Tes ing he knowledge b idging unc ion o non-mains eam
jou nals in schola ly communica ion
Vic o ia Di Césa e*, Rod igo Cos as** and Nicolas Robinson-Ga cia***
* dicesa e@ug .es
h ps://o cid.o g/0000-0001-7098-1914
EC3 Resea ch G oup, Uni o Compu a ional Humani ies and Social Sciences (U-CHASS), Uni e si y o
G anada, Spain
** [email p o ec ed].nl
h ps://o cid.o g/0000-0002-7465-6462
Cen e o Science and Technology S udies (CWTS), Leiden Uni e si y, The Ne he lands
DSI-NRF Cen e o Excellence in Scien ome ics and Science, Technology and Inno a ion Policy (SciSTIP),
S ellenbosch Uni e si y, S ellenbosch, Sou h A ica
***el obin@ug .es
h ps://o cid.o g/0000-0002-0585-7359
EC3 Resea ch G oup, Uni o Compu a ional Humani ies and Social Sciences (U-CHASS), Uni e si y o
G anada, Spain
This s udy explo es he knowledge b idging communica ion unc ion o non-mains eam jou nals, o iginally
p oposed by Cha a o e al. (2017). Based on hei de ini ion, and inco po a ing a mul idimensional concep ion
o locali y (Di Césa e & Robinson-Ga cia, 2024), we iden i y jou nals ha pe o m his unc ion wi hin a la ge
2023 OpenAlex da ase . Ou indings e eal a p edominance o such jou nals in Asian and Sou h Ame ican
coun ies. La ge scien i ic p oduce s like B azil and Russia s and ou as signi ican con ibu o s o knowledge
b idging jou nals. The Social Sciences domain leads a icle p oduc ion, ollowed closely by he Physical
Sciences. The s udy highligh s he alue o non-mains eam jou nals as knowledge b idging sou ces beyond
adi ional me ics and unde sco es he need o g ea e ecogni ion o hei ole in schola ly communica ion.
1. In oduc ion
Much has been deba ed abou he dicho omy be ween mains eam and non-mains eam
jou nals. This includes di e ing pe cep ions o hei quali y and excellence, which impac
ecogni ion du ing esea ch assessmen s (Bianco e al., 2016; Vessu i e al., 2014), as well as
hei une en isibili y and access oppo uni ies, o en shaped by economic and language
ba ie s (Tijssen e al., 2006). Such ac o s con ibu e o an unequal app aisal o he esea ch
published in non-mains eam enues ha jeopa dizes ce ain o ms o knowledge while
p omo ing o he agendas (López Piñei o & Hicks, 2015). Thei dis inc ion canno be
explained based on uni e salis ic c i e ia, like edi o ial s anda ds and scien i ic impac ,
because pa icula is ic a iables such as coun y, discipline, and language s ongly de e mine
he inclusion o jou nals in da abases like Web o Science (Cha a o e al., 2018). These
da abases, widely conside ed mains eam, a e known o being s ongly biased owa d
Wes e n, Anglophone con en (La i iè e e al., 2015; an Leeuwen e al., 2001).
Consequen ly, he accep ed assump ion ha non-mains eam, usually implying non-indexed,
jou nals lack alue in dissemina ing esea ch indings is called in o ques ion.
Cha a o e al. (2017) analyzed he ole played by non-mains eam jou nals and he easons
leading esea che s o publish in hem. They conduc ed a se ies o in e iews wi h Colombian
esea che s om which hey iden i ied h ee main communica ion unc ions. The i s
unc ion is aining, p ima ily o young esea che s and PhD s uden s. In hese jou nals, hey
9
ind a space o ini ia ion in o publishing, inco po a ion o eedback om edi o s and pee
e iewe s, iden i ica ion o ele an li e a u e in hei na i e languages, and o e all capaci y
building ha will e en ually help hem each mains eam enues. The second is knowledge
b idging, a unc ion ha connec s a icles om mains eam jou nals o communi ies wi h
limi ed o no access o hem. This is mainly done wi h dissemina ion in non-English
languages and open accessibili y pu poses, in close ela ion o eaching and in oduc ion o
concep s and me hodologies o local communi ies. Finally, he hi d unc ion is knowledge
gap- illing, used o p e en neglec o unde ep esen a ion o ce ain opics. Communica ion
unc ions wo and h ee may o e lap and e y well apply o esea che s who can publish in
he mains eam bu ins ead choose non-mains eam jou nals o ul il hese oles.
