This is a sel -a chi ed e sion o an o iginal a icle. This e sion
may di e om he o iginal in pagina ion and ypog aphic de ails.
Au ho (s):
Ti le:
Yea :
Ve sion:
Copy igh :
Righ s:
Righ s u l:
Please ci e he o iginal e sion:
CC BY 4.0
h ps://c ea i ecommons.o g/licenses/by/4.0/
Teaching sou cing du ing online inqui y – adolescen s wi h he weakes skills bene i ed
he mos
© The Au ho (s) 2022
Published e sion
Hämäläinen, Elina K.; Kiili, Ca i a; Räikkönen, Eija; Lakkala, Minna; Ilomäki, Liisa;
Toom, Auli; Ma unen, Miika
Hämäläinen, E. K., Kiili, C., Räikkönen, E., Lakkala, M., Ilomäki, L., Toom, A., & Ma unen, M.
(2023). Teaching sou cing du ing online inqui y – adolescen s wi h he weakes skills bene i ed
he mos . Ins uc ional Science, 51(1), 135-163. h ps://doi.o g/10.1007/s11251-022-09597-2
2023
ORIGINAL RESEARCH
Recei ed: 14 Decembe 2021 / Accep ed: 20 July 2022
© The Au ho (s) 2022
Elina K. Hämäläinen
[email p o ec ed]
1 Depa men o Educa ion, Uni e si y o Jy äskylä, Jy äskylä, Finland
2 Facul y o Educa ion and Cul u e, Tampe e Uni e si y, Tampe e, Finland
3 Facul y o Educa ion and Psychology, Uni e si y o Jy äskylä, Jy äskylä, Finland
4 Facul y o Educa ional Sciences, Uni e si y o Helsinki, Helsinki, Finland
5 Cen e o Uni e si y Teaching and Lea ning, Facul y o Educa ional Sciences, Uni e si y o
Helsinki, Helsinki, Finland
Teaching sou cing du ing online inqui y–adolescen s wi h
he weakes skills bene i ed he mos
Elina K.Hämäläinen1· Ca i aKiili2· EijaRäikkönen3· MinnaLakkala4·
LiisaIlomäki4· AuliToom5· MiikaMa unen1
Ins uc ional Science
h ps://doi.o g/10.1007/s11251-022-09597-2
Abs ac
Sou cing - iden i ying, e alua ing, and using in o ma ion abou he sou ces o in o ma ion
- assis s eade s in de e mining wha o us when seeking in o ma ion on he In e ne . To
su i e in he pos - u h e a, s uden s should be equipped wi h su icien sou cing skills.
This s udy in es iga ed he e icacy o a eache -led in e en ion aimed a os e ing up-
pe seconda y school s uden s’ (N = 365) sou cing du ing online inqui y. The in e en ion
(4 × 75 min) was s uc u ed in acco dance wi h he phases o online inqui y: loca ing,
e alua ing, syn hesizing, and communica ing in o ma ion. Du ing he in e en ion, each-
e s demons a ed why and how o sou ce, and s uden s p ac iced sou cing by in es iga -
ing a con o e sial opic on he In e ne . S uden s wo ked in small g oups and hei wo k
was suppo ed wi h analysis and e lec ion p omp s. S uden s’ sou cing skills we e mea-
su ed wi h a web-based online inqui y ask be o e and a e he in e en ion. Compa ed
o con ols, he in e en ion os e ed s uden s’ abili ies in h ee o he ou skills measu ed
(sou cing in sea ch que ies, c edibili y judgmen s, and w i en p oduc ). Depending on he
sou cing skill, 4–25% o s uden s showed imp o ed pe o mance. The s uden s wi h low
sou cing skills o begin wi h, bene i ed he mos om he in e en ion. The s udy demon-
s a ed ha s uden s’ sou cing skills can be suppo ed h oughou online inqui y.
Keywo ds In e en ion · Online inqui y · Sou cing · Adolescen s · Mul iple documen
comp ehension
1 3
E. K. Hämäläinen e al.
In oduc ion
One o he mo e ecen de elopmen al wa es in li e acy educa ion is he digi al wa e in
which he eade is seen as an in o ma ion explo e (Tie ney & Pea son, 2021) who engages
in online inqui y o sol e p oblems and make meaning o a ious opics (Coi o, 2021; Leu
e al., 2019). Online inqui y includes he p ocesses o speci ying in o ma ion need and
loca ing, c i ically e alua ing, syn hesizing, and communica ing online in o ma ion (Leu e
al., 2019). When engaging in success ul online inqui y, a skill ul digi al eade a ends o,
ep esen s, and e alua es he sou ces o he in o ma ion ound (B å en e al., 2018c). These
p ac ices, e med sou cing (B å en e al., 2018c; Winebu g, 1991), assis eade s o a oid
us ing misleading in o ma ion, which is widesp ead on he In e ne . A ecen s udy (Kiili
e al., 2021) showed ha sou cing can be employed h oughou online inqui y, and eade s
may engage in sou cing also in he ea lies phases o inqui y. In e es ingly, sou cing in he
ea lie phases o online inqui y suppo ed sou cing in he la e phases o inqui y, sugges ing
he impo ance o app oaching sou cing as an i e a i e p ac ice.
Despi e he impo ance o sou cing, s udies in o line and online con ex s ha e shown
ha many s uden s lack adequa e sou cing skills (e.g., Ba zilai e al., 2015; Kobayashi,
2014; McG ew e al., 2018; S ømsø & B å en, 2014). As a esul , a ious in e en ion s ud-
ies ha e been conduc ed on how s uden s’ sou cing migh bes be suppo ed (see e iews by
B and-G uwel & an S ien 2018; B an e & S ømsø, 2018; B å en e al., 2018c). Teaching
hese skills is essen ial o equip s uden s wi h s a egies o managing di e se in o ma ion in
he 21s cen u y. Howe e , in he in e en ions implemen ed in he In e ne con ex , sou c-
ing skills ha e no been sys ema ically augh and measu ed du ing all he phases o online
inqui y. This s udy ex ends p e ious wo k by examining whe he uppe seconda y school
s uden s’ sou cing can be enhanced h oughou online inqui y by a eache -led in e en ion
in an au hen ic In e ne con ex .
Sou cing du ing online inqui y
The p esen s udy on sou cing du ing online inqui y has been in o med by wo heo e i-
cal models: he Online Resea ch and Comp ehension Model (Leu e al., 2019, see also
Kiili e al., 2018) and he Documen s Model (Pe e i e al., 1999; Roue , 2006). Acco d-
ing o he Online Resea ch and Comp ehension Model (Leu e al., 2019), a p oblem-based
online inqui y comp ises i e key p ocesses: speci ying in o ma ion need and loca ing, c i i-
cally e alua ing, syn hesizing, and communica ing online in o ma ion. In he model, hese
p ocesses a e conside ed o be ecu si e and ecip ocal so ha e alua ion, o example, is
in e wined wi h he o he p ocesses. The Documen s Model (Pe e i e al., 1999; Roue ,
2006), ini ially de eloped in he con ex o in e p e ing his o ical documen s, accen ua es
he impo ance o sou ce in o ma ion in building a cohe en ep esen a ion ac oss mul iple
documen s, including con lic ing in o ma ion. This equi es eade s o connec in o ma ion
abou sou ces, such as au ho s and hei expe ise and in en ions, o he documen s’ con en s
o compa e, con as , and e alua e mul iple documen s (Pe e i e al., 1999; Roue , 2006).
Ideally, sou cing occu s du ing all online inqui y phases (Kiili e al., 2021) when eade s
g adually build a cohe en ep esen a ion o he opic hey examine. Nex , we will desc ibe
how sou cing can be applied du ing each online inqui y phase.
1 3
Teaching sou cing du ing online inqui y–adolescen s wi h he weakes …
The online inqui y begins wi h speci ying he in o ma ion need i.e., wha kind o in o -
ma ion is needed o sol e a p oblem a hand. The skill ul eade s can make use o sou ce
in o ma ion al eady in his phase o online inqui y. Fo example, hey se goals ha empha-
size he impo ance o c edible in o ma ion on he opic o in e es , and hey can conside
which sou ces p o ide he mos eliable in o ma ion (Kiili e al., 2021). These conside -
a ions can be employed when loca ing in o ma ion wi h sea ch engines (Leu e al., 2019).
When o mula ing sea ch que ies, skill ul eade s, who ame hei sea ch e ms by ci ing
eliable pe sons, o ganiza ions, o esea ch-based in o ma ion, can be conside ed o be p ac-
icing sou cing (Kiili e al., 2021). Fu he mo e, when skimming he sea ch engine esul s
page o make ex selec ions online eade s can a end o sou ce ea u es (e.g., in i les,
URLs, o example ex s) o ini ially e alua e he c edibili y and ele ance o online ex s
(Hahnel e al., 2020; Rieh, 2002). E en hough sou cing du ing selec ing po en ial online
ex s om sea ch engine esul page has been p e iously examined (e.g., Ge je s e al., 2011;
Haas & Unkel, 2017; Hau ala e al., 2018) sou cing p ac ices du ing speci ying he in o ma-
ion need and o mula ion o sea ch que ies ha e a ely been in es iga ed (Kiili e al., 2021).
