0.8) I’m no a s ick (0.3) why can’ you
see I’m S ick Man (0.4) I’m S ick Man (0.3) I
am S ick Man ha ’s me (0.4) and I wan o go
home o he amily ee
S ick Man li es in he amily ee wi h his S ick
Lady Lo e and hei s ick child en h ee (0.7) one
day he wakes ea ly (0.3) and goes o a jog (0.4)
S ick Man oh S ick Man bewa e o he dog (0.8)
a s ick ba ks he dog an excellen s ick he igh
kind o s ick o my a ou i e ick (0.3) I'll e ch
i and d op i and e ch i and hen (0.3) I'll d op
i and e ch i and d op i again (1.1) I'm no a
s ick why can' you see I'm S ick Man I'm S ick
Man I'm S ick Man ha 's me (0.3) and I wan o
go home o he amily ee
No e. Numbe s in b acke s indica e he ounded leng hs (in milliseconds) o pauses.
S12 and S22 also had o decide whe he o indi idualise he oices o he na a o , he dog
and S ick Man, and how o con ey he emo ions o hese cha ac e s as ep esen ed by
exclama ion ma ks. S12 decided o a y he oices. Fo example, he oice o he dog had a

Ma in-Rubió
Fluency in 2 asks: Lea ne pe o mance & iews
50
di e en quali y (lowe pi ch) om he o he wo oices, and she used a e y wide pi ch ange
(see Figu e 2, blue line igh ). In con as , S22 did no change oice quali y and he pi ch ange
emained unal e ed (see Figu e 2, blue line le ).
Figu e 2
S22’s Dog (le ) and S12’s Dog ( igh )
Rega ding p oduc ions o S ick Man, he con as is e en g ea e as shown in Figu e 3. S12
adop ed a low-pi ched oice and decided o ansla e he i alics and uppe case in o pi ch
a ia ions. In he i s ende ing, S12 placed he onic syllable on S ick and used a alling one.
Fo he ex in i alics she mo ed he s ess o he wo d Man and used a all- ise one. Fo he
uppe -case e sion, she dis ega ded he con ac ion in he ex and gene a ed an ex a syllable,
and also used a all- ise one. Finally, in he las one uni , she p onounced he onic syllable
me wi h a ise- all one (e en hough she w o e all- ise (F/R) in he P aa image).
S22’s e sion (see Figu e 4) is a single one uni in which he h ee imes S ick Man shou s
I’m S ick Man a e ead in exac ly he same way, deli e ing all he syllables wi h no pauses o
0.25s o mo e (as shown in Table 2), and wi h he inal alling one on he onic syllable me.
Pi ch a ia ion is also minimal, compa ed o he p oduc ions o S12.
The P aa images p esen ed he e a e om s uden s’ wo k, as hey we e augh how o use
P aa o analyse he in ona ion o hei own p oduc ions, using Tench’s (2011) ipa i e no ion
o in ona ion as onali y, onici y, and one. The analysis o Ac i i y 2 was submi ed as a ex
documen wi h P aa images on he page. The analysis o Ac i i y 3 was submi ed in a ideo,
in which hey sha ed hei sc een, and used a Powe Poin p esen a ion including P aa images
and co esponding audio iles. O e all, hey did well o mas e hese basics o P aa , despi e
ine i able e o s (as in Figu e 3).3
3 See He men (2018, 2023) on he di icul y o using P aa wi h s uden s.
Ma in-Rubió
Fluency in 2 asks: Lea ne pe o mance & iews
51
Figu e 3
S12’s S ick Man
“I’m
S ick Man.”
“
I’m S ick Man.
”
“I’M
STICK MAN.”
“ ha ’s me.”
Figu e 4
S22’s S ick Man
Ma in-Rubió
Fluency in 2 asks: Lea ne pe o mance & iews
52
4.3 S uden imp essions o he asks
In o de o unde s and how s uden s expe ienced he asks, a sho ques ionnai e was used
be o e he end o he cou se. I was answe ed by 20 o he 22 s uden s analysed o his s udy.
To he ques ion: “Did you make oices o he dog, he gi l? I no , did you conside doing i
and decided no o in he end? O you jus hough he e was no need?”, he answe s we e
e enly spli . Fi e ied o di e en ia e he cha ac e s somehow bu conside ed wha hey had
done did no amoun o making oices as illus a ed in he ollowing commen : “I ied o
change a li le bi he in ona ion bu I didn’ make oices” (S15). Ten s uden s answe ed hey
had indeed made oices, and i e ga e a nega i e answe . O hese i e esponses, wo s uden s
el i was no necessa y, S22 being one o hem. They ga e he ollowing easons: “i was
necessa y bu I decided no o do i ” (S02), “didn' hink o do i ” (S14), and “because I did no
like how hey we e u ning ou ” (S19).
5. Discussion
This s udy p esen s he esul s o a eaching app oach implemen ed in an unde g adua e cou se.
The app oach ocuses on using he di e ences be ween wo asks o make s uden s e lec on
he concep s o luency, p onuncia ion accu acy, and p osody. The s o y- eading ins uc ions
allowed he s uden s he eedom o choose whe he hey wan ed o make oices o no , because
no e e yone would be com o able making hem; his eedom was sugges ed by a colleague
who eaches op ional, p ac ical hea e wo kshops which hese s uden s can ake o c edi
wi hin he deg ee p og amme.
Gi en he esul s, including he answe s o he ques ionnai e, I would like o eplica e his
pilo s udy, encou aging s uden s mo e explici ly o y and di e en ia e he oices mo e and
o play wi h he possibili ies a hand. S uden s’ ques ionnai e esponses sugges ha a good
op ion may be o show s uden s wha people do when eading s o ies and o e lec on he
impac hese decisions ha e. Fo ins ance, some s uden s make oices ha sound e y di e en
om he na a o ’s, wi h ex eme pi ch a ia ion, o he s make enough mino a ia ions so ha
he lis ene can iden i y a new cha ac e has appea ed, and some do no hing a all. In he i s
ask, i is jus a ma e o p o iding in o ma ion in a clea way, and speech a e and pausing
should co espond o hei p o iciency le el, whe eas he second ask is an en i ely di e en
endea ou and s uden s should be awa e o his. Those who do no hing a all sound mono onous
and would p obably no cap u e child en’s a en ion, and hus ail o ul il he ask’s
en e ainmen unc ion. In addi ion, being able o widen one’s pi ch p oduc ion and o ealise
ha he in ensi y, pi ch, and onici y a ia ions can impac on he audience’s pe cep ion is o
g ea impo ance o imp o ing lea ne s’ o e all speech pe o mance. In o he wo ds, his s o y
eading ask could ep esen a golden oppo uni y o lea ne s o expe ience success in playing
wi h hei oice, in changing, o example, hei p osody and oice quali y.
This s udy p esen s insigh s de i ed om a pedagogical app oach o eaching abou
phonemes, luency, connec ed speech, and in ona ion. On he one hand, i con i med my
in ui ion ha luency measu es would change in he wo asks because, apa om he s uden ’s
le el, hei app oach o s o y- eading would ansla e in o a di e en way o deli e ing
syllables. On he o he hand, i made me ealise ha some s uden s simply see no need o a y
pi ch and di e en ia e cha ac e s, which migh mean mo e wo k is needed o make hem
unde s and he ele ance o pi ch a ia ion and cha ac e di e en ia ion.
Ma in-Rubió
Fluency in 2 asks: Lea ne pe o mance & iews
53
6. Conclusion
This pilo s udy p esen s an app oach o eaching phonology in which s uden s a e asked o
eco d hemsel es spon aneously desc ibing images and eading a child en’s book. Fo he
spon aneous pic u e desc ip ion ask, he s uden ’s p o iciency le el was a e y impo an
ac o , some hing o which hey we e well awa e. Howe e , o he s o y- eading ask, some
pa icipan s seem o be unawa e o wha hey need o do p onuncia ion-wise o make hei
speech mo e appealing. P o iding lea ne s wi h mo e oppo uni ies o play wi h and g asp he
impo ance o oice and pi ch a ia ions migh aise hei phonological awa eness and u he
imp o e hei L2 p onuncia ion.
Re e ences
Boe sma, P., & Weenink, D. (2022). P aa : Doing phone ics by compu e (Ve sion 6.2.14) [Compu e
p og am]. www.p aa .o g
De wing, T. M., Rossi e , M. J., Mun o, M.J. & Thomson, R.I. (2004). Second language luency:
Judgmen s on di e en asks. Language Lea ning, 54(4), 655–679. h ps://doi.o g/10.1111/j.1467-
9922.2004.00282.x
Donaldson, J. (2008). S ick Man. Alison G een Books.
Ejzenbe g, R. (2000). The juggling ac o o al luency: A psycho-sociolinguis ic me apho . In H.
Riggenbach (Ed.), Pe spec i es on luency (pp. 287–313). Uni e si y o Michigan P ess.
Fos e , R., & Skehan, P. (1996). The in luence o planning and ask ype on second language
pe o mance. S udies in Second Language Acquisi ion, 18(3), 299–323.
h ps://doi.o g/10.1017/S0272263100015047
Gin he , A., Dimo a, S., & Yang, R. (2010). Concep ual and empi ical ela ionships be ween empo al
measu es o luency and o al English p o iciency wi h implica ions o au oma ed sco ing. Language
Tes ing, 27(3), 379–399. h ps://doi.o g/10.1177/0265532210364407
He men , S. (2023). F om esea ch o eaching: The case o English ising con ou s. In A. Hende son
& A. Ki ko a-Nasko a (Eds.), P oceedings o he 7 h In e na ional Con e ence on English
P onuncia ion: Issues and P ac ices (pp. 83–96). Uni e si é G enoble-Alpes.
h ps://hal.science/hal-04168829
He men , S. (2018). App en issage e enseignemen de la p osodie : l’impo ance de la isualiza ion.
Re ue ançaise de linguis ique appliquée, 23(1), 73-88.
Read. K. (2014). Clues cue he smooze: Rhyme. pausing. and p edic ion help child en lea n new wo ds
om s o ybooks. F on ie s in Psychology, 5, 149. h ps://doi.o g/10.3389/ psyg.2014.00149
S a kwea he , C. W. (1987). Fluency and s u e ing. P en ice Hall.
Tench, P. (2011). T ansc ibing he sound o English: A phone ics wo kbook o wo ds and discou se.
Camb idge Uni e si y P ess.
Wenne s om, A. (2000). The ole o in ona ion in second language luency. In H. Riggenbach (Ed.),
Pe spec i es on luency (pp. 102–127). Uni e si y o Michigan P ess.
Ma in-Rubió
Fluency in 2 asks: Lea ne pe o mance & iews
54
Appendix
The wo images o he spon aneous desc ip ion ask
Image A Image B
Abou he au ho
Xa ie Ma in-Rubió holds bachelo deg ees in English Philology and in Audio isual
Communica ion om Uni e si a de Lleida, and a mas e ’s deg ee in Eu opean S udies om
Maas ich Uni e si ei . He de ended his doc o al disse a ion in 2011. He wo ked as an English
eache in a high school, wo o icial language academies and wo p i a e uni e si ies be o e
mo ing back o Uni e si a de Lleida. He s a ed wo king ull- ime a UdL in Decembe 2014,
whe e he is now an associa e p o esso . He has been in ol ed in se e al compe i i e p ojec s
as a membe o CLA, and a e a gap o se e al yea s, he is publishing again on he
in e na ionalisa ion o and EMI in highe educa ion, belie s and emo ions in addi ional
language lea ning, and luency and accu acy measu emen s ac oss asks.
Email: xa ie .ma in[email p o ec ed]

This chap e is based on he plena y alk gi en by he au ho a he 8 h
In e na ional Con e ence English P onuncia ion: Issues and P ac ices (EPIP
8) held May 8–10, 2024 a he Uni e si y o Can ab ia in San ande , Spain.
I is licensed unde he C ea i e Commons A ibu ion 4.0 In e na ional
License. To iew a copy o he license, please go o:
h p://c ea i ecommons
.o g/licenses/by/4.0/
.
Penning on, M. (2025). A c i ical examina ion o esea ch on he eaching o p onuncia ion in a second language.
In A. Ki ko a-Nasko a, P. Humánez-Be al, & A. Hende son (Eds.), P oceedings o he 8 h In e na ional
Con e ence on English P onuncia ion: Issues and P ac ices (pp. 55–70). Uni e si é G enoble-Alpes.
h ps://doi.o g/10.5281/zenodo.16695648
A c i ical examina ion o esea ch on he eaching o p onuncia ion in a
second language
Ma ha C. Penning on
Uni e si y o Reading
Bi kbeck Uni e si y o London
Abs ac
Al hough he e is a g owing esea ch base showing app oaches o eaching p onuncia ion ha
ha e p o en e ec i e, bo h he ypes o app oaches ha ha e been sys ema ically s udied and
he con ex s in which hose app oaches ha e been applied a e as ye ela i ely limi ed. The e
is s ill much o explo e ega ding wha app oaches wo k in wha con ex s and unde wha
ci cums ances. I poin ou limi a ions in he esea ch base and a eas ha ha e no ye ecei ed
much a en ion om esea che s. I hen e iew ecen s udies ha a e mo ing p onuncia ion
eaching and esea ch away om i s adi ional na ow ocus and a i icial con ex s o a b oade
concep ion and mo e au hen ic con ex s. I encou age o he esea che s o eplica e hese s udies
and o adap hei me hodologies o di e en con ex s in o de o sys ema ically in es iga e
hese p ac ices. I also encou age class oom eache s o d aw on hose me hodologies o design
ac ion esea ch o ying ou new app oaches in hei class ooms ha ha e p o en e ec i e in
o he con ex s.
Keywo ds: ac ion esea ch, applied esea ch, language eaching, p onuncia ion
Penning on
C i ical examina ion, L2 p onuncia ion eaching esea ch
56
1. In oduc ion
The inc eased in e es in p onuncia ion eaching in he cu en e a (Le is, 2019) has spu ed a
g ea expansion in he esea ch base explo ing e ec i e p ac ices (Le is, 2021a, 2022; Sai o &
Plonsky, 2019). The accumula ion o esea ch indings has made i possible o make
gene alisa ions abou he posi i e e ec s o some app oaches ha ha e been esea ched, such
as pe cep ion-o ien ed app oaches (Sakai & Moo man, 2018), especially High Va iabili y
Phone ic T aining – HVPT (Thomson, 2018), and app oaches ha add ess sup asegmen al
aspec s o p onuncia ion (Lee e al, 2015). No wi hs anding hese ad ances in ou knowledge
o app oaches ha wo k and he body o esea ch suppo ing he alue o eaching
p onuncia ion (De wing & Mun o, 2015; Lee e al., 2015), he esea ch base is s ill limi ed,
and he e is a as e i o y o explo a ion and sys ema ic s udy o he e ec i eness o speci ic
p ac ices (Penning on, 2021). In wha ollows, I e iew he p esen s a e o esea ch in he ield
and o e sugges ions o a eas ha a e ipe o u he in es iga ion by esea che s as well as
by eache s in hei own class ooms, ecommending ac ion esea ch as a way o he la e
g oup o pu esea ch in o p ac ice.
2. Limi a ions o he exis ing body o p onuncia ion esea ch
A c i ical examina ion o he exis ing body o p onuncia ion esea ch e eals a numbe o
limi a ions, including:
 es ic ed ocus o he esea ch;
 ela i ely weak esea ch design;
 insu icien a en ion o me hodological ac o s a ec ing esea ch ou comes;
 lack o sys ema ic eplica ion and compa ison ac oss di e en condi ions;
 limi ed a en ion o au hen ic language and language a ie y; and
 limi ed a en ion o au hen ic communica ion.
An examina ion o each o hese limi a ions se es o delinea e a eas in need o a en ion in
u u e esea ch.
2.1 Res ic ed ocus o he esea ch
The bulk o p onuncia ion eaching esea ch cen es on emedia ing segmen al e o s in he
speech o s uden s who ha e been lea ning English o some ime (Lee e al., 2015; Le is,
2021b; Sai o & Plonsky, 2019; Thomson & De wing, 2015). A ocus on segmen als ends o
mean a ocus on e o and accu acy ha he ield is aiming o mo e away om and owa ds a
ocus on in elligibili y (Le is, 2018). The ocus on segmen als has mean compa a i ely less
a en ion paid o sup asegmen als, discou se, and social a ibu es o p onuncia ion. Mic o-
le el ea u es o p onuncia ion ha e been a ended o a he expense o mac o-le el ea u es
ela ed o in ona ion, hy hm, s ess, and he accompanying subs an ial modi ica ions o
phonemes ha occu h ough linking and coa icula ion du ing unning speech. Because o he
ocus o esea ch on ela i ely ad anced lea ne s (Sai o & Plonsky, 2019), me hodologies o
eaching p onuncia ion o beginne s ha e ha dly been explo ed. In addi ion, he published
esea ch has cen ed mainly on lea ne s o English, wi h limi ed a en ion paid o p onuncia ion
eaching o lea ne s o o he languages (Le is, 2021b; Penning on, 2021). Mo e esea ch is
needed on p onuncia ion in con ex , on younge and ea ly-s age lea ne s, and on lea ne s o
languages o he han English.
Penning on
C i ical examina ion, L2 p onuncia ion eaching esea ch
57
2.2 Rela i ely weak esea ch design
Lee e al.’s (2015) discussion o hei me a-analy ical e iew poin s o weak esea ch design
and me hodology as a p oblem in p onuncia ion esea ch. Many s udies showing posi i e
indings need o be ollowed up by u he s udies explo ing he same app oaches wi h s onge
designs in e ms o me hodology and esea ch s anda ds. Ha dison (in p ess) lis s all he
ea u es ha cha ac e ise success ul pe cep ion aining o p onuncia ion sugges ing he
complexi y and numbe o elemen s equi ed o cons i u e a s ong esea ch design o
success ul audi o y pe cep ion aining:
1. compa ible es ing and aining s imuli wi h e i ied in elligibili y;
2. na u al e sus syn hesised speech;
3. mul iple exempla s o one o mo e a ge sounds in a ange o phone ic
en i onmen s o acili a e gene alisa ion o new s imuli;
4. s imuli p oduced by mul iple alke s (i.e., oices) o p omo e obus pe cep ual
ca ego y de elopmen and gene alisabili y;
5. an iden i ica ion ( s. a disc imina ion) ask;
6. ela i ely implici aining wi h eedback;
7. gene alisa ion o imp o ed pe cep ion o no el s imuli and un amilia oices;
8. ans e o skill o o he asks such as p oduc ion in he absence o explici
p oduc ion aining;
9. e en ion o imp o ed abili ies.
Only a small pe cen age o pe cep ion s udies inco po a e all o hese ea u es, HVPT being
he ocus o mos o hem. As explained in Thomson (2018), HVPT me hodology o pe cep ual
enhancemen has been shown o help lea ne s e ise hei men al ep esen a ions o L2 sounds
wi h po en ial ans e o p oduc ion o he sounds whose pe cep ion has been ained. In
addi ion o HVPT, many di e en ways o aining pe cep ion ha e p o en e ec i e, as shown
in he s udies e iewed by Sakai and Moo man (2018) and also he pe cep ion-o ien ed s udies
e iewed by Lee e al. (2015) and Sai o and Plonsky (2019), bu mos o he s udies epo ing
posi i e indings do no use he same me hodology. P onuncia ion eaching would bene i by
mo e eplica ions and compa a i e s udies o di e en me hodologies o es ablish he eliabili y
and gene alisabili y o he e ec s o a ious app oaches o aining pe cep ion in di e en
con ex s.
2.3 Insu icien a en ion o me hodological ac o s a ec ing esea ch ou comes
A p oblem wi h he ad ocacy o speci ic me hods is ha hose me hods may be implemen ed
in di e en ways, wi h di e en e ec s. Fo example, a sup asegmen al o ien a ion o
p onuncia ion can be as minimal as a ocus on syllables a he han phonemes o as maximal
as a ocus on discou se in ona ion (B azil, 1997; B azil e al., 1980). In be ween hese wo
ex emes, i migh ocus on wo d s ess, sen ence s ess, in ona ion, o ph ase-le el o sen ence-
le el linking be ween wo ds. I migh also be o ien ed o pe cep ion o p oduc ion, o bo h.
And he me hods o eaching sup asegmen als wi hin any o hese o ien a ions a y widely. I
would be o alue o design s udies o speci ically es he e ec s o di e en aspec s o
sup asegmen als, and especially o compa e he e ec i eness o mo e mic o-le el (e.g.,
syllable- o wo d-le el) s. mo e mac o-le el (e.g., sen ence- o discou se-le el) app oaches.
