P ac ice Pape
Recommended ci a ion: Walpi a Gamage, S., & Richa ds, A. (2025). Assessing
The Assessmen s: A Case S udy on he Rela ionship Be ween Theo e ical,
Con inuous, and P ac ical E alua ions in an Enginee ing Cou se. 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.17631501.
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.
ASSESSING THE ASSESSMENTS: A CASE STUDY ON THE
RELATIONSHIP BETWEEN THEORETICAL, CONTINUOUS, AND
PRACTICAL EVALUATIONS IN AN ENGINEERING COURSE
S H P W Gamage a,1, A Richa ds b
a The Uni e si y o Sou h Aus alia, Mawson Lakes, Aus alia, h ps://o cid.o g/0000-
0001-9209-9113
b The Uni e si y o Sou h Aus alia, Mawson Lakes, Aus alia, h ps://o cid.o g/0009-
0008-1170-5651
Con e ence Key A eas: Cu iculum De elopmen and Eme ging Cu iculum Models
in Enginee ing
Keywo ds: Enginee ing educa ion, Cu iculum de elopmen , Assessmen design,
S uden pe o mance, P ac ice-o ien ed lea ning
ABSTRACT
As enginee ing educa o s s i e o balance heo e ical knowledge wi h p ac ical
skills, unde s anding how assessmen o ma s d i e lea ning ou comes is mo e
impo an han e e . This s udy in es iga es he ela ionships be ween di e en
assessmen ypes—CSWA Tes ( heo e ical), Con inuous Assessmen (quizzes),
and Final D awing (p ac ical)—in a i s -yea unde g adua e enginee ing cou se.
Using da a om 190 s uden s, he s udy applies desc ip i e s a is ics, Pea son
co ela ion analysis, mul iple eg ession modelling, and pe o mance classi ica ion o
e alua e how hese assessmen s ela e o one ano he and p edic s uden success.
Findings e eal ha Con inuous Assessmen has he s onges co ela ion wi h Final
D awing pe o mance ( = 0.49) and was he mos eliable p edic o (β = 0.5689, R² =
0.240, p < 0.001), sugges ing ha equen , low-s akes quizzes (con inuous
assessmen s) suppo he de elopmen o p ac ical skills. In con as , he weakes
ela ionship was be ween CSWA Tes and Final D awing ( = 0.41), indica ing ha
s ong heo e ical pe o mance does no consis en ly ansla e o applied p o iciency.
These indings highligh he alue o con inuous assessmen s and applied lea ning in
enginee ing educa ion and emphasise he need o ailo ed suppo o accommoda e
di e se s uden needs. The s udy o e s guidance o cu iculum designe s and
sugges s a enues o u u e esea ch, such as explo ing he oles o p io expe ience
and adap i e lea ning ools in s uden success.
1 S H P W Gamage
si ha a.walpi [email protected]
1 INTRODUCTION
Enginee ing educa ion is inc easingly expec ed o p epa e s uden s no only o
unde s and heo e ical p inciples bu also o apply hem in eal-wo ld p oblem-sol ing
con ex s. Howe e , adi ional app oaches ha e o en leaned hea ily on summa i e,
heo y-based assessmen s such as inal exams, which p ima ily e alua e s uden s’
concep ual unde s anding and analy ical abili ies. While hese me hods ha e alue,
hey may no ully cap u e a s uden ’s abili y o ans e heo e ical knowledge in o
p ac ical enginee ing asks, which is c i ical in p o essional se ings (Dym e al.,
2005; P ince & Felde , 2006).
