Table Of ContentJournal Status
Johan Bollen⋆†, Marko A. Rodriguez†, and Herbert Van de Sompel†
February 1, 2008
6
0
0 † DigitalLibraryResearch&Prototyping Team,
2
ResearchLibrary,LosAlamosNationalLaboratory, LosAlamos,NM,87545
n
⋆Corresponding author. Tel.: +15056060030. URL:http://public.lanl.gov/jbollen,
a
J email: {jbollen, marko,herbertv}@lanl.gov
9
Abstract
]
L
D Thestatus ofanactor inasocial context iscommonly defined intermsoftwofactors: thetotal numberof
. endorsements the actor receives from other actors and the prestige of the endorsing actors. These two factors
s
c indicate the distinction between popularity and expert appreciation of the actor, respectively. We refer to the
[ former as popularity and to the latter as prestige. These notions of popularity and prestige also apply to the
1 domain of scholarly assessment. The ISI Impact Factor (ISI IF) is defined as the mean number of citations
v a journal receives over a 2 year period. By merely counting the amount of citations and disregarding the
0
prestige of the citing journals, the ISI IF is a metric of popularity, not of prestige. We demonstrate how a
3
0 weighted version ofthe popular PageRank algorithm can be used to obtain a metric that reflects prestige. We
1
contrast the rankings of journals according to their ISI IF and their weighted PageRank, and we provide an
0
6 analysisthatrevealsbothsignificantoverlapsanddifferences. Furthermore,weintroducetheY-factorwhichis
0 asimplecombinationofboththeISIIFandtheweightedPageRank,andfindthattheresultingjournalrankings
/
s correspond welltoageneralunderstanding ofjournalstatus.
c
:
v
i
X
r
a
JournalStatus-JohanBollen,MarkoA.RodriguezandHerbertVandeSompel 2
1 Introduction
Some people are popular but not prestigious and vice versa. For example, an author of pulp detectives may
sell many books, but may not have earned the respect of literary critics. Conversely, a Nobel Prize in Lit-
erature winner may be highly valued among literary experts, yet never make the New York Times bestseller
list. Inessence, these examples revealtheexistence oftwofactors thatcontribute tothestatus ofanactor ina
social context: the total number of endorsements the actor receives from other actors, and the prestige of the
endorsingactors. Intheremainderofthispaper,werefertotheformeraspopularityandtothelatterasprestige.
Similarconsiderations applytotheassessmentofscholarlycommunicationwherecitationcountsarecom-
monly used as an indication of scholarly status. For example, a journal that publishes mostly review articles
maybefrequentlycitedbygraduatestudents,yetlargelybeignoredbyexpertsinterestedinthecuttingedgeof
research. The Thomson ISI Impact Factor (ISI IF) is generally accepted as an indicator of journal status, and
isdefinedasthemeannumberofcitations toarticlespublished inajournalovera2yearperiod[10,9]. Given
that the ISI IF is based on the amount of citations to a journal, and does not take into account the prestige of
thecitingjournals, itseemstoonlyrepresent thepopularity factorofstatus, notitsprestige factor.
Many concerns have been expressed over the usefulness of the ISI IF as an indicator of journal status
[25, 11, 18, 16, 24, 1]. In fact, its focus on popularity would render it impossible to use in many other areas,
such as for example the WWW. A web page that is often linked to can indeed be of very low status and vice
versa. Forthat reason, alternatives tolink counting, developed inthe domain of social network analysis, have
beenwidelyadopted forWWWsearching.
When the Google search engine ranks webpages according totheir status itdoes so by not merely count-
ing the number of hyperlinks to apage. Google’s PageRank algorithm [7]computes thestatus of aWebpage
based on acombination ofthe number of hyperlinks that point tothe page and the status of thepages that the
hyperlinks originate from. Bytaking intoaccount boththepopularity andtheprestige factor ofstatus, Google
hasbeenabletoavoidassigning highrankstopopular butotherwiseirrelevant Webpages.