Penning on
C i ical examina ion, L2 p onuncia ion eaching esea ch
58
A ecen compu e -based s udy by Hi a a (2024), o ins ance, ound ha in aining pi ch and
du a ional con as s o English lea ne s o Japanese, sen ence-le el inpu was mo e e ec i e
han wo d-le el.
Ano he example o a ia ion in implemen a ion o p onuncia ion eaching me hodology is
he echnique o shadowing in which he lea ne aims o epea exac ly wha hey hea wi h as
li le delay as possible, almos simul aneously. Shadowing, which is based on s e ches o
unning speech, has been shown in some esea ch o imp o e in elligibili y and o he global
aspec s o p onuncia ion (Foo e & McDonough, 2017; Hamada, 2019; Mu phey, 2001). “Pu e”
shadowing is done using con inuous s e ches o speech ha he lea ne has no been exposed
o be o e (in ei he w i en o spoken e sions). The shadowing echnique can howe e be
pe o med in many di e en ways ha a ec he ask and i s ou comes (Hamada, 2019;
Mu phey, 2001), and mos s udies include mode a ing ac o s wi hou sys ema ic examina ion
o how hey a ec esul s. In any gi en case, shadowing may o may no include explici
ins uc ion, analysis o he s e ches o spoken language ha a e shadowed, and use o w i en
e sions o wha is hea d. I may also in ol e manipula ions such as a educed a e o speaking
and/o shadowing in sho s e ches wi h pauses in be ween each s e ch a he han
con inuously as speech is p oduced in longe s e ches. Shadowing may be done “li e,” ha is,
on speech as i is occu ing, o based on an audio eco ding, which is he common way in
eaching con ex s as i allows he shadowed ma e ial o be accessed mul iple imes, as well as
ia compu e deli e y. The echnique may be done h ough epea ing ei he ou loud o silen ly,
and wi h exac o comple e epe i ion o only selec i e epe i ion o some wo ds and ph ases
(Mu phey, 2001). Compa a i e esea ch on he di e en me hodological op ions o shadowing
can help o ease ou which ea u es, o combina ions o ea u es, a e mos e ec i e.
2.4 Lack o sys ema ic eplica ion and compa ison ac oss di e en condi ions
A sho coming ela ed o he insu icien a en ion o me hodological ac o s a ec ing esea ch
ou comes is he lack o sys ema ic eplica ion and compa ison ac oss di e en condi ions,
in ol ing s uden popula ions, eaching con ex s, p onuncia ion ea u es ocused on, and
ins uc ional o aining me hods. Nea ly all s udies ha e been ca ied ou on jus one class o
wi h one g oup o lea ne s (Lee e al., 2015; Sai o & Plonsky, 2019), and only in a ew a eas
o esea ch ‒ no ably, pe cep ion s udies (Sakai & Moo man, 2018) ‒ ha e la e s udies been
buil on, and speci ically compa ed o, he me hodology o p io s udies. HVPT is no ewo hy
in ha ing ca e ully buil a long line o ela ed s udies (Thomson, 2018) wi h some di e si y
ac oss he s udies in s uden popula ions, eaching con ex s, and p onuncia ion ea u es ocused
on ‒ hough he HVPT esea ch has ended o ocus on ce ain popula ions and eaching
con ex s and on a ew segmen al con as s, using essen ially he same app oach o he aining.
The esea ch on p onuncia ion eaching e ec i eness can be s eng hened by inc eased
a en ion o compa a i e esea ch and eplica ion in new con ex s.
2.5 Limi ed a en ion o au hen ic language and language a ie y
The limi ed a en ion paid o au hen ic language in eaching ma e ials and me hodologies is a
majo sho coming o p onuncia ion p ac ice and esea ch (Sai o & Plonsky, 2019). A long-
p ac ised adi ion in pedagogy o eaching and lea ning o occu in a s ep-by-s ep sequence
om simple and con olled o ee and mo e complex asks can be seen in he mains eam
app oach o p onuncia ion eaching, pa icula ly, in he o m summa ised in he
Communica i e F amewo k o Teaching P onuncia ion (Celce-Mu cia e al., 2010, pp. 44–
48), which is o en ci ed as he ins uc ional sequence ollowed in p onuncia ion eaching
Penning on
C i ical examina ion, L2 p onuncia ion eaching esea ch
65
eache s can ha e con idence ha wha hey each is suppo ed by hei own class oom
esea ch, as hey es ablish hei own linkages be ween esea ch and p ac ice.
Ac ion esea ch is a ype o applied esea ch ha has been ecommended o language
eache s and applied in second language con ex s (Bu ns, 1999, 2005; C ookes, 1993; Edge,
2001; Nunan, 1990; Wallace, 1998). I in ol es ial-e alua ion- e lec ion-adjus men cycles
o : 1) ying ou an in e en ion gea ed o a desi ed ou come o ou comes; 2) e alua ing he
e ec s and e ec i eness o he in e en ion; 3) e lec ing on he p ocesses and esul s o he
in e en ion, and hen 4) adjus ing ac ions o ei he con inue on wi h he o iginal in e en ion
o o y ou a di e en in e en ion aiming o achie e simila o be e esul s. Teache s can
pu sue ac ion esea ch as a collabo a i e ac i i y (Bu ns, 1999; Go don, 2008), wo king
oge he as esea ch pa ne s o g oups o sys ema ically in es iga e app oaches o eaching
p onuncia ion in hei classes in a compa a i e way (see Penning on, o hcoming), by:
 planning and p epa ing oge he ,
 deciding wha p onuncia ion ea u e o ea u es o ocus on and how o assess hem,
 aking no es and compa ing how hings a e going on a egula basis, i possible,
obse ing each o he ’s classes o iew he eaching-lea ning p ocess, and
 assessing he e ec s on s uden s’ pe cep ion and p oduc ion as well as on hei
mo i a ion and a i udes o he app oaches.
I encou age hose who conduc esea ch on p onuncia ion eaching o expand hei ho izons
o aid he ield in c ea ing a iche and mo e di e se eaching and esea ch agenda. The e is a
need especially o mo e compa a i e s udies—such as hose examining he e ec s o eaching
p onuncia ion a di e en s ages o language lea ning, in di e en eaching con ex s, o using
con as ing pedagogical app oaches o he same app oach bu wi h di e en cons ain s o
a o dances. P onuncia ion eaching wi h echnology also emains a as e i o y awai ing
u he explo a ion.
5. Towa ds he u u e o p onuncia ion eaching and esea ch
In he cu en e a, p onuncia ion eaching and esea ch is ansi ioning beyond i s adi ional
na ow and a i icially cons ained pedagogical ocus o inco po a e con ex ualised and
au hen ic language and o enhance pedagogy h ough use o echnology, media, and he
in e ne . In addi ion o es ablished esea ch adi ions, p onuncia ion eaching and esea ch a e
apidly de eloping new me hodologies inco po a ing linguis ic a ie y, discou se, and social
con ex in inno a i e ways o eaching p onuncia ion wi h posi i e esul s. Fu u e de elopmen
o p onuncia ion eaching and esea ch should a end o la ge con ex ea u es, including he
complex o segmen al and sup asegmen al e ec s in s e ches o speech and discou se o
di e en kinds, he kinaes he ic accompanimen s o speech, and he social dimensions o
communica ion ha shi eaching o p onuncia ion away om segmen al e o s and accen ,
and owa ds in elligibili y, speaking s yle, and deli e y. Resea che s can assis in ex ending he
esea ch agenda by eplica ing and adap ing he new me hodologies in di e en con ex s and
o di e en s uden popula ions, as class oom eache s d aw on he published esea ch o y
ou new app oaches in hei class oom h ough ac ion esea ch. In his way, bo h esea che s
and eache s will be pa icipa ing in building a u u e o p onuncia ion pedagogy ha
inco po a es he ichness o language and o he con ex s in which i occu s.

Penning on
C i ical examina ion, L2 p onuncia ion eaching esea ch
66
Re e ences
Basse i, B. (2024). O hog aphic e ec s in he phone ics and phonology o second language lea ne s
and use s. In M. Amengual (Ed.), The Camb idge handbook o bilingual phone ics and phonology
(pp. 699–720). Camb idge Uni e si y P ess.
Block, D. (2003). The social u n in second language acquisi ion. Geo ge own Uni e si y P ess.
B azil, D. (1997). The communica i e alue o in ona ion in English. Camb idge Uni e si y P ess.
B azil, D., Coul ha d, M., & Johns, C. (1980). Discou se in ona ion and language eaching. Longman.
Bu ns, A. (1999). Collabo a i e ac ion esea ch. Camb idge Uni e si y P ess.
Bu ns, A. (2005). Ac ion esea ch. In E. Hinkel (Ed.), Handbook o esea ch in second language
eaching and lea ning (pp. 241–256). E lbaum.
Celce-Mu cia, M., B in on, D. M., Goodwin, J. M., & G ine , B. (2910). Teaching p onuncia ion: A
cou se book and e e ence guide. Camb idge Uni e si y P ess.
C ookes, G. (1993). Ac ion esea ch o second language eache s: Going beyond eache esea ch.
Applied Linguis ics, 14(2), 130–144. h ps://doi.o g/10.1093/applin/14.2.130
De wing, T. M., & Mun o, M. J. (2015). P onuncia ion undamen als: E idence-based pe spec i es o
L2 eaching and esea ch. John Benjamins.
Ding, S., Libe a o e, C., Sonsaa , S., Lučić, I., Silpachai, A., Zhao, G., Chukha e -Hudilainen, E., Le is,
J., & Gu ie ez-Osuna, R. (2019). Golden speake builde – An in e ac i e ool o p onuncia ion
aining. Speech Communica ion, 115, 51–66. h ps://psi.eng . amu.edu/wp-
con en /uploads/2019/11/1-s2.0-S0167639319302675-main.pd
Ecke , P., & Rick o d, J. R. (Eds.). (2001). S yle and sociolinguis ic a ia ion. Camb idge Uni e si y
P ess.
Edge, J. (2001). Ac ion esea ch. TESOL.
Edwa ds, J. (2009). Language and iden i y: An in oduc ion. Camb idge Uni e si y P ess.
E e s, K., & Chen, S. (2022). E ec s o an au oma ic speech ecogni ion sys em wi h pee eedback on
p onuncia ion ins uc ion o adul s. Compu e Assis ed Language Lea ning, 35(8), 1869–
1889. h ps://doi.o g/10.1080/09588221.2020.1839504
Foo e, J. A., & McDonough, K. (2017). Using shadowing wi h mobile echnology o imp o e L2
p onuncia ion. Jou nal o Second Language P onuncia ion, 3(1), 34–56.
h ps://doi.o g/10.1075/jslp.3.1.02 oo
Fouz-González, J. (2017). P onuncia ion ins uc ion h ough Twi e : The case o commonly
misp onounced wo ds. Compu e Assis ed Language Lea ning, 30(7), 631–663.
h ps://doi.o g/10.1080/09588221.2017.1340309
Fouz-González, J. (2019). Podcas -based p onuncia ion aining: Enhancing FL lea ne s’ pe cep ion
and p oduc ion o ossilised segmen al ea u es. ReCALL, 31(2), 150–
169. h ps://doi.o g/10.1017/S0958344018000174
Fouz-Gonzáles, J. (2025). Teaching and lea ning p onuncia ion wi h echnology: Cu en possibili ies,
pedagogical ecommenda ions, and di ec ions o he u u e. In J. M. Le is, M. Du is, S. Sonsaa -
Hegelheime , & I. Na (Eds.), P oceedings o he 15 h P onuncia ion in Second Language Lea ning
and Teaching Con e ence (pp. 1–15). Iowa S a e Uni e si y. h ps://doi.o g/10.31274/psll .19027
Fouz-Gonzáles, J. ( o hcoming). Technology-assis ed p onuncia ion aining: B idging esea ch and
pedagogy. Uni e si y o To on o P ess.
Galimbe i, V., Mo a, J. C., & Gilabe , R. (2023). Audio-synch onized ex ual enhancemen in o eign
language p onuncia ion lea ning om ideos. Sys em, 116, 103078.
h ps://doi.o g/10.1016/j.sys em.2023.103078
Giles, H., & Robinson, W. P. (Eds.). (1990). Handbook o language and social psychology. John Wiley
& Sons.
Penning on
C i ical examina ion, L2 p onuncia ion eaching esea ch
67
Go don, J. (2021). P onuncia ion and ask-based ins uc ion: E ec s o a class oom in e en ion. RELC
Jou nal, 52(1), 94–109. h ps://doi.o g/10.1177/0033688220986919
Go don, J., & Da cy, I. (2022). Teaching segmen als and sup asegmen als: E ec s o explici
p onuncia ion ins uc ion on comp ehensibili y, luency, and accen edness. Jou nal o Second
Language P onuncia ion, 8(2), 168–195. h ps://doi.o g/10.1075/jslp.21042.go
Go don, S. P. (2008). Collabo a i e ac ion esea ch: De eloping p o essional lea ning communi ies.
Teache s College P ess.
Hamada, Y. (2019). Shadowing: Wha is i ? How o use i ? Whe e will i go? RELC Jou nal, 50(3),
386–393. h ps://doi.o g/10.1177/0033688218771380
Ha dison, D. M. (2004). Gene aliza ion o compu e assis ed p osody aining: Quan i a i e and
quali a i e indings. Language Lea ning & Technology, 8(1), 34–52. h p://dx.doi.o g/10125/25228
Ha dison, D. M. (2018). Visualizing he acous ic and ges u al bea s o emphasis in mul imodal
discou se: Theo e ical and pedagogical implica ions. Jou nal o Second Language P onuncia ion,
4(2), 231–258.
Ha dison, D. M. (in p ess). The mul imodal con ex o phonological lea ning. Uni e si y o To on o
P ess.
Ha dison, D. M., & Penning on M. C. (2021). Mul imodal second-language communica ion: Resea ch
indings and pedagogical ecommenda ions. RELC Jou nal, 52(1), 62–76.
h ps://doi.o g/10.1075/jslp.17006.ha
Hende son, A., & Rojczyk, A. (2023). Fo eign language accen imi a ion: Ma ching p oduc ion wi h
pe cep ion. In V. G. Sa degna & A. Ja osz (Eds.), English p onuncia ion eaching: Theo y, p ac ice
and esea ch indings (pp. 115–133). Mul ilingual Ma e s.
Hi a a, Y. (2024). Compu e assis ed p onuncia ion aining o na i e English speake s lea ning
Japanese pi ch and du a ional con as s. Compu e Assis ed Language Lea ning, 17(3-4), 357–376.
h ps://doi.o g/10.1080/0958822042000319629
Jenkins, J. (2000). The phonology o English as an in e na ional language. Ox o d Uni e si y P ess.
Jenkins, J. (2002). A sociolinguis ically based, empi ically esea ched p onuncia ion syllabus o
English as an in e na ional language. Applied Linguis ics, 23, 83–103.
h ps://doi.o g/10.1093/applin/23.1.83
Lan ol , J. P. (Ed.). (2000). Sociocul u al heo y and second language lea ning. Ox o d Uni e si y
P ess.
Lan ol , J. P. (2011). The sociocul u al app oach o second language acquisi ion: Sociocul u al heo y,
second language acquisi ion, and a i icial L2 de elopmen . In D. A kinson (Ed.), Al e na i e
app oaches o second language acquisi ion (pp. 24–47). Rou ledge.
LaSco e, D. K., & Ta one, E. (2022). Channeling oices o imp o e L2 English in elligibili y. Mode n
Language Jou nal, 106(4), 744 – 763. h ps://doi.o g/10.1111/modl.12812
LaSco e, D., Meye s, C, & Ta one, E. (2021). Voice and mi o ing in SLA: Top-down pedagogy o
L2 p onuncia ion ins uc ion. RELC Jou nal, 52(1), 3–21.
h ps://doi.o g/10.1177/0033688220953910
LaSco e, D., Meye s, C, & Ta one, E. (2023). Voice and mi o ing in L2 p onuncia ion ins uc ion.
Equinox / Uni e si y o To on o P ess.
Lee, B., Jang, J., & Plonsky, L. (2015). The e ec i eness o second language p onuncia ion ins uc ion:
A me a-analysis. Applied Linguis ics, 36(3), 345–366. h ps://doi.o g/10.1093/applin/amu040
Lee, B., Plonsky, L., & Sai o, K. (2020). The e ec s o pe cep ion- s. p oduc ion-based p onuncia ion
ins uc ion. Sys em, 88, 1–13. h ps://doi.o g/10.1016/j.sys em.2019.102185
Le is, J. M. (2018). In elligibili y, o al communica ion, and he eaching o p onuncia ion. Camb idge
Uni e si y P ess.
Penning on
C i ical examina ion, L2 p onuncia ion eaching esea ch
68
Le is, J. M. (2019). Cinde ella no mo e! Lea ing ic imhood behind. Speak Ou ! IATEFL
P onuncia ion SIG Jou nal, 60, 1–7.
Le is, J. M. (2021a). Connec ing he do s be ween p onuncia ion esea ch and p ac ice. In A. Ki ko a-
Nasko a, A. Hende son, & J. Fouz-Gonzáles (Eds.), P onuncia ion ins uc ion: Resea ch-based
insigh s (pp. 17–37). John Benjamins.
Le is, J. M. (2021b). Edi o ial. L2 p onuncia ion esea ch and eaching: The impo ance o many
languages. Jou nal o Second Language P onuncia ion, 7(2), 141–153.
h ps://doi.o g/10.1075/jslp.21037.le
Le is, J. M. (2022). Teaching p onuncia ion: T u hs and lies. In C. Ba del, C. Hedman, K. Rejman, &
E. Ze e holm (Eds.), Explo ing language educa ion: Global and local pe spec i es (pp. 39–72).
S ockholm Uni e si y P ess.
Le is, J. M. & Moye , A. (2014). Social dynamics in second language accen . De G uy e .
Lippi-G een, R. (2012). English wi h an accen : Language, ideology, and disc imina ion in he Uni ed
S a es (3 d ed.). Rou ledge.
Long, M. (1991). Focus on o m: A design ea u e in language eaching me hodology. In K. de Bo , R.
Ginsbe g, & C. K amsch (Eds.). Fo eign language esea ch in c oss-cul u al pe spec i e (pp. 39–
52). John Benjamins.
Long, M. (2015). Second language acquisi ion and ask-based language eaching. Wiley-Blackwell.
McC ocklin, S. (2019). ASR-based dic a ion p ac ice o second language p onuncia ion imp o emen .
Jou nal o Second Language P onuncia ion, 5(1), 98–118. h ps://doi.o g/10.1075/jslp.16034.mcc
Mompean, J. A. & Fouz-Gonzáles, J. (2021). Phone ic symbols in con empo a y p onuncia ion
ins uc ion. RELC Jou nal, 52(1), 155–168. h ps://doi.o g/10.1177/0033688220943431
Mo a, J. C., & Le kina, M. (2018). T aining owel pe cep ion h ough map asks: The ole o linguis ic
and cogni i e complexi y. In J. Le is (Ed.), P onuncia ion in Second Language Lea ning and
Teaching P oceedings (pp. 151-162). Iowa S a e Uni e si y.
h ps://www.ias a edigi alp ess.com/psll /a icle/id/15350/
Mu phey, T. (2001). Explo ing con e sa ional shadowing. Language Teaching Resea ch, 5(2), 128–
155.
Namaziandos , E., Es ahani, F. R., & Hashemi a nia, A. (2018). The impac o using au hen ic ideos
on p osodic abili y among o eign language lea ne s. In e na ional Jou nal o Ins uc ion, 11(4),
375–390. h ps://doi.o g/10.12973/iji.2018.11424a
Nunan, D. (1990). Ac ion esea ch in he language class oom. In J. C. Richa ds & D. Nunan (Eds.),
Second language eache educa ion (pp. 62–81). Camb idge Uni e si y P ess.
Penning on, M. C. (1989). Applica ions o compu e s in he de elopmen o speaking and lis ening
p o iciency. In M. C. Penning on (Ed.), Teaching languages wi h compu e s: The s a e o he a .
A hels an.
Penning on, M. C. (1996). When inpu becomes in ake: T acing he sou ces o eache s’ a i ude change.
In D. F eeman & J. C. Richa ds, Teache lea ning in language eaching (pp. 320–348). Camb idge
Uni e si y P ess.
Penning on, M. C. (2020). P onuncia ion and in e na ional employabili y. In Bocaneg a-Valle (Ed.),
Applied linguis ics and knowledge ans e : Employabili y, in e na ionalisa ion and social
challenges (pp. 225–244). Pe e Lang.
Penning on, M. C. (2021). Teaching p onuncia ion: The s a e o he a 2021. RELC Jou nal, 52(1), 3–
21. h ps://doi.o g/10.1177/00336882211002283
Penning on, M. C. ( o hcoming). The p onuncia ion book: A eache ’s guide. Uni e si y o To on o
P ess.