To add ess his, educa ional models such as design-based lea ning (DBL) and ac i e
lea ning (AL) ha e eme ged as powe ul al e na i es. DBL in eg a es i e a i e design
challenges in o cou sewo k, p omp ing s uden s o apply enginee ing p inciples in
eal- ime p oblem-sol ing scena ios (Kolodne , 2002). Fo example, Hmelo, Hol on,
and Kolodne (2000) ound ha s uden s in ol ed in design p ojec s de eloped a
mo e sys ema ic unde s anding o complex sys ems compa ed o hose in adi ional
lea ning se ings. Mehalik, Doppel , and Schunn (2008) also ound ha s uden s
engaged in DBL ac i i ies demons a ed signi ican imp o emen s in science concep
lea ning compa ed o hose in sc ip ed inqui y se ings ( eache -domina ed
en i onmen s whe e s uden s ollow p ede e mined p ocedu es o disco e p e-
de ined answe s). This app oach has been shown o p omo e deepe lea ning,
enhance concep ual unde s anding, and suppo he de elopmen o p ac ical skills
(Doppel , 2003). Simila ly, AL encou ages s uden engagemen h ough collabo a i e
ac i i ies, eal- ime eedback, and pee ins uc ion. A me a-analysis by F eeman e
al. (2014) epo ed ha AL can educe ailu e a es in STEM educa ion by o e 10%
and signi ican ly imp o e s uden achie emen .
Aligned wi h hese pedagogical inno a ions, assessmen s a egies a e also
e ol ing. Con inuous assessmen (CA)—which includes quizzes, assignmen s, and
o he o ma i e eedback ools—has been widely ecognised as a means o
ein o cing lea ning, p omo ing e en ion, and iden i ying pe o mance gaps ea ly in
he lea ning p ocess (Nicol & Mac a lane-Dick, 2006; B own e al., 2014; Li zinge e
al., 2011). CA suppo s sel - egula ed lea ning and enhances s uden mo i a ion,
pa icula ly in cou ses ha in ol e complex echnical con en (Mills & T eagus ,
2003).
Meanwhile, p ac ical assessmen s such as enginee ing d awing and Compu e -
Aided Design (CAD)-based asks e alua e essen ial p o essional skills, including
design p ecision, and isual communica ion (So by, 2009; Ches e , 2007). These
skills a e c i ical o enginee ing g adua es, ye esea ch shows ha many s uden s
who pe o m well in heo e ical exams o en s uggle wi h hese hands-on asks
(Ekwueme, Ekon, & Ezenwa-Nebi e, 2015). This disconnec sugges s a need o
be e align assessmen s wi h he compe encies expec ed in indus y.
Despi e he g owing adop ion o inno a i e pedagogies such as DBL and AL,
assessmen p ac ices in enginee ing educa ion ha e no e ol ed a he same pace.
Much o he cu en li e a u e explo es hese ins uc ional s a egies in isola ion, ye
he e is limi ed empi ical e idence examining how di e en ypes o assessmen s—
namely heo e ical exams, con inuous assessmen s, and p ac ical e alua ions—
in e ac o in luence s uden pe o mance. While s udies ha e highligh ed he
bene i s o DBL in enhancing sys ems hinking and concep ual unde s anding
(Hmelo e al., 2000; Mehalik e al., 2008) and he alue o con inuous assessmen in
ein o cing lea ning and engagemen (Nicol & Mac a lane-Dick, 2006; B own e al.,
2014), ew in es iga ions ha e quan i a i ely compa ed he ela ionships be ween
hese assessmen ypes wi hin he same lea ning con ex .
Mo eo e , esea ch o en ea s heo e ical knowledge and p ac ical applica ion as
sepa a e domains, wi h limi ed a en ion gi en o whe he pe o mance in one
assessmen a ea p edic s o aligns wi h pe o mance in ano he . As a esul , i
emains unclea whe he success in adi ional exams accu a ely e lec s a s uden 's
eadiness o applied enginee ing asks, such as CAD modelling and echnical
d awing. I is also unce ain whe he con inuous assessmen s, which a e designed o
suppo lea ning inc emen ally, can e ec i ely b idge he gap be ween concep ual
mas e y and p ac ical compe ence.
This s udy seeks o add ess hese gaps by conduc ing a compa a i e analysis o
s uden pe o mance ac oss h ee assessmen ypes— he Ce i ied SolidWo ks
Associa e (CSWA) Tes ( heo e ical), Con inuous Assessmen (quizzes), and Final
D awing (p ac ical), wi hin a i s -yea enginee ing cou se. By explo ing how hese
assessmen s ela e o each o he , he s udy con ibu es o a mo e in eg a ed
unde s anding o how assessmen design can suppo bo h heo e ical
unde s anding and p ac ical skill de elopmen .