Thesuccess ofthe PageRankalgorithm in theGoogle environment has led toPageRank becoming astan-
dard technique to assess the status of web resources. However, where the evaluation of journal status is con-
cerned the ISIIF still rules supreme. Thissituation may not be sustainable. Asan ever growing collection of
scholarlymaterialsbecomesavailableontheWeb,andhencebecomessearchable throughGoogleandGoogle
Scholar, ourperception ofarticle status (and hence ofjournal status) willchange asaresult ofthe PageRank-
driven mannerbywhich Google listsits search results. Inthefuture, PageRank, notthe ISIIF,mayvery well
startrepresenting ourperception ofarticleandjournalstatus.
A change from the ISI IF to PageRank-based metrics for journal ranking would effectively signify a shift
fromanevaluationbasedonpopularity, i.e.citationfrequency, toanevaluationbasedonprestige, i.e.thepres-
tigeofthosewhociteistakenintoaccount. Toevaluate theconsequences ofsuchachange ontheassessment
ofjournalstatus, weusedthedataset ofthe2003ISIJournal CitationReports(ISIJCR)tocomparetheISIIF
andWeighted PageRankrankings ofjournals. Wepaidspecial attention tojournals withasignificant discrep-
ancybetweentheirISIIFandWeightedPageRankvalues. Wealsointroduce arankingprinciple, theY-factor,
torankjournals according towhethertheyhavebothhighISIIFandWeightedPageRankvalues.
JournalStatus-JohanBollen,MarkoA.RodriguezandHerbertVandeSompel 3
2 Two common metrics of status
Citations are at the basis of most present attempts to assess scholarly impact. This is true for the assessment
of the impact of individual articles, journals, researchers [8, 3, 12], research departments, universities and
even countries [6, 5, 23, 13]. As articles cite one another, they define an article citation network in which
each node represents an article and each directed edge represents a citation by that article to another. By
grouping allarticles published inthesamejournal under asingle journal node, anarticle citation network can
easilybetransformed intoaJournal CitationNetwork. Inthatnetwork, thedirected edgesbetweenthejournal
nodes represent the collection of citations from one journal to another. This network can be formalized as
a set of journals V, a set of directed edges E ⊆ V2 that exist between the journals in V, and the function
W(v ,v )→ N+ whichmapseachedgebetweenthejournalv andv toapositive,integercitationfrequency.
i j i j
Arangeofjournalstatusmetricscanbeapplied tosuchaJournal CitationNetwork. Inthefollowingsections,
wediscuss twohighly commonmetrics, namelytheISIImpactFactorandGoogle’s PageRank. Thelatterhas
beenmodified totakeinto account edge weights (Weighted PageRank) sothat itcan beapplied totheJournal
CitationNetwork.
2.1 The ISI Impact Factor
TheISIIFdefinesthestatusofajournalforaspecificyearasthemeannumberofcitationsthatoccurredinthat
yeartothearticles published inthejournal duringthetwoprevious years. Moreconcretely, the2003ISIIFof
ajournalv iscalculatedbydividingthenumberofcitationsmadein2003tov ’s2001and2002articlesbythe
i i
total number ofarticles v published in 2002 and 2001. Expressed in terms ofa Journal Citation Network the
i
ISIIFcorresponds toajournal’s in-degree [2]normalized bythetotal numberofpapers thejournal published
inthatperiod. Eq. 1definestheIFofjournalv inyeart,labeledIF(v ,t),asfollows:
i i
c(v ,v ,t)
Pj j i
IF(v ,t) = (1)
i
n(v )
i
where c(v ,v ,t)corresponds tothe number ofcitations from journal v to journal v inyear t. Thenum-
j i j i
ber of publications published in journal v , denoted n(v ), during the two years previous to t, normalizes the
i i
resulting citationcount,leading toamean,2-yearcitationrateperarticle.