Penning on, M. C., & Cheung, M. (1995). Fac o s shaping he in oduc ion o p ocess w i ing in Hong
Kong. Language, Cul u e, and Cu iculum, 8, 15–34. h ps://doi.o g/10.1080/07908319509525185
Penning on
C i ical examina ion, L2 p onuncia ion eaching esea ch
69
Penning on, M. C., & Ellis, N. C. (2000). Can onese speake s’ memo y o English sen ences wi h
p osodic cues. Mode n Language Jou nal, 84(3), 372–389. h ps://doi.o g/10.1111/0026-
7902.00075
Penning on, M. C., Lau, L., & Sachde , I. (2011). Di e si y in adop ion o linguis ic ea u es o London
English by Chinese and Bangladeshi adolescen s. Language Lea ning Jou nal, 39(2), 177–199.
h ps://doi.o g/10.1080/09571736.2011.573686
Penning on, M. C., & Richa ds, J. C. (1997). Re-o ien ing he eaching uni e se: The expe ience o i e
i s -yea English eache s in Hong Kong. Language Teaching Resea ch, 1(2), 149–178.
h ps://doi.o g/10.1177/136216889700100204
Penning on, M. C., & Roge son-Re ell, P. (2019). English p onuncia ion eaching and esea ch:
Con empo a y pe spec i es. Palg a e Macmillan.
Robinson, P. (2011). Task-based language eaching: A e iew o issues. Language Lea ning, 61(1), 1–
36. h ps://doi.o g/10.1111/j.1467-9922.2011.00641.x
Robinson, W. P., & Giles, H. (Eds.). (2001). The new handbook o language and social
psychology. John Wiley & Sons.
Roge son-Re ell, P. M. (2021). Compu e -assis ed p onuncia ion aining (CAPT): Cu en issues and
u u e di ec ions. RELC Jou nal, 52(1), 189–205. h ps://doi.o g/10.1177/0033688220977406
Sai o, K., & Plonsky, L. (2019). E ec s o second language p onuncia ion eaching e isi ed: A
p oposed amewo k and me a-analysis. Language Lea ning, 69(3), 652–708.
h ps://doi.o g/10.1111/lang.12345
Sakai, M., & Moo man, C. (2018) Can pe cep ion aining imp o e he p oduc ion o second language
phonemes? A me a-analy ic e iew o 25 yea s o pe cep ion aining esea ch. Applied
Psycholinguis ics, 39(1), 187–224. h ps://doi.o g/10.1017/S0142716417000418
Sánchez-Auñón, E., Fé ez-Mo a, P. A. & Mon oy-He nández, F. (2023). The use o ilms in he
eaching o English as a o eign language: A sys ema ic li e a u e e iew. Asian Jou nal o Second
and Fo eign Language Educa ion 8(10), 1–17. h ps://doi.o g/10.1186/s40862-022-00183-0
Sa degna, V. G., & Ja osz, A. (2023). Lea ning English wo d s ess wi h echnology. In R. I. Thomson,
T. M. De wing, J. M. Le is, & K. Hiebe (Eds.), P oceedings o he 13 h P onuncia ion in Second
Language Lea ning and Teaching Con e ence (pp. 1-10). Iowa S a e Uni e si y.
h ps://doi.o g/10.31274/psll .16142
Spe i, S. (2017). Phonop agma ic dimensions o ELF in specialized immig a ion con ex s. Wo king
Pape s del Cen o di Rece ca sulle Lingue F anchenella Comunicazione In e cul u ale e
Mul imediale, Dipa imen o di S udi Umanis ici, Uni e si à del Salen o, 3.
h ps://doi.o g/10.1285/i24991449n3
Tho nbu y, S. (1996). Paying lip-se ice o CLT. EA [ELICOS Associa ion] Jou nal, 14(1), 51–63.
h p://www.sco ho nbu y.com/a icles.h ml
Thomson, R. I. (2018). High a iabili y [p onuncia ion] aining (HVPT): A p o en echnique abou
which e e y language eache and lea ne ough o know. Jou nal o Second Language
P onuncia ion, 4(2), 208–231. h ps://doi.o g/10.1075/jslp.17038. ho
Thomson, R. I. & De wing, T. (2015). The e ec i eness o L2 p onuncia ion ins uc ion: A na a i e
e iew. Applied Linguis ics, 36(3), 326–344. h ps://doi.o g/10.1093/applin/amu076
Wallace, M. (1998). Ac ion esea ch o language eache s. Camb idge Uni e si y P ess.
Wichmann, A., Dehé, N., & Ba h-Weinga en, D. (2009). Whe e p osody mee s p agma ics: Resea ch
a he in e ace. In A. Wichmann, N. Dehé, & D. Ba h-Weinga en (Eds.), Whe e p osody mee s
p agma ics, S udies in P agma ics 8 (pp. 1–20). Eme ald.
Yu, L. & Odlin, T. (2016). New pe spec i es o ans e in second language lea ning. Mul lingual
Ma e s.
Penning on
C i ical examina ion, L2 p onuncia ion eaching esea ch
70
Wisniewska, N., & Mo a, J. (2020). Can cap ioned ideo bene i second language p onuncia ion?
S udies in Second Language Acquisi ion, 42(3), 599–624.
h ps://doi.o g/10.1017/S0272263120000029
Abou he au ho
Ma ha C. Penning on (PhD Linguis ics, Uni e si y o Pennsyl ania) is a Resea ch Fellow a
Bi kbeck Uni e si y o London in he School o C ea i e A s, Cul u e and Communica ion
and is cu en ly a Visi ing P o esso a he Uni e si y o Reading in he Depa men o English
Language and Applied Linguis ics. He books include Phonology in English Language
Teaching: An In e na ional App oach (Longman, 1996), The Powe o CALL (A hels an,
1996), Phonology in Con ex (Palg a e Macmillan, 2007), English P onuncia ion Teaching
and Resea ch: Con empo a y Pe spec i es (Palg a e Macmillan, 2019, co-au ho ed wi h
Pamela Roge son-Re ell), and The P onuncia ion Book: A Teache ’s Guide, o appea in he
se ies Applied Phonology and P onuncia ion Teaching, which has ecen ly been acqui ed om
Equinox by he Uni e si y o To on o P ess. She has also published widely in li e acy,
including Why Reading Books S ill Ma e s: The Powe o Li e a u e in Digi al Times
(Rou ledge, 2018, co-au ho ed wi h Robe P. Waxle ), and in language educa ion, including
Language P og am Leade ship in a Changing Wo ld: An Ecological Model (De G uy e B ill,
2010, co-au ho ed wi h Ba ba a J. Hoekje), olume 1 in he se ies Inno a ion and Leade ship
in English Language Teaching.
Email: m.[email p o ec ed]

This chap e is based on he o al p esen a ion gi en by he au ho a he 8 h
In e na ional Con e ence English P onuncia ion: Issues and P ac ices (EPIP
8) held May 8–10, 2024 a he Uni e si y o Can ab ia in San ande , Spain.
I is licensed unde he C ea i e Commons A ibu ion 4.0 In e na ional
License. To iew a copy o he license, please go o:
h p://c ea i ecommons.o g/licenses/by/4.0/
.
Pie aszek, M. (2025). Li e o lea e! Vowel quali y and owel leng h in he in elligibili y o ad anced Spanish-
accen ed English. In A. Ki ko a-Nasko a, P. Humánez-Be al, & A. Hende son (Eds.), P oceedings o he 8 h
In e na ional Con e ence on English P onuncia ion: Issues and P ac ices (pp. 71–83). Uni e si é G enoble-
Alpes. h ps://doi.o g/10.5281/zenodo.16696002/
Li e o lea e! Vowel quali y and owel leng h in he in elligibili y o
ad anced Spanish-accen ed English
Ma eusz Pie aszek
Complu ense Uni e si y o Mad id
Abs ac
O en cha ac e ised as sho and long in bo h esea ch and eaching p ac ice (C u enden, 2008;
Hancock, 2012, Walke , 2010), he KIT and FLEECE owels a e commonly con used by Spanish
lea ne s o English as hei mo he ongue lacks a simila dis inc ion (RAE, 2011). Leng h is
also adi ionally insis ed upon in ins uc ion and e en mo e ecen in elligibili y-based (Le is,
2005) app oaches o in e na ional English (Jenkins, 2000; Walke , 2010) ad oca e i s
impo ance p io i ising i o e owel quali y. The goal o his pape is o examine he ole o
owel leng h and quali y in Spanish-accen ed English om a quan i a i e and quali a i e
explo a o y pe spec i e. Six y ad anced English speake s wi h L1 Spanish we e eco ded and
he acous ic p ope ies o hei FLEECE and KIT owels we e analysed in he wo ds li e and
lea e. The o e all in elligibili y (Kang e al., 2018; Wang, 2007), comp ehensibili y, and
o eign-accen edness (De wing & Mun o, 1997, 2015) o he pa icipan s’ speech we e
assessed online by an in e na ional sample o 330 lis ene s om 28 L1 backg ounds. No
associa ion was de ec ed be ween he o e all p onuncia ion sco es and owel leng h. Howe e ,
some e idence sugges ed ha dis inguishing a leas one o man migh be associa ed wi h
inc eased comp ehensibili y and dec eased accen edness. Addi ionally, he analysis o
phonological in elligibili y b eakdowns showed ha hese could only be explained wi h a
e e ence o o man alues bu no leng h dis inc ions. Thus, i is a gued ha owel leng h in
KIT and FLEECE may be dis ega ded and he ocus in ecep i e and p oduc i e ins uc ion should
be on quali y dis inc ions.
Keywo ds: accen edness, in elligibili y, owel leng h, owel quali y
Pie aszek
Vowel quali y & leng h, Spanish-accen ed English
72
1. In oduc ion
The English FLEECE /iː/ and KIT /ɪ/ owels a e mos o en cha ac e ised as long and sho
segmen s in bo h dic iona ies and lea ning ma e ials o L2 English s uden s (C u enden, 2008;
Hancock, 2012; Wells, 2008), especially – bu no exclusi ely – hose eaching B i ish English.
In educa ional con ex s, his migh lead o he conclusion ha con as ing hese owels mainly
based on leng h is su icien in o de o ease hem apa , as isola ed d ills e y o en ocus on
leng h as he mos in eg al ea u e o be ained when con as ing hose owels (see Mompeán-
González, 2001). Howe e , his simplis ic assump ion poses se e al p oblems agg a a ed by
he e y ac ha owel leng h is a ela i e phenomenon. Fi s ly, he FLEECE and KIT owels
ha e di e en a icula ions ega dless o hei leng h, he la e being bo h mo e open and back
han he o me .1 Secondly, analyses o S anda d Sou he n B i ish English (Lindsey, 2019;
Wells, 1982) and Gene al Ame ican (Celce Mu cia e al., 2010) end o desc ibe he FLEECE
owel as a diph hong [ɪj], which glides om a mo e open and cen alised a icula ion owa ds
a highe and on e poin o a icula ion. Thi dly, owel leng h in English is con ex -dependen
(Lade oged, 2001; Odden, 2011) and i is commonly used as a cue ega ding he oicing o he
ollowing consonan (Rojczyk, 2010) o e en he p eceding consonan (Viswana han e al.,
2020) by na i e speake s o English. As a esul o p e- o is clipping (Roach, 2009), long
owels may e en be p onounced sho e han sho owels, e.g., when compa ing a long owel
be o e a oiceless s op wi h a sho owel be o e a long one. Leng h also a ies depending on
he numbe o syllables in a wo d (Lade oged, 2001), hence he i s syllable o eading will
be sho e han he wo d ead in he same con ex . Finally, pe cep ual s udies sugges leng h is
insu icien when dis inguishing owel quali ies among na i e speake s (Hillenb and, 1995;
Liu e al, 2014).
I is he e o e e iden ha owels a e ne e dis inguished by leng h alone in English, and
he desc ip ions o some a ie ies o English – including he mos widely spoken, Gene al
Ame ican – make no e e ence o leng h as a ele an owel ea u e (see A e y & Eh lich,
1992). Howe e , e en ela i ely ecen eaching ecommenda ions – in all likelihood
encou aged by canonical phonemic ansc ip ion in dic iona ies (Wells, 2008) – insis on
p ese ing leng h con as s. No only do hey posi leng h as a ea u e necessa y o
in e na ionally in elligible English, bu hey may e en p io i ise i o e quali y dis inc ions.
This is he case, o example, o he Lingua F anca Co e o ea u es necessa y o in e na ional
communica ion in English as a global language (Jenkins, 2000, 2002; Walke , 2010). Howe e ,
many languages do no ha e phonological leng h dis inc ions whe eas all languages, including
English and Spanish, dis inguish be ween owel quali ies, which migh be su icien o basic
owel dis inc ions in English as an in e na ional language. In he pa icula case o Spanish,
al hough owel leng h di e ences exis in he language (Ma ín Gál ez, 1994), hei
phonological unc ions a e no compa able o hose in English (Fox e al., 1995). This ea u e,
coupled wi h he ac ha Spanish only has one high on owel /i/, indeed poses p onuncia ion
p oblems o L1-Spanish speake s o English (Finch & O iz Li a, 1982; Gómez-González &
Sánchez-Rou a 2016; Hancock, 2012; Walke , 2010). P e ious s udies ha e shown ha , unlike
English na i e speake s, Spanish speake s a e la gely insensi i e o leng h di e ences (Fox e
al., 1995; Viswana han e al., 2020) o ha hei sensi i i y o du a ion is lowe han ha o
na i e speake s (Kondau o a & F ancis, 2008). O he e idence sugges s ha ins uc ion
p ac ices may lead Spanish speake s o ely on leng h e en mo e han na i e speake s do in
disc imina ion asks (Mompeán-González, 2001). Finally, ega ding in e na ional
1 See Du and (2005) o a comp ehensi e summa y o he deba e on he ele ance o owel leng h and
quali y in English phonology.
Pie aszek
Vowel quali y & leng h, Spanish-accen ed English
73
in elligibili y, he e idence in suppo o leng h dis inc ions o p ese e in elligibili y has so a
been sca ce and inconclusi e (Ju ado-B a o, 2018).
1.1 Aims and esea ch ques ions
The main goal o his explo a o y s udy is o examine he u ili y o insis ing on leng h e sus
quali y dis inc ions in eaching he FLEECE : KIT con as in Peninsula Spanish, applying he
in elligibili y p inciple (Le is, 2005) in English as an in e na ional lingua anca (Jenkins,
2000). Wi h his aim in mind, he ollowing esea ch ques ions (RQ) will be add essed:
RQ1: Do Spanish speake s o English in he sample p oduce he FLEECE and KIT
owels wi h di e en leng h and/o quali y?
RQ2: Is a con as be ween FLEECE and KIT in L2 English associa ed wi h imp o ed
gene al p onuncia ion pe o mance sco es?
RQ3: Wha phonological in elligibili y b eakdowns may occu when FLEECE and KIT
owels a e misp onounced?
2. Me hodology
2.1 Pa icipan s, eco dings, and sco ing
This mixed-me hods (C eswell e al., 2008) explo a o y s udy is pa o a b oade esea ch
p ojec (Pie aszek, 2024a, Pie aszek, 2024b). The sample consis ed o 60 uni e si y s uden s
(26 emales and 34 males; aged 19–26, M = 21.2) majo ing in disciplines un ela ed o
linguis ics and ha ing B2–C2 le els o English p o iciency. The speake s we e eco ded using
a se o seman ically unp edic able sen ences (SUS) (Benoî e al., 1996; Pie aszek, 2024a)
and an elici a ion pa ag aph (Pie aszek, 2024b). SUS we e selec ed as he bes ins umen o
es phonemic in elligibili y, because mos seman ic in o ma ion is elimina ed (Benoî e al.,
1996; Kang e al., 2018). Subsequen ly, 330 in e na ional lis ene s wi h 28 di e en L1s
e alua ed he speake s’ comp ehensibili y and o eign-accen edness, and ansc ibed hei SUS
o in elligibili y measu emen s. The comp ehensibili y (COM) and o eign-accen edness (FA)
(De wing & Mun o, 1997, 2015) sco es a e a e aged a ings he speake s ecei ed om he
lis ene s on a 1–6 seman ic di e en ial scale using exce p s om he pa ag aph (Pie aszek,
2024b). In elligibili y (INT) was ope a ionalised as phonological u e ance decoding a he
segmen al le el (c . Jenkins, 2000; Kang e al., 2018). Thus, he INT sco e was calcula ed as a
mean sco e based on he numbe o wo ds co ec ly in e p e ed by he lis ene s in he es
ollowing a specially designed p o ocol (Pie aszek, 2024a). Simila ly, he quali a i e pa o
he s udy, i.e., in elligibili y b eakdown analysis, is based on lis ene ansc ip ions o SUS in
he online es (Pie aszek, 2024a).
2.2 Acous ic analyses
The acous ic analyses a e based on wo wo ds ex ac ed om wo syn ac ically compa able
SUS: li e o he KIT owel and lea e o he FLEECE owel. The owel leng h o he LIVE and
LEAVE owels was measu ed o each speake using P aa (Boe sma & Weenink, 2019). As
“ he leng hs o segmen s depend on hei posi ion in he wo d, hei posi ion in he ph ase and
he whole u e ance, whe e he s esses occu in he u e ance, and many o he ac o s”
(Lade oged, 2003, p. 102), in o de o he measu emen s o be compa able, he ins ances we e
aken om SUS numbe 3 and 4 in se ies 3 o he minimal pai lea e–li e, i.e., Lea e he spo
and he hough and Li e he spo and he und. The ac ha he owels came om wo
Pie aszek
Vowel quali y & leng h, Spanish-accen ed English
74
consecu i e sen ences ensu ed a ela i ely s able pace o speech pe speake . Addi ionally, he
syn ac ic posi ion o bo h wo ds was he same, in o de o minimise con ex ual, phono ac ic
in luence on owel leng h. The exac p ocedu e and he selec ion o s a ing and ending poin s
in leng h measu emen s was pa icula ly impo an owing o he phonological con ex o he
a ge ed owels. These we e si ua ed be ween a la e al app oximan and a ica i e, whose
bounda ies a e mo e luid han hose o plosi es, o example, as bo h la e als and ica i es
a e close o owels on he sono i y scale (see C u enden, 2008). The bounda ies we e se
using a spec og am (see Figu e 1) showing he in ensi y o owel o man s. The leng h was
ex ac ed using he accompanying wa e o m (see Figu e 2) by selec ing he i s alley and he
Figu e 1
Spec og am o /liː /
Figu e 2
Oscillog am o /liː /
Pie aszek
Vowel quali y & leng h, Spanish-accen ed English
81
While his explo a o y esea ch canno eliably answe he ques ion o whe he leng h plays
a ole in in e na ional English in elligibili y, i p o ides some e idence ha changing he
quali y o a owel may lead o misunde s andings. Thus, dismissing owel quali y as a co e
in e na ional English ea u e while insis ing on leng h seems un ounded and he e o e isky. I
also places an addi ional bu den on he lea ne s who a e asked o de elop a phonological owel
ca ego y o eign o hei linguis ic sys em, while hei a ge language uses i mainly as a ma ke
o inal consonan oicing. Thus, i seems easonable, as a pedagogical implica ion o he
esul s p esen ed in his s udy, o aise lea ne s’ awa eness o he a ying du a ion due o p e-
o is clipping ega dless o he exac owel quali y, a he han o ea du a ion as he mos
ele an dis inc i e ea u e o he FLEECE and KIT phonemes.
Re e ences
A e y, P., & Eh lich, S. (1992). Teaching Ame ican English p onuncia ion. Ox o d Uni e si y P ess.
Benoî , C., G ice, M., & Hazan, V. (1996). The SUS es : A me hod o he assessmen o ex - o-speech
syn hesis in elligibili y using Seman ically Unp edic able Sen ences. Speech Communica ion, 18(4),
381–392. h p://dx.doi.o g/10.1016/0167-6393(96)00026-X
Boe sma, P., & Weenink, D. (2025). P aa : Doing phone ics by compu e [Compu e p og am]. Ve sion
6.4.36. h ps://p aa .o g
Celce-Mu cia, M., B in on, D., Goodwin, J. M. & G ine , B. (2010). Teaching p onuncia ion: A cou se
book and e e ence guide (2nd ed.). Camb idge Uni e si y P ess.
Cohen, J. (1988). S a is ical powe analysis o he beha io al Sciences. Psychology P ess.
C eswell, J. W., Plano Cla k, V. L., Gu mann, M. L., & Hanson, W. E. (2008). An expanded ypology
o classi ying mixed me hods esea ch in o designs. In V. L. Plano Cla k, & J. W. C eswell
(Eds.), The Mixed Me hods Reade (pp. 161–196). SAGE Publica ions Inc.
C u enden, A. (2008). Gimson's p onuncia ion o English. Hodde Educa ion.