This s udy is guided by he ollowing esea ch ques ions:
1. How do di e en assessmen ypes (CSWA Tes , Con inuous Assessmen , and
Final D awing) co ela e wi h each o he in e alua ing s uden pe o mance?
2. To wha ex en does Con inuous Assessmen p edic success in Final D awing
compa ed o CSWA Tes sco es?
3. Wha a e he p ima y pe o mance pa e ns among s uden s, and how do hey
di e be ween hose excelling in heo e ical e sus p ac ical assessmen s?
Th ough hese ques ions, he s udy con ibu es o a deepe unde s anding o how
assessmen design can be op imised o be e e lec and suppo s uden lea ning in
enginee ing educa ion, pa icula ly in cou ses ha aim o balance heo e ical
ins uc ion wi h hands-on echnical compe ence.
2 CONTEXT AND PRACTICAL WORK
2.1 Backg ound
This s udy ocuses on i s -yea enginee ing cou ses designed o de elop s uden s’
enginee ing d awing skills. The cou se in eg a es heo e ical and p ac ical
componen s o c ea e a comp ehensi e lea ning expe ience. Theo e ical ins uc ion
is deli e ed h ough lec u es and u o ials, wi h lec u es p ima ily in ol ing passi e
lea ning and u o ials inco po a ing c i ical hinking asks o os e ac i e
engagemen . The p ac ical componen cen es on a design-and-build ac i i y, whe e
s uden s c ea e and manu ac u e a p oduc , applying enginee ing p inciples o
p oduce a unc ional a i ac . Al hough in e media e s eps such as pa d awings and
machining a e no di ec ly assessed, s uden s submi a inal se o enginee ing
d awings o e alua ion. This hands-on app oach highligh s he eal-wo ld
impo ance o p ecision and accu acy in echnical d awings, enabling s uden s o
expe ience he consequences o imp ecise wo k while de eloping key p oblem-
sol ing skills.
Assessmen in he cou se includes an in igila ed, compu e -based d awing exam
( he CSWA es , 25% o he inal g ade), weekly CAD-based quizzes (15% o al), a
inal echnical d awing submission (10%), and p o o ype building, p esen a ions, and
o he ac i i ies (50%, no de ailed he e). Con inuous assessmen s happen in he i s
pa o he semes e , hen he CSWA es , hen he inal d awings, in ha o de . The
quizzes designed o p o ide immedia e eedback o suppo i e a i e lea ning. All 12
quizzes con ibu e o he g ade: six allow unlimi ed a emp s, while he o he six a e
limi ed o wo a emp s wi hin a 30-minu e window o encou age p ecision and ime
managemen . Fo each quiz, only he highes sco e is eco ded.
2.2 Me hodology
This s udy adop s a quan i a i e app oach o examine he ela ionships be ween
di e en assessmen ypes and s uden pe o mance in a i s -yea enginee ing
cou se. The analysis is based on anonymised da a om 190 s uden s who
comple ed h ee key assessmen s.
Desc ip i e s a is ics, including mean, median, and s anda d de ia ion, we e
calcula ed o explo e o e all pe o mance ends and iden i y a eas whe e s uden s
ypically excelled o s uggled. Sco e dis ibu ions we e isualised using his og ams
and densi y plo s o gain u he insigh in o he sp ead and concen a ion o s uden
pe o mance ac oss each assessmen ype. Following his, Pea son’s co ela ion
analysis was conduc ed o assess he s eng h and di ec ion o he ela ionships
be ween he CSWA es , con inuous assessmen quizzes, and he inal d awing
submission. To u he explo e he p edic i e powe o heo e ical and con inuous
assessmen s on p ac ical pe o mance, a mul iple eg ession analysis was
conduc ed.