Insocialnetworkanalysisterms,in-degreecanbeconsideredametricofpopularitybecauseitcorresponds
tothenumberofendorsements receivedbyaparticularactorinthenetwork. And,indeed,whenassumingthat
acitationtoajournalindicatesanendorsementofthejournal’scontent,wefindthat,intermsofsocialnetwork
analysis,theISIIFisameasureofpopularitybecauseajournalhasahigherISIIFifitsarticlesaremoreoften
cited.
2.2 Journal PageRank
This aspect of the ISI IF has been known and studied for decades. In particular, Pinski et al. (1977) [22]
propose an algorithm that evaluates the influence ofjournals by taking into account not simply the number of
citationsfromonejournaltotheother,butalsotheprestigeofthecitingjournal. Journalsthatreceivemanyci-
tations from prestigious journals areconsidered highly prestigious themselves. Byiteratively passing prestige
from one journal tothe other, astable solution is reached which reflects the relative prestige ofjournals. This
procedure is highly related to efforts in social network science to define status in terms of "inherited" status,
e.g. eigenvectorcentrality[4],systemstoseparatewebpagesinto"authoraties"and"hubs"[14,15]andrecent
JournalStatus-JohanBollen,MarkoA.RodriguezandHerbertVandeSompel 4
investigations oftheroleofjournalsasknowledgesources orstorers[19].
Within this long lineage of social network metrics of status, the founders of the Google search engine
outline an algorithm to assess the prestige of web pages based on similar principles in their 1998 paper "The
anatomyofalarge-scalesearchengine"[7]. Thisconceptisfurtherdevelopedinlaterpublications[20]interms
of random walk models of web navigation. Much like the proposal by Pinski et al., PageRank is calculated
by an iterative algorithm which propagates prestige values from one web page to another and converges to a
solution [21]. The PageRank equation that governs the iterative transfer of PageRank values from one web
pagetotheotherisshowninEq. 2:
(1−λ) 1
PR(vi)= +λXPR(vj)× (2)
N O(v )
j
j
InEq. 2,itisassumedthatacollectionofpagesv linktoarecipientpagev andeachtransfersaproportion
j i
of their PageRank, denoted PR(v ), to v . It is also assumed that PageRank values are equally distributed
j i
along a page’s out-links, i.e. if a page v has 3 out-links each recipient page v receives only one-third of
j i
v ’s PageRank. Transfered PageRank values are therefore normalized by the number of out-links from page
j
v which is denoted O(v ). The parameter λ, which can take values between zero and one, represents the
j j
attenuation of prestige values as they are transferred from one web page to the other. The parameter (1−λ)
N
represents theminimalamountofprestige assigned toeachwebpage. N represents thetotalnumberofpages
inthenetwork.
2.3 Weighted PageRank for Journal Citation Networks
PageRank has become astandard toevaluate the status ofweb pages. It isour objective toapply it toJournal
Citation Networkssothat wecancompare twohighly commonmetricofstatus, i.e. theISIIFandPageRank,
in terms of their ability to evaluate the relative popularity or prestige of journals. The PageRank definition
above assumes that prestige is distributed equally across all of a web page’s hyperlinks. This is appropriate
sincehyperlinks arenotweighted, i.e. eachhyperlink indicates anequal degreeofrelationship between apair
of linked pages. In the Journal Citation Network, however, not all edges are created equal; some journals are
connected by more citations than others. The PageRank equation when applied to journal citation networks
shouldthereforebeadaptedtotakeintoaccountjournalcitationfrequenciesinitstransferofPageRankvalues.
Indeed, ifajournal v cites journal v 10timesmorefrequently thananyotherjournal, theamount ofprestige
j i
transferred from v to v should be ten times as high. More generally, a journal that receives many citations
j i
fromaspecificotherjournalshould receiveamatchingproportion ofthatjournal’s prestige.
This is in fact a common problem encountered in applications of PageRank to weighted networks. Mod-
ifications of the PageRank equation have therefore been proposed to take into account link weights. A Web-
basedWeightedPageRankalgorithm hasearlierbeendefined[29]tocalculate aggregatewebsiteprestige and
a weighted PageRank algorithm has been to used to rank authors in a weighted co-authorship network [17].