In his s udy we e isi he knowledge b idging unc ion as desc ibed by Cha a o e al.
(2017), si ua ing i wi hin a mul idimensional amewo k o local esea ch we ecen ly
de eloped (Di Césa e & Robinson-Ga cia, 2024). This amewo k concep ualizes locali y
h ough mul iple dimensions, allowing o a mo e nuanced unde s anding o how knowledge
ci cula es ac oss geog aphical and linguis ic bounda ies. Building on his, we de ine
knowledge b idging jou nals as hose ha se e local audiences –ope a ionalized by he
geog aphic concen a ion o ci a ions ecei ed– while d awing om globally dis ibu ed
knowledge sou ces –as indica ed by he geog aphic o igin o hei ci ed e e ences. We also
limi ou analysis o jou nals publishing in non-English languages. Acco dingly, we iden i y
knowledge b idging jou nals by a combina ion o global e e encing, local ci ing, and
non-English publishing. Wi h his esea ch, we in end o deepen ou unde s anding o he ole
di e en jou nals play in a scien i ic communica ion s ill deeply in e sec ed by he
mains eam/non-mains eam di ide, while g an ing hem ecogni ion as ele an sou ces
beyond adi ional impac indica o s.
2. Da a and me hods
We wo ked wi h a la ge OpenAlex da ase comp ising 59,230 jou nals ha published a icles
du ing 2023, en iched wi h Web o Science Mas e Jou nal Lis (MJL), Scopus, and DOAJ
language da a. We use he OpenAlex e sion p o ided by he InSySPo p og am h ough
Google BigQue y (public da abase las upda ed Ma ch 2025) (Mazoni & Cos as, 2024). We
es ablished a h eshold o e ain only hose jou nals wi h a p opo ion o a icles wi h
e e ences equal o o g ea e han 50% o a oid skewed e e ence p ocessing. All
compu a ions we e un in R ( e sion 4.4.0). Bo h he R and SQL codes can be ound in Di
Césa e (2025).
Figu e 1 p o ides an o e iew o how we ope a ionalize he knowledge b idging unc ion. We
combine he use o e e ences, ci a ions, and language o iden i y knowledge b idging
jou nals. The in e ening a iables, as well as hei de ini ions, measu emen me hods, and
h esholds a e based on Di Césa e & Robinson-Ga cia (2024). The e e enced p opo ion
a iable was de ined as he p opo ion o appea ance o he maximum e e enced coun y pe
jou nal acco ding o he geog aphical o igin o e e ences. Simila ly, he ci ing p opo ion
a iable cap u es he maximum ci ing coun y based on he o igin o ci a ions. In addi ion, he
non-English publishing a iable conside s whe he he jou nal publishes in languages o he
han English acco ding o MJL, Scopus, DOAJ, and OpenAlex da a. In Appendix 1 we lis
examples o sou ces ha mee he condi ions o be iden i ied as knowledge b idging jou nals.
2
Figu e 1: Ope a ionaliza ion o he knowledge b idging communica ion unc ion ac oss local
esea ch dimensions.
To build he jou nal subse s, we applied he hi d qua ile o each con inuous a iable’s
dis ibu ion as a cu -o poin . Fo e e enced p opo ion, we selec ed jou nals in which he
mos equen ly e e enced coun y accoun ed o less han 42% o all ci ed e e ences
–indica ing a mo e global e e encing pa e n. Fo ci ing p opo ion, we included jou nals
whe e o e 86% o ci a ions came om a single coun y, e lec ing a locally concen a ed
eade ship. In cases whe e he same coun y is bo h he mos equen ly e e enced and he op
ci ing coun y o a gi en jou nal, bu i s p opo ions emain wi hin he es ablished h esholds,
he jou nal is s ill conside ed o exhibi a global e e encing pa e n. In o he wo ds, a sha e o
e e ences om he op- e e enced coun y below 42% is indica i e o geog aphic dispe sion,
e en i ha same coun y also domina es he jou nal’s ci ing pa e n (Appendix 1). Language
was ea ed as a dicho omous a iable, wi h all non-English ou pu s conside ed local. We
iden i ied language combining OpenAlex in o ma ion wi h ha o o he da abases,
p io i izing he in o ma ion ob ained om Scopus, Web o Science, and DOAJ, when
a ailable. This me hodological decision was made in ligh o he accu acy issues iden i ied by
Céspedes e al. (2025). As a esul , we ob ained h ee subse s o 44,270 jou nals wi h global
e e ences, 13,424 jou nals wi h local ci a ions, and 7,984 jou nals wi h non-English language
publica ions.
