P ac ice Pape
Recommended ci a ion: Kon o, I., & Kuulu ainen, H. (2025). Use O Video
Ma e ials on a Flipped-Class oom Physics 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.17631381.
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.
USE OF VIDEO MATERIALS ON A FLIPPED-CLASSROOM PHYSICS
COURSE
I. Kon o a,
1
, H. Kuulu ainen b
a Physics Uni , Tampe e Uni e si y, Tampe e, Finland, 0000-0002-1874-3756
b Physics Uni , Tampe e Uni e si y, Tampe e, Finland, 0000-0002-6003-3547
Con e ence Key A eas: Teaching ma hema ics and physics in enginee ing
educa ion, Digi al ools and AI in enginee ing educa ion
Keywo ds: physics, lipped class oom, ideo
ABSTRACT
We s udied he use o ideo ma e ials on a lipped class oom enginee ing physics
cou se. The cou se’s ideo ma e ials consis ed o heo y ideos, wo ked examples
and mo i a ional ideos. We ound ha mos s uden s wa ched he heo y ideos
comple ely o almos comple ely (median pe cen age wa ched 96%), bu only
pa ially he ideos wi h wo ked examples (median 40%). Mos s uden s did no
wa ch he mo i a ional ideos (median 0%). The e we e no s a is ically signi ican
di e ences in he iewing me ics o s uden s o di e en majo s o gende s. The
di e en ideo ypes we e used di e en ly: he iewing da a o heo y ideos had
mo e sea ching and sho iews whe eas he mo i a ional ideo, when wa ched, was
iewed o longe consecu i e amoun s o ime. Viewing heo y ideos co ela ed
mode a ely wi h exam poin s (Spea man’s 𝜌 = 0.43). Viewing he o he ypes o
ideo ma e ials had weak o no co ela ions wi h exam poin s.
1
Co esponding Au ho
I. Kon o
inke i.kon o@ uni. i
1 INTRODUCTION
The lipped class oom is an ins uc ional model, in which adi ional class oom
ac i i ies such as lec u ing a e ans e ed o indi idual s udy a home, and he
p ac ice o he new ma e ial, such as comple ing assignmen s, a e done wi h eache
guidance (Akçayı & Akçayı , 2018).
Flipped-class oom eaching o en u ilizes lec u e ideos. Sho ideos a e
ecommended (B ame, 2017). Fo example, based on da a om a massi e online
open cou se (MOOC), Guo e al. (2014) ecommend ha ideo leng h does no
exceed 6 minu es. Howe e , esul s om MOOCs may no be gene alizable o
deg ee s uden s, whose mo i a ion and engagemen a e likely highe han o non-
deg ee s uden s s udying online. Recen ly, he demand o e y sho ideos has
indeed been c i icized. Seidel (2024) epo ed ha using segmen s in longe ideos
p omo es lea ning compa ed o non-segmen ed ideos, and ha ideo leng h is o
lesse impo ance. Ahn and Chan (2025) showed ha playback speed did no a ec
lea ning, bu s uden s ha e indi idual p e e ences o he s yle o he ideos, which
a ec s hei engagemen o ideos. In in oduc o y uni e si y physics, many opics
a e no easily p esen ed in six minu es. Calcula ions and de i a ions ake ime, and
adding con ex and discussing he esul s leng hens he ideo bu is impo an o
linking he esul s o p io knowledge.
Focusing on concep ual lea ning is impo an in uni e si y-le el physics. T adi ional,
lec u e-based ins uc ion p oduces lowe concep ual unde s anding ha in e ac i e
me hods (Hake, 1998, Fo mic e al., 2010). Many ac i e lea ning s a egies such as
lipped class oom o Jus -in-Time-Teaching, which in oduces p e-assignmen s,
imp o e concep ual knowledge (Fo mic e al., 2010). Howe e , in e ac i e me hods
a e no necessa ily be e a eaching p oblem-sol ing skills (McDaniel e al., 2016).
A a ia ion o he lipped class oom model, p ime ime lea ning, also in oduces
ac i e g oup wo k and he g oup’s e lec ion wi h a eache (Koskinen e al, 2018).
Flipped class oom in gene al and p ime ime lea ning in pa icula has ad an ages
o lea ning. P ime ime lea ning p omo es concep ual lea ning and imp o es pass-
a es in physics (Koskinen e al, 2018).
In his wo k, we s udy how enginee ing s uden s use ideo ma e ials on a lipped-
class oom cou se, which u ilizes ideos on heo y, wo ked examples and
mo i a ional ideos. Speci ically, ou esea ch ques ions (RQ) a e:
• RQ1: How do he s uden s make use o he ideo ma e ial?