In ecen yea s, s uden s’ sou cing has been inc easingly examined in he la e phases
o online inqui y in ela ion o e alua ing he c edibili y o online ex s and using sou ce
in o ma ion o syn hesize and communica e in o ma ion in w i en p oduc s (e.g., Lis e
al., 2017; Salme ón e al., 2018; S ømsø e al., 2013). When skill ul eade s explo e he
selec ed online ex s, hey can e alua e ex s’ sou ce in o ma ion, including he au ho ’s
expe ise and in en ions as well as he enue’s a ea o expe ise and publishing p ac ices
(c . Pe e i e al., 1999; Roue , 2006). E alua ion o sou ces in o ms he eade s’ judgmen s
o he accu acy o in o ma ion. The ela ion be ween he sou ce and con en e alua ion is
ecip ocal, hus, he judgmen s o he con en alidi y can also in o m he judgmen s o
sou ce us wo hiness (Ba zilai e al., 2020). Howe e , he impo ance o sou ce e alua ion
is highligh ed i eade s lack p io knowledge abou he opic (B å en e al., 2018b; B omme
& Goldman, 2014).
The las phases o online inqui y conce n syn hesizing and communica ing in o ma ion
du ing which s uden s comple e and communica e hei ep esen a ion o he examined
opic. The Documen s Model (Pe e i e al., 1999; B i e al., 2018) is pa icula ly use ul o
unde s and how eade s syn hesize selec ed in o ma ion in hei w i en p oduc s. Acco ding
o he Documen s Model, eade s can cons uc wo ypes o ep esen a ions when eading
mul iple ex s: an in e ex model and an in eg a ed men al model. The in e ex model pos-
i s ha sou ce in o ma ion (e.g., au ho / enue and hei expe ise/in en ions) is connec ed o
he documen ’s con en and o he in o ma ion sou ces (Pe e i e al., 1999; Roue , 2006).
These links a e o wo di e en ypes: sou ce- o-con en and sou ce- o-sou ce.
Sou ce- o-con en links show how a eade combines in o ma ion abou he sou ce o
a documen wi h i s con en whe eas sou ce- o-sou ce links show how a eade connec s
sou ces om mul iple documen s by showing he ela ionships be ween hem, such as sup-
po ing, complemen ing, o opposing. The in e ex model is pa icula ly use ul in si ua ions
whe e eade s con on con lic ing in o ma ion ha p e en s hem om cohe en ly in eg a -
ing he con en o mul iple documen s, he eliabili y o which needs o be ensu ed (B i e
al., 2014). The in eg a ed men al model, in u n, ocuses on he con en o documen s and
desc ibes eade s’ unde s anding o he opic discussed ac oss hem. The ull documen s
model is ealized when eade s in e connec he in e ex and in eg a ed men al models (Pe -
1 3
E. K. Hämäläinen e al.
e i e al., 1999) by acking who said wha and by using his in o ma ion o in e p e and
e alua e he documen s’ con en (B i e al., 2014).
P e ious sou cing in e en ions
In ecen yea s, in e en ions o imp o e s uden s’ sou cing skills ha e been conduc ed a
di e en educa ional le els (see e iews by B and-G uwel & an S ien, 2018; B an e &
S ømsø, 2018). Modeling e ec i e s a egies, use o wo kshee s, p omp s, guided p ac ice,
and g oup discussions ha e been common ins uc ional me hods in mos o hese in e en-
ions (see also Hämäläinen e al., 2020; McG ew & By ne, 2020). Fu he , du ing in e en-
ions, s uden s ha e been asked o ead mul iple documen s including con o e sies (see
B an e & S ømsø, 2018; B å en e al., 2019). A he lowe educa ional le els, iden i ica ion
o sou ce in o ma ion and c edibili y e alua ion ha e been emphasized whe eas olde s u-
den s ha e been augh o ci e sou ces mo e p ecisely and use sou ce ea u es in in e p e -
ing documen s’ con en (see B an e & S ømsø, 2018). E en hough some o he longe
eache -led in e en ions (e.g., A gelagós & Pi a e, 2012; Kingsley e al., 2015) conduc ed
in he In e ne con ex ha e co e ed he whole p ocess o online inqui y (de ining ques ions,
sea ching, e alua ing, syn hesizing, and p esen ing in o ma ion), sou cing has no been
augh o s uden s when speci ying hei in o ma ion need o o mula ing sea ch que ies.
Nex , we p esen h ee in e en ion s udies ca ied ou a he uppe seconda y school
le el ha ha e aimed di ec ly a imp o ing s uden s’ sou cing skills. Thus, hese in e en-
ions in o med he ways sou cing was augh in he p esen s udy e en hough hey we e no
conduc ed using an au hen ic In e ne con ex .
B i and Aglinskas (2002) conduc ed one o he i s s udies, comp ising h ee sho
in e en ions (2 × 40 min), ocused di ec ly on s uden s’ sou cing skills. They designed a
compu e -based en i onmen ha p omp ed high school s uden s o iden i y and a end o
sou ce ea u es in his o y ex s. The en i onmen was designed based on p inciples o each-
ing h ough si ua ed p oblem sol ing, suppo ing expe ep esen a ions, decomposing he
ask, suppo ing ans e , p o iding explici ins uc ion, and mo i a ing engagemen . The
e icacy o he in e en ions was es ed wi h a sou cing es in which s uden s ead exce p s
om six au hen ic ex s ha add essed con o e sial his o ical opics. While eading, hey
we e allowed o make no es on he ex s ha hey could la e use when answe ing ques ions
on he iden i ica ion and e alua ion o he sou ces and he cen al na a i e, pe spec i e on
he con o e sial issue, and a gumen s used in he ex s. Fo sou cing sco es, co ec in o -
ma ion abou he sou ces in s uden s’ no e shee s was also coun ed. In all h ee in e en ions,
he in e en ion g oup showed g ea e imp o emen in hei sco es han he con ols. When
compu e -based and ex book-based eaching we e compa ed, he essays p oduced by he
g oup using a specially designed compu e -based en i onmen con ained mo e sou ce in o -
ma ion and ci a ions o sou ces han he essays o he ex book-based g oup.
Simila ly, B aasch e al. (2013) examined he e icacy o a sho (60 min) esea che -led
sou cing in e en ion among uppe seconda y school s uden s (N = 130). The in e en ion
used a con as ing cases app oach whe e wo hypo he ical adolescen s, one wi h less and
one wi h mo e sophis ica ed s a egies, e alua ed exce p s o online ex s on he heal h isks
o cell phone use. A e amilia izing hemsel es wi h he cases, s uden s we e p omp ed o
independen ly iden i y, compa e, and con as he s a egies used by he hypo he ical adoles-
1 3

Teaching sou cing du ing online inqui y–adolescen s wi h he weakes …
cen s. They hen discussed wi h a pa ne he s a egies hey had iden i ied o decide which
o hese we e he bes and why. Finally, he bes s a egies we e collec ed and sha ed in a
whole-class session. S uden s who pa icipa ed in he in e en ion included mo e scien i ic
concep s ela ed o El Niño in hei essays, displayed be e ankings o he use ulness o he
ex s, ga e mo e sou ce-based jus i ica ions o hei ankings, and mo e o en a ibu ed he
us wo hiness o he ex s o sou ce ea u es han hose o con ols.
B å en e al. (2019) ecen ly conduc ed a comp ehensi e sou cing in e en ion in na u-
al sciences among uppe seconda y school s uden s (N = 250). Compa ed o he s udies
desc ibed abo e, he in e en ion was eache -led and ma kedly longe (9 × 90 min). In he
sc ip ed lessons (3 × 90 min), eache s used a con as ing cases app oach (see also B aasch
e al., 2013) and ex s ha a ied in hei sou ce in o ma ion. A e hese lessons, he s u-
den s p ac iced he p inciples o adap i e sou cing h ough an indi idual w i ing assignmen
(3 × 90 min) and a g oup-based o al assignmen (3 × 90 min). S uden s’ pe o mance was
measu ed by immedia e and delayed pos - es s. In bo h es s, he s uden s in he in e en-
ion g oup p oduced mo e sou ce-based jus i ica ions o hei ex selec ions han con ols.
They also spen mo e ime eading he selec ed ex s and e isi ed he ex s mo e o en han
con ols. Fu he , s uden s who pa icipa ed in he in e en ion included mo e e e ences o
sou ce ea u es in hei w i en p oduc s han con ols.