To gain mo e unde s anding o s uden lea ning p o iles, a pe o mance classi ica ion
app oach was also employed. S uden s we e g ouped based on a 50% h eshold
in o ou ca ego ies: 1) hose who pe o med s ongly in he CSWA es bu poo ly in
he Final D awing; 2) hose who pe o med well in he Final D awing bu no in he
CSWA es ; 3) hose who pe o med well in bo h; and 4) hose who s uggled in bo h
a eas. This classi ica ion helped o highligh di e ing s eng hs among s uden s and
iden i y whe e addi ional ins uc ional suppo migh be needed.
3 RESULTS AND INSIGHTS
The analysis o s uden pe o mance ac oss CSWA Tes s, Con inuous Assessmen ,
and Final D awing e eals key ends in heo e ical and p ac ical lea ning. The
dis ibu ion o sco es (Fig.1) highligh s signi ican a ia ion in s uden pe o mance.
CSWA Tes sco es we e widely dis ibu ed, wi h a subs an ial numbe o s uden s
sco ing below 50%, indica ing challenges in heo e ical exam pe o mance.
Con inuous Assessmen sco es, howe e , we e skewed owa ds highe alues,
sugges ing ha s uden s gene ally ound quizzes mo e manageable and bene icial
o lea ning e en ion. In con as , Final D awing sco es we e he lowes , wi h many
s uden s sco ing below 50%, ein o cing he di icul y o p ecision enginee ing
d awing. These indings sugges ha s uden s bene i om equen , s uc u ed
assessmen o ma s bu may equi e addi ional ins uc ional suppo in enginee ing
d awing and CAD-based asks o b idge he gap be ween heo e ical knowledge and
p ac ical applica ion.
Fig. 1. Dis ibu ion o Sco es in Th ee Assessmen s
The co ela ion analysis (Table 1) p o ides insigh s in o he ela ionships be ween
assessmen ypes. The s onges co ela ion was obse ed be ween Con inuous
Assessmen and Final D awing ( = 0.49), sugges ing ha equen , low-s akes
assessmen s suppo s uden s in applying enginee ing concep s p ac ically. The
co ela ion be ween CSWA Tes and Con inuous Assessmen ( = 0.47) was sligh ly
weake , indica ing ha s uden s who pe o m well in quizzes also end o do well in
heo e ical exams. In con as , he weakes co ela ion was be ween CSWA Tes and
Final D awing ( = 0.41), ein o cing he no ion ha heo e ical exam success does
no necessa ily ansla e o s ong enginee ing d awing pe o mance.
Table 1. The pai wise Pea son co ela ion coe icien s
Sca e plo s (Fig. 2) illus a e hese ela ionships and e eal conside able da a
dispe sion, e lec ing mode a e co ela ions and no able a ia ion in indi idual
pe o mance. This a iabili y could be due o addi ional ac o s such as lea ning
s yles, p io expe ience, and p oblem-sol ing app oaches, which in luenced s uden
success.
CSWA
Tes (%)
Con inuous
Assessmen (%)
Final D awing
(%)
CSWA Tes (%)
1
0.47
0.41
Con inuous Assessmen (%)
0.47
1
0.49
Final D awing (%)
0.41
0.49
1
Fig. 2. Sca e Plo s Be ween Each Assessmen Componen s
To u he analyse s uden pe o mance, s uden s we e classi ied based on a 50%
h eshold in o ou ca ego ies (Fig. 3): S ong in CSWA bu Weak in Final D awing
(36 s uden s), S ong in Final D awing bu Weak in CSWA (32 s uden s), S ong in
Bo h (48 s uden s), and Weak in Bo h (74 s uden s). The la ges ca ego y comp ised
s uden s who s uggled in bo h assessmen s, highligh ing he need o a ge ed
ins uc ional in e en ions. A no able end was ha 19% o s uden s pe o med well
in CSWA bu s uggled in Final D awing, sugges ing ha heo e ical knowledge alone
does no gua an ee p o iciency in enginee ing d awing. Con e sely, 17% s uden s
excelled in Final D awing bu unde pe o med in CSWA, indica ing ha some
s uden s h i e in applied lea ning en i onmen s bu may s uggle wi h heo e ical
assessmen s. These indings emphasise he impo ance o di e si ied eaching and
assessmen me hods ha in eg a e heo e ical ins uc ion wi h hands-on lea ning
expe iences o suppo di e en s uden lea ning p e e ences.