De wing, T. M., & Mun o, M. J. (1997). Accen , in elligibili y, and comp ehensibili y: E idence om
ou L1s. S udies in Second Language Acquisi ion, 19(1), 1–16.
h ps://doi.o g/10.1017/S0272263197001010
De wing, T.M., & Mun o, M.J. (2015). P onuncia ion undamen als: E idence-based pe spec i es o
L2 eaching and esea ch. John Benjamins.
Du and, J. (2005). Tense/Lax, he owel sys em o English and phonological heo y. In P. Ca , J.
Du and, & C. J. Ewen (Eds.), Headhood, elemen s, speci ica ion and con as i i y: Phonological
pape s in honou o John Ande son (pp. 77–97). John Benjamins.
h ps://doi.o g/10.1075/cil .259.08du
Finch, D. F., & O iz Li a, H. (1982). A cou se in English phone ics o Spanish speake s. Heinemann
Educa ional Books.
Fox, R. A., Flege, J. E., & Mun o, M. J. (1995). The pe cep ion o English and Spanish owels by na i e
English and Spanish lis ene s: A mul idimensional scaling analysis. The Jou nal o he Acous ical
Socie y o Ame ica, 97(4), 2540–2551. h ps://doi.o g/10.1121/1.411974
Gómez González, M. d. l. A., & Sánchez Rou a, T. (2016). English p onuncia ion o speake s o
Spanish. De G uy e .
Hancock, M. (2012). English p onuncia ion in use. Camb idge Uni e si y P ess.
Hillenb and, J., Ge y, L. A., Cla k, M. J., & Wheele , K. (1995). Acous ic cha ac e is ics o Ame ican
English owels. The Jou nal o he Acous ical Socie y o Ame ica, 97(5), 3099–3111.
h ps://doi.o g/10.1121/1.411872
Hisagi, M., Higby, E., Zandona, M., Acos a, A. P., Ken , J., & Tajima, K. (2024). Impac o speech a e
on pe cep ion o owel and consonan du a ion by bilinguals and monolinguals. JASA Exp ess
Le e s, 4(5), 055201. h ps://doi.o g/10.1121/10.0025862
Jenkins, J. (2000). The phonology o English as an in e na ional language. Ox o d Uni e si y P ess.
Jenkins, J. (2002). A sociolinguis ically based, empi ically esea ched p onuncia ion syllabus o
English as an in e na ional language. Applied Linguis ics, 23. 83–103.
h p://dx.doi.o g/10.1093/applin/23.1.83

Pie aszek
Vowel quali y & leng h, Spanish-accen ed English
82
Ju ado-B a o, M. Á. (2018). Vowel quali y and owel leng h in English as a lingua anca in Spain.
Miscelánea, 57(57), 13–34. h ps://doi.o g/10.26754/ojs_misc/mj.20186310
Kang, O., Thomson, R. I., & Mo an, M. (2018). Empi ical app oaches o measu ing he in elligibili y
o di e en a ie ies o English in p edic ing lis ene comp ehension. Language Lea ning, 68(1),
115–146. h ps://doi.o g/10.1111/lang.12270
Kondau o a, M. V., & F ancis, A. L. (2008). The ela ionship be ween na i e allophonic expe ience
wi h owel du a ion and pe cep ion o he English ense/lax owel con as by Spanish and Russian
lis ene s. The Jou nal o he Acous ical Socie y o Ame ica, 124(6), 3959–3971.
h ps://doi.o g/10.1121/1.2999341
Lade oged, P. (2001). Vowels and consonan s. Blackwell.
Lade oged, P. (2003). Phone ic da a analysis: An in oduc ion o ieldwo k and ins umen al
echniques. Blackwell.
Le is, J. M. (2005). Changing con ex s and shi ing pa adigms in p onuncia ion eaching. TESOL
Qua e ly, 39(3), 369–377. h p://dx.doi.o g/10.2307/3588485
Lindsey, G. (2019). English a e RP. Ox o d Uni e si y P ess.
Liu, C., Jin, S., & Chen, C. (2014). Du a ions o Ame ican English owels by na i e and non-na i e
Speake s: Acous ic analyses and pe cep ual e ec s. Language and Speech, 57(2), 238–253.
h ps://doi.o g/10.1177/0023830913507692
Ma ín Gál ez, R. (1994). La du ación ocálica en español. Es udios de lingüís ica, 10(10), 213–226.
h ps://doi.o g/10.14198/ELUA1994-1995.10.11
Mompeán-González, J. A. (2001). A compa ison be ween English and Spanish subjec s' ypicali y
a ings in phoneme ca ego ies: A i s epo . In e na ional Jou nal o English S udies, 1(1), 115–
155. h ps:// e is as.um.es/ijes/a icle/ iew/47641
Odden, D. (2011). The ep esen a ion o owel leng h. In Oos endo p, M. e al. (Eds.), Companions o
linguis ics: The Blackwell companion o phonology (pp. 465–490). John Wiley & Sons, L d.
Pe e son, G. E., & Lehis e, I. (1960). Du a ion o syllable nuclei in English. The Jou nal o he
Acous ical Socie y o Ame ica, 32(6), 693–703. h ps://doi.o g/10.1121/1.1908183
Pie aszek, M. (2024a). Associa ing speake a iables wi h English p onuncia ion a ings in Spanish
e ia y educa ion. Po a Lingua um An In e na ional Jou nal o Fo eign Language Teaching and
Lea ning, (41), 189–207. h ps://doi.o g/10.30827/po alin. i41.27083
Pie aszek, M. (2024b). ‘The sun was ising’: A new elici a ion pa ag aph o English p onuncia ion
esea ch and assessmen . In D. Ma ín González (Ed.), English linguis ics mee s he 21s cen u y.
Dykinson.
RAE (Real Academia Española). (2011). Nue a g amá ica de la lengua española: oné ica y onología.
Espasa Calpe.
Roach, P. (2009). English phone ics and phonology (4 h ed.). Camb idge Uni e si y P ess.
Rojczyk, A. (2010). P eceding owel du a ion as a cue o he consonan oicing con as : Pe cep ion
expe imen s wi h Polish-English bilinguals. In E. Waniek-Klimczak (Ed.), Issues in accen s o
English: Va iabili y and no m (pp. 342–360). Camb idge Schola s.
Viswana han, N., Olms ead, A. J., & Ai a , M. P. (2020). The use o owel leng h in making oicing
judgmen s by na i e lis ene s o English and Spanish: Implica ions o a e no maliza ion. Language
and Speech, 63(2). h ps://doi.o g/10.1177/0023830919851529
Walke , R. (2010). Teaching he p onuncia ion o English as a lingua anca. Ox o d Uni e si y P ess.
Wang, H. (2007). English as a lingua anca: Mu ual in elligibili y o Chinese, Du ch and Ame ican
speake s o English. LOT.
Wells, J. (1962). A s udy o he o man s o he pu e owels o B i ish English. [Mas e ’s hesis].
Uni e si y o London.
Wells, J. C. (1982). Accen s o English. Camb idge Uni e si y P ess
Wells, J. C. (2008). Longman p onuncia ion dic iona y. Pea son Longman.
Pie aszek
Vowel quali y & leng h, Spanish-accen ed English
83
Abou he au ho
Ma eusz Pie aszek is an Assis an P o esso a he Complu ense Uni e si y o Mad id. He
holds a PhD in English Linguis ics and an MA in Spanish Linguis ics. He has been eaching
English o Speci ic Pu poses, Gene al English, and English Phone ics and Phonology a he
uni e si y le el o 16 yea s. His esea ch in e es s include English and Spanish phone ics
(in elligibili y, accen , and a i udes o p onuncia ion), English as a Lingua F anca (ELF),
compa a i e mo phosyn ax, English-medium ins uc ion (EMI), and highe educa ion. A
passiona e language lea ne , he speaks se en languages (English, Spanish, Polish, F ench,
Ca alan, Ge man, and Espe an o) and is a membe o he In e na ional Hype polyglo
Associa ion (HYPIA).
Email: mp[email p o ec ed]
This chap e is based on he o al p esen a ion gi en by he au ho a he 8 h
In e na ional Con e ence English P onuncia ion: Issues and P ac ices (EPIP
8) held May 8–10, 2024 a he Uni e si y o Can ab ia in San ande , Spain.
I is licensed unde he C ea i e Commons A ibu ion 4.0 In e na ional
License. To iew a copy o he license, please go o:
h p://c ea i ecommons.o g/licenses/by/4.0/
.
Rasmussen, S. H. (2025). Wo d-ini ial oicing in L2 English: E alua ions o /z/ p oduc ions be o e and a e
pe cep ual aining. In A. Ki ko a-Nasko a, P. Humánez-Be al, & A. Hende son (Eds.), P oceedings o he
8 h In e na ional Con e ence on English P onuncia ion: Issues and P ac ices (pp. 84–96). Uni e si é
G enoble-Alpes. h ps://doi.o g/10.5281/zenodo.16737662
Wo d-ini ial oicing in L2 English: E alua ions o /z/ p oduc ions be o e and
a e pe cep ual aining
Sidsel Holm Rasmussen
Aa hus Uni e si y
Abs ac
This s udy examines he e ec o pe cep ual aining on he p oduc ion o syllable onse L2
English /z/. L1 Danish pa icipan s’ p oduc ions o English /z/ and /s/ we e e alua ed in e ms
o in elligibili y, de ined as he abili y o na i e English lis ene s o co ec ly iden i y he
in ended phone in a Two-Al e na i e Fo ced-Choice ask. O e all, he aining g oup was mo e
in elligible han he con ol g oup. A e aining, ainees’ /z/ p oduc ions we e iden i ied wi h
61.3% accu acy (gain sco e o 12 pe cen age poin s), whe eas he gain sco e o he con ol
g oup a e aged 4.5 pe cen age poin s ( om 52.9% o 57.4%). Acous ic measu emen s we e
examined om he ainees whose /z/ okens imp o ed he mos a e aining. Compa ison
be ween he wo ypes o e alua ions sugges s ha na i e lis ene s’ iden i ica ion o he English
oiced al eola sibilan co esponds o a g ea e pe cen age (≥ 25%) o pe iodici y in he
ica i e. Me hodological conside a ions conce ning di e en ypes o p oduc ion e alua ions
a e discussed.
Keywo ds: English sibilan s, e alua ion me hods, o eign accen , ica i e oicing, L2
phone ic aining, pe cep ion and p oduc ion
Rasmussen
Wo d-ini ial oicing o /z/, impac o aining
85
1. In oduc ion
A numbe o ecen e iew pape s del e in o High Va iabili y Phone ic T aining (HVPT) as a
widely used me hod o e icien ly acili a ing he acquisi ion o no el speech sounds (Ba iuso
& Hayes-Ha b, 2018; Thomson, 2018). This line o esea ch has a long adi ion, wi h ea ly
s udies demons a ing ha exposu e o se e al exempla s o he English / – l/ con as
imp o ed L1 Japanese lis ene s’ pe cep ion o he non-na i e speech sounds (Logan e al.,
1991; Pisoni e al., 1994). Fu he mo e, he e ec s o pe cep ion aining on L1 Japanese
pa icipan s’ p oduc ion o hese English consonan s we e explo ed using di e en me hods o
e alua ions which demons a ed ha lea ning ( o some ainees) led o imp o ed p oduc ions
(B adlow e al., 1997). Me a- e iews ha e u he sc u inised his po en ial o L2 pe cep ion
aining o lead o imp o emen s in L2 p oduc ion, and ind ha small- o-medium p oduc ion
gains a e epo ed in he li e a u e (Sakai & Moo man, 2018; Uchiha a e al., 2024). Despi e
he wide ange o esul s in s udies examining he p oposed link be ween pe cep ion and
p oduc ion in non-na i e speech lea ning, as well as he ac ha he wo modali ies a e now
gene ally conside ed o be co-e ol ing in a bidi ec ional manne (Flege & Bohn, 2021), HVPT
seems o o e a use ul me hod o lea ne s o ecei e a ge ed exposu e o speci ic non-na i e
con as s. This pape p o ides esul s om some o he indings om an HVPT expe imen ,
explo ing he e icacy o pe cep ion aining on L1 Danish pa icipan s’ p oduc ion o English
/z/.
1.1 L2 pe cep ion and p oduc ion o English /s/ and /z/
The pai o English al eola sibilan s is known o cause p oblems o many lea ne s whose L1
does no ha e oiced ica i es, and issues ela ed o con usion be ween he wo English phones
a e epo ed in bo h pe cep ion and p oduc ion. Fo example, i was ound ha L1 Spanish
speake s ail o p oduce a dis inc ion (as measu ed acous ically) be ween wo d medial and wo d
inal English /z/ and /s/ (Mai ano e al., 2021). Wi h ocused pe cep ual aining, howe e , L1
Spanish s uden s o English imp o ed bo h iden i ica ion and p oduc ion o /z/ (Fouz-
González, 2019). Simila ly, in a longi udinal ansc ip ion s udy, Kap ano (2022) ound ha
No wegian EFL lea ne s consis en ly mishea d /z/ as /s/ e en a e a ending a cou se in English
phone ics. Kap ano akes his esul o be indica i e o L1 No wegian lea ne s’ pe sis en
(pe cep ual) di icul ies wi h he simila sounding non-na i e /z/. Fo L1 Swedish speake s o
L2 English, McAllis e (2007) also epo s di icul ies ela ed o he /s – z/ con as in codas;
while all p oduc ions o English /s/ we e judged accep able by a na i e English lis ene , he
same was only ue o abou 22% o he Swedish pa icipan s’ /z/ p oduc ions, and hal o he
success ul /z/ i ems we e p oduced by he same wo mos success ul pa icipan s. E iden ly,
many di e en EFL speake s encoun e he same ype o p oblems wi h he English oiced
al eola ica i e.
Acquisi ional di icul ies o he English /s – z/ con as ha e also been explo ed o L1
Danish pa icipan s in p e ious s udies. Examining he con as in wo d- inal posi ion, Ege
and Bohn (2015) ound ha L1 Danish speake s pe cei ed he con as less ca ego ically han
did na i e English lis ene s, ye pe o med mo e like na i e English pa icipan s in pe cep ually
dis inguishing wo d- inal /s/ and /z/ han hey did in p oduc ion. These indings a e in line wi h
hose epo ed by T app and Bohn (2002) who ained nine L1 Danish adolescen s o
pe cep ually iden i y coda /s/ and /z/ om minimal pai s. While he ainees’ mean
iden i ica ion sco es had imp o ed a e aining, he e was no signi ican imp o emen in hei
measu ed p oduc ion accu acy. In pa icula , “i was he p oduc ion o he wo d- inal /z/ a he
han /s/ ha c ea ed p oblems o he Danes” (p. 349). Mo eo e , acous ic analyses
demons a ed ha L1 Danish ainees’ owel- o- ica i e a io was equally sho be o e and
Rasmussen
Wo d-ini ial oicing o /z/, impac o aining
86
a e aining, pa icipan s hus ailing o employ he du a ion cue common in na i e English
p oduc ions o coda /z/.
Whe eas hese ea ly s udies we e conce ned wi h /s/ o /z/ in wo d inal posi ion, Ho slund
and Bohn (2022) examined how Danes pe cei e English consonan s in ini ial posi ion. Fi s ,
hey ound ha ini ial /s/ and /z/ assimila e o Danish /s/ as a Ca ego y Goodness (CG)
assimila ion ype (Bes & Tyle , 2007). The au ho s hus hypo hesised ha English /s/ is easily
iden i iable o L1 Danish lis ene s, while iden i ica ion o /z/ depends on p e ious exposu e o
English. In e es ingly, expec a ions om his hypo hesis we e no bo ne ou in hei ollowing
iden i ica ion s udy – a he , Ho slund and Bohn obse ed lowe iden i ica ion sco es o /s/
han o /z/, and lowe sibilan iden i ica ion sco es o he mo e expe ienced o wo lea ne
g oups included. These unexpec ed esul s caused he au ho s o specula e abou po en ial
hype co ec ion ha may lead o such esponse biases a ou ing he less amilia phone in a
CG pai such as /s/~/z/.
The su p ising esul s om Ho slund and Bohn’s s udy (2022), coupled wi h p e ious
esea ch and epo s o o he L1 EFL lea ne s’ issues, mo i a es his s udy o L1 Danish
lis ene s’ pe cep ion and p oduc ion o ini ial English /s/ and /z/ be o e and a e aining.1 The
scope o he cu en pape is conce ned only wi h he p oduc ions o he English sibilan
con as in onse posi ion p oduced by L1 Danish speake s, as he aim o his pape is wo old:
Fi s , we examine he ex en o which pe cep ual aining o syllable ini ial English /s/ and /z/
has an e ec on non-na i e p oduc ion o he same speech sounds. Since Danish does no ha e
any oiced ica i es, bo h English sibilan s a e assimila ed o he na i e /s/ (Ho slund & Bohn,
2022), and p oduc ion o he un oiced English /s/ is expec ed o be unp oblema ic (o
su icien ly simila o be iden i ied co ec ly by L1 English lis ene s), while p oduc ion o he
oiced /z/ is expec ed o be less na i elike. Secondly, his esea ch also aims o explo e he
insigh s we may gain om di e en ypes o e alua ions o non-na i e p onuncia ion.
Speci ically, his pape add esses he ollowing esea ch ques ions:
RQ1: How accu a ely can L1 Danish speake s p oduce he syllable ini ial sibilan con as
/s – z/ (as judged by na i e English lis ene s) be o e aining?
RQ2: Do L1 Danish speake s’ p oduc ion o /z/ ( he mo e challenging o he sibilan pai )
become mo e in elligible o na i e lis ene s a e pe cep ual aining?
RQ3: Which p oduc ion s a egies a e deployed by he speake s o whom co ec /z/
iden i ica ion by na i e e alua o s is highes /mos imp o ed?
2. Me hod
2.1 Pa icipan s: Na i e Danish speake s
Reco dings om 45 (F = 34, M = 11) L1 Danish speake s we e included in his s udy.
Pa icipan s we e andomly assigned o he expe imen al aining g oup (TG, n = 33) o he
con ol g oup (CG, n = 12). Thei p o ile da a a e p esen ed in Table 1. Thei ages anged om
20–76 (M = 44.6, SD = 21.9)2 and hey had s a ed lea ning English a he mean age o 10.2
(SD = 1.91).3 Pa icipan s comple ed a simple language backg ound ques ionnai e om which
1 Mo e in o ma ion abou he p ojec and ou g oup’s esea ch on pe cep ual aining is a ailable on he
websi e: h ps://cc.au.dk/en/phone ic- lexibili y-in-old-age/ou - esea ch
2 No e ha he pa icipan s we e ec ui ed o ma ch one o ou speci ied age g oups (ei he 18–35 yea s
o 60+)
3 Younge pa icipan s had on a e age s a ed lea ning English ea lie (8.7 y s. old) han he olde
pa icipan s (11.7 y s.)

Rasmussen
Wo d-ini ial oicing o /z/, impac o aining
87
we con i med ha hey had no spen mo e han 6 mon hs in an English-speaking coun y. On
a scale om 1–5 (poo –good) pa icipan s also epo ed hei sel -assessed le el o “abili y o
unde s and spoken English” and hei “abili y o speak English”.
Table 1
Backg ound In o ma ion o L1 Danish Pa icipan s
TG (n = 33)
M (SD)
CG (n = 12)
M (SD)
Age 44.6 (21.7) 44.7 (23.4)
AOL English 10.2 (1.91) 10.3 (1.97)
Sco e, unde s and 4.18 (0.81) 4.08 (1.08)
Sco e, speaking 4.03 (0.95) 3.83 (1.53)
2.2 P oduc ion ask
P oduc ion o bo h ini ial and inal /s/ and /z/ we e elici ed in a delayed epe i ion ask (Flege
e al., 1995). Pa icipan s i s lis ened o a eco ding o a na i e speake who p oduced 46
wo ds4 sepa a ely, inse ed in he ca ie ph ase: The nex wo d is _____, a e which hey hea d
he same oice ask: Wha is he nex wo d? They we e hen p omp ed o epea he wo d in he
ca ie ph ase: Le i be _____. Ta ge wo ds examined in he cu en s udy we e 15 (nea )
minimal monosyllabic pai s o English wo ds which di e ed in oicing in he ini ial ica i e
(e.g., sink  zink). P oduc ions we e elici ed om pa icipan s be o e he pe cep ion asks (Lab1),
and he same p oduc ion ask was comple ed again a e h ee weeks a a mee ing in he lab o
pos - aining es (Lab2). The ocus o his pape is limi ed o he p oduc ions o ini ial /s/ and
/z/.