3. Resul s
In Figu e 2, we iden i y and cha ac e ize he knowledge b idging jou nals wi hin he con ex
o his OpenAlex da ase (N = 59,230). Acco ding o he condi ions and h esholds ou lined
be o e, 1,271 jou nals (S1) simul aneously exhibi global e e ences, local ci a ions, and
non-English language a icles (A). O he possible in e media e subse s include hose
combining global e e ences and non-English languages (S2, n = 4883), jou nals wi h local
ci a ions and non-English a icles (S3, n = 3090), and hose con aining global e e ences bu
local ci a ions (S4, n = 6192). When compa ing he knowledge b idging jou nals o he o he
subse s in e ms o hei a e age p oduc ion (B) and a e age impac (C), we ind ha S2
published he mos a icles in 2023 (x
p = 27) and ecei ed he mos ci a ions (x
i = 28). In
3
con as , he a e age p oduc ion and impac o knowledge b idging jou nals (x
p = 20, x
i = 8),
S3 (x
p = 18, x
i = 9), and S4 (x
p = 19, x
i = 7) we e app oxima ely one- hi d lowe han he
o me mean alues.
Figu e 2: In e sec ions o he (A) iden i ica ion o knowledge b idging jou nals (S1), hei (B)
a e age scien i ic p oduc ion and (C) a e age impac compa ed o subse s S2, S3 and S4.
Figu e 3 shows he o al and pe cen age dis ibu ion o knowledge b idging jou nals’
p oduc ion in he op 20 coun ies. In e ms o o al publica ions wi hin jou nals ha pe o m
his unc ion (A), B azil clea ly domina es he lis wi h 5,260 a icles, ollowed by Russia (n =
3,594) and Indonesia (n = 3,573). F om Tu key (n = 1,363) and Spain (n = 1,298) onwa ds,
alues d op by hal , wi h a ied coun ies like Japan, Mexico, F ance, A gen ina, China, and
he Uni ed S a es each accumula ing be ween 800 and 600 publica ions each. Mo e han hal
o hese op 20 coun ies belong o No h Ame ica and Eu ope, wi h Eu ope con ibu ing he
mos cases. Howe e , hei combined p oduc ion (n = 9,285) does no su pass he o al ou pu
o he Sou h Ame ican and Asian coun ies in he lis (n = 13,406). (A) also shows ha , when
conside ing hese coun ies’ o al 2023 ou pu , hei sha e o publica ions wi hin knowledge
b idging jou nals is e y low, anging om 0.08% in he case o China o 3.7% o B azil.
As o he sha e o publica ions o each coun y o al ou pu ha is published in knowledge
b idging jou nals (B), we obse e ha he dis ibu ion o coun ies changes signi ican ly. The
A ican Republic o Guinea-Bissau epo s he highes p opo ion o a icles in knowledge
b idging jou nals (24%) ela i e o i s 2023 o al p oduc ion, ollowed by Nica agua (9%).
Beyond hese wo, he emaining coun ies display qui e homogenous sha es below 5% and
dec easing. Mos o hem a e om Sou h and Cen al Ame ica and he Ca ibbean, wo egions
ha in oduce coun ies a he highes places o his op 20 which a e no p esen in (A)
because o hei mo e limi ed o e all p oduc ion o scien i ic pape s. Such a e he cases o
Hondu as, Pa aguay, Boli ia, and Gua emala, whose 2023 p oduc ion anges be ween 20 and
50 a icles. The middle sec ion o his skewed dis ibu ion includes only a ew Asian and
Eu opean coun ies (e.g. Timo -Les e and Russia), while A ica con ibu es one mo e owa ds
he end (Mozambique). B azil and Russia s and ou in bo h (A) and (B) because hey a e la ge
4
coun ies in e ms o hei o e all scien i ic ou pu ha also con ibu e signi ican ly o
knowledge b idging jou nals.
Figu e 3: Top 20 coun ies wi hin knowledge b idging jou nals a anged acco ding o hei (A)
o al p oduc ion and (B) publica ions sha e.
