Wo kshop
Recommended ci a ion: Decke , C., & Söhni z, S. (2025). Wo ds Don’ Come
Easy o Me – Teaching Academic W i ing o Enginee ing S uden s in he Age O
AI. In Kangaslampi, R., Langie, G., Jä inen, H.-M., & Nagy, B. (Eds.), SEFI 53 d
Annual Con e ence. Eu opean Socie y o Enginee ing Educa ion (SEFI),
Tampe e, Finland. DOI: 10.5281/zenodo.17631224.
This Con e ence Pape is b ough o you o open access by he 53 d Annual Con e ence
o he Eu opean Socie y o Enginee ing Educa ion (SEFI) a Tampe e Uni e si y in
Tampe e, Finland. This wo k is licensed unde a C ea i e Commons
A ibu ion-NonComme cial-Sha e Alike 4.0 In e na ional License.
WORDS DON’T COME EASY TO ME –
TEACHING ACADEMIC WRITING TO ENGINEERING STUDENTS IN
THE AGE OF AI
C. Decke a,
1
, S. Söhni z b
a Hochschule Düsseldo Uni e si y o Applied Sciences, Düsseldo , Ge many,
0000-0001-6883-566X
b Hochschule Düsseldo Uni e si y o Applied Sciences, Düsseldo , Ge many,
0000-0003-3988-1768
Con e ence Key A eas: Enginee as a social deba e – new skills needed?; Digi al
ools and AI in enginee ing educa ion; Enginee ing skills, p o essional skills, and
ans e sal skills
Keywo ds: li e acy, eading compe ence, w i ing compe ence, academic w i ing,
c i ical hinking
ABSTRACT
The compe ence o academic w i ing is cu en ly s ill a cen al equi emen o
ob aining an academic deg ee. Howe e , in eaching academic w i ing, lec u e s ha e
o deal wi h he wo challenges o dec easing eading compe ence in young people
and inc easing use o a i icial in elligence (AI) in eading and w i ing assignmen s. In
enginee ing in pa icula , li e acy unde s ood as eading and w i ing compe ence is
essen ial - whe he o unde s anding complex asks, analysing case s udies o dealing
wi h echnical documen s such as he p oduc equi emen s documen . Agains his
backg ound, i is c ucial o de elop s a egies and discuss mode n app oaches o
speci ically p omo e s uden s' eading and w i ing skills and hus c ea e an essen ial
basis o success ul scien i ic wo k and p oblem-sol ing. This wo kshop aims o
s imula e an in e ac i e exchange and de elop p ac ical solu ions in he h ee co e
a eas s eng hening eading compe ence, de eloping academic w i ing skills, and
subs i u ing o eading and w i ing assessmen s.
1
Co esponding Au ho
C. Decke
ca s en.decke @hs-duesseldo .de
1 BACKGROUND AND RATIONALE
1.1 Li e acy as Reading and W i ing Compe ence
Besides nume acy and skills in he na u al sciences, he enginee ing p o essions a e
also in need o a ce ain amoun o li e acy skills. The e m li e acy is no always clea ly
de ined: In a na ow sense i comp ises only eading compe ence, in a somewha
wide sense i also includes eading and w i ing compe ence, and some imes he
de ini ion o li e acy ex ends eading and w i ing by speaking and lis ening compe ence
(Camb idge Assessmen , 2013). In his wo kshop we ocus on li e acy as eading and
w i ing compe ence, as we see eading compe ence as a main p e equisi e o
academic w i ing.
In hei Lea ning Compass 2030, he O ganisa ion o Economic Co-ope a ion and
De elopmen (OECD, 2020) de ines li e acy as “The abili y o iden i y, unde s and,
in e p e , and c ea e wi h w i en, spoken, isual and mul i-modal ex s. Li e a e
s uden s a e able o decode and cons uc di e en ypes o ex s app op ia e o li e
in and ou o school. These ex s will include subjec -speci ic ex s (such as hose
equi ed in science o his o y) and isual ex s such as diag ams and g aphs”.
Acco ding o he OECD (2020), li e acy is he ounda ion o co e cogni i e skills o
communica ion, c ea i i y, and c i ical hinking.
Enginee ing s uden s need eading compe ence o unde s and e.g. calcula ion
exe cises (which a e o en p esen ed in ex o m), case s udies, and s uden p ojec
desc ip ions (e.g. a p oduc equi emen s documen ). Apa om hese unc ional
conside a ions, s uden s should also be able o ead and unde s and complex
a gumen a ions o de elop hei hinking and easoning skills. In a sense, you a e wha
you ead: “Reading well awakens and b oadens he mind. I p o ides a as ealm o
inne expe ience ha ex ends a beyond e e yday li e” (DiYanni, 2022, p. ix).