• RQ2: Do s uden s wi h di e en backg ound (gende , majo subjec ) di e in
how hey use he ideos?
• RQ3: How does wa ching he ideo ma e ial co ela e wi h lea ning ou comes,
as measu ed by exam esul s?
2 CONTEXT AND PRACTICAL WORK
2.1 Physics 3
Physics 3 is a calculus-based physics cou se o enginee ing s uden s who majo in
echnical physics, chemis y o ma hema ics a Tampe e Uni e si y. Physics 3
co e s in oduc o y elec omagne ism and is based on he book Unde s anding
Physics (Mans ield and O’Sulli an, 2020). The ins uc ion on Physics 3 loosely
ollows he p ime ime model o Koskinen e al. (2018). The s udy week s a s wi h
ideo lec u es, which include a longe ideo on heo y, sho e ideos on wo ked
examples, and a mo i a ional ideo called ”Physicis s’ able”.
Based on he ma e ials, s uden s sol e p e-assignmen s. S uden s also u n in wo
concep s, which hey a e willing o explain, and wo ques ions, which hey ha e no
ound he answe o, in p epa a ion o he p ime ime sessions. This p ocess is
simila o Jus -in-Time-Teaching (Fo mic e al., 2010). Du ing p ime ime, he
s uden s discuss p e ious weeks’ exe cises and model solu ions wi h hei g oup o
4-6 s uden s. They discuss he concep s and ques ions o all pa icipan s, and inally
he solu ions o he p e-exe cises. The eache s pa icipa e in he discussion as
needed.
A e he p ime ime sessions, s uden s sol e calcula ion exe cises ela ing o he
week’s ma e ial ei he in u o ial sessions o indi idually. The exe cises a e handed
in a he end o he u o ial session, a e which he s uden s s a s udying o he
nex week (Fig. 1, op).
The cou se’s g ading includes he inal exam, exe cises and p e-assignmen s,
pa icipa ion in he p ime ime sessions and labo a o y wo k (Fig. 1, bo om). The
inal exam mos ly con ained calcula ion p oblems, bu also one essay ques ion.
Fig. 1. Top: he weekly schedule o he cou se. Bo om: he weigh s o he di e en g ade
componen s.
2.2 Video ma e ials
The ideo ma e ials o he cou se a e di ided in o h ee ca ego ies: 1) Theo y ideos,
2) Wo ked example ideos, and 3) Mo i a ional ideos. Each week o he cou se
includes one lec u e ideo, 2–6 example ideos, and one mo i a ional ideo.
In he heo y ideos, he heo y is explained and isualized by using slides wi h bulle
poin s and igu es as well as isualiza ions and sho de i a ions demons a ed wi h
a ouch sc een. The heo y ideos co e all he cou se opics, and hey ollow he
s uc u e o he cou se handou . Each heo y ideo is di ided in o pa s acco ding o
he sec ion i les and numbe ing o he cou se handou . When wa ching he ideos,
he s uden s can see he con en and sec ion i les and hey can easily jump o he
beginning o each sec ion. The leng h o he heo y ideos a ies om 1h 22 min o
2h 37 min and he slides o he heo y ideos a e a ailable o he s uden s as pd
iles.
Wo ked example ideos demons a e he key p oblem-sol ing echniques and
applica ions o he co e concep s o he cou se. One o he main pu poses o hese
ideos is o each how o apply he Maxwell’s equa ions in p oblem-sol ing, which
includes unde s anding he ole o di e en symme ies, o ming in eg als, and ec o
ield p ope ies. The p oblems a e sol ed in he ideos mainly based on ske ching
igu es and de i ing equa ions by hand on a ouch sc een. The leng h o he example
ideos a ies om 7 min o 29 min and he ma e ial p oduced in he ideos a e
a ailable o he s uden s as pd iles.
Mo i a ional ideos also called “Physicis s’ able” ideos a e addi ional ma e ial in
which some o he challenging concep s a e discussed. In hese ideos, wo physics
eache s discuss 2–3 concep s pe week in a class oom by using a blackboa d and
some demons a ion equipmen o isualiza ion. The ideo o ma is mean o be
somewha en e aining and mo i a ing. The pu pose o hese ideos is o deepen he
s uden s’ unde s anding o he challenging concep s and encou age he s uden s o
discuss by hemsel es du ing he p ime ime. The leng h o he mo i a ional ideos
a ies om 24 min o 38 min and he con en o he ideos is isible o he s uden s
so ha hey can jump di ec ly o he beginning o ce ain concep s o ques ions.