Thenotionofweightedlinkweightsisinfactanintegral partofPinskiandNarin’sapproach todefinejournal
prestige. Wewillbriefly discuss these common modifications ofthe PageRank equation in termsof weighted
journalcitationnetworksbelow.
AssumeweneedtorewriteEq. 2toaccountforthetransmission ofjournalprestige relativetothenumber
ofcitationsthatexistbetweenpairsofjournalsintheJournalCitationNetwork. First,wedefineapropagation
JournalStatus-JohanBollen,MarkoA.RodriguezandHerbertVandeSompel 5
proportion w(v ,v ) between journals v and v by normalizing the link weights emanating from a particular
j i i j
journalv asfollows:
j
W(v ,v )
i j
w(v ,v ) = (3)
j i
W(v ,v )
Pk j k
.
Foranyparticular journalv ,allw(v ,v )nowsumuptooneanditcanthereforebeusedtodeterminethe
j i j
fractionofajournal’s PageRankittransfers tothejournalsitcites.
WenowobtaintheWeightedPageRankequation forjournalv asfollows:
j
(1−λ)
PRw(vi) = +λXPRw(vj)×w(vj,vi) (4)
N
j
According to Eq. 4, the transfer ofprestige from onejournal to theother ismodulated by thepropagation
proportion w(v ,v ). In effect, the equal distribution of PageRank values in Eq. 2, as given by the factor
j i
1 , has been replaced by the propagation proportion w(v ,v ) thereby allowing Weighted PageRank to be
O(vi) j i
calculated forJournalCitationNetworks.
2.4 Product of ISI IF and Weighted PageRank
Wenowhavetwodifferent, buthighlycommon, metricsofstatusatourdisposal. TheISIIFreliesoncitation
frequencies andtherefore stresses thepopularity aspect ofjournal status. TheWeighted PageRank, asdefined
above,reliesonapropagationofprestigevaluesfromonejournaltotheother,andthereforecorrespondsbetter
to our intuitive notion that prestige is not only a matter of the number of endorsements, but who is actually
endorsing.
Thus defined, the ISI IF and the Weighted PR represent highly common, but possibly different, facets of
journalstatus. Aswillbedemonstrated insection3.2,therecanindeedexistsignificantdiscrepancies between
a journal’s ISI IF and Weighted PageRank values, i.e. some journals can have high ISI IF and low Weighted
PageRankvalues,andviceversa. Torankjournalsonthebasisofbothmetricscombinedwedefinedaproduct
oftheISIIFandtheWeightedPageRank,labeledY-factor, asshowninEq. 5below.
Y(v ) = ISIIF(v )×PR (v ) (5)
j j w j
Journals that score highly on the Y-factor will be ranked highly by either or both the ISIIF and Weighted
PageRank. Theresulting rankings areincluded inthefollowingsection forinformational purposes.
3 Indicators of Journal Status
Inthissection, wecompare three indicators ofstatus intheJournal Citation Network: thepopularity-oriented
ISI IF, the prestige-oriented Weighted PageRank and a product of both, namely the Y-factor. We do so based
onthedatasetprovided bythe2003ISIJournalCitationReports.
JournalStatus-JohanBollen,MarkoA.RodriguezandHerbertVandeSompel 6
3.1 Comparing the ISI IF and the Weighted PageRank
In order to obtain Weighted PageRank values for journals that have an ISI Impact Factor, a Journal Citation
Network was constructed on the basis of the 2003 ISI JCR data set which contains 2003 journal citations to
2001and2002publications. Thisjournalcitation information wasrepresented asamatrixinwhichbothrows
and columns represent journals, and in which cells represent the amount of times a journal in a row cites a
journal in a column. Not surprisingly, a sparse matrix resulted, with 5710 journals having non-zero citation
counts.