The sou cing in e en ions desc ibed abo e ha e led o impo an unde s andings o how
o each sou cing skills o uppe seconda y school s uden s, and younge and olde s uden s
as well. Fo example, ask assignmen s and eading ma e ials applied in he lessons ha e
included con o e sies ela ed o he in es iga ed opic (B i & Aglinskas, 2002; B å en e
al., 2019) and/o con as ing cases app oach (B aasch e al., 2013; B å en e al., 2019) which
bo h elici s uden s’ sou cing beha io when eading mul iple documen s. Fu he , in e en-
ions ha e highligh ed explici ins uc ion o sou cing s a egies as well as s uden s’ guided
p ac ice a e whole-class ins uc ion. In addi ion, p omp s o ques ions in he wo kshee s
ha e been applied o enable s uden s’ independen wo k and o guide hei a en ion o he
speci ic sou ce ea u es a he ime. In wo o he s udies (B aasch e al., 2013; B å en e al.,
2019), discussions wi h pee s and in he whole class we e seen as impo an in sha ing s u-
den s’ ideas and lea ning. Du ing he las lesson o he s udy by B å en e al. (2019), s uden s
ga e p esen a ions in small g oups by d awing on sou ces hey had selec ed and e lec ing
hei sou cing ac i i ies du ing he ask. In o med by p e ious s udies, we applied se e al
ins uc ional me hods in designing he in e en ion o p omo e s uden s’ sou cing h ough-
ou online inqui y, such as s uc u ing he online inqui y ask, using con as ing opics and
ask p omp s, explici eaching o sou cing s a egies, and collabo a i e wo k (see Me hod:
Design and implemen a ion o he in e en ion).
The p esen s udy
The p esen s udy in es iga es he e icacy o a eache -led in e en ion ha aimed a
enhancing uppe seconda y school s uden s’ sou cing du ing online inqui y. The design o
he in e en ion ollowed he online inqui y phases (Leu e al., 2019). To acili a e s uden s’
sou cing du ing di e en phases o online inqui y and build a cohe en ep esen a ion o he
examined issue (Pe e i e al., 1999), we applied ins uc ional me hods ha ha e been used
in p e ious sou cing in e en ions. Du ing he in e en ion (4 × 75 min), s uden s wo ked
1 3
E. K. Hämäläinen e al.
collabo a i ely o sol e a con o e sial heal h- ela ed p oblem wi h au hen ic online in o -
ma ion. S uden s’ wo k was suppo ed wi h explici ins uc ion and a join , digi al wo king
documen , including ask p omp s. S uden s’ lea ning o sou cing skills was compa ed o
ha o con ol s uden s by using a quasi-expe imen al p e-pos design.
The ollowing esea ch ques ions we e se :
RQ1. Did uppe seconda y school s uden s’ sou cing in di e en phases o an online
inqui y h ough a eache -led in e en ion inc ease compa ed o con ols?
RQ2. How did s uden s’ sou cing pe o mance change du ing he in e en ion?
RQ3. How we e s uden s’ p e-in e en ion sou cing skills, eading luency, p io opic
knowledge, and opic o de in he asks associa ed wi h changes in hei sou cing pe o -
mance du ing he in e en ion?
In e ms o RQ1, we assumed ha he in e en ion g oup would ou pe o m he con-
ol g oup in sou cing in c edibili y judgmen s and w i en p oduc s when hei p e-sou c-
ing skills, eading luency, p io opic knowledge, and opic o de we e con olled o .
The assump ions a e in line wi h p e ious sou cing in e en ions ha ha e success ully
enhanced uppe seconda y school s uden s’ sou cing in hei c edibili y e alua ions, such as
sou ce-based jus i ica ions o hei ex selec ions (B å en e al., 2019) and use ulness ank-
ings (B aasch e al., 2013). Fu he , i could be assumed ha s uden s will in eg a e mo e
sou ces o hei essays a e he in e en ion (B å en e al., 2019; B i & Aglinskas, 2002).
Because p e ious in e en ions ha e no examined sou cing in speci ying in o ma ion need
o in sea ch que ying, we did no se speci ic hypo heses on hese sou cing p ac ices.
In ou analysis (RQ1), we con olled o s uden s’ p e-sou cing skills, hei p io opic
knowledge, eading luency, and opic o de . S uden s’ p e-sou cing skills we e con olled
o because hey a e impo an p edic o s o hei pos -in e en ion pe o mance (e.g.,
Hämäläinen e al., 2020; McG ew & By ne, 2020). P io opic knowledge and eading lu-
ency we e con olled o because o hei undamen al ole in eading comp ehension. The
eading comp ehension models accen ua e he ole o p io knowledge when eade s make
meaning om he ex s (Ce e i & W igh , 2020), whe eas he lowe -le el eading skills,
such as eading luency, se e as a ounda ion o eading comp ehension (Duke & Ca -
w igh , 2021). Acco dingly, he ecen e iew by Anma k ud e al. (2021) shows ha he
mos examined cogni i e skills in ela ion o sou cing a e p io knowledge (e.g., Mason
e al., 2014; S ang-Lund e al., 2019) and eading skills (e.g., Macedo-Roue e al., 2020;
Po ocki e al., 2020), e en hough he esul s ha e been somewha mixed.
In addi ion, he opic o de o he ex s was con olled o (RQ1) because in es iga ed
opics may elici s uden s’ sou cing di e en ly (B å en e al., 2018b). Fo example, s uden s
ha e alued au ho expe ise o a g ea e ex en when he opic has been less amilia o
hem (e.g., B å en e al., 2018b; McC udden e al., 2016). I also seems ha he ela ionship
be ween indi idual di e ences and sou cing may a y wi h he opic add essed in eading
ma e ials (Anma k ud e al., 2021).
In e ms o RQ2, we assumed ha s uden s would di e in how hei sou cing pe o -
mance changed du ing he in e en ion. We expec ed ha he subs an ial po ion o he
s uden s, bu no all, would imp o e hei sou cing pe o mance. Fo example, McG ew
and By ne (2020) conduc ed a sou cing in e en ion s udy among high school s uden s, and
obse ed s uden s who inc eased, did no change, o dec eased hei sou cing on he online
con en e alua ion ask. Finally, we did no se any hypo hesis abou RQ3, as p e ious s ud-
ies ha e no in es iga ed how he abo e-in oduced ac o s (p e-in e en ion sou cing skills,
1 3
Teaching sou cing du ing online inqui y–adolescen s wi h he weakes …
eading luency, p io opic knowledge, and opic o de ) a e associa ed wi h changes in
s uden s’ sou cing pe o mance du ing he in e en ion. As hese ac o s a e ela ed o mul-
iple documen li e acy and sou cing (see Anma k ud e al., 2021; B å en e al., 2018c),
hei associa ions wi h changes in s uden s’ sou cing pe o mance we e wo h sol ing in he
p esen s udy.
Me hod
Pa icipan s
Pa icipan s comp ised 365 s uden s (Mage = 17.35; SD = 0.40) om eigh uppe seconda y
schools in Finland. Females accoun ed o 58.6%, which is equi alen o he p opo ion
o emales g adua ing om uppe seconda y school in Finland (Suomen i allinen ilas o
[O icial S a is ics o Finland], 2020). In e ms o pa en al educa ion, 75.2% o s uden s’
mo he s and 66.1% o hei a he s had a e ia y le el deg ee. Da a we e collec ed in 2018–
19, be o e he COVID19 pandemic, du ing an obliga o y language a s cou se “Tex s and
in luence”. While all s uden s comple ed he es s and asks, only he esponses o hose who
ga e hei in o med consen we e used in his s udy. I a s uden was unde age, consen was
also eques ed om his/he gua dian(s).
Resea ch design
We applied a quasi-expe imen al p e-pos design wi h a nonequi alen con ol g oup (see
Handley e al., 2018). Fo p ac ical easons, he in e en ion g oup eache s (N = 5) we e
ec ui ed based on hei oppo uni y and willingness o implemen he in e en ion lessons.
The con ol g oup eache s (N = 6) we e no om he same schools as he in e en ion g oup
eache s and we e ec ui ed a e he in e en ion g oup eache s. The in e en ion g oup
comp ised 196 s uden s (56.1% emales) in nine cou ses and he con ol g oup o 169 s u-
den s (61.5% emales) in se en cou ses.
As p e- and pos - es s, he s uden s pe o med an online inqui y ask. We coun e balanced
he opic o de ( accina ion and a s) in bo h condi ions. Be ween he es s, he in e en ion
g oup pa icipa ed in a eache -led in e en ion (4 × 75 min lessons) on online inqui y as a
pa o hei Tex s and in luence cou se ( o al o 23 × 75 min lessons) while he con ol g oup
pa icipa ed in a egula Tex s and in luence cou se. The con ol g oup eache s ecei ed
in e en ion ma e ials a e he comple ion o he s udy. Thus, du ing he s udy, he con ol
g oup was no exposed o any o he eaching ma e ials used in he in e en ion.