These indings highligh he impo ance o di e en ia ed ins uc ional s a egies
ailo ed o s uden s’ speci ic lea ning needs. Fo hose s uggling wi h Final D awing,
a ge ed in e en ions—such as CAD-based design wo kshops, s uc u ed eedback
cycles, and s ep-by-s ep u o ials inco po a ing spa ial isualisa ion echniques o
suppo men al modelling—can enhance d awing p ecision and p ac ical applica ion.
Con e sely, s uden s who pe o m well in d awing bu ace challenges wi h CSWA
Tes s may bene i om concep ual ein o cemen h ough p oblem-sol ing asks,
in e ac i e simula ions, and guided discussions aimed a deepening hei
unde s anding o enginee ing design p inciples.
Fig. 3. Pe o mance Classi ica ion o S uden s (Th eshold 50%)
To quan i y hese ela ionships, eg ession analysis (Table 2) was conduc ed o
de e mine he p edic i e s eng h o di e en assessmen s. The model examining
Con inuous Assessmen and CSWA Tes pe o mance esul ed in an R² alue o
0.225 and a eg ession coe icien (β) o 0.5413 (p < 0.001), indica ing ha a 1%
inc ease in Con inuous Assessmen sco es is associa ed wi h a 0.54% inc ease in
CSWA Tes sco es. This sugges s ha while con inuous assessmen s ein o ce
heo e ical lea ning, addi ional ac o s such as o he exam s a egies also con ibu e
o CSWA pe o mance. The s onges ela ionship was obse ed be ween
Con inuous Assessmen and Final D awing, wi h an R² alue o 0.240, β = 0.5689, p
< 0.001. This sugges s ha s uden s who pe o m well in con inuous assessmen s
also end o excel in p ac ical applica ions. Howe e , he sca e plo shows
conside able a ia ion, meaning ha while quizzes ein o ce lea ning, hey do no
ully gua an ee s ong enginee ing d awing pe o mance. A 1% inc ease in
Con inuous Assessmen sco es is associa ed wi h a 0.57% inc ease in Final D awing
sco es, making his he mos p edic i e ac o o s uden s' p ac ical success,
ein o cing he impo ance o s uc u ed, i e a i e lea ning in de eloping applied
enginee ing skills. In con as , he weakes co ela ion was ound be ween Final
D awing and CSWA Tes sco es (R² = 0.169, β = 0.4032, p < 0.001), indica ing ha
s ong pe o mance in inal enginee ing d awing does no consis en ly align wi h high
sco es in heo e ical assessmen s. This inconsis ency may also e lec ex e nal
ac o s, such as s uden s ecei ing assis ance on hei inal d awing asks. The da a
dispe sion obse ed in he sca e plo suppo s his in e p e a ion, con i ming ha
p ac ical d awing skills alone a e no a eliable p edic o o heo e ical exam
pe o mance, and ice e sa. These indings highligh he impo ance o a balanced
cu iculum ha in eg a es bo h heo e ical p oblem-sol ing and hands-on lea ning
expe iences.
Table 2. Reg ession Analysis
Reg ession Model
Coe icien (𝛽𝛽)
𝑅𝑅2 Value
P Value
Con inuous Assessmen s
CSWA
0.5413
0.225
< 0.001
Con inuous Assessmen s
Final D awing
0.5689
0.24
< 0.001
Final D awing s CSWA Tes
0.4032
0.169
< 0.001
4 CONCLUSIONS AND IMPLICATIONS
This s udy examined he ela ionships be ween in igila ed CSWA Tes sco es, non-
in igila ed Con inuous Assessmen , and Final D awing pe o mance o e alua e how
di e en assessmen ypes in luence s uden success in a i s -yea enginee ing
cou se. The indings highligh he alue o con inuous, low-s akes assessmen s in
suppo ing applied lea ning ou comes. S uden s pe o med bes in Con inuous
Assessmen bu s uggled signi ican ly wi h he Final D awing ask, sugges ing ha
while s uc u ed quizzes help ein o ce heo e ical knowledge, applying ha
knowledge h ough p ecise echnical d awings emains a challenge o many.