To provide an indication of the overall characteristics of the ISI IF and the Weighted PageRank, Table 1
showsthetenhighestranking journals forbothstatusmetrics. Clearly, therankings divergesignificantly, with
only three journals, Nature, Science and The New England Journal of Medicine being represented in both
lists. Weobserve thatthejournals withthehighest ISIIFarestrongly positioned intheareaofmedicine, with
reviewjournalsbeingheavilyrepresented. ThelatterconfirmsthecharacterizationoftheISIIFasapopularity-
orientedmetric,sincereviewjournalstypicallypublishbackgroundmaterialthatislikelytobecitedfrequently.
Overall, thelisting according to Weighted PageRank shows morevariations inscholarly discipline, andmany
ofthetop-ranked journals suchasScience,Nature,CellandtheJournalofBiologicalChemistryaregenerally
considered highlyprestigious journals.
A more quantitative analysis of the overlap and discrepancies between the two status metrics is provided
by the scatter plot of Fig. 1. Despite the strong discrepancies in the top ten listings according to both sta-
tus metrics, the plot reveals asignificant overall correlation, confirmed by aPearson correlation coefficient of
r = 0.48,p < 0.01 between the ISI IF and the Weighted PageRank. We note that the journals Nature and
Science are positioned in the top-right corner of the scatter plot, reflecting the fact that they have both high
ISI IF and high Weighted PageRank values. This means that both journals are often cited and are cited by
prestigious journals.
Since it is common knowledge in bibliometrics that comparisons of ISI IF values across scholarly disci-
plines are problematic due to differences in the publication and citation process, we decided to focus on a
subset of our Journal Citation Network that pertains to Physics journals. We selected Physics journals in the
2003 ISIJCR dataset on the basis of the ISI subject categories listed in Table 2. The resulting Physics subset
oftheJournalCitationNetworkcontained229journals. ArankingofthissubsetaccordingtoISIIF,Weighted
PageRankandtheY-factorisshowninTable5.
We can detect a pattern similar to that found for the complete Journal Citation Network. Again, only 2
journals, namely Physical Review Letters and Journal of High Energy Physics, are amongst the highest rank-
ing according to both the ISI IF and the Weighted PageRank. In addition, the ISI IF rankings can again be
characterized byapreponderance ofappliedphysics journals thatfrequently publish background materialthat
islikelytobecited. TheWeightedPageRankrankingseemstofocusonasetofjournals typically appreciated
bydomainexperts, asthejournals oftheAmericanPhysicalSociety: PhysicalReviewA,DandE.ThePear-
soncorrelation coefficient between ISIIFandWeighted PageRankvalues wasfound tobelowerthan wasthe
casewiththecompleteJournalCitationNetwork,namelyr = 0.24,p < 0.01. Thislowercorrelation indicates
a lesser degree of intra-discipline overlap between both metrics. We leave the interpretation of the Y-factor
rankingstothereader.
Proceeding along the same lines, we also compared the ISI IF and the Weighted PageRank for journals
JournalStatus-JohanBollen,MarkoA.RodriguezandHerbertVandeSompel 7
ISIIF PRw Y-factor
rank value Journal value(x103) Journal value(x102) Journal
1 52.28 ANNUREVIMMUNOL 16.78 NATURE 51.97 NATURE
2 37.65 ANNUREVBIOCHEM 16.39 JBIOLCHEM 48.78 SCIENCE
3 36.83 PHYSIOLREV 16.38 SCIENCE 19.84 NEWENGLJMED
4 35.04 NATREVMOLCELLBIO 14.49 PNAS 15.34 CELL
5 34.83 NEWENGLJMED 8.41 PHYSREVLETT 14.88 PNAS
6 30.98 NATURE 5.76 CELL 10.62 JBIOLCHEM
7 30.55 NATMED 5.70 NEWENGLJMED 8.49 JAMA
8 29.78 SCIENCE 4.67 JAMCHEMSOC 7.78 LANCET
9 28.18 NATIMMUNOL 4.46 JIMMUNOL 7.56 NATGENET
10 28.17 REVMODPHYS 4.28 APPLPHYSLETT 6.53 NATMED
Table1: Thehighestrankingjournalsaccording toISI IF, WeightedPageRank and Y-factor
ISICategory ISICategoryName
UB PHYSICS,APPLIED
UF PHYSICS,FLUIDS&PLASMAS
UH PHYSICS,ATOMIC,MOLECULAR&CHEMICAL
UI PHYSICS,MULTIDISCIPLINARY
UK PHYSICS,CONDENSEDMATTER
UN PHYSICS,NUCLEAR
UP PHYSICS,PARTICLES&FIELDS
UR PHYSICS,MATHEMATICAL
Table2: ISISubject Categoriesfor PhysicsJournals.