Design and implemen a ion o he in e en ion
To p omo e s uden s’ sou cing du ing online inqui y, we designed a eache -led in e en ion
ha was in o med by se e al ins uc ional p inciples (see also Kiili e al., 2022). Fi s , we
designed an online inqui y ask ha was s uc u ed in o manageable sequences (Van Me -
iënboe & Ki schne , 2007) ollowing he phases o online inqui y (Leu e al., 2019) and
ela ed lea ning objec i es (see Table 1). I is no able ha o p ac ical easons, we we e able
o design a 4 × 75 min uni . As a consequence, we combined he ins uc ion o he i s wo
1 3
E. K. Hämäläinen e al.
phases o online inqui y, i.e., speci ying he in o ma ion need and sea ching o in o ma ion,
in o he i s lesson. Mo e emphasis was pu on sea ching o in o ma ion han on speci ying
Table 1 Phases o Online Inqui y, Lea ning Objec i es, Desc ip ion o he Sub-Tasks, and E alua ion C i e ia
o In e en ion Lessons (4 × 75 min)
Lessons o
online inqui y
Lea ning objec i es Desc ip ion o he
sub- asks *
E alua ion c i e ia **
Lesson 1:
Task assignmen
Planning sea ch
Loca ing
in o ma ion
S uden s a e able o speci y
hei in o ma ion need.
S uden s a e able o
selec pu pose ul sea ch
s a egies.
S uden s a e able o o -
mula e sea ch que ies by
u ilizing co e concep s and
sou ce in o ma ion.
In you small g oup,
selec one o he ou
con o e sial heal h
opics.
Explo e on he
In e ne wha kinds
o s akeholde s w i e
abou he issue.
Selec wo di e en
s akeholde s whose
iews you will exam-
ine mo e closely. Se-
lec wo online ex s
ha ep esen each o
he s akeholde s ( o al
ou ex s).
S uden s’ sea ch plan includes
main concep s abou he in es-
iga ed opic and ela ed au ho s
and enues.
S uden s ha e speci ied pu pose-
ul sea ch que ies ela ed o hei
opic and ela ed enues.
S uden s ha e selec ed wo s ake-
holde s di e ing in e.g., expe ise,
mo i es and poin o iews o he
opic.
S uden s ha e selec ed online
ex s ha a e sui able o he ask.
Lesson 2:
E alua ing
in o ma ion
S uden s a e able o e alu-
a e mul iple aspec s o
online ex s.
S uden s a e able o
iden i y sou ce ea u es
and e alua e hem when
in e p e ing he quali y o
con en .
E alua e and analyze
he selec ed ou
online ex s.
S uden s ha e ecognized di e en
sou ce ea u es and ealized how
hose ea u es a ec he c edibili y
and plausibili y o online ex s.
S uden s ha e ecognized he
main claim in each online ex and
conside ed how well i is jus i ied
in he ex .
S uden ha e u ilized hei no ions
abou online ex ’s sou ce ea u es
when e alua ing he c edibili y
and plausibili y o ex ’s con en .
Lesson 3:
Syn hesizing
in o ma ion
om mul iple
online ex s
S uden s a e able o com-
pose a ex ha compa es
he di e en iews o
sou ces and mo i es and
e idence hese iews a e
based on.
S uden s a e able o ci e
he sou ces by p o id-
ing he eade wi h an
adequa e amoun o sou ce
in o ma ion.
Compa e he iews
o he s akeholde s,
conside po en ial
easons o hei di -
e en iews, and con-
side whose iews a e
he mos plausible.
S uden s’ syn hesis includes
insigh ul conside a ions o simi-
la i ies and di e ences in selec ed
online ex s (no jus lis ed).
S uden s ha e ealized why
c i ical eading on he in e ne is
impo an and wha kind o online
ex s should be elied on when
making impo an decisions.
Lesson 4:
Communica ing
o o he s he
esul s o he
inqui y
S uden s a e able o
communica e he main
indings o he inqui y o
o he s uden s and engage
in discussions abou he
indings.
P esen you indings
o o he small g oups
in he concluding
semina .
Discuss wha you
ha e lea ned abou
c i ical eading
online.
* S uden s’ wo king documen including p omp s o each lesson can be ound as Appendix 1, see also
Kiili e al. (2022).
** E alua ion c i e ia we e gi en o s uden s be o e hey engaged in he online inqui y ask.
1 3
Teaching sou cing du ing online inqui y–adolescen s wi h he weakes …
pos - es se ed as dependen a iables and we e analyzed sepa a ely. In each analysis, we
con olled o he co esponding p e- es sco e. G oup (0 = con ol, 1 = in e en ion) was
used as he independen a iable, whe eas Reading luency (0–100), Topic o de (0 = acci-
na ion– a s, 1 = a s– accina ion), and P io opic knowledge (0–6) we e also con olled o .
To examine he in e en ion e ec on Sou cing in speci ying in o ma ion need, Sou c-
ing in c edibili y judgmen s, and Sou cing in w i en p oduc , we applied linea eg ession
analysis. Because Sou cing in sea ch que ies was a non-no mally dis ibu ed coun a i-
able wi h la ge o e -dispe sion, we examined i s in e en ion e ec wi h nega i e binomial
eg ession analysis (Coxe e al., 2009).
The nega i e binomial eg ession analysis models he log o he expec ed coun o Sou c-
ing in sea ch que ies in he pos - es (dependen a iable) as a unc ion o independen /con-
ol a iables (Coxe e al., 2009). We p esen eg ession coe icien s as inciden a e a ios
(IRRs) which we e ob ained by exponen ia ing eg ession coe icien s using base e. Fo a
dicho omous independen a iable (i.e., G oup), IRR ep esen s he change in he expec ed
a e o Sou cing in sea ch que ies in he pos - es when he alue o he independen a i-
able changes om 0 o 1. An IRR > 1 indica es how many imes g ea e he expec ed a e o
Sou cing in sea ch que ies in he pos - es is o s uden s in he in e en ion g oup han hose
in he con ol g oup. In con as , an IRR < 1 indica es ha he expec ed a e o Sou cing in
sea ch que ies in he pos - es is g ea e o s uden s in he con ol g oup han hose in he
in e en ion g oup.
Wi h con inuous con ol a iables (i.e., Reading luency), he IRR ep esen s he change
in he expec ed a e o Sou cing in sea ch que ies in he pos - es when he alue o he
con ol a iable inc eases by one uni . We de e mined he s a is ical signi icance o all IRRs
by compu ing hei 95% con idence in e als (CI). An IRR di e s s a is ically signi ican ly
om he alue 1 i i s con idence in e al does no include he alue 1.
All eg ession analyses we e conduc ed using Mplus s a is ical package ( e sion 7.4;
Mu hén & Mu hén, 1998–2017) wi h he ull in o ma ion maximum likelihood p ocedu e
(Ende s, 2010), as missing da a (0.00–0.17%) we e assumed o be missing a andom.
Fu he , we es ima ed model pa ame e s by using maximum likelihood es ima ion wi h
non-no mali y obus s anda d e o s. In he da a, s uden s we e nes ed wi hin 16 cou ses.
Al hough in a-class co ela ions a he cou se le el we e small (0.01–0.11) o all a iables,
we used he cou se as a clus e ing a iable and es ima ed unbiased s anda d e o s.
Ou eg ession analyses o RQ1 p o ide mo e gene al agg ega e-le el in o ma ion on
he di e ences be ween he in e en ion and con ol g oups in hei sou cing pe o mance
du ing he in e en ion. Howe e , agg ega e da a do no necessa ily apply o any speci ic
s uden because he g oup mean may conceal indi idual de e io a ion despi e imp o e-
men on a e age. Mo eo e , indi idual pa e ns o change a e no e ealed in he agg ega e,
al hough i is in o ma ion applicable o indi idual s uden s ha is needed o unde s and who
bene i s om he in e en ion (i.e., he e icacy o he in e en ion).
The e o e, we supplemen he analyses o RQ1 wi h a mo e indi idual-le el examina-
ion o he e ec s o he in e en ion on s uden s’ sou cing pe o mance (RQ2) by calcu-
la ing he Reliable Change Index sepa a ely o each sou cing a iable (RCI; Jacobson &
T uax, 1991) o each s uden in he in e en ion g oup. RCI de e mines, o each s uden ,
i a change in he sou cing a iables can be a ibu ed o he in e en ion a he han chance
o measu emen e o a p < .05, which co esponds o he alue o 1.96 in he s anda dized
no mal dis ibu ion.
