Does the type of study on Covid-19 influence the value of Altmetrics?
Abstract
More than half of the News mentions are not analyzed because period studied is very recent, while volume of information about Covid-19 is high. Scientific evidence Pyramid is well reflected by the Altmetric Attention Score. At the forefront we find Consensus development conferences & guidelines, given their great utility in a period of such uncertainty about the pandemic and the fact that they are usually backed by international organizations. Consensus development conferences are a way to bring together citizens, decision-makers and experts to address issues of public importance. Involve a series of experts who deliver scientific evidence on a subject. This category is followed by Clinical trials, at the peak of the pyramid of scientific evidence, since their results are highly valid and it can be considered the flame of scientific knowledge. Behind these two are the Systematic reviews and Meta-analyses, with similar values. On the other hand, Systematic review attempts to gather all available empirical research by using clearly defined systematic methods to obtain answers to a specific question. A meta-analysis is the statistical process of analyzing and combining results from several similar studies. Reviews, in addition to their educational component, are hypothesis generators, which is very important to analize a new topic such as Covid-19. The lowest values in all the altmetrics were Case reports; although they are useful to start a clinical investigation, is the base of the pyramid. The graphic trends of Twitter mentions and the Altmetric Attention Score are nearly identical, due to AAS collects the 95% of Twitter mentions. We can conclude that altmetric in biomedical research, specifically in the covid research front, could be highly determined by the type of study, and that altmetric can capture the utility or altmetric at attention marker.
Full text
Resea ch in p og ess
STI 2022 Con e ence P oceedings
P oceedings o he 26 h In e na ional Con e ence on Science and Technology Indica o s
All pape s published in his con e ence p oceedings ha e been pee e iewed h ough
a pee e iew p ocess adminis e ed by he p oceedings Edi o s. Re iews we e
conduc ed by expe e e ees o he p o essional and scien i ic s anda ds expec ed o
a con e ence p oceedings.
P oceeding Edi o s
Nicolas Robinson-Ga cia
Daniel To es-Salinas
Wenceslao A oyo-Machado
Ci a ion: Valde ama, P., & To es-Salinas, D.
(2022). Does he ype o s udy on Co id-19
in luence he alue o Al me ics?. In N. Robinson-
Ga cia, D. To es-Salinas, & W. A oyo-Machado
(Eds.), 26 h In e na ional Con e ence on Science
and Technology Indica o s, STI 2022 (s i22233).
h ps://doi.o g/10.5281/zenodo.6957471
Copy igh : © 2022 he au ho s, © 2022 Facul y o Communica ion
and Documen a ion, Uni e si y o G anada, Spain. This is an open
access a icle dis ibu ed unde he e ms o he C ea i e Commons
A ibu ion License.
Collec ion: h ps://zenodo.o g/communi ies/s i2022g x/
STI 2022
F om Global Indica o s o Local Applica ions
STI 2022 | h ps://doi.o g/10.5281/zenodo.6957471
1 / 6
Does he ype o s udy on Co id-19 in luence he alue o Al me ics?
Pila Valde ama* and Daniel To es-Salinas*
* [email p o ec ed]; o ess[email p o ec ed]
Depa men o In o ma ion & Communica ion, Uni e si y o G anada, Campus Ca uja, G anada 18071 (Spain)
In oduc ion
Co id-19 has become a global esea ch on ha is being s udied om di e en bibliome ic
pe spec i es. Some o hese wo ks ha e ocused on de e mining he a e o g ow h (To es-
Salinas, 2020), on o e ing a gene al bibliome ic pe spec i e o ac o s and hemes (Cola izza
e al. 2021) o on analyzing he deba e gene a ed by scien i ic esul s on co id in socie y (Nane
e al., 2021). This las line, ha is, he measu emen o he social in luence o a icles on
Co id19, is ano he one o he pe spec i es ha has gene a ed he mos wo k, especially when
using al me ic indica o s (To es-Salinas e al., 2020). Al me ics a e al e na i e indica o s
de i ed om he men ions made in he media, web 2.0 and social ne wo ks o scien i ic esul s
(To es-Salinas e al., 2020) and a e acked o explo e he social ecep ion o esea ch indings
(P iem e al., 2010). These new me ics ha e eme ged ha a e helping o unde s and aspec s o
science communica ion beyond adi ional channels (A oyo-Machado e al. 2021), which is
why hey ha e been impo an when s udying he co id-19 phenomenon.