ISICategory ISICategoryName
EP COMPUTERSCIENCE,ARTIFICIALINTELLIGENCE
ER COMPUTERSCIENCE,CYBERNETICS
ES COMPUTERSCIENCE,HARDWARE&ARCHITECTURE
ET COMPUTERSCIENCE,INFORMATIONSYSTEMS
EV COMPUTERSCIENCE,INTERDISCIPLINARYAPPLICATIONS
EW COMPUTERSCIENCE,SOFTWARE,GRAPHICS,PROGRAMMING
EX COMPUTERSCIENCE,THEORY&METHODS
Table3: ISI SubjectCategories forComputerScience Journals.
JournalStatus-JohanBollen,MarkoA.RodriguezandHerbertVandeSompel 8
1
+0 ANNU REV IMMUNOL
5e NAT REV MOL CELL BIO NEW ENGL J MED NATURE
NAT GENET
REV MOD PHYS JAMA
ANNU RAENV NPUH ARREMV APCHOYLSIOL ANNU REV CELL DEV BI NAT CELL BIOL MOL CELL LANCCEELTL SCIENCE
TRENDS PHARMACOL SCI ACCOUNTS CHEM RES CIRCULATION P NATL ACAD SCI USA
ANNU REV ASTRON ASTR
Q REV BIOPHYS PHYS REP CURR BIOL
CURR OPIN STRUC BIOL EMBO J
0 PROG RETIN EYE RES PSYCHOL BULL AM J PSYCHIAT PHYS REV LETT
e+0 REV PHYSIOL BIOCH P J MOL BIOL J AM CHAEPMP SL OPCHYS LETTJ BIOL CHEM
5 CRIT REV SOLID STATE FEBS LETT BIOCHEMIJS TGREYO−PUHSYS RES
J CHEM PHYS
J PHYS CHEM AJ APPL PHYS
ELECTRON LETT
T AM MATH SOC
1
SI IF 5e−0 J MATER PROCESS TEJC AHLGEBRA
I P AM MATH SOC
J STAT PLAN INFER DISCRETE MATH
TOPOL APPL
Z ANGEW MATH MECH
INDIAN J PURE AP MAT
TRANSPORT RES REC
IF = 2043.39 PRw + 1.46
OIL GAS J
02 T I MIN METALL A r= 0.48
−
e
5 INDIAN VET J
J PETROL TECHNOL
ANN ARID ZONE
03 NAV ARCHIT
−
e
5
2e−05 5e−05 1e−04 2e−04 5e−04 1e−03 2e−03 5e−03 1e−02 2e−02
PRw
Figure1: ScatterplotoftheISI IFversustheWeightedPageRank (PRw).
ISICategory ISICategoryName
DS CRITICALCAREMEDICINE
FF EMERGENCYMEDICINE
FY DENTISTRY,ORALSURGERY&MEDICINE
OI INTEGRATIVE&COMPLEMENTARYMEDICINE
OP MEDICINE,LEGAL
PY MEDICINE,GENERAL&INTERNAL
QA MEDICINE,RESEARCH&EXPERIMENTAL
VY RADIOLOGY,NUCLEARMEDICINE&MEDICALIMAGING
YU TROPICALMEDICINE
Table4: ISI Subject CategoriesforMedicineJournals.