Academic w i ing unde s ood as p o essional w i ing used in a uni e si y se ing is
needed by enginee ing s uden s o w i e e m pape s, in e nship epo s, lab epo s,
s uden p ojec epo s, and – inally – hei opus magnum: he hesis. Again, unc ional
conside a ions aside, he e is an educa ional adi ion ha desc ibes “w i ing as
hinking”. W i ing can be unde s ood as “ hinking ha uses pape o o he media o
ex e nalize and manipula e symbolic exp essions” (Oa ley & Djikic, 2008, p. 9). E en
i his migh be o e s a ed, w i ing usually is a p e equisi e o highe -o de hinking and
w i ing and c i ical hinking ein o ce each o he (Kaag & an Belle, 2024). In his sense
“W i ing is no simply a ma e o exp essing p e-exis ing hough s clea ly. I ’s he
p ocess h ough which ideas a e p oduced and e ined” (Min z, 2021).
Acco ding o he in e na ional P og amme o In e na ional S uden Assessmen
(PISA) eading pe o mance o mos coun ies d opped be ween 2018 and 2022, bo h
in highes -sco ing and weakes young s uden s. I eland – hos coun y o he SEFI
2023 con e ence – has he highes mean sco e among Eu opean coun ies (OECD,
2023). In Ge many, he eading compe ence o young s uden s – as measu ed by
PISA – has allen signi ican ly be ween 2012 and 2022 and is now nea he OECD
a e age. The coun y has a compa a i ely high numbe o adolescen s wi h weak
eading compe ence o a coun y nea he OECD a e age (Heine e al., 2023, p. 151).
The esul s o he P og amme o he In e na ional Assessmen o Adul Compe encies
(PIAAC) show ha li e acy has declined o emained s able in compa ison o he las
su ey and has declined mo e s ongly o men han o women. Abou a qua e o he
adul popula ion sco es a he wo lowes le els in li e acy. Finland – enue o he SEFI
2025 con e ence – has he highes mean sco e o li e acy (OECD, 2024).
1.2 Li e acy o Enginee ing S uden s
Enginee s need eading and w i ing skills o unde s and and/o gene a e e.g. lab and
p ojec epo s, p oduc equi emen s documen s, ope a ing manuals o echnical
equipmen , and ma ke o echnology end s udies in hei p ac ical wo k. I hey
pu sue a job in science and esea ch, hey need o be able o ead and w i e schola ly
publica ions such as jou nal a icles, esea ch epo s and su eys as well as
labo a o y s udies. Fu he mo e, in hei socie al ole as in o med ci izens, hey need
o be able o unde s and e.g. poli ical news and gene al s a is ics on economic and
echnological de elopmen s.
To be e unde s and he eading and w i ing habi s o ou enginee ing s uden s, we
conduc ed a su ey in a cou se on academic w i ing isi ed by uni e si y s uden s om
he six h semes e . Mos o he s uden s pa icipa ing in he cou se (25 ou o 29)
answe ed he su ey a he beginning o he semes e . Mos o ou s uden s (in o al
72%) don’ ead books a all (32%) o only one book pe qua e o less (40%). Only
28% o he s uden s ead one book pe mon h o one book e e y wo weeks o mo e.
64% o ou s uden s ead no newspape s o magazines on a egula basis.
Wi h ega d o w i ing habi s, 44% o he s uden s answe ed ha hey ha e p i a ely
ne e w i en any long ex . S uden s who ha e w i en long ex s w o e (in descending
o de ) p o essional a icles (28%), sho s o ies (28%), newspape a icles (20%), and
poe y (poem, song ly ics, ap, …) (16%), and one s uden e en w o e a no el. As pa
o hei s udies, all s uden s had o w i e a ex . Mos o he s uden s w o e lab epo s
(68%) and in e nship epo s (64%) ollowed by e m pape s (40%), s uden p ojec
epo s (36%), and p o essional a icles (12%).
Main p oblems in academic w i ing acco ding o ou s uden s (see ig. 1) a e he
echnical aspec s o w i ing such as li e a u e esea ch and e alua ion o sou ces
(52%) and ci a ion (44%), he in e nal s uc u al aspec s o w i ing such as a icula ing
hei own hough s (44%), summa izing o he sou ces (32%) and a gumen a ion
(24%), as well as de eloping he ex e nal ex s uc u e (32%). Mo i a ion is seen as
a challenge by 28% o ou s uden s, mainly he ones who do no ead o w i e much.