Co ela ion analysis con i med ha Con inuous Assessmen had he s onges
associa ion wi h Final D awing pe o mance ( = 0.49), while he ela ionship
be ween CSWA and Final D awing was he weakes ( = 0.41). These pa e ns,
along wi h high da a dispe sion, indica e ha s uden pe o mance could be shaped
by mul iple ac o s.
The esul s ca y se e al impo an implica ions o cu iculum design and s uden
suppo in enginee ing educa ion. Since Con inuous Assessmen eme ged as he
bes p edic o o applied pe o mance, educa o s should embed s uc u ed, hands-on
ac i i ies, such as CAD exe cises, spa ial- isualisa ion ac i i ies, mini design
p ojec s, and simula ions h oughou he e m o ein o ce heo y and build
ans e able skills. Se en y- ou s uden s unde pe o med ac oss bo h heo e ical
and p ac ical asks, unde sco ing he need o ea ly, a ge ed suppo . D awing on
Ha ie and Timpe ley’s (2007) indings ha imely, ac ionable eedback d ama ically
accele a es lea ning, o e ing emedial CAD wo kshops, pe sonalised u o ials, and
pee -lea ning coho s du ing he i s hal o semes e can help a - isk lea ne s be o e
hey ace high-s akes assessmen s. A he same ime, balancing assessmen modes
is c ucial. S uden s who excel a Final d awing bu unde pe o m on CSWA es s
could gain om mo e a ge ed concep ‐based quizzes and p oblem-sol ing
exe cises. Con e sely, s uden s who a e s ong in heo y would bene i om applied
design challenges, ensu ing ha bo h analy ical easoning and c ea i e skills a e
assessed equi ably (C uz, Saunde s-Smi s, & G oen, 2019).
While he s udy p o ides aluable insigh s, i is limi ed by i s ocus on a single coho
wi hin one cou se, which may cons ain he gene alisabili y o esul s. Fu he mo e,
he s udy did no con ol o ex e nal a iables ha could a ec s uden pe o mance,
and he absence o quali a i e da a limi s unde s anding o s uden pe spec i es.
Fu u e esea ch should in eg a e in e iews o su eys o unco e how lea ne s
expe ience di e en assessmen ypes and examine addi ional p edic o s o
pe o mance. Mo eo e , e alua ing adap i e, AI-d i en u o ing sys ems (Gligo ea e
al., 2023) could e eal how eal- ime eedback and pe sonalised con en impac bo h
heo e ical unde s anding and p ac ical skill de elopmen . Ul ima ely, a well-balanced
enginee ing cu iculum—combining equen con inuous assessmen s, s uc u ed
design asks, and ea ly pe sonalised suppo —can b idge he gap be ween
concep ual knowledge and p ac ical compe ence, os e ing g ea e s uden
engagemen and success.
ETHICS APPROVAL
This esea ch was app o ed by he Human E hic Commi ee, Uni e si y o Sou h
Aus alia unde e e ence numbe 206958.
REFERENCES
Black, P., & Wiliam, D. (2009). De eloping he heo y o o ma i e assessmen .
Educa ional Assessmen , E alua ion and Accoun abili y, 21(1), 5–31.
h ps://doi.o g/10.1007/s11092-008-9068-5
B own, P. C., Roedige , H. L., & McDaniel, M. A. (2014). Make i s ick: The science
o success ul lea ning. Ha a d Uni e si y P ess.
Ches e , I. (2007). Teaching o CAD expe ise. In e na ional Jou nal o Technology
and Design Educa ion, 17(1), 23–35. h ps://doi.o g/10.1007/s10798-006-9015-z