2.3 Pe cep ion es s and pe cep ion aining
Pe cep ion o he English sibilan con as was assessed in wo sepa a e Two-Al e na i e
Fo ced-Choice (2AFC) iden i ica ion asks in which L1 Danish pa icipan s lis ened o CV and
VC syllables, espec i ely. Token a iabili y was in oduced by wo di e en alke s and by
a ious owel con ex s: ini ial [si], [sa], [su], [zi], [za], [zu], and inal [is], [as], [us], [iz], [az],
[uz]. In each ask pa icipan s we e asked o de e mine which sibilan began o ended he
syllable. Response bu ons we e labelled o hog aphically wi h “S” and “Z”, and no eedback
was p o ided in he es s. Pa icipan s in he expe imen al g oup we e hen ained o pe cei e
onse /s/ and /z/ in a simila iden i ica ion ask, using he same s imuli, bu in which co ec i e
eedback was p o ided a e each ial. T ainees we e asked o comple e a o al o 10 aining
sessions, each o 120 ials, in he cou se o 3–4 weeks a e which he pos - aining es
(iden ical o he i s es ) was adminis e ed in ou lab.
4 15 monosyllabic wo d pai s beginning wi h /s/ o /z/; 8 wo d pai s ending wi h /s/ o /z/
Rasmussen
Wo d-ini ial oicing o /z/, impac o aining
88
2.4 Na i e lis ene e alua ions
Non-na i e p oduc ions o /s/ and /z/ we e assessed audi o ily by 40 na i e English judges who
we e ec ui ed in o nea Aa hus, Denma k. Ta ge wo ds we e cu om he ca ie ph ase and
p esen ed o lis ene s who iden i ied he ini ial sibilan o hog aphically as ei he “S” o “Z” in
a 2AFC ask. Each lis ening session included 30 English wo ds eco ded by h ee o he L1
Danish pa icipan s a Lab1 and Lab2 (30 wo ds x 3 speake s x 2 eco dings = 180 i ems)
p esen ed in andomised o de , and hal o he judges ag eed o comple e a second session on
ano he day, on which hey a ed okens om 3 o he speake s. P oduc ions om each o he
Danish speake s we e e alua ed by 4 di e en na i e judges, which means ha e alua ions o
each speake ’s i ems a e based on he mean iden i ica ion sco es om ou lis ene s. A
calcula ed In aclass Co ela ion Coe icien (ICC) o 0.84 indica ed good eliabili y be ween
na i e lis ene judges in he p oduc ion e alua ion expe imen .5 Judges epo ed using English
50–100% in hei daily li es (M = 88.5, SD = 15.3), and hey had on a e age spen eigh yea s
in a non-English speaking coun y (SD = 8.15).
3. Resul s
The na i e lis ene s’ e alua ions o sibilan p oduc ions we e compu ed in R s udio (R Co e
Team, 2021) using he Tidy e se package (Wickham e al., 2019). T ials o which he a ge
ica i e was hea d as in ended ecei ed he sco e o 1 while ials in which he sibilan was no
iden i ied ecei ed he sco e o 0. A e age e alua ion sco es o each pa icipan we e
calcula ed as p opo ions.
Sibilan p oduc ions om all 45 na i e Danish speake s we e on a e age iden i ied as
in ended in 65.8% o all ins ances when eco ded be o e pe cep ual aining. These e alua ion
esul s we e based on 81.4% p oduc ions o in ended /s/ and 50.3% p oduc ions o in ended /z/
as summa ised in Table 2. Abou hal o he eco ded okens o in ended /z/ p oduced be o e
pe cep ion aining we e iden i ied as /s/ by na i e lis ene s. Thi y- h ee pa icipan s ained
iden i ica ion o he /s – z/ con as audi o ily. In he emaining analysis and isualisa ions, da a
om he TG and CG a e sepa a ed and compa ed.
Table 2
Na i e English (NE) Lis ene s’ Iden i ica ion o In ended /s/ and /z/ by all Danish (DA)
Speake s be o e T aining
F ica i e in ended by DA speake
Iden i ied by
NE
/s/ /z/
/s/
81.4 % 49.7 %
/z/
18.6 % 50.3 %
5 The ICC es ima es we e calcula ed using R s a is ical so wa e (R Co e Team, 2021) wi h he lme4
package (Ba es e al., 2015), based on a mean- a ing (k = 4), absolu e-ag eemen , one-way andom
e ec s model.
Rasmussen
Wo d-ini ial oicing o /z/, impac o aining
89
While iden i ica ion sco es o he in ended /s/ p oduc ions a e consis en ly high o bo h
g oups a bo h imes o eco ding (app oxima ely 80% o ainees a Lab 1 and Lab 2; 85–86%
o con ols), he /z/ p oduc ions a e less consis en ly hea d as in ended. Boxplo s in Figu e 1
u he mo e e eal g ea a ia ion among he e alua ions o speake s’ eco dings o ini ial /z/
o ainees and con ols alike. A logis ic mixed-e ec s model was un using he lme4 package
(Ba es e al., 2015) o examine he e ec o aining on he e alua ion o sibilan p oduc ion.
Th ee ac o s we e ea men coded, i.e., he expe imen g oup (TG/CG) wi h TG as he
e e ence, p oduc ion session wi h Lab1 as he e e ence and in ended ica i e wi h “z” as he
e e ence. F om Lab1 o Lab2 he e alua ed sco e o /z/ was signi ican ly highe o ainees’
p oduc ion, β = 0.93, z = 5.85, p < 0.001, and a pos -hoc z- es compa ison using emmeans
(Len h, 2025) wi h Tukey HSD co ec ions o mul iple compa isons shows ha ha was no
he case o he p oduc ion o /z/ by he con ol g oup, z = -1.99, p = 0.49. The in e ac ion
be ween p oduc ion sessions and he wo g oups was no s a is ically signi ican , β = -0.34, z =
-1.02, p = 0.31, sugges ing ha i is unce ain i he ainees’ p oduc ion ac ually does become
mo e in elligible han hose o he con ols.
Figu e 1
In elligibili y o Sibilan s as Judged by Na i e English lis ene s
No e. E alua ions (EVAL) o in elligibili y o he T aining G oup – TG (le ) and Con ol G oup – CG
( igh )
3.1 Indi idual a ia ion
Indi idual in elligibili y sco es o he 33 speake s in he TG we e also calcula ed be o e and
a e aining (see Appendix o o e iew). Figu e 2 depic s he indi idual sco es o each
speake , and i clea ly demons a es he a ied na u e o he da a. Resul s om i e selec ed
speake s a e highligh ed (in yellow) o illus a e his poin . Fo example, he /z/ p oduc ions o
one speake (S29) we e e y consis en ly iden i ied a bo h imes o es ing ( op yellow line),
while /z/ okens om ano he speake (S02) we e ha dly e e hea d as in ended (bo om yellow
line). P oduc ions o h ee speake s imp o ed by a leas 35 pe cen age poin s a e aining as
ep esen ed in he s eepe slopes in Figu e 2. In elligibili y sco es om hese i e selec ed
speake s a e also summa ised in Table 3.
Rasmussen
Wo d-ini ial oicing o /z/, impac o aining
90
Figu e 2
P opo ion o Co ec ly Iden i ied /z/ I ems
No e. In elligibili y sco es o each indi idual pa icipan in TG be o e and a e aining. The
highligh ed (yellow) lines ep esen esul s om i e example speake s.
Table 3
In elligibili y Sco es pe cen age o /z/
Speake numbe Lab1 sco e Lab2 sco e Gain sco e
S29 98.3 100 1.67
S30 56.7 95 38.3
S08 45 80 35
S01 1.69 48.3 46.7
S02 1.67 1.67 0
3.2 Explo ing he acous ic p ope ies o /z/
To u he p obe he na u e o success ul /z/ p oduc ions, okens om he i e highligh ed
speake s in Figu e 2 we e analysed o du a ion o and pe iodici y in he ica i e po ion. The
h ee speake s wi h he highes gain sco es (S30; S08; S01) we e hose whose okens we e mos
imp o ed in e ms o a ge like p onuncia ion. Tokens om he wo speake s wi h consis en ly
highes (S29) and lowes (S02) in elligibili y sco es we e also included. Al oge he , acous ic
measu emen s we e ob ained o 150 okens (15 /z/ wo ds x 2 imes o eco ding x 5 alke s)
by selec ing he ica i e po ion and using he oice epo in P aa (Boe sma & Weenink,
This chap e is based on he o al p esen a ion gi en by he au ho a he 8 h
In e na ional Con e ence English P onuncia ion: Issues and P ac ices (EPIP
8) held May 8–10, 2024 a he Uni e si y o Can ab ia in San ande , Spain.
I is licensed unde he C ea i e Commons A ibu ion 4.0 In e na ional
License. To iew a copy o he license, please go o:
h p://c ea i ecommons.o g/licenses/by/4.0/
.
Rich e , K. (2025). P epa ed o escue Cinde ella? Explo ing Aus ian EFL s uden eache s’ belie s abou
p onuncia ion eaching and lea ning. In A. Ki ko a-Nasko a, P. Humánez-Be al, & A. Hende son (Eds.),
P oceedings o he 8 h In e na ional Con e ence on English P onuncia ion: Issues and P ac ices (pp. 97–108).
Uni e si é G enoble-Alpes. h ps://doi.o g/10.5281/zenodo.16696158
P epa ed o escue Cinde ella? Explo ing Aus ian EFL s uden eache s’
belie s abou p onuncia ion eaching and lea ning
Ka in Rich e
Uni e si y o Vienna
Abs ac
O e he pas decade, he e has been a g owing in e es in esea ching EFL p onuncia ion
lea ning and eaching. Despi e his inc eased schola ly a en ion, p onuncia ion emains a
challenging a ea o many EFL eache s, la gely due o he lack o specialised pedagogy cou ses
in eache educa ion. This pape highligh s indings om a s udy on he iews o 78 EFL
s uden eache s a he Uni e si y o Vienna, Aus ia. U ilising a quan i a i e app oach, he
esea ch explo es s uden eache s’ belie s abou hei pe cei ed p onuncia ion p o iciency and
eaching compe ence. The esul s e eal ha while mos pa icipan s a e sa is ied wi h hei
own p onuncia ion skills, abou hal o hem lack con idence in hei abili y o each
p onuncia ion e ec i ely. Fu he mo e, he e is a clea demand o a dedica ed didac ics cou se
on p onuncia ion ins uc ion, as he cu en cu iculum p io i ises o he linguis ic a eas.
Al hough an exis ing p onuncia ion class os ensibly helps imp o e hei con idence in hei
own phonological compe ence, i does no adequa ely p epa e hem o eaching p onuncia ion.
These indings con i m a need o eache educa o s o o e p onuncia ion eaching cou ses
wi hin EFL aining p og ams, ensu ing u u e educa o s a e equipped o add ess his o en
neglec ed a ea.
Keywo ds: EFL eache educa ion, eache belie s, English p onuncia ion eaching, explici
p onuncia ion ins uc ion

Rich e
Aus ian EFL s uden eache s’ p onuncia ion belie s
98
1. In oduc ion
O e he pas decade, nume ous esea che s ha e ad oca ed o a enewed emphasis on EFL
p onuncia ion eaching, a ield p e iously ma ginalised and me apho ically desc ibed as he
“Cinde ella” o language eaching (Celce-Mu cia e al., 1996, p. 323) o “ he neglec ed o phan”
(Deng e al., 2009, p. 1; Gajewska, 2021, p. 20). This g owing ecogni ion o p onuncia ion as
a legi ima e esea ch discipline has led some linguis s o obse e ha Cinde ella has seemingly
ans o med in o he “Belle o he Ball” (De wing, 2019, p. 27). Howe e , he ques ion a ises
whe he in ensi ied scien i ic a en ion ansla es in o angible imp o emen s in p onuncia ion
eaching p ac ices in EFL class ooms. Indeed, his shi does no ye seem o ha e eached i s
happily-e e -a e s age. As Le is (2021) poin s ou , he e emains a no iceable gap be ween
esea ch and p ac ice, wi h many in e es ing s udies lacking clea implica ions o eaching and
p ac ically-o ien ed publica ions showing insu icien g ounding in esea ch. In o he wo ds,
p onuncia ion pedagogy con inues o s uggle o keep pace wi h ad ancemen s in esea ch
(Ki ko a-Nasko a, 2023).
Recen s udies ha e con i med ha p onuncia ion s ill emains a challenging a ea o
nume ous EFL eache s (e.g., Schae e , 2023; Tegne ed & Ren ne , 2021). Commonly ci ed
among he co e easons o his neglec is he sca ci y o p onuncia ion pedagogy cou ses in
eache educa ion p og ammes (e.g., Ba an-Łuca z, 2022; Bu i, 2015; Yokomo o, 2016). This
is due o many uni e si ies s ill elying on gene al linguis ics cou ses a he han o e ing
a ge ed ins uc ion on how o e ec i ely each p onuncia ion.
2. The p onuncia ion eaching pa adox
P onuncia ion is wi hou doub a undamen al aspec o language compe ence, playing a pi o al
ole in bo h in elligibili y and communica i e e ec i eness in EFL con ex s (Celce-Mu cia e
al., 2010; Jenkins, 2000). Despi e clea e idence suppo ing he bene i s o explici
p onuncia ion eaching and i s posi i e impac on s uden pe o mance (e.g., Koma , 2019;
Mu phy & Bake 2019; Rao, 2019; Sai o & Plonsky, 2019), many p ac i ione s con inue o
bypass p onuncia ion in hei lessons (Da cy, 2018; Yoshida, 2016), a phenomenon Da cy
(2018) e e s o as he “p onuncia ion eaching pa adox” (p. 16).
Resea ch has consis en ly demons a ed ha many EFL eache s con ess o being eluc an
o each p onuncia ion (e.g., Hende son e al., 2012; Mu phy, 2018; Schä e , 2023). E en
expe ienced eache s equen ly exp ess unease when co ec ing o e alua ing hei s uden s’
p onuncia ion skills, wi h some a oiding p onuncia ion ins uc ion al oge he (Bake & Bu i,
2016; Nangimah, 2020). This esis ance is pa icula ly s ong among no ice eache s, who,
con on ed wi h he complexi y o di e se class ooms and e ol ing cu icula, end o iew
p onuncia ion as he leas impo an aspec o language eaching (Bake & Bu i, 2016). The
li e a u e o en a ibu es his eluc ance o he absence o emphasis in cu icula (Jenkins, 2000),
inapp op ia e eaching ma e ials (Bake & Mu phy, 2018), igh schedules and igid syllabi
(Le is & Sonsaa , 2019), as well as a lack o aining and con idence (Hende son e al., 2012;
Mu phy, 2014; Mu phy & Bake , 2015).
Agains his backd op, s udies by Bake (2011), Bu i (2015), o Yokomo o (2016) ha e
shown ha EFL eache s’ pe sonal expe iences wi h p onuncia ion lea ning and eaching
g ea ly impac hei con idence in eaching p onuncia ion, as well as hei belie s abou
phonological ins uc ion in gene al. Despi e he c ucial ole hese expe iences play, many EFL
eache educa ion p og ams ail o p o ide adequa e aining in his espec (Ba an-Łuca z,
2022; De wing, 2019; Mu phy, 2014), o en lea ing eache s ill-p epa ed o add ess
p onuncia ion issues in hei class ooms. This de iciency in eache p epa a ion can pe pe ua e
ine ec i e p onuncia ion eaching p ac ices, which in u n, hampe s lea ne s’ p og ess in
Rich e
Aus ian EFL s uden eache s’ p onuncia ion belie s
99
de eloping p onuncia ion skills. Con e sely, a numbe o empi ical s udies ha e ound ha a
dedica ed p onuncia ion pedagogy cou se can in ac signi ican ly boos EFL eache s’
con idence, leading hem o add ess p onuncia ion mo e sys ema ically in hei class ooms,
compa ed o hose wi hou a ge ed aining (e.g., Ba an-Łuca z, 2022; Nagle e al., 2018;
Kochem & Le is, 2022).
I is c ucial o conside he iews o u u e EFL eache s o ailo eache educa ion
p og ammes o he needs o 21s language lea ne s, and p e en p onuncia ion om being
elega ed o he sides o abandoned in English language class ooms. Gi en hese conce ns, i is
impe a i e o explo e he ole o p onuncia ion in EFL eache educa ion mo e ho oughly.
This esea ch aims o shed ligh on he e ec s an explici p onuncia ion cou se has on
s uden eache s’ pe cep ions o p onuncia ion ins uc ion. Unlike many Eu opean EFL eache
aining p og ams, which o en lack a ocus on p ac ical phone ics (Hende son e al., 2012), he
Depa men o English and Ame ican S udies a he Uni e si y o Vienna o e s an explici
p onuncia ion cou se as pa o i s Bachelo o Educa ion (BEd) p og amme. A p esen , no
cou ses a e a ailable ha specialise in p onuncia ion pedagogy. This p omp s he ques ion i a
single language compe ence class is su icien o build s uden s’ con idence in eaching
p onuncia ion while, a he same ime, equipping hem wi h a comp ehensi e epe oi e o
eaching ools and me hods o e ec i ely add ess p onuncia ion in hei u u e class ooms.
2.1 Resea ch ques ions
G ounded in he p emise ha s uden eache s’ iews and expe iences can exe a signi ican
impac on hei p ospec i e pedagogical p ac ices, his s udy aims o in es iga e he belie s
ad anced s uden eache s o English a he Depa men o English and Ame ican s udies hold
abou p onuncia ion eaching and lea ning. To his end, he ollowing h ee main esea ch
ques ions we e o mula ed:
RQ1: Wha do EFL s uden eache s hink abou hei own p onuncia ion skills?
RQ2: How con iden do EFL s uden eache s eel abou hei p onuncia ion eaching
skills?
RQ3: Wha a e EFL s uden eache s school- ela ed expe iences wi h explici
p onuncia ion ins uc ion?
3. Resea ch con ex and me hodology
3.1 EFL eache educa ion a he Uni e si y o Vienna
As pa o he Bachelo o Educa ion p og amme, s uden s en olled in he eache educa ion
p og amme a he Depa men o English and Ame ican S udies a he Uni e si y o Vienna a e
equi ed o comple e an explici p onuncia ion cou se. This cou se, P ac ical Phone ics and
O al Communica ion Skills (PPOCS), is designed o assis s uden s in de eloping hei own
p onuncia ion skills (Rich e , 2021). In addi ion o a weekly 90-minu e in e ac i e class wi h a
lec u e , he cou se also includes a 90-minu e compulso y language lab session e e y week o
u he p ac ice. A close look a he syllabus o PPOCS 1 e eals ha each session ocuses on
a small numbe o speci ic segmen al and sup asegmen al ea u es, which a e p ac ised
p oduc i ely and ecep i ely. As a as p onuncia ion eaching pedagogy is conce ned,
p onuncia ion is add essed only ma ginally, i a all, in in oduc o y EFL eaching cou ses.
Rich e
Aus ian EFL s uden eache s’ p onuncia ion belie s
100
3.2 Pa icipan s
The s uden eache s (N =78) who pa icipa ed in his s udy we e all en olled in he Mas e o
Educa ion p og amme a he Depa men o English and Ame ican S udies a he Uni e si y o
Vienna. Da a was collec ed om he win e semes e 2021/22 o he win e semes e 2022/23.
The s uden s we e in o med abou he pu pose o he s udy, and hei pa icipa ion was
olun a y, ensu ing adhe ence o e hical esea ch guidelines. All he pa icipan s had aken
PPOCS as pa o hei BA s udies. A he ime o da a collec ion, 40% o he esponden s we e
al eady eaching English a school o ano he educa ional ins i u ion. Th ee qua e s (76%) o
he esponden s we e emale. This 3:1 a io la gely co esponds o he a e age gende
dis ibu ion in he eache educa ion p og amme a he depa men . The a e age age o he
pa icipan s was 24, and he le el o English language compe ence he pa icipan s a e expec ed
o ha e eached a his s age is C1, acco ding o he Common Eu opean F amewo k o
Re e ence (Council o Eu ope, 2001). In o de o ensu e con iden iali y, pa icipan s we e
coded as R1 h ough R78.
3.3 Ins umen s
In o de o elici he s uden eache s’ iews, a quan i a i e app oach was adop ed. A ca e ully
cons uc ed online su ey wi h closed and open ques ions was adminis e ed using Google
Fo ms. Pa icipa ion was op ional and anonymous. The online ques ionnai e encompassed
i ems ega ding s uden s’ biog aphical da a, hei pe cep ion o hei own p onuncia ion skills,
and hei iews on and expe iences wi h p onuncia ion eaching. The majo i y o he ques ions
consis ed o 4-poin Like -scale i ems. Fo some o hese i ems, open ques ions asking he
s uden s o p o ide easons o hei choices we e gi en. Howe e , no all he esponden s
added e bal commen s in he p o ided spaces.