No s uden pe cei es spelling and g amma as one o he main p oblems in academic
w i ing, al hough he w i e s con i med h ough p oo eading hei pape s ha hey
de ini ely exis as a di icul y o some o he s uden s.
The in luence o a i icial in elligence (AI) ools based on la ge language models
(LLMs), pa icula ly Cha GPT, on uni e si y lea ning is no ye ully unde s ood. While
a me a-analysis o expe imen al s udies by Deng e al. (2025) inds a posi i e impac
o Cha GPT on academic pe o mance and highe -o de hinking, o he s udies ind
nega i e e ec s on easoning and a gumen a ion (S adle e al., 2024) and c i ical
hinking (Ge lich, 2025; Valcea e al., 2024). Howe e , i seems o be uncon es ed ha
he use o AI ools such as Cha GPT has an impac on bo h eading and w i ing habi s
o s uden s, as i can summa ize complica ed ex s and w i e academic o a leas
academic-sounding ex s on a ious opics. Ra selj e al. (2025, p. 3) ind in hei
su ey o 23,218 s uden s om 109 coun ies and e i o ies ha s uden s ind
Cha GPT use ul o “simpli ying complex in o ma ion and summa izing con en ” and
al eady widely use i o eading and w i ing ex s, while belie ing i will inc ease hei
chance o achie ing be e g ades. Valcea e al. (2024) a gue ha he use o AI o
lowe -le el asks which o me ly equi ed eading and w i ing skills by s uden s leads
o weakened highe -o de hinking skills, so ha AI boos s he pe o mance o expe s
bu es ains he lea ning e o s o no ices. Fo uni e si y eache s, his means ha a
ask in ol ing a simple li e a u e analysis is now o limi ed alue, as i migh be easily
pe o med wi h he help o Cha GPT o simila AI ools.
Fig. 1. Main P oblems in Academic W i ing (n = 25, up o h ee possible answe s)
2 WORKSHOP OBJECTIVES
2.1 Ta ge audience
Faced wi h he challenges o dec easing eading and w i ing abili ies and he h ea s
posed by Cha GPT, his wo kshop add esses he possibili ies o eaching academic
w i ing o enginee ing s uden s in o de o inc ease hei li e acy. Thus, he a ge
audience in a na ow sense a e educa o s and lec u e s who a e aced wi h he
challenge o each academic w i ing unde he al e ed condi ions. Howe e , since
many enginee ing cou ses ely on li e acy, he wide a ge audience includes
educa o s and lec u e s wi h cou ses including some o m o eading assignmen o
w i ing assessmen . In he wides sense, he a ge audience includes e e y lec u e
conce ned abou he quali y and esul s o he inal hesis o hei s uden s which up
ill now equi es in ensi e eading and w i ing compe ence.
2.2 Expec ed lea ning ou comes
The wo kshop is a quali a i e esea ch app oach o ga he he expe iences and
pe spec i es o he pa icipan s, which can be seen as a i s s ep and a p epa a ion
o possible u he quan i a i e esea ch. The lea ning ou comes o his wo kshop a e
wo- old: The i s a ge is o ga he expe iences om educa o s om di e en
coun ies as a ounda ion o lea n om he di e se expe iences and o be e
unde s and he challenge and i s implica ions o enginee ing educa ion. The second
a ge is o collec and discuss possible solu ion app oaches o each academic w i ing
o enginee ing s uden s unde he al e ed ci cums ances o declining li e acy and he
dis up ion by Cha GPT. The possible solu ions should ocus on h ee co e a eas:
• S eng hening eading compe ence
• De eloping academic w i ing skills
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Li e a u e esea ch and e alua ion o sou ces
Ci a ion
A icula ing my own hough s
Summa izing o he sou ces
Tex s uc u e
Mo i a ion
A gumen a ion
Time
Academic/ echnical language
Spelling and g amma
O he s
• Subs i u ing o eading and w i ing assessmen s
These ca ego ies co e he e ec s on lea ning ou comes as well as he impac on he
me hod o examina ion.
3 WORKSHOP DESIGN
3.1 Time plan
The ime able o he wo kshop is displayed in able 1. A e a sho in oduc ion o gi e
an impulse o he opic consis ing o key e ms, cu en da a, and a de ini ion o he
challenge, he pa icipan s will discuss hei expe iences wi h academic w i ing in small
g oups. The esul s will be sha ed and discussed in he en i e g oup a e wa d o
es ablish a backg ound o he discussion abou possible solu ions. This discussion
will, again, ake place as a small g oup exe cise, be o e he main esul s will be sha ed
wi h he en i e g oup o a inal discussion and conclusion o he wo kshop.