Howe e , no all esul s, and especially hose published in scien i ic jou nals, ecei e he same
a en ion om an al me ic poin o iew. In his sense, i could be a gued ha in he medical
ield di e en documen a y ypologies ( o example, case epo s, clinical ials, e iews...)
could ecei e di e en a en ion. Cu en ly da abases such as Pubmed o e de ailed
in o ma ion on he ypology, speci ically classi ying hei wo ks in o 7 ca ego ies: 1. Case
epo s; 2. Clinical ials; 3. Consensus De elopmen Con e ences & Guidelines; 4. Re iews;
5. Sys ema ic Re iews; 6. Me a-Analyses; 7. Obse a ional s udies. Conside ing all o he
abo e, he main objec i e o his communica ion is o es ablish whe he he a en ion, measu ed
h ough al me ic indica o s, ecei ed by scien i ic publica ions on co id-19 published in
biomedical jou nals depends on he documen a y ype. Fo his, he wo ks published du ing he
yea 2021 on Co id19 in he Pubmed da abase will be conside ed as a sample, which will be
analyzed h ough s a is ical in e ence me hods.
Me hodology
Fi s ly, he eco e y o wo ks on Co id19 was p oceeded. A e sea ching o Co ona i us as
he Mesh e m, in he page o esul s we used il e s om PubMed Clinical Que ies o e ine
he sea ch, selec ing “Co id-19” o he il e Ca ego y wi hin he il e ield “Gene al”, hus
ob aining a o al o 124,141 esul s. The yea selec ed o ob ain a icles was 2021; he ypes o
STI 2022
F om Global Indica o s o Local Applica ions
STI 2022 | h ps://doi.o g/10.5281/zenodo.6957471
2 / 6
publica ion selec ed o he s udy we e hose 8 men ioned in he In oduc ion. A o al o 21,218
publica ions we e inally downloaded o ou pape . To e ie e he al me ics, Al me ic.com
was used as a sou ce, using he DOI as an iden i ie o he e ie al o wo ks. In o al 17,259
we e e ie ed h ough hei DOI. As i is a wo k in p og ess, no all he al me ic indica o s
ha e been used and some o hem we e disca ded based mainly on h ee c i e ia (a) Pla o ms
wi h an i ele an numbe o men ions (eg. You ube, Policy epo s, Wikipedia, Blog men ions,
e c.), (b) Pla o ms wi h a s ong geog aphic componen (e.g. Weibo o Reedi ) and Pla o ms
ha no longe exis o ope a e (eg. Google Plus, Syllabi). Finally, he al me ic indica o s
conside ed we e: Al me ic A en ion Sco e, News men ions, Twi e men ions and Mendeley
eade s. Likewise, Dimension ci a ions we e conside ed o make a compa ison wi h adi ional
bibliome ic indica o s. Once he o al da a se had been classi ied, he i s s ep was o calcula e
he mean alue o he indica o s o each ype o s udy, and a g aphic ep esen a ion was made.
Nex , conside ing he eigh ypes o s udy as e e ence a iables, he no mali y hypo hesis was
con as ed using he Kolmogo o -Smi no es . When i was signi ican , compa ison be ween
he a iables was made using he non-pa ame ic K uskal-Wallis es . Finally, he possible
associa ion be ween he ype o s udy and he di e en al me ics was app aised by applying
he chi-squa e es o he con ingency able ob ained.
Resul s
Table 1. F equency, median and in e qua ile in e als o al me ics o he eigh ypes o s udy
Gene al Indica o s
Median alues (in e qua ile in e al)
S udy ype acco ding o
ype PubMed da abase
Num.
Publ.
To al
men ions
Al me ic
A en. Sco e
News
men ions
Twi e
men ions
Mendeley
eade s
Dim. ci a ions
1. Case Repo
2052
1597
4
(1-12)
0
(0-1)
3
(1-14)
25
(15-41)
2
(1-6)
2. Clinical T ial
848
772
11
(3-68)
0
(0-3)
10
(3-54)
64
(34-115)
7
(2-24)
3. Consensus De . Con .
Guideline
83
72
11.5
(3-46)
0
(0-2)
14.5
(3-49.75)
49.5
(30-97.25)
6.5
(2-12)
4. Re iews
7454
5637
6
(2-17)
0
(0-1)
5
(2-19)
46
(25-86)
5
(1-13)
5. Sys ema ic Re .
554
504
8
(2-25)
0
(0-1)
9
(3-26.75)
60
(32.25-102)
7
(2-17)
6. Me a-Analysis
153
126
8
(2-29.25)
0
(0-1)
10
(2-32)
47.5
(27.75-77.25)
8
(3-20.25)
7. Obse . S udy
1821
1512
7
(2-22)
0
(0-1)
6
(2-22)
41
(23-71)
5
(1-12)
The able o equencies and he median alues o he al me ics o he se en ypes o s udy
a e shown in Table 1, while he co esponding g aphic ep esen a ion o he means is displayed
in Figu e 1. When he da a does no i a no mal dis ibu ion, i is mo e co ec o use he median.