4. Da a analysis and esul s
The collec ed da a was analysed using Excel. Pa icipan esponses we e analysed desc ip i ely
using mean sco es, pe cen ages, and equencies. Fo he quali a i e pa , esponses o open-
ended i ems we e g ouped hema ically h ough con en analysis. The hema ic analysis
in ol ed manually iden i ying, coding, and g ouping esponses in o key ca ego ies based on
ecu ing pa e ns and concep s. These ca ego ies, guided by he esea ch ques ions, included
pa icipan s’ posi i e and nega i e pe cep ions and expe iences, challenges encoun e ed, and
sugges ions o imp o emen .
4.1 S uden eache s’ own p onuncia ion skills
In o de o add ess he i s esea ch ques ion pe aining o he s uden eache s’ iews abou
hei own p onuncia ion skills, wo i ems om he ques ionnai e we e conside ed. Fi s ly, hey
we e asked o indica e he ex en o which hey ag eed wi h he s a emen ha EFL eache s
need o ensu e ha hey ha e good p onuncia ion skills. Figu e 1 shows ha he majo i y o all
he esponden s ei he s ongly ag eed (52%) o sligh ly ag eed (42%).
The pa icipan s we e also asked o a e hei sa is ac ion wi h hei own p onuncia ion on a
4-poin scale wi h he op ions: 1 ( e y unhappy) o 4 ( e y happy). As Figu e 2 shows, he
pa icipan s’ o e all sa is ac ion wi h hei own p onuncia ion was e y high. The majo i y o
he esponden s indica ed ha hey we e e y happy (35.9%) o mode a ely happy (48.7%).
Less han 16% chose he op ion ‘sligh ly unhappy’. No a single esponden op ed o he op ion
‘ e y unhappy’.
Rich e
Aus ian EFL s uden eache s’ p onuncia ion belie s
101
Figu e 1
Impo ance o Good P onuncia ion Skills o Teache s
Figu e 2
Sa is ac ion wi h Own P onuncia ion Skills
0,0%
10,0%
20,0%
30,0%
40,0%
50,0%
60,0%
S ongly ag ee Sligh ly ag ee Sligh ly disag ee S ongly disag ee
EFL eache s ha e o make su e hey ha e good
p onuncia ion skills.
35,9%
48,7%
15,4%
0,0%
How happy a e you wi h you p onuncia ion a he
momen ?
e y happy mode a ely happy sligh ly unhappy e y unhappy
Rich e
Aus ian EFL s uden eache s’ p onuncia ion belie s
102
To gain u he insigh in o he da a, he e bal commen s some o he s uden s p o ided
we e examined (n = 57). Among hose who claim ha hey a e happy wi h hei p onuncia ion,
mos esponses included e e ences o he PPOCS cou se: “PPOCS eally de eloped my
p onuncia ion skills” (R9) o “I ha e lea ned du ing PPOCS how o imp o e my skills” (R24).
A he same ime, many esponden s acknowledged oom o imp o emen . as no ed in he
commen s: “I could be be e i I had mo e p onuncia ion classes and eedback” (R35) o “I
wish I spen mo e ime p ac ising and ge ing be e ” (R72).
4.2 S uden eache s’ con idence o each p onuncia ion
Fo he second esea ch ques ion abou pa icipan s’ iews ega ding hei abili ies o each
p onuncia ion, esponses o he i em How do you eel abou you p onuncia ion eaching skills
we e analysed. The pa icipan s chose hei answe s on a 4-poin scale om 1 ( e y insecu e)
o 4 ( e y con iden ). As Figu e 3 shows, he da a displays a clea di ide. Sligh ly mo e han
hal o he pa icipan s eel e y con iden (9%) o qui e con iden (46.1%), whe eas 41% epo
eeling a he insecu e, and 3.9% e y insecu e.
Figu e 3
P onuncia ion Teaching Skills
As a as hei e bal commen s a e conce ned, many esponden s (n = 47) claim o ha e
lea ned how o each p onuncia ion om he PPOCS cou se: “[...] because o he
eaching/didac ic s a egies I ha e acqui ed h ough PPOCS” (R9) o “[...] because I wen
h ough PPOCS” (R34). A he same ime, many o he s uden eache s poin ou ha hey lack
he necessa y skills, as shown in he ollowing esponse: “We ne e lea n how o each i .
Whe e should I s a ? Should I in oduce he IPA alphabe a i s o wha and when do I each
p onuncia ion?” (R5). Ano he pa icipan ci ed he lack o p onuncia ion eaching con en in
gene al: “This has ne e been he ocus o any o he cou ses a uni” (R53). In e es ingly, among
hose who exp ess insecu i y, qui e a ew asse ha hey hemsel es expe ienced p oblems in
he PPOCS cou se: “I don’ ha e any expe ience wi h doing so and s uggled wi h
9,0%
46,1%
41,0%
3,9%
How con iden do you eel abou you p onuncia ion
eaching skills?
e y con iden qui e con iden a he insecu e e y insecu e

Rich e
Aus ian EFL s uden eache s’ p onuncia ion belie s
103
p onuncia ion mysel ” (R59). A u he heme which eme ged om he da a ela es o eache s’
iews on wha p onuncia ion ins uc ion ac ually en ails, as illus a ed in he commen s: “I eel
con iden ha I can each p onuncia ion when mis akes come up” (R33) o “Teaching
p onuncia ion a school would mos ly be done indi ec ly. Focusing on wo ds ha a e di icul
o p onounce, as hey come up and ocus on speci ic sounds ha way” (R69). These ema ks
e eal a common mispe cep ion acco ding o which p onuncia ion eaching solely consis s o
eache s co ec ing misp onounced wo ds.
4.3 S uden eache s’ expe ience wi h explici school-based p onuncia ion eaching
The hi d esea ch ques ion in ended o gauge he pa icipan s’ expe iences wi h explici
p onuncia ion eaching in a school con ex . To his end, h ee di e en i ems e e ing o hei
expe ience as lea ne s a school, as obse e s du ing school p ac ice, and as eache s we e
analysed. As can be seen in Figu e 4, a s agge ing 80% o he s uden eache s epo ha hey
did no ha e any expe iences wi h explici p onuncia ion eaching when hey we e lea ne s a
school. Eigh y- h ee pe cen asse ha hey had no obse ed any lessons wi h explici
p onuncia ion eaching elemen s deli e ed by expe ienced eache s in he cou se o hei
s udies. Al hough 33% claim ha hey hemsel es ha e augh p onuncia ion explici ly (in he
cou se o hei school-based p ac ica), he pe cen age o hose who ha e no is s ill e y high
(almos 70%). Mos o he commen s ela ed o his i em e e o he lack o aining, o
ins ance: “No expe ience; no class augh me how o do ha ” (R53) and “I ha e he necessa y
backg ound knowledge ( heo y) bu am missing he didac ic knowledge on how o ansmi i
in a un way o he s uden s” (R34).
Figu e 4
Expe ience wi h P onuncia ion Teaching
0,0%
10,0%
20,0%
30,0%
40,0%
50,0%
60,0%
70,0%
80,0%
90,0%
as a lea ne a school as an obse e du ing eache
aining
as a eache
Expe ience wi h p onuncia ion eaching
yes no can' emembe
Rich e
Aus ian EFL s uden eache s’ p onuncia ion belie s
104
In o de o ind ou whe he he s uden eache s eel he need o mo e suppo ega ding
p onuncia ion eaching, hey we e asked whe he hey would be in e es ed in aking a cou se
ocusing on p onuncia ion pedagogy. Figu e 5 shows ha app oxima ely 80% o he
pa icipan s exp essed an in e es in he cou se (58% de ini ely, 28% yes, maybe). Some o he
commen s in a ou o such a cou se p o ide u he insigh in o he s uden eache s’ iews,
wi h many o hem poin ing a he c ucial ole o p onuncia ion, o ins ance: “I is an impo an
skill o ha e. P onuncia ion makes o b eaks a language. Lea ning e ec i e me hods on
eaching p onuncia ion could be use ul o my u u e eaching ca ee ” (R10) o “I hink I could
bene i om ac i i ies o games ha can be used in class o also mo i a e my s uden s. I ha e
also had he expe ience ha especially younge lea ne s ind i un o mimic he eache and
explo e p onuncia ion (R37)”.
Figu e 5
In e es in a P onuncia ion Teaching Cou se
Among he 14% (11% I don’ hink so, 3% ce ainly no ) who claimed no o be in e es ed
in such a cou se, he opinion ha p onuncia ion is no as impo an as o he skills p e ailed:
“In my opinion, he e a e mo e impo an hings han eaching p onuncia ion. You can su i e
wi hou he pe ec p onuncia ion” (R77). A p e e ence o o he language skills could also be
no ed in he esponses, o ins ance: “I migh be in e es ing, bu in compa ison o eaching
o he skills, such as ocabula y, speaking e c., I don’ hink i ’s as impo an ” (R55). Ano he
heme ha eme ged ela es o how some s uden eache s ques ion he necessi y o eaching
p onuncia ion, gi en he lea ne s’ equen exposu e o online media, as illus a ed in he
ollowing obse a ion: “Because I don’ belie e i ’s eally ha necessa y, mos s uden s speak
wi h pe ec p onuncia ion due o compu e games, TV shows o podcas s anyways” (R62).
0,0%
10,0%
20,0%
30,0%
40,0%
50,0%
60,0%
70,0%
yes, de ini ely Yes, maybe I don' hink so Ce ainly no
I he English depa men o e ed a p onuncia ion
pedagogy cou se, would you be in e es ed?
Rich e
Aus ian EFL s uden eache s’ p onuncia ion belie s
105
5. Discussion o esul s and key indings
This s udy se ou o explo e EFL s uden eache s’ pe cep ions o and expe iences wi h
p onuncia ion ins uc ion. The indings show ha he o e whelming majo i y o he
esponden s a e sa is ied wi h hei own p onuncia ion abili ies a e comple ing a single
explici p onuncia ion cou se as pa o hei unde g adua e p og amme. Howe e , his
con as s sha ply wi h he ac ha almos hal o hem epo lacking con idence in eaching
p onuncia ion. Gi en ha he esponden s a e nea ing he comple ion o hei academic
educa ion, his de ici is conce ning, hough no en i ely unexpec ed. In ac , simila esul s
we e obse ed in a la ge-scale s udy conduc ed by Hende son e al. (2012), which su eyed
640 p ac ising EFL eache s ac oss eigh di e en Eu opean coun ies. They also ound ha
hei pa icipan s a ed hei own p onuncia ion p o iciency as a ou able bu hei
p onuncia ion eaching skills as poo . This end seems o pe sis e en mo e han a decade la e ,
no only in Eu ope bu also in o he EFL con ex s such as U uguay (Coupe , 2016) o B azil
(Buss, 2016) whe e eache s do no eel con iden eaching p onuncia ion due o a lack o skills
and knowledge. The ambi alence ound in he s uden s’ app ecia ion o he explici
p onuncia ion class was use ul bu addi ional aining would ha e been e en mo e bene icial,
as obse ed in o he esea ch s udies (e.g., Go don & Ba an es-Elizondo, 2024).
Ano he impo an obse a ion om he da a is he pa icipan s’ epo ed lack o expe ience
wi h explici school-based p onuncia ion ins uc ion. The majo i y o he s uden eache s ha e
had e y li le o no expe ience wi h p onuncia ion eaching as lea ne s, obse e s, o eache s.
Abou one hi d o hem, howe e , epo ed some expe ience eaching p onuncia ion in an EFL
class oom. The admi edly limi ed con idence o conduc such lessons wi hou o mal aining
migh pe haps s em om he skills and he knowledge gained in a compulso y language
compe ence cou se ocusing on p ac ical phone ics, o om a misconcep ion o p onuncia ion
eaching as me ely co ec ing misp onounced wo ds on he spo (Foo e e al., 2016). This
limi a ion o co ec i e eedback in he o m o ecas s has also been obse ed in a numbe o
o he s udies (Foo e e al., 2016; Mu phy, 2011; Nguyen & New on, 2020; Schä e , 2023).
On a mo e posi i e no e, he as majo i y o s uden s exp essed in e es in a p onuncia ion
pedagogy cou se should i be o e ed as pa o hei uni e si y s udies. O he empi ical
in es iga ions ha e also highligh ed he need o specialised p onuncia ion aining in eache
educa ion p og ams o adequa ely p epa e s uden eache s o e ec i e p onuncia ion
ins uc ion in L2 class ooms. Fo ins ance, Bu i (2015) demons a ed ha a ge ed
p onuncia ion ins uc ion signi ican ly impac s eache s’ con idence and e ec i eness in he
class oom, and he eby posi i ely shapes hei a i udes owa d p onuncia ion eaching. Simila
esul s ha e been ound in o he s udies (Bake , 2014; Kochem, 2022) and e en in p o essional
de elopmen ypes o aining (e.g., Nguyen & New on, 2021).
Taken oge he , hese esul s co obo a e ea lie esea ch demons a ing ha p onuncia ion
eaching s ill does no ecei e he a en ion i dese es in oday’s EFL eache educa ion. Thus,
hese indings unde sco e he need o a sys ema ic e-e alua ion o eache aining p og ams
o ensu e ha p onuncia ion is in eg a ed as a co e componen .
6. Conclusion and implica ions
The objec i e o he p esen s udy was o shed ligh on whe he EFL s uden eache s en olled
a he Mas e o Educa ion p og amme a he Uni e si y o Vienna eel adequa ely p epa ed o
escue Cinde ella and each p onuncia ion in hei ( u u e) class ooms. The indings indica e
ha an explici language compe ence cou se dedica ed o enhancing he s uden s’ own
p onuncia ion skills can ha e a posi i e impac on hei sel -con idence ega ding hei own
Rich e
Aus ian EFL s uden eache s’ p onuncia ion belie s
106
p onuncia ion skills. Howe e , i does no esul in a co espondingly high le el o con idence
in hei abili y o each p onuncia ion.
Though limi ed in hei ocus on a e y speci ic g oup o s uden eache s, he insigh s
gained in his p ojec emain aluable in ea i ming a key end: he pe sis en gap be ween
esea ch and p ac ice. To ensu e high-quali y p onuncia ion ins uc ion in EFL class ooms,
s uden eache s need o be equipped wi h he me hodological expe ise and he necessa y ools
in o de o help L2 lea ne s o pe cei e and p oduce a new phonological sys em. Teache
educa o s ha e o ealise ha a lec u e on phone ics and phonology o a single explici
p onuncia ion class does no inhe en ly esul in he seamless ans o ma ion o his knowledge
in o ac ual eaching p ac ices.
By making p onuncia ion ins uc ion an obliga o y componen o EFL eache aining
educa ion, u u e gene a ions o eache s will be be e equipped o deli e e ec i e
p onuncia ion ins uc ion, leading o imp o ed lea ning ou comes and g ea e o e all
p o iciency in English as a o eign language. Such a cou se o s uden eache s could hen help
lay he g oundwo k o u u e gene a ions o EFL ins uc o s o de elop he con idence needed
o each a skill as i al o communica i e language compe ence as p onuncia ion.
Re e ences
Bake , A. (2011). ESL eache s and p onuncia ion pedagogy: Explo ing he de elopmen o eache s’
cogni ions and class oom p ac ices. In J. Le is & K. LeVelle (Eds.), P oceedings o he 2nd
p onuncia ion in second language lea ning and eaching con e ence (pp. 82–94). Iowa S a e
Uni e si y. h ps://apling.engl.ias a e.edu/wp-
con en /uploads/si es/221/2015/05/PSLLT_2nd_P oceedings_2010.pd
Bake , A. (2014). Explo ing eache s’ knowledge o second language p onuncia ion echniques: Teache
cogni ions, obse ed class oom p ac ices, and s uden pe cep ions. TESOL Qua e ly, 48(1), 136–
163. h ps://doi.o g/10.1002/ esq.99
Bake , A., & Bu i, M. (2016). Feedback on second language p onuncia ion: A case s udy o EAP
eache s’ belie s and p ac ices. Aus alian Jou nal o Teache Educa ion, 41(6), 1–19.
h ps://doi.o g/10.14221/aj e.2016 41n6.1
Ba an-Łuca z, M. (2022). ‘The cou se should be obliga o y!’: A i udes o Polish u u e EFL eache s
owa ds a cou se on p onuncia ion eaching. In J. Le is & A. Guska oska (Eds.), P oceedings o he
12 h P onuncia ion in Second Language Lea ning and Teaching Con e ence (pp.1-10). Iowa S a e
Uni e si y. h ps://doi.o g/10.31274/psll .13260
Bu i, M. (2015). S uden eache s’ cogni ion abou L2 p onuncia ion ins uc ion: A case s udy.
Aus alian Jou nal o Teache Educa ion, 40(10), 66–87.
h ps://doi.o g/10.14221/aj e.2015 40n10.5
Buss, L. (2016). Belie s and p ac ices o B azilian EFL eache s ega ding p onuncia ion. Language
Teaching Resea ch, 20, 619–637. h ps://doi.o g/10.1177/1362168815574145
Celce-Mu cia, M., B in on, D., & Goodwin, J. (1996). Teaching p onuncia ion: A e e ence o eache s
o English o speake s o o he languages. Camb idge Uni e si y P ess.
Celce-Mu cia, M., B in on, D. M., & Goodwin, J. M. (2010). Teaching p onuncia ion: A cou se book
and e e ence guide (2nd ed.). Camb idge Uni e si y P ess.
Council o Eu ope. (2001). Common Eu opean amewo k o e e ence o languages: Lea ning,
eaching, assessmen . Camb idge Uni e si y P ess.
Coupe , G. (2016). Teache cogni ion o p onuncia ion eaching amongs English language eache s in
U uguay. Jou nal o Second Language P onuncia ion, 2(1), 29–55.
h ps://doi.o g/10.1075/jslp.2.1.02cou
Da cy, I. (2018). Powe ul and e ec i e p onuncia ion ins uc ion: How can we achie e i ? The
CATESOL Jou nal, 30(1), 13–45. h ps://doi.o g/10.5070/B5.35963
Deng, J., Hol by, A., Howden-Wea e , L., Nessim, L., Nicholas, B., Nickle, K., Pannekoek, C., S ephan,
S., & Sun, M. (2009). English p onuncia ion esea ch: The neglec ed o phan o second language
Rupp & Hende son
P onuncia ion ea u es no iced by Vie namese MOOC use s
113
English speake who pa icipa ed in he men o p og amme o he MOOC.4 The coding was
documen ed in a code book.
4. Da a analysis and esul s
In he 149 selec ed commen s, Vie namese MOOC use s exp ess challenges o conce ns
ega ding hei English p onuncia ion. Some o hese commen s (N = 109) we e p omp ed. Fo
example, in an exe cise on long and sho owels, use s a e asked: Does you na i e language
ha e long and sho owels?, o in an exe cise on wo d endings: Do you omi wo d endings?
O he commen s we e no p omp ed (N = 40), e.g., a he beginning o he MOOC, use s a e
asked: Which English p onuncia ion ea u es a e di icul o you? and Wha a e you needs
ega ding English p onuncia ion? Mos o he unp omp ed commen s (28/40) we e abou
sup asegmen al ea u es. Table 1 p o ides an o e iew o he ea u es which Vie namese
speake s commen ed upon:
Table 1
Fea u es Men ioned by Vie namese EPGW Use s
Type o ea u e English ea u e # o
men ions
To al # o
men ions
Segmen al
owel leng h;
including in diph hongs and clipping
27
87
consonan s,
e.g., / /, /ʃ/, /ʒ/, /d
ʒ/, /θ, ð/, /p, b/
21
owels,
e.g., NURSE
19
educ ion
o consonan clus e s
17
ho ici y
3
Sup asegmen al
wo d s ess
31
62 in ona ion
16
connec ed speech
15
Segmen al and sup asegmen al ea u es a e lis ed, ollowed by he numbe o men ions o
each ea u e. The ac ha bo h segmen al and sup asegmen al ea u es we e commen ed upon
may be ea ed as e idence o salience, wi h segmen als being sligh ly mo e equen ly
men ioned. Ac oss he wo ypes o ea u es, owel leng h and wo d s ess a e men ioned a
simila numbe o imes (27 and 31 men ions, espec i ely) and hus seem mo e salien o hese
4 Fo an ou line o his p og amme, see Rupp and Hende son ( o hcoming).

Rupp & Hende son
P onuncia ion ea u es no iced by Vie namese MOOC use s
114
EPGW use s. Rho ici y, in con as , is so a ely men ioned (3 imes) ha i does no seem salien
o hem.