Table 1. Time plan
Run ime
Ac i i y
No es
4 min
Welcome & In oduc ion
In oducing key e ms
Cla i y wo kshop objec i es.
7 min
Sho Inpu & bes p ac ice
B ie p esen a ion on he issue: Wha
de iciencies exis ? Why a e hese
compe encies c ucial? Examples om
di e en coun ies (P esen a ion o da a
om PISA/PIAAC)
12 min
In e ac i e ac i i y 1
Small g oup exe cise (3-5 pa icipan s):
Discuss expe ience wi h academic
w i ing om you uni e si y
Resul s documen ed on lipcha s o
digi al boa d (e.g. Mi o)
7 min
Discussion
Sha e expe iences wi h he whole g oup
20 min
In e ac i e ac i i y 2
Small g oup exe cise: Possible solu ions
o each s uden s academic w i ing
Guiding ques ions: Wha me hods ha e
been e ec i e? Wha new app oaches
could help?
10 min
Discussion and conclusions
Sha e solu ions wi h he g oup
3.2 In e ac i i y
The wo kshop con ains wo in e ac i e ac i i ies in he o m o small g oup exe cises.
In he i s ac i i y expe iences wi h academic w i ing a e discussed in small g oups o
lay he ounda ion o he nex phase. In he second in e ac i e phase, possible
solu ions o each s uden s academic w i ing will be collec ed and discussed.
4 WORKSHOP RESULTS
In he wo kshop, 24 people om he con e ence pa icipa ed. The pa icipan s mainly
came om Eu ope (Denma k, Finland, Ge many, he Ne he lands, Swi ze land,
Tu key, he UK), bu a small mino i y came om ou side o Eu ope (USA, Aus alia).
In he i s discussion abou he expe iences, i became clea ha all pa icipan s ha e
simila expe iences as he ones exp essed in chap e 1.2. The ollowing esul s
eme ged om he g oup discussions abou possible solu ions in he h ee gi en
ca ego ies:
• S eng hening eading compe ence: Possible solu ion app oaches a he
cou se le el a e o discuss he impo ance o eading and he connec ion o
w i ing wi h s uden s, o each eading s a egies, and o make s uden s ead
mo e du ing lec u e ime. To assis his pu pose, he e can be equen
eedback, eading quizzes, o sho checks o comp ehension o de e mine ex
unde s anding and mo i a e eading. Ou side o cou ses, i could help o o m
eading g oups o o ganized eading sessions o o e come he ime cons ain s
du ing lec u es, and o o e speci ic AI aining o lec u e s.
• De eloping academic w i ing skills: One possible solu ion a he s udy p og am
le el is o de elop guidelines o p o essional AI use and o le s uden s decla e
hei AI use in pape s. Possible solu ion app oaches a he cou se le el a e o
discuss he impo ance o w i ing (“w i ing as hinking”), o le s uden s w i e in
class o p ac ice hei skills, o use AI in he cou se as an assis i e ool o e en
o co-c ea ion du ing he w i ing p ocess as well as o le s uden s co ec he
mis akes o AI-w i en ex s so ha s uden s lea n he possibili ies and
limi a ions o AI use, and o use na a i e explana ions on calcula ions o
(science) ic ion s o ies o each s uden s w i ing skills.
• Subs i u ing o eading and w i ing assessmen s: Possible solu ions in his
ca ego y a e o mo e om w i en o o al assessmen s o p esen a ions, o
conduc pape exams in class wi hou AI assis ance, and o ha e au oma ed o
pee eedback on w i en ex s. A downside o his app oach is ha some
compe ences migh ge los i he me hod o examina ion changes. As one
g oup o pa icipan s w i es: “Can’ eplace all w i ing assessmen s because
s uden s a e s ill equi ed o w i e a hesis a he end o hei deg ees”.
In he ensuing discussion he ollowing h ee h eads eme ged:
• See he impac o AI ools such as Cha GPT as analogous o he impac o he
pocke calcula o on ma hema ics o he Global Posi ioning Sys em (GPS) on
na iga ion and de elop academic w i ing assessmen s ha ake his ac in o
accoun (e.g. allow AI use o spelling and ph asing).
• Go in o he o ensi e and inco po a e Cha GPT in he lea ning p ocess o
academic w i ing o each s uden s a basic unde s anding as well as he p os
and cons o he ool in his ega d.
• Focus mo e on he s uden s who a e willing o lea n academic w i ing and don’
design he cou se a ound he unwilling ones o close loopholes (As one
pa icipan pu i : “I don’ wan o spend my ime policing s uden s who don’
wan o lea n.”).