The median is much mo e obus , which means ha i is less a ec ed by he p esence o biases
in he dis ibu ion o ex eme alues.
STI 2022
F om Global Indica o s o Local Applica ions
STI 2022 | h ps://doi.o g/10.5281/zenodo.6957471
3 / 6
Figu e 1. Means o he i e indica o s
conside ed o he di e en s udy ypes
o Co id19 esea ch p oduc ion in
PubMed. The indica o s a e Ci a ions in
Dimensions, Al me ic A en ion Sco e,
News Men ions, Twi e and Mendeley
Reade s. Da a om Al me ic.com
The applica ion o he Kolmogo o -Smi no es was signi ican in all cases (p<0.001), which
indica es ha he s udy ype a iables we e no dis ibu ed acco ding o he Gauss model.
The e o e, compa isons be ween he al me ics o he eigh s udy ypes we e ca ied ou , in
o de o analyse how he selec ed indica o s, beha e acco ding o each s udy ype and as a
unc ion o he le el o he “py amid” o scien i ic e idence (Figu e 2), by means o he
K uskal-Wallis es , esul ing in p<0.001, meaning he e we e signi ican di e ences among
he al me ics o each s udy ype. The Twi e pla o m ha es ed he highes numbe o
men ions, ollowed by he Al me ic A en ion Sco e, Mendeley eade s, and a behind,
Dimension ci a ions and News men ions. A c oss- abula ion be ween S udy ype and al me ics,
collec ing he join equencies, is shown in Table 2. The chi-squa e es o independence was
also signi ican (χ2=136,128.95; p<0.001), so he null hypo hesis was ejec ed, admi ing ha
he e is an associa ion be ween he wo.
STI 2022
F om Global Indica o s o Local Applica ions
STI 2022 | h ps://doi.o g/10.5281/zenodo.6957471
4 / 6
Table 2. C oss- abula ion o join equencies be ween al me ic and s udy ype
S udy
ype
Indica o s
Al me ic
A en . Sco e
News
men ions
Twi e
men ions
Mendeley
eade s
Dim.
ci a ions
1. Case Repo
2. Clinical T ial
3. Consensus Guideline
4. Re iews
5. Sys ema ic Re
6. Me a-Analysis
7. Obse . S udy
84,410
229,425
5,035
330,780
47,562
22,268
157,331
2,371
14,002
243
13,128
2,929
1,016
9,993
138,225
320,017
6,315
526,306
58,451
32,419
201,869
51,500
85,071
5,252
424,562
44,617
8,901
88,071
9,723
30,616
1,243
77,842
8,209
2,578
20,918
Conclusion
Mo e han hal o he News men ions a e no analyzed because pe iod s udied is e y ecen ,
while olume o in o ma ion abou Co id-19 is high. Scien i ic e idence Py amid (Kowalcyk
and T uluck, 2013; Mu ad e al., 2016) is well e lec ed by he Al me ic A en ion Sco e. A
he o e on we ind Consensus de elopmen con e ences & guidelines (3), gi en hei g ea
u ili y in a pe iod o such unce ain y abou he pandemic and he ac ha hey a e usually
backed by in e na ional o ganiza ions.
Consensus de elopmen con e ences a e a way o b ing oge he ci izens, decision-make s and
expe s o add ess issues o public impo ance. In ol e a se ies o expe s who deli e scien i ic
e idence on a subjec . This ca ego y is ollowed by Clinical ials (2), a he peak o he py amid
o scien i ic e idence, since hei esul s a e highly alid (Lazcano-Ponce e al., 2004) and i
can be conside ed he lame o scien i ic knowledge. Behind hese wo a e he Sys ema ic
e iews (6) and Me a-analyses (7), wi h simila alues.
On he o he hand, Sys ema ic e iew a emp s o ga he all a ailable empi ical esea ch by
using clea ly de ined sys ema ic me hods o ob ain answe s o a speci ic ques ion. A me a-
analysis is he s a is ical p ocess o analyzing and combining esul s om se e al simila s udies
(Ha is e al., 2014). Re iews (5), in addi ion o hei educa ional componen , a e hypo hesis
gene a o s, which is e y impo an o analize a new opic such as Co id-19 (Valde ama e al.,
2021). The lowes alues in all he al me ics we e Case epo s (1); al hough hey a e use ul o
s a a clinical in es iga ion, is he base o he py amid. The g aphic ends o Twi e men ions
and he Al me ic A en ion Sco e a e nea ly iden ical, due o AAS collec s he 95% o Twi e
men ions.