1 3

E. K. Hämäläinen e al.
The RCI o an indi idual s uden was compu ed by di iding he di e ence be ween his/
he p e- and pos - es sco es by he pooled s anda d de ia ion o he co esponding p e- es
sou cing a iable. When compu ing he pooled s anda d de ia ion, we used in o ma ion
om bo h he in e en ion and he con ol g oups in o de o ake in o accoun he po en-
ial di e ences be ween he g oups in he a ia ion. The RCI alue o he indi idual s u-
den desc ibes how many s anda d de ia ions his/he p e- and pos - es sco es di e in each
sou cing a iable. Nex , we de e mined he cu -o alue by coun ing he weigh ed midpoin
be ween he p e- es means o he in e en ion and con ol g oups (A kins e al., 2005).
We used indi idual RCI and cu -o alues o classi y s uden s in o hose who showed
a nega i e change du ing he in e en ion (RCI < -1.96), hose who showed no change
(-1.96 ≤ RCI ≤ 1.96), hose who showed a eliable posi i e change (RCI > 1.96 bu did no
pass he cu -o c i e ion), and hose who also passed he cu -o c i e ion, hus showing a
clea posi i e change (RCI > 1.96 + cu -o ) in hei sou cing skills.
To answe RQ3, we in es iga ed how he con ol a iables (P e- es sco es, Reading
luency and P io opic knowledge) we e associa ed wi h he in e en ion g oup s uden s’
sou cing pe o mance acco ding o he RCIs. As he a iable Sou cing in sea ch que ies was
non-no mally dis ibu ed and he e we e only a ew s uden s in some RCI classes, we used
boo s ap analysis wi h 95% CIs o mean di e ences (E on, 1987). When 95% CI does
no include he alue 0, he di e ence be ween he means o he RCI classes is s a is ically
signi ican . We simula ed 2 000 boo s ap samples by using bias-co ec ed accele a ed con-
idence in e als (E on, 1987) and s a i ied sampling acco ding o he s uden s’ cou ses.
Fu he , we in es iga ed how opic o de was associa ed wi h he in e en ion g oup s u-
den s’ sou cing pe o mance acco ding o he RCI by using c oss abula ion and χ2 es wi h
C ame ’s V o e ec size.
Resul s
Desc ip i e s a is ics
Desc ip i e s a is ics o s uden s’ pe o mance in sou cing and con ol a iables a e p e-
sen ed in Table 3. In he p e es , he in e en ion g oup ou pe o med he con ol g oup only
in Sou cing in c edibili y judgmen s ( (342.03) = -2.05, p = .041, d = 0.22). In all he o he
p e- es sou cing a iables and he es s o Reading luency and P io opic knowledge, he
in e en ion and he con ol g oups pe o med equally, indica ing no ema kable g oup di -
e ences a baseline.
E icacy o he in e en ion
Wi h espec o RQ1, he eg ession analyses (see Table 4) showed ha he in e en ion os-
e ed s uden s’ a en ion o sou ce ea u es in hei c edibili y judgmen s as well as hei use
o sou ces in hei w i en p oduc s. Fu he mo e, he in e en ion g oup used sou ce ea-
u es in hei sea ch que ies 2.23 imes mo e o en in he pos - es han con ols. Howe e ,
he in e en ion did no enhance s uden s’ use o sou ce ea u es and e alua i e s a emen s
in speci ying he in o ma ion need. Addi ionally, in he pos - es , he accina ion ask s u-
den s pe o med be e in all he sou cing a iables han he a s ask s uden s.
1 3
Teaching sou cing du ing online inqui y–adolescen s wi h he weakes …
Table 3 Sco es o he Sou cing and Con ol Va iables o he In e en ion and Con ol G oups
In e en ion g oup
(N = 175–191)
Con ol g oup
(N = 143–162)
P e- es measu es (obse ed ange) M SD Md M SD Md
Sou cing in speci ying in o ma ion need (0–3) 1.05 0.96 1 1.01 1.01 1
Sou cing in sea ch que ies (0–7) 0.37 0.87 0 0.32 0.72 0
Sou cing in c edibili y judgmen s (0–8) 3.50 1.62 3 3.16 1.47 3
Sou cing in w i en p oduc (0–7) 2.96 1.93 3 2.62 1.74 3
Con ol a iables (obse ed ange)
Reading luency (4–100) 71.42 16.52 72 71.77 16.46 73
P io opic knowledge (in he pos - es ) (0–6) 4.34 1.17 4 4.26 1.03 4
Pos - es measu es (obse ed ange)
Sou cing in speci ying in o ma ion need (0–3) 0.79 0.94 1 0.63 0.84 0
Sou cing in sea ch que ies (0–8) 0.58 1.11 0 0.24 0.54 0
Sou cing in c edibili y judgmen s (0–9) 4.41 1.77 4 3.55 1.71 4
Sou cing in w i en p oduc (0–7) 3.59 2.21 4 2.68 1.88 3
Table 4 Resul s o Linea (β) and Nega i e Binomial Reg ession analysis (IRR; 95% CI) o he Associa ions
Be ween P edic o s, Independen Va iable (G oup) and S uden s’ Sou cing Pe o mance in he Pos -Tes
Dependen a iables
P edic o s Pos - es :
Sou cing in
speci ying
in o ma ion need
Pos - es :
Sou cing in
c edibili y
judgmen s
Pos - es :
Sou cing in
w i en
p oduc
Pos - es :
Sou cing in
sea ch
que ies
β β β IRR [95%
CI]
P e- es : Sou cing in speci ying in o -
ma ion need
0.44***
P e- es : Sou cing in c edibili y
judgmen s
0.43***
P e- es : Sou cing in w i en p oduc 0.39***
P e- es : Sou cing in sea ch que ies 1.32 [1.03;
1.68] *
Reading luency − 0.02 0.10 0.08 1.02 [1.00;
1.04]
P io opic knowledge (in he pos - es ) 0.02 0.03 0.10 1.28 [1.09;
1.49] *
Topic o de (0 = accina ion- a s,
1 = a s- accina ion)
0.11** 0.21*** 0.16** 2.41 [1.65;
3.53] *
G oup (0 = con ol, 1 = in e en ion) 0.08 0.19*** 0.18** 2.23 [1.28;
3.88]*
Cohens’d o g oup e ec [95%CI]0.16 [-0.05; 0.38] 0.39 [0.17;
0.61]
0.37 [0.15;
0.58]
R² = 0.20*** R² = 0.29*** R² = 0.24***
No es:
* The associa ion is s a is ically signi ican when 95% CI o IRR (Inciden Ra e Ra io) does no include
he alue 1.
**p < .01; ***p < .001
The ows highligh ed in bold p esen he esul s be ween in e en ion g oup and con ol g oup wi h
Cohen’s d [95% CI] o e ec s.
1 3
E. K. Hämäläinen e al.
The RCI classes o he sou cing pe o mance o he in e en ion g oup s uden s a e
p esen ed in Table 5. Wi h espec o RQ2, i is no able ha he numbe o s uden s showing
no change was high in all he sou cing a iables. Fu he , 4.1% o he s uden s showed a eli-
able o clea posi i e change in Sou cing in speci ying in o ma ion need, 24.6% in Sou cing
in sea ch que ies, 18.5% in Sou cing in c edibili y judgmen s, and 20.5% in Sou cing in
w i en p oduc . Fo Sou cing in sea ch que ies, all he s uden s demons a ing a posi i e
change, eliable o clea , imp o ed subs an ially; howe e , almos one- i h o he s uden s
showed a nega i e change. In compa ison, he changes in Sou cing in c edibili y judgmen s
and w i en p oduc we e mos ly posi i e.
RQ3 ega ded he associa ions be ween con ol a iables (p e- es sou cing a iables,
eading luency, p io opic knowledge, and opic o de ) and s uden s’ RCI classes. As shown
in Table 6, he in e en ion g oup s uden s who showed a clea posi i e change (RCI class
4) in hei sou cing pe o mance sco ed he lowes in all he p e- es sou cing a iables. Fu -
he mo e, he s uden s who showed a nega i e change (RCI class 1) sco ed he highes in all
he p e- es sou cing a iables. Mo eo e , he s uden s showing a nega i e change di e ed
om he s uden s in he o he RCI classes in all he p e- es sou cing a iables (see Table 7).
Fu he , in Sou cing in c edibili y judgmen s and Sou cing in w i en p oduc , he s uden s
showing a eliable change (RCI class 3) o a clea change (RCI class 4), had lowe p e- es
sco es in co esponding sou cing a iables han hose showing no change (RCI class 2). In
addi ion, s uden s showing a clea change in Sou cing in w i en p oduc had lowe p e- es
sco es in he co esponding sou cing a iable han hose showing a eliable change.