We analysed he commen s o de e mine whe he hey cons i u ed e idence o one o he
o he ype o salience. Fo example, one Vie namese en olee is awa e ha hei p onuncia ion
di icul ies a ec how success ully hey communica e: “I ha e di icul y in connec ed speech,
speaking unna u ally and making i ha d o unde s and wha I am saying” (VMa23ad). They
e e o a speci ic linguis ic ea u e o English bu hey also assess hei own speech as unna u al
and di icul o deciphe . Ano he en olee exp esses awa eness o a no m, o ue p onuncia ion:
“I can speak sen ences wi h ue p onuncia ion bu no in ona ion” (VMdb4d8). They claim hei
achie emen – wi h ‘ ue’ p esumably meaning being able o p onounce well a he segmen al
le el o co ec ly in ela ion o a s anda d – bu hen hey also s a e a pe cei ed sho coming
(i.e., no in ona ion). Bo h ypes o commen seem o hin a language-ex e nal, i.e.,
sociolinguis ic, salience: how hei co-locu o s po en ially judge hei speech and he
imposi ion o a no m.
No e in his ela ion ha many o he ea u es men ioned by he en olees ha e been
ea ma ked and esea ched om a con as i e analysis o a pedagogical pe spec i e, hin ing a
sociolinguis ic salience in educa ional p ac ices. S udies o ypical di icul ies aced by
Vie namese English use s ha e a gued ha owel leng h is challenging o hei in elligibili y
(Cunningham, 2009a, 2009b). The monoph hongisa ion o diph hongs has also been desc ibed
as a common ea u e (Swan & Smi h, 2001). Single consonan s, oo, may be modi ied, as
poin ed ou by Cunningham (2009a): “… inal / / as in he wo d i is o en p onounced as [ip].
This p onuncia ion is iewed as cha ac e is ic o Vie namese accen ed English in Vie nam –
eache and hus lea ne awa eness o his is high, and he ea u e is s igma ised” (p. 3).
Consonan s in inal posi ion can also be pa icula ly di icul (Ngo, 2011; Nguyen, 2020)
leading o he elision o he simpli ica ion o clus e s, pe haps in pa because Vie namese does
no allow consonan clus e s in any posi ion (A e y & Eh lich, 1992).
In e ms o sup asegmen als, Vie namese is a onal language, whe e pi ch ca ies lexical o
g amma ical meaning, i.e., dis inguishes wo ds o hei in lec ions, whe eas English can use
wo d s ess o he addi ion o ano he wo d such as his and ha ins ead o a one (Tang, 2007,
11). En olees’ commen s on wo d s ess (n = 31) cons i u e hal o all commen s on
sup asegmen al ea u es. These commen s seem o esona e wi h adul Vie namese ESL
speake s in a s udy by Zielinski (2006), whose non-s anda d wo d s ess pa e ns made hem
less in elligible o h ee na i e Aus alian lis ene s. Fo example, as Cunningham (2009a) no es,
“ he use in English o he ising sac one on syllables ha ha e a oiceless s op in he coda …
can esul in a pi ch p ominence ha may be in e p e ed as s ess by lis ene s” (p. 2). Connec ed
speech is ano he a ea whe e Vie namese and English di e . English exhibi s a ied sandhi
a ia ion in all egis e s and speech a es, wi h widesp ead linking, assimila ion and
dissimila ion, dele ion, and epen hesis. In con as , Vie namese-accen ed English a ely
exhibi s any such ea u es (Cunningham, 2009b).
5. Discussion o esul s and key indings
In ou sea ch o e idence o sociolinguis ic salience in Vie namese EPGW en olees, h ee key
indings eme ge:
1) Vie namese EPGW en olees pos unp omp ed commen s, so hey do ha e some
awa eness o pa icula English p onuncia ion ea u es;
2) Bo h segmen al and sup asegmen al ea u es a e commen ed upon;
Rupp & Hende son
P onuncia ion ea u es no iced by Vie namese MOOC use s
115
3) Gi en 1), a MOOC has po en ial as a sui able ool o enqui ing in o he salience o
p onuncia ion ea u es.
The esul s o ou s udy demons a e ha Vie namese en olees in EPGW show awa eness
o a ange o bo h segmen al and sup asegmen al English p onuncia ion ea u es, wi h a sligh
majo i y o he unp omp ed commen s (28/40) being abou sup asegmen al ea u es. They lag
bo h as ea u es ha a e ei he challenging o hem o which hey wish o de elop in hei
English p onuncia ion. Does his awa eness s em om linguis ic o sociolinguis ic salience?
Two issues a ise he e. The issue wi h linguis ic salience is ha whils ela i e dis ance in
p onuncia ion may be qui e s aigh o wa dly measu ed o owels, i is no s aigh o wa d o
sup asegmen al ea u es such as s ess and in ona ion. How does one measu e he dis ance in
s ess be ween a monosyllabic language like Vie namese and a s ess- imed language like
English? O be ween a onal language and a non- onal language? An issue wi h sociolinguis ic
salience is ha en olees in EPGW a e no asked abou he easons o lagging he English
p onuncia ion ea u es ha hey do. Is i because hey we e old abou hese ea u es in hei
educa ion? Ou su ey o he esea ch li e a u e on Vie namese English speake s sugges s ha
sup asegmen al ea u es a e add essed o a lesse ex en han segmen al ea u es. Howe e , Ha
Noi Open Uni e si y p o ided us wi h he syllabi o wo English Phone ics and Phonology
cou ses ha hey un, which showed ha hey pay a en ion o segmen al and sup asegmen al
ea u es o an equal numbe o weeks. Pe haps he Vie namese en olees ha e had pa icula
social expe iences ega ding he English p onuncia ion ea u es ha hey men ioned in hei
daily li es; because he lea ning uni s in EPGW do no ask en olees o he mo i a ion o hei
commen s, we canno ell. None heless, he ac ha Vie namese en olees lagged he ange o
ea u es obse ed e eals ha hese a e impo an o hem, ega dless o whe he he a ached
impo ance is due o linguis ic o sociolinguis ic salience.
6. Conclusion and implica ions/applicabili y
The solu ion o p obing sociolinguis ic salience in EPGW would seem o be o explici ly ask
en olees o mo i a e hei commen s ega ding English p onuncia ion ea u es ha hey ind
challenging o would like o de elop in hei English p onuncia ion. This would wo k bes in
s eps a he beginning o EPGW, when en olees ha e no ye engaged wi h any s eps ha
add ess pa icula p onuncia ion ea u es (which may p omp hem o pos a commen abou
hose ea u es).
Mo e b oadly, we would like o ad oca e an app oach o English p onuncia ion eaching
ha akes accoun o and add esses English p onuncia ion ea u es ha may ha e social alue
o people de eloping hei English p onuncia ion. Knowledge o language-in e nal ac o s
(linguis ic salience) cons i u es a sound ounda ion o p onuncia ion eache s, ye hey should
also be knowledgeable abou he language-ex e nal ac o s (sociolinguis ic salience) which
impac upon hei lea ne s. Teache s may p e e o no p io i ise a ea u e i i con ibu es li le
o in elligibili y, e.g., he English den al ica i es, ye lea ne s may be pa icula ly mo i a ed
o wo k on such ea u es in o de , o example, o a oid s igma isa ion. P onuncia ion eache s
could ake he oppo uni y o expe ly balance hei objec i es wi h hose o hei lea ne s.
Following Boswijk and Cole (2020, p. 714), salience is no a s a ic p ope y o phone ic o ms
o an inhe en a ibu e o uni s o his kind. The deg ee o salience o a o m can di e be ween
speake s.
Rupp & Hende son
P onuncia ion ea u es no iced by Vie namese MOOC use s
116
Re e ences
Aue , P., Ba den, B., & G osskop , B. (1998). Subjec i e and objec i e pa ame e s de e mining
‘salience’ in long- e m dialec accommoda ion. Jou nal o Sociolinguis ics, 2(2), 163–187.
h ps://doi.o g/10.1111/1467-9481.00039
A e y, P., & Eh lich, S. (1992). Teaching Ame ican English: Ox o d handbook o language
eache s. Ox o d Uni e si y P ess.
Boswijk, V., & M. Cole (2020). Wha is salience? Open Linguis ics, 6(1), 713–22.
h ps://doi.o g/10.1515/opli-2020-0042
Cunningham, U. (2013). Teachabili y and lea nabili y o English p onuncia ion ea u es o
Vie namese-speaking lea ne s. In E. Waniek-Klimczak & L. R. Shockey (Eds.), Teaching and
esea ching English accen s in na i e and non-na i e speake s (pp. 3–14). Sp inge .
Cunningham, U. (2009a). Quali y, quan i y and in elligibili y o owels in Vie namese accen ed
English. In Ewa Waniek-Klimczak (Ed.), Issues in accen s o English II: Va iabili y and no m
(pp. 3–20). Camb idge Schola s Publishing.
Cunningham, U. (2009b). Phone ic co ela es o unin elligibili y in Vie namese-accen ed English.
FONETIK 2009 The XXII h Swedish Phone ics Con e ence (pp. 108–111). S ockholm.
h ps://u n.kb.se/ esol e?u n=u n:nbn:se:du-4086
Ellis, N. (2018). Salience in usage-based SLA. In S. M. Gass, P. Spinne , & J. Behney (Eds.),
Salience in second language acquisi ion (pp. 21–40). Rou ledge.
Jansen, S. (2014). Salience e ec s in he No hwes o England. Linguis ik Online, 64(4), 91–110.
h ps://doi.o g/10.13092/lo.66.1574
Ke swill, P., & Williams, A. (2002). ‘Salience’ as an explana o y ac o in language change:
E idence om dialec le elling in u ban England. In M. C. Jones & E. Esch (Eds.), Language
change (pp. 81–110). De G uy e . h ps://doi.o g/10.1515/9783110892598.81
Le is, J. M. (2005). Changing con ex s and shi ing pa adigms in p onuncia ion eaching. TESOL
Qua e ly, 39(3), 369–377. h ps://doi.o g/10.2307/3588485
Llamas, C., Wa , D., & Johnson, D. E. (2009). Linguis ic accommoda ion and he salience o
na ional iden i y ma ke s in a bo de own. Jou nal o Language and Social Psychology 28(4),
381–407. h ps://doi.o g/10.1177/0261927X09341962
Ngo, P. A. (2011). L1 in luence on Vie namese accen ed English. Kajian Linguis ik dan Sas a,
21(2),108–125. h ps://doaj.o g/a icle/5 7dd263e 12446d9d25ba3335948b5a
Nguyen, D. D. (2020). Common e o s in English speaking lessons o second-yea English majo
s uden s a Haiphong Technology and Managemen Uni e si y [Unpublished doc o al
disse a ion]. Haiphong Technology and Managemen Uni e si y (Đại học Quản lý à Công
nghệ Hải Phòng).
Podes a, R. J. (2011). Salience and he social meaning o decla a i e con ou s: Th ee case s udies
o gay p o essionals. Jou nal o English Linguis ics, 39(3), 233–64.
h ps://doi.o g/10.1177/0075424211405161
Rupp, L. (2019). English P onuncia ion in a Global Wo ld. [MOOC]. Fu u eLea n.
h ps://www. u u elea n.com/cou ses/english-p onuncia ion/
Rupp, L. & A. Hende son ( o hcoming). Explo ing he use ulness o MOOC English
P onuncia ion in a Global Wo ld o L2 lea ning, eaching and esea ch. In E. Vasu & A.
Ki ko a-Nasko a (Eds.), Achie emen s in second language p onuncia ion: Good p ac ices o
L2 eaching and lea ning. Camb idge Uni e si y P ess.
Rupp, L., Das, A., Kamps A., & Acos a, E. (2023). English p onuncia ion in a Global Wo ld: A
MOOC cou se o p onuncia ion eaching. In A. Hende son & A. Ki ko a-Nasko a (Eds.),
P oceedings o he 7 h In e na ional Con e ence on English P onuncia ion: Issues and
P ac ices (pp. 225–235). Uni e si é G enoble-Alpes. hal-04178893
Siegel, J. (2010). Second dialec acquisi ion. Camb idge Uni e si y P ess.
Swan, M., & Smi h, B. (Eds.). (2001). Lea ne English: A eache ’s guide o in e e ence and o he
p oblems (2nd ed). Camb idge Uni e si y P ess.
Rupp & Hende son
P onuncia ion ea u es no iced by Vie namese MOOC use s
117
Tang, G. (2007). C oss-linguis ic analysis o Vie namese and English wi h implica ions o
Vie namese language acquisi ion and main enance in he Uni ed S a es. Jou nal o Sou heas
Asian Ame ican Educa ion and Ad ancemen , 2(3). h ps://doi.o g/10.7771/2153-8999.1085
T udgill, P. (1986). Dialec s in con ac . Blackwell.
Abou he au ho s
Lau a Rupp is a Senio Lec u e in English Language and Linguis ics and he di ec o o he
Cen e o Global English a he V ije Uni e si ei Ams e dam. She c ea ed he online cou se
(MOOC) English P onuncia ion in a Global Wo ld. He esea ch expe ise is in g amma ical
language a ia ion and change, and in English p onuncia ion lea ning and eaching. She has
published in e na ionally on bo h subjec s. He wo k has been unded by he B i ish Academy
and The Ne he lands Ini ia i e o Educa ion Resea ch. He p ojec Language S o ies was
g an ed pa onage by he Du ch UNESCO Commission.
Email: l.[email p o ec ed]
Alice Hende son is a Full P o esso a G enoble – Alpes Uni e si y, F ance, whe e she eaches
English o Speci ic Pu poses o STEM s uden s and is cu en ly he assis an di ec o o he
LIDILEM esea ch g oup. She augh English phone ics and phonology o English majo s o
24 yea s a ano he F ench uni e si y and has been in ol ed in eache aining in F ance,
No way, Poland, and Spain. In 2009 she ini ia ed he bi-annual in e na ional con e ence
English P onuncia ion: Issues & P ac ices (EPIP). She has published in e na ionally
on English p onuncia ion lea ning and eaching, he pe cep ion o o eign-accen ed speech,
and English Medium Ins uc ion (EMI).
Email: alice.hende son@uni -g enoble-alpes.
This chap e is based on he o al p esen a ion gi en by he au ho s a he 8 h
In e na ional Con e ence English P onuncia ion: Issues and P ac ices (EPIP
8) held May 8–10, 2024 a he Uni e si y o Can ab ia in San ande , Spain.
I is licensed unde he C ea i e Commons A ibu ion 4.0 In e na ional
License. To iew a copy o he license, please go o:
h p://c ea i ecommons.o g/licenses/by/4.0/
.
Whi e-Hau zinge , V. e al. (2025). The ole o p onuncia ion in he assessmen o lis ening skills in he Polish
Ma u a exam. In A. Ki ko a-Nasko a, P. Humánez-Be al, & A. Hende son (Eds.), P oceedings o he 8 h
In e na ional Con e ence on English P onuncia ion: Issues and P ac ices (pp. 118–128). Uni e si é G enoble-
Alpes. h ps://doi.o g/10.5281/zenodo.16696495
The e ec s o explici p onuncia ion ins uc ion: A s udy o ad anced
Aus ian EFL s uden s
Vale ie Whi e-Hau zinge , Ma ija Djenadic, Ca olin Rumple , and Mi iam Fiala
Uni e si y o Vienna
Abs ac
As English consolida es i s ole as a globalised language and a c ucial medium o
communica ion ac oss na ions, p onuncia ion eme ges as a con ibu ing ac o o achie ing
in e na ional in elligibili y. This, in u n, emphasises he impo ance o p onuncia ion eaching
o English as a Fo eign Language (EFL) lea ne s. S uden s a he Uni e si y o Vienna
comple e he obliga o y explici p onuncia ion ins uc ion (EPI) cou se P ac ical Phone ics and
O al Communica ion Skills 1 (PPOCS1) and i s accompanying language lab, ocusing on one
o he wo o e ed a ge a ie ies. In his cou se, s uden s ecei e EPI on bo h segmen al and
sup asegmen al ea u es. This s udy examines how such EPI a ec s 82 second-yea EFL
uni e si y s uden s’ p onuncia ion, speci ically analysing h ee segmen al ea u es (/ɒ – ɑː/, /əʊ
– oʊ/, / /) assessed by ou ained a e s using a a ing scale. Responden s we e eco ded
eading he same ex a he beginning and he end o he 14-week cou se. The audi o y analysis
o he da a e eals a signi ican shi owa ds hei chosen a ge a ie y a e EPI, mos no ably
in he consis en use o /oʊ/ in Gene al Ame ican English (GAE) and he diph hong /əʊ/ in
S anda d B i ish English (SBE). Gi en hese indings, his esea ch aims o inspi e cu iculum
designe s o ecognise he alue o in eg a ing EPI in o hei p og ammes.
Keywo ds: audi o y analysis, explici p onuncia ion ins uc ion, segmen al ea u es, e ia y
EFL p og ammes

Whi e-Hau zinge e al.
Explici ins uc ion & ad anced Aus ian EFL lea ne s
119
1. In oduc ion
As globalisa ion u ns English in o a key ool o communica ion, p onuncia ion eme ges as a
ac o o pe sonal and p o essional success. Resea ch shows ha explici p onuncia ion
ins uc ion (EPI), in ol ing di ec s uc u ed eaching o p onuncia ion ea u es, imp o es
in elligibili y and educes lis ene e o (Da cy, 2018) ac oss a ious language backg ounds,
including Indonesian (Pa dede, 2018), Japanese (Sai o, 2011), Ku dish (Mahmood, 2023), and
Tunisian A abic (Bouchhioua, 2016). Howe e , he e is a esea ch gap ega ding how such
ins uc ion a ec s EFL lea ne s om Ge man backg ounds, pa icula ly in app oxima ing
speci ic a ie ies o English.
The exis ence o his gap is unsu p ising as p onuncia ion ins uc ion i sel emains
unde ep esen ed in language eaching, pa icula ly in e ia y language p og ammes (Da cy,
2018). This is especially conce ning gi en ha many uni e si y s uden s o English aspi e o
become English eache s hemsel es, who unc ion as ole models o hei s uden s bu o en
ecei e nei he a su icien amoun o ins uc ion o imp o e hei own accen s no aining o
each p onuncia ion (Hende son e al., 2015). Fo hose p e-se ice eache s, clea and
s uc u ed eaching o phone ic and phonological aspec s (i.e., EPI) can be e ec i e o
modi ying p onuncia ion. The manda o y P ac ical Phone ics and O al Communica ion Skills
1 (PPOCS1) cou se a he Uni e si y o Vienna p o ides such explici aining in bo h
segmen al and sup asegmen al ea u es, helping EFL s uden s imp o e hei o al p o iciency
in ei he S anda d B i ish English (SBE) o Gene al Ame ican English (GAE) (Rich e , 2021).
This s udy examines how his EPI cou se impac s he de elopmen o h ee segmen al ea u es
(/ɒ – ɑː/, /əʊ – oʊ/, and pos ocalic / /) in ad anced EFL s uden s who a e mainly L1 Ge man
speake s (N = 82). Using p e- and pos - eco dings, his s udy p o ides insigh s in o he e ec s
o EPI on lea ne s wi h a Ge man language backg ound ac oss wo di e en English a ie ies.
2. P e ious esea ch in he ield
P e ious esea ch highligh s he posi i e impac o EPI on L2 lea ne s’ p onuncia ion skills.
Fo example, i enhances consis ency a bo h he segmen al and sup asegmen al le els, along
wi h imp o ing wo d segmen a ion skills (Da cy, 2018). Beyond speaking skills, EPI has been
ound o s eng hen audi o y language pe cep ion (Linebaugh & Roche, 2015), spelling (P a o ,
1971), and owel pe cep ion (Gho bani e al., 2016). These indings highligh he b oad
bene i s o p onuncia ion ins uc ion, hough none ocus on speci ic a ge a ie ies, such as
SBE and GAE, o Ge man EFL lea ne s. These lea ne s o en p esen dis inc i e challenges,
such as monoph hongisa ion o diph hongs (Mende, 2009) and s uggling wi h owels a he
han consonan s (Ta zl, 2011), highligh ing he need o a ge ed esea ch on his lea ne g oup.
Agains his backd op, a ious ypes o p onuncia ion ins uc ion ha e been shown o
imp o e ad anced EFL lea ne s’ p onuncia ion skills (Go don & Da cy, 2022), wi h some
app oaches mo e e ec i e han o he s. A a (2018) emphasises he impo ance o EPI in
meaning ul con ex s, ocusing no only on pe cep ion bu also on p oduc ion o enhance
luency. To c ea e such a meaning ul con ex , Galan e and Picca do (2022) in eg a ed
con olled speech p ac ice, along wi h weekly audio eco dings, ex ensi e eedback, and sel -
e lec ions, which p o ed highly success ul. S udies compa ing explici and implici
app oaches gene ally show ha explici ins uc ion can ha e la ge e ec s on s uden s’
p onuncia ion. Fo ins ance, s uden s who ecei ed eigh weeks o EPI showed g ea e
imp o emen in hei English p onuncia ion compa ed o hose who ecei ed implici
ins uc ion (Yaku 2020), a inding also suppo ed by Go don e al. (2013), whose s udy used
EPI in combina ion wi h a p esen a ion-p ac ice-p oduc ion sequence o e he cou se o h ee
weeks. Bo h o hese s udies used p e- and pos - eco dings o s uden s eading aloud. These
Whi e-Hau zinge e al.