JournalStatus-JohanBollen,MarkoA.RodriguezandHerbertVandeSompel 9
inComputer Science and Medicine. Again, subsets of the Journal Citation Network wereextracted by means
of ISI category codes; Table 3 and Table 4 list these codes for Computer Science and Medicine, respectively.
The results for Computer Science reveal an even greater discrepancy between the ISI IF and the Weighted
PageRank. Indeed, ascanbeseeninTable6,onlythejournalBioinformatics ranksinthetoptenaccording to
both the ISIIF and the Weighted PageRank. Again, it seems that many of the top-ranking journals according
totheWeighted PageRankspecialize inafocused research area. Thescatterplot inFig. 3further confirmsthe
greaterdivergence betweentheISIIFandtheWeightedPageRankvalues. ThePearsoncorrelation coefficient
was found to be r = 0.5,p < 0.01. The Medicine subset of the Journal Citation Network follows a different
pattern than that of the Physics and Computer Science subsets. As can be seen in Table 7, 9 journals appear
in the top ten according to both the ISI IF and the Weighted PageRank. And, the scatterplot in Fig. 4 further
confirms the higher degree of overlap between the two metrics in Medicine. Indeed, the Pearson correlation
coefficient between the ISI IF and the Weighted PageRank values was found to be r = 0.91,p < 0.01,
indicating thatthenotions ofprestige andpopularity aremorestrongly intertwined forMedicine thantheyare
for the other explored domains. Overall, it seems that the level of discrepancy between the ISI IF and the
WeightedPageRankacrossdisciplinesrelatestovariationsinthecharacteristics ofthepublication andcitation
practicesindifferentdomains.
3.2 Popular and Prestigious Journals
Intrigued bythesignificant correlation betweentheISIIFandtheWeightedPageRankasshowninFig. 1and
thesignificantdiscrepanciesrevealedinTable1andTable5,wesetouttoinspecttheJournalCitationNetwork
forjournalsthathavestronglydivergingISIIFandWeightedPageRankvalues. Twotypesofdivergenceswere
explored:
PopularJournals arejournals that arecited frequently byjournals withlittle prestige. Thesejournals havea
veryhighISIIFandaverylowWeightedPageRank.
PrestigiousJournals arejournalsthatarenotfrequentlycited,buttheircitationscomefromhighlyprestigious
journals. Thesejournals haveaverylowISIIFandaveryhighWeightedPageRank.
We identified Popular and Prestigious Journals in the full Journal Citation Network, but were unable to
recognize ameaningfulpatternintheresults. Thiswasnotunexpected astheexerciseamountedtocomparing
ISIIFvaluesacrossdisciplines. Hence,wedecidedtorefocusourattentiononthePhysicssubsetoftheJournal
Citation Network. Weempirically decided onthreshold values fortheISIIFandtheWeighted PageRank that
guaranteed asufficientnumberofjournals inboththePopularandPrestigious category. First,wedecided that
anyWeightedPageRankvaluebelowthe40thpercentilewasverylow,andanyvalueabovethe90thpercentile
was very high. These choices are represented by the vertical lines in the scatter plot of Fig. 2. Second, to
determine the low and high threshold values for the ISI IF, we generated a linear regression model for the
relationship between the ISIIF and the Weighted PageRank. Theresult is visualized inFig. 2 asthe line that
cutsacrossthecloudofphysicsjournals. Figure2outlinestheregionsofthescatterplotthatcorrespondtoour
categorization of Popular and Prestigious Journals and to our chosen threshold values. The former category
is shown as the top-left region, the latter as the bottom-right region. Popular, prestigious and high Y-factor
ranking journals are labeled by their abbreviated journal titles in the graph. Table 8 shows the ten top-ranked
journals in both the Popular and Prestigious Journals category ranked by the degree to which their actual ISI
IFdeviates fromthevaluepredicted bythelinearregression model,labeledIF .