Wi h espec o he o he con ol a iables, opic o de was associa ed wi h RCI classes
in Sou cing in sea ch que ies (χ2(2) = 15.32, p < .001, V = 0.22) and Sou cing in c edibili y
judgmen s (χ2(3) = 10.59, p = .014, V = 0.18). The s uden s who explo ed a s in he p e- es
demons a ed a clea posi i e change (RCI class 4) in bo h a iables signi ican ly mo e
o en han he s uden s who explo ed accina ion in he p e- es . Con e sely, he s uden s
who explo ed accina ion in he p e- es demons a ed a clea posi i e change (RCI class 4)
in bo h a iables mo e a ely han s uden s who explo ed a s in he p e- es . Fu he mo e,
he s uden s who explo ed a s in he p e- es , demons a ed a nega i e change (RCI class
1) in Sou cing in sea ch que ies mo e a ely han he s uden s who explo ed accina ion in
he p e- es and ice e sa. Howe e , opic o de was no associa ed wi h RCI classes o
Table 5 F equencies ( ) and Pe cen ages (%) o S uden s in he In e en ion G oup Demons a ing Nega i e
Change, No Change, Reliable Posi i e Change and Clea Posi i e Change in Sou cing Va iables
Sou cing a iables Nega i e
change
(RCI ≤ -1.96)
(%)
No change
(-1.96 ≤ RCI ≤ 1.96)
(%)
Reliable
posi i e change
(RCI > 1.96)
(%)
Clea
posi i e change
(RCI > 1.96 + cu -
o )
(%)
Sou cing in speci ying
in o ma ion need (N = 172)
20 (11.6) 145 (84.3) 0 (0.0) 7 (4.1)
Sou cing in sea ch que ies
(N = 171)
31 (18.1) 98 (57.3) 0 (0.0) 42 (24.6)
Sou cing in c edibili y
judgmen s (N = 179)
5 (2.8) 141 (78.7) 18 (10.1) 15 (8.4)
Sou cing in w i en p oduc
(N = 180)
14 (7.8) 129 (71.7) 20 (11.1) 17 (9.4)
No e. RCI = Reliable Change Index (Jacobson & T uax, 1991)
1 3
Teaching sou cing du ing online inqui y–adolescen s wi h he weakes …
Table 6 Means (SD) o In e en ion G oup S uden s’ RCI Classes (1 = Nega i e Change, 2 = No Change,
3 = Reliable Posi i e Change, 4 = Clea Posi i e Change) Acco ding o Con ol Va iables Based on Boo s ap
Analysis
Sou cing a iables RCI classes Numbe o
s uden s (N)
P e- es
sco e
M (SD)
Reading luency
( ange 0–100)
M (SD)
P io opic
knowledge
( ange 0–6)
M (SD)
Sou cing in speci ying in o -
ma ion need
( ange 0–3)
1
2
4
20
145
7
2.40 (0.50)
0.88 (0.88)
0.57 (0.54)
70.89 (15.35)
72.68 (15.62)
68.29 (18.20)
4.20 (1.06)
4.43 (1.10)
3.86 (2.04)
Sou cing in sea ch que ies
( ange 0– )
1
2
4
31
98
42
1.42 (0.72)
0.12 (0.50)
0.05 (0.22)
73.90 (13.38)
69.21 (16.03)
76.71 (16.16)
4.65 (1.05)
4.19 (1.14)
4.60 (1.23)
Sou cing in c edibili y
judgmen s
( ange 0–9)
1
2
3
4
5
141
18
15
5.60 (1.14)
3.67 (1.53)
2.67 (1.24)
2.27 (1.34)
70.40 (14.86)
72.09 (15.84)
75.22 (14.25)
68.93 (18.92)
4.80 (0.84)
4.32 (1.17)
4.06 (1.06)
4.67 (1.29)
Sou cing in w i en p oduc
( ange 0–7)
1
2
3
4
14
129
20
17
5.29 (1.44)
3.07 (1.82)
1.95 (1.28)
1.18 (1.02)
71.43 (17.72)
73.11 (15.97)
69.80 (16.30)
67.29 (12.43)
4.07 (0.62)
4.34 (1.22)
4.35 (1.27)
4.47 (1.01)
No e. RCI = Reliable Change Index (Jacobson & T uax, 1991). In a iables Sou cing in speci ying
in o ma ion need and Sou cing in sea ch que ies none o he s uden s we e classi ied in RCI class 3.
Table 7 Compa isons o he In e en ion G oup S uden s’ RCI Classes (1 = Nega i e Change, 2 = No Change,
3 = Reliable Posi i e Change, 4 = Clea Posi i e Change) Acco ding o Con ol Va iables
Sou cing a iables Compa i-
sons o
RCI
classes
P e- es sco e Reading luency P io opic
knowledge
Mean di e ence [95% CI] *, Cohens’ d
Sou cing in
speci ying
in o ma ion need
(N = 170–172)
1 s. 2
1 s. 4
2 s. 4
1.52 [1.27; 1.78], d= -1.80
1.83 [1.39; 2.28], d= -3.58
0.31 [-0.11; 0.72]
-1.79 [-8.97; 5.25]
2.61 [-11.94; 17.51]
4.40 [-9.07; 18.56]
-0.23 [-0.72; 0.27]
0.34 [-1.06; 2.09]
0.57 [-0.75; 2.27]
Sou cing in
sea ch que ies
(N = 170–171)
1 s. 2
1 s. 4
2 s. 4
1.30 [1.06; 1.57], d= -2.31
1.37 [1.16; 1.64], d= -2.76
0.07 [-0.04; 0.19]
4.70 [-0.88; 10.48]
-2.81 [-9.27; 4.07]
-7.51 [-13.05; -2.09],
d = 0.47
0.46 [-0.00; 0.86]
0.05 [-0.48; 0.56]
-0.41 [-0.83; 0.03]
Sou cing in
c edibili y
judgmen s
(N = 177–179)
1 s. 2
1 s. 3
1 s. 4
2 s. 3
2 s. 4
3 s. 4
1.93 [0.89; 2.88], d= -1.27
2.93 [1.74; 4.03], d= -2.40
3.33 [2.07; 4.50], d= -2.57
1.00 [0.41; 1.61], d= -0.67
1.40 [0.63; 2.09], d= -0.93
0.40 [-0.47; 1.27]
-1.69 [-21.06; 17.68]
-4.82 [-26.33; 16.69]
1.47 [-20.51; 23.44]
-3.13 [-13.79; 7.53]
3.16 [-8.40; 14.72]
6.29 [-8.59; 21.17]
0.48 [-0.26; 1.21]
0.74 [-0.05; 1.54]
0.13 [-0.77; 1.10]
0.26 [-0.23; 0.75]
-0.35 [-1.01; 0.32]
-0.61 [-1.39; 0.21]
Sou cing in
w i en p oduc
(N = 178–180)
1 s. 2
1 s. 3
1 s. 4
2 s. 3
2 s. 4
3 s. 4
2.22 [1.38; 2.99], d= -1.24
3.34 [2.39; 4.25], d= -2.49
4.11 [3.18; 4.89], d= -3.36
1.12 [0.52; 1.76], d= -0.64
1.89 [1.32; 2.40], d= -1.08
0.77 [0.10; 1.41], d= -0.66
-1.68 [-11.10; 8.06]
1.63 [-9.86; 13.40]
4.13 [-6.34; 15.12]
3.31 [-3.97; 10.64]
5.82 [-0.66; 11.99]
2.51 [-6.46; 11.35]
-0.27 [-0.66; 0.09]
-0.28 [-0.89; 0.33]
-0.40 [-1.00; 0.14]
-0.01 [-0.60; 0.57]
-0.13 [-0.69; 0.41]
-0.12 [-0.84; 0.60]
No e. RCI = Reliable Change Index (Jacobson & T uax, 1991) In a iables Sou cing in speci ying
in o ma ion need and Sou cing in sea ch que ies none o he s uden s we e classi ied in RCI class 3.
* The mean di e ence is s a is ically signi ican i 95% CI does no include 0. CI was calcula ed by
boo s ap analysis.
1 3
E. K. Hämäläinen e al.
Sou cing in speci ying in o ma ion need and Sou cing in w i en p oduc . In addi ion o
opic o de , we also ound an associa ion be ween Reading luency and Sou cing in sea ch
que ies (see Table 7). Namely, s uden s showing a clea change (RCI class 4) in Sou cing in
sea ch que ies sco ed highe on Reading luency han hose showing no change (RCI class
2). P io opic knowledge was no associa ed wi h RCI classes.
Discussion
This s udy epo s a sou cing in e en ion (4 × 75 min) wi h in e en ion and con ol g oups
comp ising a o al o o e 360 uppe seconda y school s uden s. Whe eas p e ious in e en-
ions ha e measu ed s uden s’ sou cing only in one o wo phases o inqui y (see B an e
& S ømsø, 2018), ou s udy ocused on eaching and measu ing sou cing on he In e ne
du ing he di e en phases o online inqui y. The uniqueness o he p esen s udy also lies
in examining he cha ac e is ics o he s uden s whose sou cing skills imp o ed o did no
imp o e du ing he in e en ion (c . McG ew & By ne, 2020). We i s discuss he main
indings and limi a ions o he s udy and conclude wi h he ins uc ional implica ions o he
indings.