We can conclude ha al me ic in biomedical esea ch, speci ically in he co id esea ch on ,
could be highly de e mined by he ype o s udy, and ha al me ic can cap u e he u ili y o
al me ic a a en ion ma ke .
STI 2022
F om Global Indica o s o Local Applica ions
STI 2022 | h ps://doi.o g/10.5281/zenodo.6957471
5 / 6
Figu e 2. Py amid adap ed om E idence-
Based Medicine by Uni e si y o Washing on
Heal h Sciences Lib a y (Kowalczyk and
T uluck, 2013) and om Mu ad e al. (2016)
* I has been conside ed ha he consensus
con e ences and he guidelines, i hey ha e been
based on sys ema ic e iews and me a-analyses,
could be placed in he highes pa o he scien i ic
py amid, since hey a e o ms o syn hesis o
in o ma ion o clinical applica ion.
Re e ences
A oyo-Machado, W., To es-Salinas, D. & Robinson-Ga cia, N (2021). Iden i ying and
cha ac e izing social media communi ies: a socio-seman ic ne wo k app oach o al me ics.
Scien ome ics, 126, 9267–9289
Cola izza, G., Cos as, R., T aag, V.A., an Eck, N.J., an Leeuwen, T., y Wal man, L. (2021)
A scien ome ic o e iew o CORD-19. PLoS ONE, 16(1), e0244839.
Ha is, J.D., Qua man, C.E., Man ing, M.M., Sis on, R.A. y Flanigan, D.C. (2014). How o
w i e a sys ema ic e iew. Ame ican Jou nal o Spo s Medicine, 42(11), 2761-2768.
Kowalczyk, N. y T uluck, C. (2013): Li e a u e e iews and sys ema ic e iews: Wha is he
di e ence? Radiologic Technology, 85(2), 219-222.
Lazcano-Ponce, E., Salaza -Ma ínez, E., Gu ié ez-Cas ellón, P., Angeles-Lle enas, A.,
He nández-Ga duño, A. y Vi amon es, J.L. (2004). Ensayos clínicos alea o izados: a ian es,
mé odos de alea o ización, análisis, conside aciones é icas y egulación. Salud Pública de
México, 46(6), 559-584.
Mu ad, M.H., Asi, N., Alsawas, M. y Alahdab, F. (2016). New e idence py amid. BMJ
E idence-Based Medicine, 21, 125–127.
Nane, G.F., an Schalkwyk, F., Dudek, J., To es-Salinas, D., Cos as, R., Robinson-Ga cia,
N. (2021). The Role o Scien i ic Ou pu in Public Deba es in Times o C isis: A Case S udy
o he Reopening o Schools Du ing he COVID-19 Pandemic. Communica ion and
Resilience du ing Pandemics. Da id M. Be ube (Ed.).
P iem, J., Ta abo elli, D., G o h, P. y Neylon, C. (2010), Al me ics: A mani es o, 26 Oc obe
2010. h p://al me ics.o g/mani es o
To es-Salinas, D. (2020). Ri mo de c ecimien o dia io de la p oducción cien í ica sob e Co id-
19. Análisis en bases de da os y eposi o ios en acceso abie o. P o esional De La in o mación,
29(2), e290215.
STI 2022
F om Global Indica o s o Local Applica ions
STI 2022 | h ps://doi.o g/10.5281/zenodo.6957471
6 / 6
To es-Salinas, D., A oyo-Machado, W., & Robinson Ga cía, N. (2020). Una mi ada
al e na i a a la p oducción cien í ica de la Sociología Española ¿Qué nos dicen las al me ics?
Tech. Rep (Digibug).
To es-Salinas, D., Robinson-Ga cía, N., Cas illo-Valdi ieso, P. Open Access and Al me ics
in he pandemic age: Fo escas analysis on COVID-19 ela ed li e a u e.
Yu, Y., Li, Y., Zhang, Z., Gu, Z., Zhong, H., Zha, Q., Yang, L., Zhu, C., & Chen, E. (2020). A
bibliome ic analysis using VOS iewe o publica ions on COVID-19. Annals o ansla ional
medicine, 8(13), 816.
Valde ama, P., Baca, P., Solana, C., & Fe e -Luque, C. M. (2021). Roo Canal Disin ec ion
A icles wi h he Highes Rela i e Ci a ion Ra ios. A Bibliome ic Analysis om 1990 o 2019.
An ibio ics, 10(11), 1412.