2.3 Pa icipan s and da a
The s udy comp ised he h ee inal weeks o he cou se Physics 3. The s uden s
we e asked o in o med consen o pa icipa e in he s udy. O he app oxima ely 80
s uden s on he cou se, 56 s uden s consen ed o pa icipa e, and only hei da a was
collec ed.
The exam poin s and he ideo iew da a (including ime s amps, minu es wa ched,
numbe o iews and pe cen age comple ed by ideo) we e collec ed. A e
combining he da a, he da a was pseudonymized and analyzed wi hou iden i ying
de ails. All da a analysis was done in R. As he numbe o pa icipan s was small and
he da a gene ally no no mally dis ibu ed, we used non-pa ame ic s a is ical es s:
he Mann-Whi ney U- es o compa ing wo samples, he K uskal-Wallis es o
compa isons o mo e han wo samples, and Spea man’s co ela ion o s udying
co ela ions be ween a iables.
3 RESULTS AND INSIGHTS
3.1 Use o ideos
Fi s we s udied how s uden s made use o he di e en ideos (RQ1). Mos
s uden s wa ched a leas a pa o he ideos. As can be seen om Fig. 2, he
median pe cen age wa ched was 68%. When he da a is sepa a ed by ideo
ype, he heo y ideos ha e a median pe cen age wa ched o 96%. The leas
wa ched ideo ype is he mo i a ional Physicis s’ able, which mos s uden s did
no wa ch. Howe e , many s uden s also wa ched hese ideos in hei en i e y.
Fig. 2. Boxplo o he pe cen age o ideos wa ched by ideo ype.
We also looked a how many imes s uden s wa ched he ideos. Da a om wo
ideos on week 4 show ha no only did mo e s uden s wa ch he heo y ideo, bu
he iewing pa e ns show di e en quali ies (Fig. 3). The hea map o he heo y
ideo has a pa chie appea ance, as s uden s ha e wa ched ce ain pa s again.
The Physicis s’ able ideo has a mo e uni o m wa ching pa e n, bu mos
s uden s ha e skipped his ideo en i ely.
Fig. 3. Hea maps o he di e en s uden ’s wa ch coun o ideo segmen s on week 4. A:
heo y lec u e ideo, B: mo i a ional physicis s’ able ideo. Dashed lines indica e he
posi ion o sea ch ma ks.
I is appa en om he hea maps ha s uden s click a ound pa icula ly in he heo y
ideo. To s udy he du a ion o each consecu i e iew, we ex ac ed he minu es
deli e ed o each click o he heo y ideo on week 4. The his og am o he du a ion
o he deli e y is shown in Fig. 4.
The majo i y o clicks esul in e y sho iews o a ew seconds (Fig. 4, le panel).
Howe e , when we il e ou he sho e iews below 6 minu es (Fig. 4, igh ), we see
ha he du a ion a ies a lo . The cu -o o 6 minu es was chosen, as his was he
mode o iewing ime in Guo e al. (2014). A ew s uden s wa ch he ideo in one go,
bu he majo i y o s uden s wa ch he ideo in pieces. Howe e , as seen in bo h Fig.
2 and Fig. 3, mos s uden s co e a majo i y o he heo y ideos, e en hough hey
do no necessa ily wa ch i in o de o du ing he same session.
Fig. 4. The du a ion o each in e ac ion wi h he heo y ideo on week 4. Le : all in e ac ions
ollowing a click, igh : in e ac ions exceeding 6 minu es. The o al leng h o he ideo is 149
min.
Guo e al. (2014) obse ed in a MOOC ha he mode o iewing ime on ideos o
any leng h was 6 minu es and p oceeded o ecommend ha lec u e ideos be e y
sho . Howe e , ou da a shows ha deg ee s uden s mos ly do wa ch he ull lec u e
ideos, he longes o which a e o e wo hou s.
3.2 Video use o di e en s uden g oups
To see whe he s uden s o di e en gende o majo used he ideos di e en ly
(RQ2), we spli he iew da a by he backg ound a iables. The e was no signi ican
di e ence in he dis ibu ion o pe cen age wa ched ei he by s uden gende o
majo . As seen in Table 1, men had a highe a e age pe cen wa ched han women,
bu acco ding o he Mann-Whi ney U es , he di e ence was no s a is ically
signi ican (W = 1.48, p = 0.13). Simila ly, while p ospec i e physics majo s had a
highe a e age pe cen age wa ched han s uden s o o he majo s, he di e ence
was no s a is ically signi ican acco ding o he K uskal-Wallis es (H (4, N = 56) =
6.50, p = 0.16).
Table 1. Mean and s anda d de ia ion o he wa ched pe cen age o heo y ideos.