As we expec ed, compa ed o con ols, he in e en ion g oup s uden s employed sou ce
in o ma ion mo e o en when hey e alua ed he c edibili y o online ex s and composed a
w i en p oduc in he pos - es . These esul s a e in line wi h ea lie indings showing ha
e en qui e sho in e en ions can be e ec i e in os e ing uppe seconda y school s uden s’
sou cing skills in c edibili y judgmen s and w i en p oduc s (e.g., B aasch e al., 2013; B i
& Aglinskas, 2002). Fu he , he in e en ion enhanced s uden s’ use o sou ce in o ma ion
when hey o mula ed sea ch que ies.
Howe e , sou cing in speci ying in o ma ion need did no inc ease du ing he in e en-
ion. This was no wholly su p ising as he alue o sou cing in speci ying in o ma ion need
was no augh as explici ly as ha o sou cing in he o he phases o online inqui y (c .
Heijl jes e al., 2014; Ma in & Halpe n, 2011). This esul sugges s ha eaching sou cing in
one phase o online inqui y does no necessa ily ans e o o he phases o online inqui y,
highligh ing he impo ance o eaching sou cing in all he inqui y phases. Teaching why
and how o sou ce in he ea lie phases o online inqui y would be impo an because sou c-
ing in he ea lie phases seems o suppo sou cing in he la e phases o online inqui y (Kiili
e al., 2021).
In he p e- es , s uden s did no commonly make use o sou ces o sou ce ea u es (e.g.,
o ganiza ions, c eden ials) in hei sea ch que ies. Thus, i is impo an o inc ease s uden s’
awa eness and p ocedu al knowledge abou sou cing in sea ch que ies o help hem b oaden
hei s a egic sea ch epe oi e. A he g oup le el, ou in e en ion p omo ed sou cing in
sea ch que ies o some ex en , al hough he s uden s’ pos - es sco es emained low. No ably,
one- ou h o he s uden s showed a clea posi i e change in hei pe o mance o sou c-
ing in sea ch que ies. As hese s uden s had ha dly engaged in sou cing when o mula ing
sea ch que ies a he beginning o he in e en ion, his esul sugges s ha hey may ha e
adop ed a new sou cing p ac ice. Abou one- i h o he s uden s pe o med wo se in he
pos - es han p e- es . This may pa ly be explained by he opic (c . Anma k ud e al., 2021;
B å en e al., 2018b). I seems ha i was easie o loca e use ul online ex s on he a s opic
(see Hämäläinen e al., 2021) and his did no equi e he s uden s o add sou ce in o ma ion
1 3

Teaching sou cing du ing online inqui y–adolescen s wi h he weakes …
in hei que ies. In sum, ou s udy ex ends ou unde s anding o sou cing du ing in o ma ion
sea ch (see also Kiili e al., 2021) as mos o he p e ious s udies ha e ocused on s uden s’
sea ch s a egies and e o mula ion o que ies wi hou paying speci ic a en ion o he use o
sou ces in sea ch que ies (e.g., Wildemu h e al., 2018).
When p omp ed o e alua e he c edibili y o online ex s, he in e en ion g oup s uden s
a ended o and e alua ed sou ce ea u es mo e o en han con ols. Likewise, he in e en-
ions by B aasch e al. (2013) and B å en e al. (2019) enhanced uppe seconda y school
s uden s’ use o sou ce-based jus i ica ions o hei ex selec ions o ankings. When we
examined changes in s uden s’ sou cing in c edibili y judgmen s, we ound ha almos one-
i h o he s uden s showed imp o ed pe o mance, whe eas he emainde (79%) showed
no change. The s uden s who imp o ed had pe o med a he poo ly in he p e- es , a end-
ing, on a e age, o only one sou ce ea u e pe online ex . The in e en ion helped hem o
mo e owa ds mo e e sa ile sou cing when judging he c edibili y o online ex s. In e es -
ingly, he s uden s showing no change did no pe o m pa icula ly well in he p e- es ei he ,
indica ing ha he e was no ceiling e ec . These esul s sugges ha o enhance s uden s’
c i ical online eading skills, he e is a need o egula ly each sou cing when s uden s ead
online ex s a ying in quali y.
In addi ion, i seems ha di e en ex s elici di e en kinds o sou cing beha io (c .
B å en e al., 2011; 2015). Fo example, in he p esen s udy, some au hen ic online ex s
missed he name o he au ho and in some ex s, he au ho ’s mo i es we e mo e ob ious
han in o he s. E en hough s uden s esponded o he sepa a e ques ions ega ding aspec s
ha s eng hened and aspec s ha weakened he c edibili y o online ex s, hey did no
a end o and e alua e consis en ly sou ce ea u es (au ho , enue, in en ions) h ough di e -
en ex s, no e en in he pos - es . Howe e , paying a en ion o he au ho expe ise should
be egula ly used sou cing p ac ice (e.g., B å en e al., 2018b).
Fu he , he in e en ion enhanced s uden s’ use o sou ce in o ma ion in hei w i en
p oduc s when jus i ying hei s ance on accina ing a child o a oiding sa u a ed a s (see
also B å en e al., 2019; B i & Aglinskas, 2002). I should be no ed ha he sco es o s u-
den s’ w i en p oduc s included he men ioned sou ces bu also he use o e alua i e s a e-
men s, sou ce-sou ce links, and sou ce-con en links (see Pe e i e al., 1999). Again, he
s uden s wi h he weakes skills in he p e- es we e mos ly hose who showed imp o emen
(al oge he 20.5% imp o ed) in he pos - es . This means ha hey had ha dly used he links
o e alua i e s a emen s in hei w i en p oduc s be o e he in e en ion and ha he in e -
en ion guided hem owa ds he mo e sophis ica ed sou cing p ac ices ha a e equi ed o
build an in e ex model (see Pe e i e al., 1999).
I is no able ha he s uden s we e allowed o consul hei sel -selec ed online ex s
when composing he w i en p oduc (c . B å en e al., 2019), a p ocedu e which makes his
sub ask easie han when based solely on memo y and men al ep esen a ions, as in some
ea lie s udies (e.g., B aasch e al., 2013; B i & Aglinskas, 2002). Howe e , ou ask also
esembles basic school assignmen s as well as expe p ac ices, whe e documen s a e usu-
ally a ailable when composing a w i en syn hesis (c . Vande meulen e al., 2020).
Despi e ou expec a ions, he numbe o s uden s whose sou cing pe o mance imp o ed
was limi ed (4–25% ac oss di e en sou cing p ac ices). Howe e , he in e en ion espe-
cially os e ed he pe o mance o he s uden s wi h he weakes sou cing skills in he p e-
es . This esul is impo an as e y limi ed sou cing skills may esul in he ecu ing use
o dis- and misin o ma ion (Sina a & Lomaba di, 2020). Thus, he s uden s whose pe o -
1 3
E. K. Hämäläinen e al.
mance did no change du ing he in e en ion had be e sou cing skills o s a wi h han
hose whose pe o mance imp o ed.
Some o he mo e ad anced s uden s also pe o med wo se in he pos - es han p e- es .
This may pa ly be explained by he es opics, which seemed o elici sou cing ac i i y
somewha di e en ly (c . Anma k ud e al., 2021). I is also possible ha some s uden s
we e no su icien ly mo i a ed o pu e o in o he pos - es assignmen (see B å en e al.,
2018a; Lis & Alexande , 2018). Al e na i e explana ions may ela e o he small g oup
wo k. Teache s epo ed a ia ion in s uden s’ engagemen , some small g oups we e mo e
engaged han o he s. I may also well be ha some g oups did no ha e an op imal con-
s uc ion o lea ning. Acco dingly, small g oups including s uden s wi h weake and be e
skills, may se e s uden s wi h be e skills i hey a e he ones gi ing he elabo a ed help
o pee s wi h weake skills (see e iew by Wilkinson & Fung, 2002).
S uden s’ p io opic knowledge and eading luency we e no associa ed, wi h one
excep ion, wi h hei sou cing skills in he pos - es and he changes in hei sou cing pe -
o mance du ing he in e en ion. The ecen e iew by Anma k ud e al. (2021) epo ed
mixed esul s on he con ibu ion o eading luency and p io knowledge o s uden s’ sou c-
ing skills. The au ho s sugges ed ha mixed esul s may be ela ed o he used measu es
(Anma k ud e al., 2021). Ou esul s ega ding he ole o p io opic knowledge a e in line
wi h he s udy by Kamme e e al. (2016), who likewise applied ue/ alse i ems, and did no
ind an associa ion be ween s uden s’ p io opic knowledge and hei sou cing skills. Fu -
he , in ou s udy, he p io knowledge measu e only included six i ems. In e ms o eading
luency, uppe seconda y school s uden s ha e p obably eached a easonable le el so ha i
does no hinde hem in acqui ing sou cing skills. I is no able ha in Finland, a e 9 yea s
o compulso y comp ehensi e school, abou hal o he s uden s selec academic-o ien ed
uppe seconda y school.