Figu e 4 displays he dis ibu ion o knowledge b idging jou nals acco ding o OpenAlex’s
ca ego iza ion in o ields and domains. I is impo an o no e ha OpenAlex assigns hese
5
ca ego ies o indi idual a icles. As a esul , when we agg ega e hem a he jou nal le el, a
single jou nal may be associa ed wi h mul iple unique ields and domains simul aneously. The
Social Sciences ield o he homonymous domain p esen s he highes p opo ion o
knowledge b idging jou nals (88%, n = 1119 jou nals), ollowed by Heal h P o essions (54%,
n = 685) om he Heal h Sciences domain, and Compu e Science (53%, n = 676) om he
Physical Sciences. The lowes p opo ions o knowledge b idging jou nals a e ound in
Chemical Enginee ing (2%, n = 25) om he Physical Sciences, Ve e ina y (3%, n = 42), and
Den is y (5%, n = 68), bo h ields belonging o Heal h Sciences.
The Social Sciences domain leads he dis ibu ion wi h 94% (n = 1195) o knowledge
b idging jou nals con aining a icles assigned o one o mo e o i s ields. No a om hose
alues, we ind ha 83% (n = 1049) o he knowledge b idging jou nals belong o he Physical
Sciences, wi h ields like En i onmen al Science (52%, n = 666) and Enginee ing (35%, n =
442) being also qui e p edominan in hem. The Heal h and Li e Sciences domains a e
ep esen ed mo e mode a ely in hese jou nals (70%, n = 888, and 59%, n = 744 espec i ely),
wi h Medicine (51%, n = 651) closely ollowing he Heal h P o essions ield wi hin he i s
domain, and Ag icul u al and Biological Sciences (46%, n = 591) clea ly domina ing he
second one.
Figu e 4: Dis ibu ion o he sha e o knowledge b idging jou nals by OpenAlex ields and
domains ca ego ies.
4. Discussion
In his wo k in p og ess, we ha e ope a ionalized he knowledge b idging communica ion
unc ion o es Cha a o e al.'s (2017) de ini ion. The esul s ob ained so a indica e ha
knowledge b idging jou nals ha e a p edominance o Asian and Sou h Ame ican coun ies as
a icle p oduce s. We obse e ha hese Global Sou h egions ha e a s onge endency o
publish in hei own languages in his ype o jou nals, which a e also la gely g ounded in
knowledge om o he pa s o he wo ld bu p ima ily ci ed locally. The cases o B azil and
Russia a e qui e singula , as hey a e la ge scien i ic p oduce s which also con ibu e
6
signi ican ly o knowledge b idging jou nals. The Social Sciences s and ou as he main
domain in o which hese jou nals a e ca ego ized. Howe e , he Physical Sciences also play a
p ominen ole, meaning ha concep s and me hodologies om i s ields could be
communica ed, ansla ed, o adap ed o local communi ies h ough he knowledge b idging
jou nals.
While ou indings align wi h he model p oposed in Cha a o e al. (2017), u he esea ch
is needed o assess whe he con ibu o s o knowledge b idging jou nals a e indeed p ima ily
ea ly-ca ee esea che s, as well as o add ess he pe sis en challenge o iden i ying
knowledge b idging jou nals wi hin anglophone communi ies. Mo ing o wa d, we also
expec o del e deepe in o he p ecise o igin o e e ences and ci a ions, wi h he goal o
es ablishing a b idge no only be ween mains eam and non-mains eam jou nals bu also
be ween scien i ic publica ions om he Global No h and he Global Sou h. Ul ima ely, we
in end o con inue wo king on he ope a ionaliza ion o he aining and knowledge gap- illing
communica ion unc ions, whe e he dimensions o he local in e sec wi h epis emic
app oaches and academic ajec o ies.
Open science p ac ices
We used open da a om he OpenAlex da abase o conduc his esea ch. I s o igin and
cha ac e is ics a e ully desc ibed in he Da a and me hods sec ion o allow o eplicabili y.
Fu he mo e, we sha ed he comple e R and SQL code h ough an open eposi o y on Gi Hub,
which is also ci ed in his pape .
Au ho con ibu ions
VDC: Concep ualiza ion, Da a cu a ion, Fo mal analysis, In es iga ion, Me hodology,
So wa e, Visualiza ion, W i ing - o iginal d a , W i ing - e iew & edi ing.
NRG: Concep ualiza ion, Funding acquisi ion, Me hodology, P ojec adminis a ion,
Resou ces, Supe ision, Valida ion, W i ing - e iew & edi ing.