∆
AcloseexaminationoftheresultingPopularJournalcategoryrevealsthatitcontainseitherreviewjournals
or journals that frequently publish data tables. Such journals are likely to be cited as background material,
JournalStatus-JohanBollen,MarkoA.RodriguezandHerbertVandeSompel 10
rank IF Title PRw×103 Title Y ×102 Title
1 28.17 REVMODPHYS 8.41 PHYSREVLETT 5.91 PHYSREVLETT
2 13.09 ADVPHYS 4.28 APPLPHYSLETT 1.73 APPLPHYSLETT
3 11.98 PHYSREP 2.59 JAPPLPHYS 1.50 REVMODPHYS
4 10.03 MATSCIENGR 2.38 PHYSREVD 1.09 PHYSREVD
5 8.67 ANNUREVNUCLPARTS 2.34 PHYSREVE 0.69 JCHEMPHYS
6 8.41 REPPROGPHYS 2.32 JCHEMPHYS 0.66 JHIGHENERGYPHYS
7 7.04 PHYSREVLETT 1.56 PHYSLETTB 0.63 PHYSLETTB
8 7.00 SOLIDSTATEPHYS 1.55 PHYSREVA 0.57 NUCLPHYSB
9 6.06 JHIGHENERGYPHYS 1.22 CHEMPHYSLETT 0.56 JAPPLPHYS
10 5.97 PROGNUCLMAGRESSP 1.09 JHIGHENERGYPHYS 0.56 PHYSREP
Table5: Thehighestranking Physicsjournalsaccordingto ISIIF, WeightedPageRank (PRw) and Y-factor.
rank IF Title PRw×104 Title Y ×104 Title
1 7.50 ACMCOMPUTSURV 10.08 IEEETINFORMTHEORY 62.27 BIOINFORMATICS
2 6.70 BIOINFORMATICS 9.29 BIOINFORMATICS 22.64 IEEETINFORMTHEORY
3 4.54 VLDBJ 5.90 COMPUTMETHODAPPLM 21.68 IEEETPATTERNANAL
4 3.87 IEEENETWORK 5.67 IEEETPATTERNANAL 12.11 ACMCOMPUTSURV
5 3.82 IEEETPATTERNANAL 5.48 JCOMPUTPHYS 11.78 IEEETIMAGEPROCESS
6 3.76 IEEETMEDIMAGING 4.98 COMMUNACM 11.47 IEEETMEDIMAGING
7 3.73 IEEEINTELLSYSTAPP 4.95 THEORCOMPUTSCI 10.37 JACM
8 3.61 IBMJRESDEV 4.46 IEEETIMAGEPROCESS 9.65 JCOMPUTPHYS
9 3.33 INFORMSYST 4.35 COMPUTER 8.59 IEEEINTELLSYSTAPP
10 3.32 JACM 3.36 IEEETNEURALNETWOR 8.20 ARTIFINTEL
Table 6: The highest ranking Computer Science journals according to ISI IF, Weighted PageRank (PRw) and
Y-factor.
rank IF Title PRw×103 Title Y ×102 Title
1 34.83 NEWENGLJMED 5.70 NEWENGLJMED 19.84 NEWENGLJMED
2 30.55 NATMED 4.25 LANCET 8.49 JAMA
3 21.46 JAMA 3.96 JAMA 7.78 LANCET
4 18.32 LANCET 2.27 JCLININVEST 6.53 NATMED
5 15.30 JEXPMED 2.23 JEXPMED 3.42 JEXPMED
6 14.31 JCLININVEST 2.14 NATMED 3.25 JCLININVEST
7 12.42 ANNINTERNMED 1.40 AMJRESPCRITCARE 1.44 ANNINTERNMED
8 11.38 ANNUREVMED 1.16 ANNINTERNMED 1.24 AMJRESPCRITCARE
9 8.88 AMJRESPCRITCARE 0.91 NEUROIMAGE 0.59 ARCHINTERNMED
10 6.76 ARCHINTERNMED 0.87 ARCHINTERNMED 0.57 NEUROIMAGE
Table7: Thehighestranking Medicinejournalsaccordingto ISIIF, WeightedPageRank (PRw) andY-factor.