Limi a ions and u u e esea ch
The s udy also has i s limi a ions. Fi s , we a anged a h ee-hou p o essional de elopmen
session o he eache s o he in e en ion g oup a couple o weeks be o e he in e en ion.
Al hough his included an in oduc ion o c i ical online eading skills, he ime was qui e
sho o eache s o each a p o ound unde s anding o sou cing in online eading. In u u e
s udies, a longe and mo e ecu en aining p og am (c . B å en e al., 2019) could be e
equip eache s o each sou cing du ing online inqui y and also challenge he compe encies
o s uden s possessing be e sou cing skills.
Second, sou cing in speci ying in o ma ion need was no augh as explici ly du ing he
in e en ion as sou cing in he o he phases o online inqui y. I was only implici ly embed-
ded in he ask assignmen and in he wo king documen when s uden s planned hei in o -
ma ion sea ch. In he u u e, s udies should imp o e he e icacy o hei in e en ions by
including mo e explici eaching on sou cing when de ining in o ma ion need.
Thi d, because he con en o language a s cou ses in uppe seconda y school is e y
b oad, he eache s we e no able o ind mo e ime o us o in es iga e he sus ainabili y
o he esul s wi h a delayed pos - es . As ou esul s showed di e en -le el changes in s u-
den s’ sou cing pe o mance du ing he in e en ion, in u u e esea ch, i would be impo -
an o asce ain how pe manen hese changes a e. I should be no ed ha he simila i y o
1 3
Teaching sou cing du ing online inqui y–adolescen s wi h he weakes …
he in e en ion g oup s uden s’ ou come means ac oss he di e en ly imed pos - es s does
no ell us how sus ainable he lea ned skills a e i changes a he indi idual le el om one
pos - es o ano he a e no also measu ed (c . B å en e al., 2019).
Ins uc ional implica ions
Ou esul s sugges ha he designed sou cing in e en ion has he po en ial o p omo e
uppe seconda y school s uden s’ sou cing skills. This equi es he explici eaching o
sou cing p ac ices and he sequenced p ac icing o s a egies ha ollow he ou online
inqui y phases. Ou s udy e ealed ha di e ging om hese p inciples is no wo hwhile.
Thus, educa o s applying he de eloped in e en ion should ensu e o explici ly each all
inqui y p ac ices, including sou cing in speci ying in o ma ion need (c . Heijl jes e al.,
2014; Ma in & Halpe n, 2011).
The ins uc ional me hods used in his s udy seemed pa icula ly bene icial o he s u-
den s wi h he weakes sou cing skills. Thus, highligh ing he a en ion o, e alua ion, and
use o sou ce in o ma ion h ough modeling, lec u ing, and sca olding s uden s wi h guid-
ing ques ions ep esen e icien me hods o eaching sou cing (c . B an e & S ømsø, 2018).
S uden s wi h he weake skills may also p o i om discussing and explo ing a con o e -
sial opic in small g oups, as his p o ides hem wi h oppo uni ies o disco e mo e ways
o e alua e, use, and in e p e sou ce in o ma ion in online ex s (e.g., Kiili e al., 2019).
Al hough he s uden s in he p esen s udy we e allowed o o m he small g oups by hem-
sel es, he sca olded small-g oup wo k combined wi h explici eaching seemed o be an
e icien me hod o s uden s wi h he weakes skills o lea n sou cing skills du ing online
inqui y (c . Wilkinson & Fung, 2002).
Despi e hese p omising esul s, ou in e en ion did no se e as e ec i ely he s uden s
who pe o med be e in he p e- es han s uden s who had he weakes skills in he p e- es .
This sugges s ha mo e a en ion should be pu o di e en ia ing ins uc ion, o example,
by ensu ing a su icien di icul y le el o he asks. This need was suppo ed by eache s’
commen s in hei dia ies. They epo ed ha e en hough he p omp s o e ed oppo uni ies
o s uden s o p ac ice sou cing a hei own le el, hey obse ed ha some asks we e oo
easy o oo di icul o pa o he s uden s.
Al hough he p esen in e en ion was designed o uppe seconda y school s uden s,
eaching sou cing h oughou online inqui y could be scaled down o seconda y and e en
uppe p ima y school s uden s. This would equi e he use o mo e conc e e concep s
h oughou he ask. Fo younge s uden s, sou cing in sea ch que ies could be limi ed o
p o essions and selec ed ex s o wo con adic o y ones w i en by a p o essional and a
laype son. As sou cing in w i en ex s is pa icula ly challenging o p ima y and second-
a y school s uden s (Kiili e al., 2020; Pé ez e al., 2018), s uden s’ composi ion o a w i en
p oduc could be sca olded wi h sen ence s a e s equi ing in eg a ion o sou ces in hei
w i ing. Wha e e he means o acili a ion, i is c i ical ha also younge s uden s also
expe ience sou cing when engaging in online inqui y.
The e a e se e al ways how ou in e en ion can be imp o ed. Fi s , p o iding eedback
on s uden s’ sou cing du ing online inqui y could sca old s uden s owa ds mo e sophis-
ica ed sou cing p ac ices. In he p esen s udy, ou design did no include any sys ema ic
eedback p ocedu es o guidelines o he eache s e en hough eedback plays a c ucial ole
1 3
E. K. Hämäläinen e al.
in s uden s’ lea ning (Ha ie & Timpe ley, 2007; Van de Kleij e al., 2015). The ex e nal
eedback om a eache is essen ial (Huisman e al., 2019), bu in some ci cums ances,
pee eedback can be as e ec i e as eache eedback (Huisman e al., 2019). Impo an ly,
pee eedback no only bene i s he ecei e bu also he p o ide , as i equi es s uden s o
ac i ely conside he c i e ia o ad anced sou cing (Huisman e al., 2018) and helps hem
o e lec on hei own sou cing skills (Van Pop a e al., 2017).
Secondly, mo e a en ion could be paid o designing engaging asks. In he p esen s udy,
we designed ou al e na i e ask scena ios on heal h issues ha we e connec ed o young
people’s li es. Acco ding o eache and s uden eedback, he opics did no , howe e , ini i-
a e in e es among some s uden s (see Kiili e al., 2022). This accen ua es he impo ance
o selec ing online inqui y opics ha a e bo h opical and no el among young people (c .
Anma k ud e al., 2021). A hei bes , opics will s imula e p oduc i e emo ions, such as
cu iosi y and enjoymen (Chinn e al., 2021).
Conclusions
When eading and lea ning h ough online in o ma ion, sou cing is one o he key p ac-
ices suppo ing he e alua ion o in o ma ion, comp ehension o mul iple iewpoin s, and
decision-making (Scha e & Salme ón, 2016). Sou cing is also an o e a ching p ac ice ha
can occu h oughou online inqui y, s a ing om he poin when eade s u n o he In e ne
o sol e a p oblem and ending when hey communica e hei indings o o he s (Kiili e al.,
2021). Ou s udy sugges s ha sou cing can be augh h oughou he online inqui y p ocess
by ca e ully designing sou cing p ac ices as an in eg al pa o online inqui y.
The apid sp ead o alse in o ma ion online has inc eased conce ns abou he ulne -
abili y o child en and adolescen s wi h low c i ical eading skills (Howa d e al., 2021).
Fo example, adolescen s who use social media equen ly end o o e look sou ces’ c ed-
ibili y (e.g., Macedo-Roue e al., 2020) which may lead hem o sp ead disin o ma ion
unin en ionally. Encou agingly, he in e en ion implemen ed he e succeeded in enhancing
he sou cing skills o he s uden s wi h he weakes skills. Howe e , sou cing is no e o -
less o adolescen s o easy o each o hem and hus, p omo ion o sou cing should be a
con inuous e o and implemen ed in di e en school subjec s.
Supplemen a y In o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps://doi.
o g/10.1007/s11251-022-09597-2.
Acknowledgemen s This esea ch was unded by he Academy o Finland (P ojec Numbe : 285806). The
au ho s would like o hank Timo Salminen and Tuulikki Alame älä o he da a collec ion, Ja i Hämäläinen,
Symcode Oy o he so wa e de elopmen , Michael F eeman o his aluable commen s on he language,
and he s uden s and eache s who pa icipa ed in he s udy.
Au ho con ibu ions Elina K. Hämäläinen: Concep ualiza ion, Me hodology, Resou ces, Fo mal analysis,
W i ing - o iginal d a p epa a ion.
Ca i a Kiili Concep ualiza ion, Me hodology, Resou ces, W i ing – e iew and edi ing, Supe ision.
Eija Räikkönen Me hodology, Fo mal analysis, W i ing – e iew and edi ing, Supe ision.
Minna Lakkala Concep ualiza ion, Resou ces, In es iga ion, W i ing – e iew and edi ing.
1 3