RC: Supe ision, W i ing - e iew & edi ing.
Compe ing in e es s
The au ho s decla e no compe ing in e es s.
Funding in o ma ion
This wo k is pa o he COMPARE p ojec (Re : PID2020-117007RA-I00) unded by he
Spanish Minis y o Science (Re : MCIN/AEI/10.13039/501100011033 FSE in ie e en u
u u o). Vic o ia Di Césa e is cu en ly suppo ed by a FPI g an om he Spanish Minis y o
Science (Re : PRE2021-097022). Nicolas Robinson-Ga cia is cu en ly suppo ed by a
Ramón y Cajal g an om he Spanish Minis y o Science (Re : RYC2019-027886-I).
Re e ences
Bianco, M., G as, N., & Su z, J. (2016). Academic E alua ion: Uni e sal Ins umen ? Tool o
De elopmen ? Mine a, 54(4), 399–421. h ps://doi.o g/10.1007/s11024-016-9306-9
Céspedes, L., Kozlowski, D., P adie , C., Sain e‐Ma ie, M. H., Shokida, N. S., Benz, P.,
Poi as, C., Ninko , A. B., Eb ahimy, S., Ayeni, P., Filali, S., Li, B., & La i iè e, V.
(2025). E alua ing he linguis ic co e age o OpenAlex: An assessmen o me ada a
7
accu acy and comple eness. Jou nal o he Associa ion o In o ma ion Science and
Technology, asi.24979. h ps://doi.o g/10.1002/asi.24979
Cha a o, D., Rà ols, I., & Tang, P. (2018). To wha ex en is inclusion in he Web o Science
an indica o o jou nal ‘quali y’? Resea ch E alua ion, 27(2), 106–118.
h ps://doi.o g/10.1093/ ese al/ y001
Cha a o, D., Tang, P., & Rà ols, I. (2017). Why esea che s publish in non-mains eam
jou nals: T aining, knowledge b idging, and gap illing. Resea ch Policy, 46(9),
1666–1680. h ps://doi.o g/10.1016/j. espol.2017.08.002
Di Césa e, V. (2025). Local jou nals code (Ve sion 1.0.0) [R, SQL].
h ps://gi hub.com/ dicesa e/Local.Jou nals (O iginal wo k published 2025)
Di Césa e, V., & Robinson-Ga cia, N. (2024). Wha is local esea ch? Towa ds a
mul idimensional amewo k linking heo y and me hods. Zenodo.
h ps://doi.o g/10.5281/zenodo.14033472
La i iè e, V., Haus ein, S., & Mongeon, P. (2015). The Oligopoly o Academic Publishe s in
he Digi al E a. PLOS ONE, 10(6), e0127502.
h ps://doi.o g/10.1371/jou nal.pone.0127502
López Piñei o, C., & Hicks, D. (2015). Recep ion o Spanish sociology by domes ic and
o eign audiences di e s and has consequences o e alua ion. Resea ch E alua ion,
24(1), 78–89. h ps://doi.o g/10.1093/ ese al/ u030
Mazoni, A., & Cos as, R. (2024, June 6). Towa ds he democ a isa ion o open esea ch
in o ma ion o scien ome ics and science policy: The Campinas expe ience. Leiden
Mad ics.
h ps://www.leidenmad ics.nl/a icles/ owa ds- he-democ a isa ion-o -open- esea ch-i
n o ma ion- o -scien ome ics-and-science-policy- he-campinas-expe ience
8
Tijssen, R. J. W., Mou on, J., an Leeuwen, T. N., & Bosho , N. (2006). How ele an a e
local schola ly jou nals in global science? A case s udy o Sou h A ica. Resea ch
E alua ion, 15(3), 163–174. h ps://doi.o g/10.3152/147154406781775904
an Leeuwen, T. N., Moed, H. F., Tijssen, R. J. W., Visse , M. S., & Van Raan, A. F. J. (2001).
Language biases in he co e age o he Science Ci a ion Index and i s consequences
o in e na ional compa isons o na ional esea ch pe o mance. Scien ome ics, 51(1),
335–346.
Vessu i, H., Guédon, J.-C., & Ce o, A. M. (2014). Excellence o quali y? Impac o he
cu en compe i ion egime on science and scien i ic publishing in La in Ame ica and
i s implica ions o de elopmen . Cu en Sociology, 62(5), 647–665.
h ps://doi.o g/10.1177/0011392113512839
9