One inal insigh o he wo kshop is ha he echnological de elopmen canno be
e e sed and ha he use o AI in academic w i ing will no diminish. The e o e, he e
is an u gen need o he academic communi y o ind solu ions, i we s ill wan s uden s
o be able o unde s and and w i e academic ex s.
5 ACKNOWLEDGEMENTS
The au ho s would like o hank he pa icipan s o he wo kshop o hei ac i e
coope a ion and hei aluable inpu .
REFERENCES
Camb idge Assessmen (2013). Wha is li e acy? An in es iga ion in o de ini ions o
English as a subjec and he ela ionship be ween English, li e acy and ‘being li e a e’.
h ps://www.camb idgeassessmen .o g.uk/Images/130433-wha -is-li e acy-an-
in es iga ion-in o-de ini ions-o -english-as-a-subjec -and- he- ela ionship-be ween-
english-li e acy-and-being-li e a e-.pd
Deng, R., Jiang, M., Yu, X., Lu, Y., & Liu, S. (2025). Does Cha GPT enhance s uden
lea ning? A sys ema ic e iew and me a-analysis o expe imen al s udies. Compu e s
& Educa ion, 227, A icle 105224. h ps://doi.o g/10.1016/j.compedu.2024.105224
DiYanni, R. (2022). You A e Wha You Read. A P ac ical Guide o Reading Well.
P ince on Uni e si y P ess.
Ge lich, M. (2025). AI Tools in Socie y: Impac s on Cogni i e O loading and he Fu u e
o Thinking. Socie ies, 15(1), A icle 6. h ps://doi.o g/10.3390/soc15010006
Heine, J.-H., Heinle, M., Hahnel, C., Lewal e , D., & Becke -M o zek, M. (2023).
Lesekompe enz in PISA 2022: E gebnisse, Ve ände ungen und Pe spek i en
[Reading Compe ence in PISA 2022: Resul s, Changes and Pe spec i es]. In D.
Lewal e , J. Died ich, F. Goldhamme , O. Kölle , & K. Reiss (eds.), PISA 2022:
Analyse de Bildungse gebnisse in Deu schland (pp. 139-162). Waxmann.
Kaag, J., & an Belle, J. (2024). Thinking Th ough W i ing: A Guide o Becoming a
Be e W i e and Thinke . P ince on Uni e si y P ess.
Min z, S. (2021). W i ing Is Thinking. Inside Highe Ed.
h ps://www.insidehighe ed.com/blogs/highe -ed-gamma/w i ing- hinking
Oa ley, K., & Djikic, M. (2008). W i ing as Thinking. Re iew o Gene al Psychology,
12(8), 9-27.
OECD (2020). OECD Lea ning Compass 2030 – Glossa y.
h ps://www.oecd.o g/con en /dam/oecd/en/abou /p ojec s/edu/educa ion-
2040/publica ions/OECD%20Lea ning%20Compass%202030%20-%20Glossa y.pd
OECD (2023). PISA 2022 Resul s (Volume I): The S a e o Lea ning and Equi y in
Educa ion. OECD Publishing. h ps://doi.o g/10.1787/53 23881-en
OECD (2024). Do Adul s Ha e he Skills They Need o Th i e in a Changing Wo ld?:
Su ey o Adul Skills 2023. OECD Publishing. h ps://doi.o g/10.1787/b263dc5d-en
Ra selj, D., Ke zic, D., Tomaze ic, N., Umek, L. B ezo a , N., Iahad, N. A., Abdulla,
A., Akopyan, A., Segu a, M. W. A., AlHumaid, J., Allam, M. F., Alló, M., Andoh, R. P.
K. A., And onic, O., A hu , Y. A., Aydin, F., Bad an, A., Balbon in-Al a ado, R., Saad,
H. B., …, A is o nik, A. (2025). Highe educa ion s uden s’ pe cep ions o Cha GPT:
A global s udy o ea ly eac ions. PLoS ONE, 20(2), A icle e0315011.
h ps://doi.o g/10.1371/jou nal.pone.0315011
S adle , M., Banne , M., & Saile , M. (2024). Cogni i e ease a a cos : LLMs educe
men al e o bu comp omise dep h in s uden scien i ic inqui y. Compu e s in Human
Beha io , 160, A icle 108386. h ps://doi.o g/10.1016/j.chb.2024.108386
Valcea, S., Hamdani, M. R., & Wang, S. (2024). Explo ing he Impac o Cha GPT on
Business School Educa ion: P ospec s, Bounda ies, and Pa adoxes. Jou nal o
Managemen Educa ion, 48(5), 915–947.
h ps://doi.o g/10.1177/10525629241261313