Explici ins uc ion & ad anced Aus ian EFL lea ne s
120
indings sugges ha he meaning ul and con olled con ex o EPI could play a c ucial ole in
imp o ing ad anced EFL lea ne s’ p onuncia ion skills.
To conclude, while exis ing esea ch has es ablished he posi i e e ec s o EPI ac oss
a ious lea ne popula ions and app oaches, subs an ial gaps emain. In pa icula , lea ne s
om Ge man language backg ounds ha e been la gely o e looked, as ha e s udies ha ocus
on app oxima ing speci ic EPI a ge a ie ies o English.
2.1 Resea ch ques ion and hypo hesis
In ligh o hese gaps, he ollowing esea ch ques ion was in es iga ed: How does he
p onuncia ion o EFL uni e si y s uden s o selec ed segmen al ea u es – in pa icula , /ɒ – ɑː/,
/əʊ – oʊ/, and pos ocalic / / – de elop o e a semes e o EPI? The ollowing hypo hesis was
o mula ed: EFL uni e si y s uden s show signi ican imp o emen in hei p onuncia ion o
he selec ed segmen al ea u es a e a semes e o EPI.
3. Me hodology
3.1 Pa icipan s and se ing
The s udy was conduc ed a he Depa men o English and Ame ican S udies o he Uni e si y
o Vienna. Pa icipan s we e ou h-semes e bachelo s uden s en olled in PPOCS1, who a e
expec ed o ha e eached an uppe in e media e B2 p o iciency le el (Council o Eu ope,
2020). PPOCS1 is a weekly wo-hou class o e ing EPI. P io o en olmen , s uden s choose
be ween one o he wo o e ed a ie ies, SBE and GAE. PPOCS1 is complemen ed by a
manda o y weekly wo-hou language lab, ocusing on he chosen a ie y.
A o al o 82 s uden s pa icipa ed in he s udy. O hese, 45 s uden s a ended he BE
PPOCS1 cou se, while he emaining 37 ook he GA PPOCS1 cou se. 84% (n = 73) o he
pa icipan s epo ed Ge man as hei i s language (L1), 6% (n = 5) we e bilingual Ge man
speake s, and he emaining 10% (n = 9) had o he L1s, including Albanian, A abic,
Bosnian/C oa ian/Se bian, Czech, Du ch, Hunga ian, Polish, Romanian, and Tu kish. Despi e
a ying language backg ounds, all pa icipan s spoke Ge man. None o he pa icipan s
iden i ied English as hei L1.
3.2 P onuncia ion ins uc ion
Pa icipan s a ended wo EPI classes pe week (4 hou s o al), ecei ing speci ic English
a ie y ins uc ion on segmen al and sup asegmen al ea u es. PPOCS1 and he language lab
combine heo y wi h p ac ical class oom ac i i ies, such as lis ening o audio guides and
p oducing a ge ed ea u es, which we e shown o be e ec i e p ac ices (Galan e & Picca do,
2022). Acco ding o Rich e (2021), he wo cou ses aim a de eloping s uden s’ heo e ical
and p ac ical knowledge o segmen al and sup asegmen al ea u es and help hem app oxima e
one o he wo o e ed a ie ies o English. In hei design, he cou ses o e no only hands-on
oppo uni ies o p ac ise spoken language communica i ely bu also egula eedback on he
p oduc ion o segmen als and sup asegmen als (Rich e , 2021).
3.3 Da a collec ion
P io o da a collec ion, he es ing p ocedu es we e ho oughly explained o he s uden s, and
any ques ions we e add essed o ensu e cla i y and comp ehension. Pa icipan s p o ided hei
in o med consen o pa icipa e in he s udy and ag eed o he p ocessing o he da a. A he
Whi e-Hau zinge e al.
Explici ins uc ion & ad anced Aus ian EFL lea ne s
121
beginning o he semes e , s uden s comple ed a ques ionnai e o p o ide in o ma ion on hei
language backg ound and hei chosen a ie y o PPOCS1. Speech samples we e collec ed a
he beginning o he cou se in week 1 (Ma ch 2023, p e- es ) and a he end o he cou se in
week 14 (June 2023, pos - es ) in he language lab. A bo h es ing poin s, pa icipan s we e
equi ed o eco d hemsel es indi idually eading a ex passage aloud. While eading aloud
is no an op imal measu e o p onuncia ion change, i acili a es compa abili y ac oss
eco dings and a mo e consis en analysis o he a ge ed ea u es.
Speci ically, he ex “The No h Wind and he Sun” was employed due o i s es ablished
use in p e ious esea ch s udies (e.g., Gass & Va onis, 1994) and i s inclusion o all ele an
ea u es unde in es iga ion. This s udy ocuses on h ee segmen al ea u es: wo con as s /ɒ –
ɑː/ and /əʊ – oʊ/, and he pos ocalic / /, which a e ecognised as dis inguishing cha ac e is ics
be ween GAE and SBE and ypical challenging ea u es o Ge man speake s (Schmi , 2016).
The ex passage includes 30 wo ds (see Table 1) ea u ing one o mo e o he a ge ed
phonemes, which we e embedded in he s o y o ensu e ha s uden s ead he wo ds in con ex
a he han in isola ion. All speech s imuli we e eco ded using he so wa e KALTURA1 and
sa ed in MP4 o ma . A o al o 164 eco ded samples we e analysed (82 a p e- es and 82 a
pos - es ).
Table 1
Dis ibu ion o 33 Selec ed Segmen al Fea u es
Selec ed
segmen als
To al numbe
o okens
Example
wo ds
/ɒ – ɑː/ 6 s onge , o , s opped
/əʊ – oʊ/ 7 cloak, closely, old, so
/ / 20
No h, we e, s onge , a elle ,
di y, i s , conside ed, o he ,
ha d, mo e, wa mly
4. Da a analysis and esul s
Da a we e analysed using R e sion 4.3.3 (R Co e Team, 2024) and RS udio e sion 2024.04.2
(RS udio Team, 2024); speci ically, he R packages “s a s” (R Co e Team, 2024),
“ companion” (Mangia ico, 2024), and “ cd” (Meye e al., 2023).
4.1 Ra e s and a ing scale
To in es iga e s uden s’ de elopmen in p onuncia ion a e ecei ing EPI, his s udy used an
audi o y analysis, as in p io esea ch (e.g., Sai o, 2011; Zhang & Yuan, 2020). Du ing he
1 KALTURA h ps://co p.kal u a.com/
Whi e-Hau zinge e al.
Explici ins uc ion & ad anced Aus ian EFL lea ne s
122
a ing s age, ou ained lis ene s a ed he pa icipan s’ pe o mances. These lis ene s we e
conside ed ained due o hei oles as language lab u o s a he Uni e si y o Vienna, whe e
hey a e egula ly exposed o a wide ange o EFL uni e si y s uden s’ speech and expec ed o
p o ide eedback. Each lis ene was assigned 41 s uden eco dings (i.e., ou se s) pe a ing
cycle, wi h wo cycles in o al o ensu e in e - a e eliabili y. In he i s cycle, each a e
assessed one se , and in he second cycle, hey assessed a di e en se , ensu ing ha each
eco ding ecei ed wo independen a ings.
Subsequen ly, a balanced a ing app oach was adop ed, which included assigning codes o
each s uden eco ding. This coding sys em ensu ed pa icipan anonymi y and p e en ed a e s
om knowing whe he hey we e e alua ing p e- o pos -ins uc ion eco dings. Addi ionally,
each a e assessed eco dings om bo h SBE and GAE a ie ies, which ensu ed a ully c ossed
design and u he minimised po en ial a ing e ec s. Weigh ed Cohen’s Kappa (κw)2 was
calcula ed o each se o a e s o all h ee a ge ea u es. The ange o κw o each ea u e is
p esen ed he e o summa ise he o e all ag eemen le els ac oss all ou se s: Fo /ɒ – a:/, κw
alues anged om .32 o .70, indica ing a ela i ely b oad le el o ag eemen om ai o
subs an ial; /əʊ – oʊ/ showed mode a e o subs an ial ag eemen , wi h κw alues be ween .50
and .74. Finally, ega ding he pos ocalic / /, he ag eemen was consis en ly highe , wi h κw
alues anging om .74 o .90, e lec ing subs an ial o almos pe ec ag eemen ac oss he
se s.
Mo eo e , all a e s u ilised a s anda dised a ing scale (see Appendix) de eloped
speci ically o his s udy o main ain consis ency. Each segmen al was assessed ac oss a se
numbe o okens. Fo each oken o a ea u e, a e s de e mined whe he he p onuncia ion
app oxima ed he a ge a ie y, assigning a bina y sco e o 0 (no app oxima ed) o 1
(app oxima ed). The o al numbe o co ec okens o each ea u e was calcula ed by summing
he sco es ac oss all ins ances. This alue was hen con e ed in o a pe cen age o accu a e
app oxima ions ela i e o he o al numbe o a ge ins ances o ha ea u e. Based on he
pe cen age o co ec app oxima ions, pa icipan s we e assigned a sco e on a 5-poin scale: no
app oxima ed o a ge a ie y ( alue 1), sligh ly app oxima ed o a ge a ie y ( alue 2),
mode a ely app oxima ed o a ge a ie y ( alue 3), mos ly app oxima ed o a ge a ie y
( alue 4), and ully app oxima ed o a ge a ie y ( alue 5).
4.2 Resul s
To in es iga e he signi icance o changes om p e- es o pos - es , a Wilcoxon signed- ank
es was compu ed o each a ge ea u e. The analysis also included he e ec sizes o
signi ican di e ences; his s udy epo s he ank bise ial co ela ion coe icien ( b), which is
app op ia e o non-pa ame ic es s o di e ence (King e al., 2011). To in e p e he
magni ude o he b, he analysis ollowed ield-speci ic benchma ks o second language
esea ch p oposed by Plonsky and Oswald (2014), who classi y s anda dised - alues o .25,
.40, and .60 as small, mode a e, and la ge e ec sizes, espec i ely.3 To educe he isk o alpha
e o accumula ion, he signi icance h eshold was se a .01.
The Wilcoxon signed- ank es e ealed signi ican imp o emen s in s uden s’
p onuncia ion om he p e- es o he pos - es ac oss all a ge ea u es. Table 2 p esen s he
2 To in e p e he magni ude o ag eemen , Landis and Koch’s (1977) benchma ks we e employed.
3 While Plonsky and Oswald’s (2014) ield-speci ic benchma ks o in e p e ing Pea son we e
de eloped based on pa ame ic da a, he concep ual simila i y be ween Pea son and ank bise ial ₍ b₎
(Rosen hal, 1991) allows o a en a i e applica ion o hese h esholds. As a gued by Ki by (2014), b
is mo e app op ia e o non-pa ame ic es s as i akes in o accoun ank sum di e ences. The e o e,
b was expec ed o yield mo e nuanced esul s.
Index
accen i, ii, xiii, 17, 19, 21, 22,
23, 40, 46, 60, 61, 84, 85,
112, 119, 125, 130
accen edness i, ii, xi, 70, 73, 92
accen s xii, 22, 34, 41, 60, 119, 125
accommoda ion 111
accu acy ix, x, 43, 44, 46, 47, 48, 52,
56, 61, 84, 85, 124
ac ion esea ch 55, 56, 64, 65
a ec i e 3, 61
age 17, 31, 41, 42, 43, 60, 62,
73, 93, 100, 112
AI 64
Ame ican iii, 17, 34, 41, 72, 75, 77,
99, 100, 118, 119, 120
applied esea ch 55, 65
A abic iii, ix, 119, 120
ASR 64
assessmen ix, x, 1, 2, 29, 59, 61, 92, 93
a i ude(s) 17, 31, 32, 42, 43, 62, 66,
70, 110
audi o y analysis 118, 122
Aus ian iii, ix, 97,118, 125
au hen ic 11, 15, 16, 17, 22, 23, 24,
54, 56, 58, 59, 62, 63, 64,
65
au hen ic speech 11, 16, 17, 22, 23
awa eness 15, 16, 22, 28, 53, 63, 81,
109, 110, 114, 115
B i ish iii, 17, 29, 34, 41, 72, 75,
118, 119
Ca alan iii, 43, 45, 63
cha bo s 64
Cha GPT 63
Chinese iii, 16
CLT 59
clus e s 3, 112, 113, 114
cogni i e load 59
compe ence 1, 2, 10, 17, 30, 97, 98, 99,
100, 105, 106
compu e -gene a ed isuals 64
con idence x, 28, 30, 34, 35, 37, 40, 65,
97, 98, 99, 102, 105, 106
connec ed speech x, 3, 5, 10, 15, 16,
17, 18, 19, 20, 22, 23, 44,
52, 113, 114
consonan s 33, 34, 85, 86, 112, 113,
114, 119, 124
C oa ia iii, 120
CSP 15, 16
Czech iii, 120
Danish iii, ix, 84, 85, 86, 87, 88,
92, 93, 94
deli e y 3, 16, 22, 58, 60, 61, 65
diph hongs 113, 114, 119, 124
discou se-le el 57, 61
Du ch iii, 120
e ec i eness x, 11, 40, 56, 57, 58, 59, 61,
65, 98, 105, 124
EFL lea ne s 15, 16, 17, 22, 23, 85, 86,
119, 120, 124, 125
EFL s uden s 118, 119, 124
elision x, 4, 5,9, 10, 15, 17, 18, 19,
21, 22, 23, 114
EPI 118, 119, 120, 124
e o (s) xi, 16, 30, 33, 50, 56, 62,
65, 122
e alua ion(s) 3, 65, 84, 85, 86, 88, 89, 91,
92, 93, 94, 105
explici p onuncia ion eaching 30,
98, 103
acial exp ession 60
eedback i, ii, xi, 30, 32, 36, 57,
62, 87, 102, 105, 119, 120,
122, 124
ilms 59, 64
luency ix, xi, 22, 23, 43, 44, 46,
47, 48, 49, 52, 61, 64, 92,
119
F ench iii, 16, 60
ica i e(s) x, 74, 84, 85, 86, 87, 88, 89,
90, 91, 92, 93
den al 28, 33, 34, 36, 37,
62, 115
Gene al Ame ican English 75, 118, 119
gene alisabili y 57
Ge man iii, 119, 120, 121, 124,
125
ges u e(s) i, ii, ix, xi, 60, 61
Global English 17, 22, 117
Golden Speake 64
Haussa iii, ix
hesi a ion 2
Hunga ian iii, 28, 30, 31, 32, 33, 34,
36, 37, 40, 41, 42, 120
HVPT 56, 57, 58, 59, 85
iden i y 60, 61, 62, 111
Indian iii, ix

130
in elligibili y i, ii, ix, x, xi, 3, 16, 17,
29, 56, 57, 58, 61, 62, 65,
71, 73, 78, 80, 81, 84, 89,
90, 91, 93, 98, 111, 114,
118, 119
in e en ion x, 17, 65, 110, 123, 124
in ona ion xiii, 42, 44, 50, 52, 56, 57,
60, 61, 64, 113, 114, 115
in usi e 5, 7, 9, 10, 15, 17, 18, 19,
21, 22, 23
IPA 33, 35, 36, 37, 102
I alian iii
Japanese iii, 58, 85, 119
L2 1, 2, 3, 4, 11, 16, 17, 44, 53,
57, 60, 61, 62, 63, 64, 72,
73, 84, 85, 92, 93, 94, 105,
106, 119
language
acquisi ion 109, 110, 111
a i udes 110, 111
lab 99, 118, 120, 121, 122
lis en and epea 33, 35, 36
lis ene (s) ii, xiii, 2, 3, 30, 44, 46, 52,
62, 63, 73, 74, 78, 80, 84,
85, 86, 88, 89, 92, 93, 94,
115, 119, 122
lis ening comp ehension xi, 1, 2, 3, 4,
5, 11, 16, 17, 24, 41
Macedonian iii, ix, 15, 17, 18, 23, 24
men o 113
me hodological ac o s 56, 57, 58
mi o ing 60, 61, 65
mobile media 59
monosyllabic 87, 115
MOOC ii, ix, x, xi, xiii, 109, 110,
111, 113, 115
mo emen 33, 43, 61, 62
na i e speech x, 16, 17, 18, 22, 24, 85
na u alness xi, 15, 16, 17, 22, 23
non-na i e 2, 16, 17, 20, 23, 30, 85, 86,
88, 92, 93, 94
No wegian ix, 85
open access i,
o hog aphic ix, 64, 88
ou comes x, 43, 56, 57, 58, 66, 107,
125
Pakis an ix
pedagogy x, 30, 58, 59, 60, 63, 65, 97,
98, 99, 104, 105
pe cep ion and p oduc ion x, 22, 23, 84,
85, 86
pe cep ions x, xi, 15, 16, 17, 22, 23, 44,
99, 100, 105
pe cep ual aining x, 84, 85, 86, 88, 92,
94
Philippines iii
phone ic aining 56, 84
pi ch 43, 44, 46, 50, 52, 53, 58,
61, 62, 68, 114
Polish iii, 1, 2, 120
p ac ical phone ics 99, 105, 118, 119
p o iciency xi, 4, 9, 44, 52, 53, 60, 62,
73, 93, 97, 105, 106, 119,
120
p ominence 3, 18, 61, 109, 114
p osodic 44, 61, 62, 64
p osody ii, ix, x, 43, 44, 46, 49, 52,
59, 61, 62, 64
quan i a i e ix, 18, 32, 68, 71, 76, 97,
100
a ing scale 118, 121, 122
eading aloud 46, 48, 119, 121
eliabili y 32, 57, 88, 122, 125
eplica ion 56, 57, 58, 64
esea ch base 55, 56, 64
esea ch design ix, 56, 57
hy hm x, 16, 19, 22, 23, 33, 34, 41,
44, 56, 61
Romanian ix, 120
salience x, 109, 110, 111, 112, 113,
114, 115
Sco ish 15, 17, 18
sc ip ed ex s 1, 2, 3, 9, 11
seconda y school ix, 28, 29, 31, 34, 35
segmen als 56, 62, 113, 120, 124
shadowing i, ii, ix, xi, 58, 59, 64
sibilan s 84, 85, 86, 89, 93, 94
silen le e s 28, 33, 34, 36, 41, 44
Sindhi iii
social meaning 60, 61, 62, 111
social alue 109, 115
Spain i , iii, ix, 44
Spanish iii, 30, 63, 71, 72, 73, 78,
80, 85
speaking s yle 3, 61, 65
speech a e 2, 3, 6, 8, 9, 15, 16, 18, 22,
44, 46, 52, 76, 84, 114
spoken ex (s) 1, 2, 4, 5, 9, 11
spon aneous ex s 9
S anda d B i ish English 118, 119
131
s e eo yped 62, 111
s igma isa ion 115
s o y 42, 43, 44, 45, 46, 48, 49,
52, 53, 121
s ess x, xi, 11, 31, 33, 34, 41, 42,
44, 45, 56, 57, 60, 61, 63,
64, 73, 112, 113, 114, 115
wo d 31, 34, 37, 41, 57, 112, 113,
114
s uden eache s ix, 97, 99, 100, 102,
103, 104, 105, 106
s yling 60
sup asegmen al(s) 10, 44, 56, 57, 60,
61, 62, 64, 65, 99, 109, 113,
114, 115, 118, 119, 120
Swedish iii, 85
a ge ii, x, xi, 4, 21, 30, 37, 40,
57, 81, 87, 88, 90, 92, 93,
110, 118, 119, 120, 122,
123, 124, 127
ask-based 59
eache educa ion x, 29, 37, 97, 98, 99,
100, 105
echnology 60, 63, 64, 65
ele ision 59, 64
e ia y 118, 119, 124
hough g oups 61
iming 61
onal 114, 115
ans e 57, 60, 63
Uk ainian iii, ix
U du iii
a ia ionis 109, 110
a ie y x, 17, 29, 30, 56, 59, 64, 65,
111, 118, 120, 121, 122,
127
ideo 33, 42, 50, 63, 64
Vie namese iii, ix, 109, 110, 112, 113,
114, 115
owel(s) xi, 71, 73, 74, 75, 76, 77,
78, 79, 113, 115, 119, 124
leng h 71, 72, 73, 74, 76, 113, 114
educ ion 17, 18, 19, 20, 22, 23, 25,
34
oice quali y 50, 52, 61, 71, 76, 78, 81
whole-body con ex 60
YouGlish 64
YouTube 64
130