Gende
N
Mean (%)
S.D. (%)
Male
29
75.2
39.8
Female
26
64.1
42.2
Majo
N
Mean (%)
S.D. (%)
Physics
17
85.4
32.7
Chemis y
13
51.4
47.4
Ma hema ics
9
72.8
30.9
Undecided
7
46.1
49.7
O he
10
75.0
42.4
As we ound no s a is ically signi ican di e ences in he da a, examining he de ails
o iewing du a ions o hea maps by s uden majo o gende is unlikely o yield
s a is ically signi ican esul s wi h his numbe o s uden s. Howe e , as he e is a
di e ence in means (Table 1), i would be impo an o collec mo e da a o s udy his
u he .
3.3 Co ela ion wi h exam esul s
Finally, we wan ed o see whe he wa ching ideos co ela es wi h lea ning
ou comes on he cou se (RQ3). To see how ideo use co ela es wi h physics
knowledge a he end o he cou se, we calcula ed he Spea man co ela ion
coe icien s o he pe cen age wa ched o ideos and exam poin s. A isualiza ion o
he co ela ion ma ix is in Fig. 5. F om he igu e, i is clea ha while bo h he
pe cen age wa ched o example and heo y ideos co ela e wi h exam poin s, he
co ela ion be ween heo y ideos and exam poin s is s onge . Fo his pai , 𝜌 =
0.43 (𝑝 = 0.0012), which indica es a mode a e co ela ion (Dancey & Reidy, 2020, p.
183). Fo example ideos and exam poin s, he co ela ion coe icien 𝜌 =
0.27 (𝑝 = 0.048), which indica es a weak co ela ion.
Fig. 5. Spea man co ela ion able o he a e age pe cen age o ideo wa ched by ideo ype
(All, Theo y, Examples, Table) and exam poin s.
4 CONCLUSIONS AND IMPLICATIONS
Ou s udy shows ha s uden s use ideo ma e ials in di e en ways. Mos s uden s
wa ched he majo i y o he heo y ideos. S uden s gene ally clicked a ound heo y
ideos and skipped o e-wa ched segmen s. Fewe s uden s used he mo i a ional
“A he physicis s’ able” ideos, and hese ideos we e wa ched in longe segmen s.
We did no obse e s a is ically signi ican di e ences be ween s uden s o di e en
majo subjec s o be ween men o women, bu his may be due o he small sample
size.
As s uden s clicked a ound he ideos, he du a ion o each consecu i e iew was
e y sho . Howe e , we did no obse e ha s uden s would only wa ch he
beginning o he ideo, much less ha he mode o minu es wa ched would ha e
been 6 minu es, as Guo e al. (2014) obse ed in a MOOC. The likely easons o
his di e ence a e he be e mo i a ion o deg ee s uden s and he inclusion o
sea ch ma ks in he ideo, as also ecommended by Seidel (2024).
We saw a mode a e co ela ion be ween he pe cen age o heo y ideos wa ched
and exam poin s. The co ela ion be ween he pe cen age o example ideos
wa ched and exam poin s was weak, and wa ching he “Physicis s able” ideos did
no co ela e wi h he lea ning ou come. Howe e , a limi a ion o he s udy is ha we
did no measu e ini ial knowledge o he s uden s. I is possible ha wa ching heo y
ideos is linked o be e ini ial knowledge o o he a iables in he lea ning p ocess,
which we we e no able o measu e.
As we only included da a om willing pa icipan s, i is possible ha he da a is
skewed. We we e also no able o measu e engagemen wi h he ideos beyond
clicks and iewing du a ion. Howe e , he iewing da a p o ides ich and in e es ing
da a, which can be used o unde s and which pa s o he ideo s uden s wa ch. The
hea maps can be used o s udy ideo lec u es and unde s and which pa s a e
impo an o s uden s in hei lea ning p ocesses. S udying s uden s’ in e ac ion wi h
he ma e ials has alue o he p agma ic de elopmen o eaching ma e ials.
Fu he s udies a e needed o be e unde s and he ole o ideo ma e ials o
di e en s uden s’ lea ning p ocesses.
ACKNOWLEDGEMENTS
Mic oso Copilo (GPT-4o) was used o gene a e a pa o he R code o he igu es,
in pa icula he da a p ocessing and he plo ing o he hea maps (Figu e 3). We
hank p o . Jo ma Keskinen o discussions ega ding he s udy.
REFERENCES
Ahn, D., & Chan, J. C. (2025). Wha D i es S uden Engagemen and Lea ning in
Video Lec u es? An In es iga ion o Ins uc o Visibili y, Playback Speed, and
S uden P e e ences. Applied Cogni i e Psychology, 39(2), e70026.
