Sti 2014 sivertsen

41
Scholarly publication patterns in the social sciences and humanities and their coverage in Scopus and Web of Science Gunnar Sivertsen Nordic Institute for Studies in Innovation, Research and Education, Oslo, Norway 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Health Sciences Natural Sciences Engineering Social Sciences Humanities Scopus Web of Science Indexed journals Journals and series Books

Transcript of Sti 2014 sivertsen

Scholarly publication patterns in the social sciences and humanities and their coverage in Scopus and Web of Science

Gunnar Sivertsen

Nordic Institute for Studies in Innovation, Research and Education, Oslo, Norway

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Health

Sciences

Natural

Sciences

Engineering Social

Sciences

Humanities

Scopus Web of Science

Indexedjournals

Journals and series

Books

Overview

1. Introduction: Why?

2. Methods: WoS, BCI, and Scopus compared to National CRIS-data

3. Results: Large variations among subfields in the SSH

4. Discussion:

a. Major differences between Elsevier and Thomson Reuters?

b. What types of publications are not covered?

c. Same pattern across countries?

d. What will happen to book publishing and the use of native languages in the SSH?

e. Will it become possible to achieve comprehensive coverage of scholarly publishing in the SSH?

Overview

1. Introduction: Why?

2. Methods: WoS, BCI, and Scopus compared to National CRIS-data

3. Results: Large variations among subfields in the SSH

4. Discussion:

a. Major differences between Elsevier and Thomson-Reuters?

b. What types of publications are not covered?

c. Same pattern across countries?

d. What will happen to book publishing and the use of native languages in the SSH?

e. Will it become possible to achieve comprehensive coverage of scholarly publishing in the SH?

Why?

The purpose of the study is to discuss the options for an improved coverage of the SSH in international bibliographic data sources in the future.

How?

A dataset of more than 70,000 scholarly publications from the eight years 2005-2012, 44 per cent of which are in the SSH, will be used here to study the publication patterns in the SSH and their coverage in Scopus and Web of Science.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Health Sciences Natural Sciences Engineering Social Sciences Humanities

Scopus Web of Science

How?

Names of publication channels are standardized in the Cristin-data:

Journals or series (ISSN)

Publishers of books (ISBN-)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Health Sciences Natural Sciences Engineering Social Sciences Humanities

Scopus Web of Science

How?

Only peer-reviewed scholarly publications are included in the comparison:

Articles in journals or series

Articles in books

Books

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Health Sciences Natural Sciences Engineering Social Sciences Humanities

Scopus Web of Science

Cristin-data are matched by publication channel to the channels covered by Scopus, BCI, and WoS in 2013

Overview

1. Introduction: Why?

2. Methods: WoS, BCI, and Scopus compared to National CRIS-data

3. Results: Large variations among subfields in the SSH

4. Discussion:

a. Major differences between Elsevier and Thomson Reuters?

b. What types of publications are not covered?

c. Same pattern across countries?

d. What will happen to book publishing and the use of native languages in the SSH?

e. Will it become possible to achieve comprehensive coverage of scholarly publishing in the SH?

Coverage in WoS and Scopus of the HumanitiesScholarly journal publishingBased on complete data from the Norwegian HE sector 2005-2012

All publications Journal publications

Subfield Total % in journals % in int. language % in WoS % in Scopus

Classical Studies 259 66 % 48 % 23 % 26 %

Theatre and Drama 129 66 % 61 % 14 % 29 %

Linguistics 1057 61 % 93 % 39 % 41 %

Ethnology 392 57 % 47 % 12 % 16 %

Literature 764 57 % 31 % 16 % 18 %

Archaeology and Conservation 765 56 % 52 % 26 % 30 %

Slavonic Studies 231 56 % 84 % 17 % 43 %

Architecture and Design 424 54 % 38 % 11 % 24 %

Philosophy and History of Ideas 1121 54 % 45 % 28 % 33 %

Art History 278 54 % 45 % 21 % 25 %

Musicology 403 50 % 43 % 28 % 26 %

Theology and Religion 2126 50 % 42 % 16 % 34 %

History 1645 45 % 44 % 40 % 44 %

Media and Communication 1073 39 % 73 % 19 % 47 %

Asian and African Studies 237 39 % 99 % 42 % 49 %

Germanic Studies 238 38 % 100 % 39 % 37 %

Romance Studies 304 35 % 100 % 47 % 55 %

Scandinavian Studies 1777 35 % 17 % 2 % 2 %

English Studies 329 32 % 100 % 39 % 60 %

Total humanities 13551 49 % 52 % 23 % 32 %

Coverage in WoS and Scopus of the Social Sciences Scholarly journal publishingBased on complete data from the Norwegian HE sector 2005-2012

All publications Journal publications

Subfield Total % in journals % in int. language % in WoS % in Scopus

Library and Information Science 389 83 % 98 % 56 % 80 %

Psychology 1940 79 % 79 % 66 % 72 %

Geography 853 78 % 86 % 72 % 78 %

Economics 1081 75 % 83 % 73 % 77 %

Business & Administration 2904 63 % 76 % 39 % 57 %

Law 2108 61 % 31 % 6 % 13 %

Anthropology 597 53 % 65 % 32 % 82 %

Gender Studies 358 48 % 38 % 19 % 37 %

Sociology 1157 46 % 60 % 40 % 48 %

Political Science 1655 45 % 76 % 64 % 73 %

Education & Educational Research 4861 43 % 51 % 22 % 35 %

Total social sciences 17903 58 % 66 % 42 % 54 %

Coverage in WoS and Scopus of the Social Sciences Scholarly book publishingBased on complete data from the Norwegian HE sector 2005-2012

Book publications

Subfield Publications in books BCI Scopus

Library and Information Science 58 14 % 5 %

Psychology 158 33 % 8 %

Geography 161 14 % 4 %

Economics 2275 8 % 3 %

Business & Administration 720 20 % 9 %

Law 228 48 % 22 %

Anthropology 244 29 % 9 %

Gender Studies 511 23 % 5 %

Sociology 773 40 % 7 %

Political Science 890 35 % 12 %

Education & Educational Research 345 28 % 8 %

Total social sciences 6363 22 % 7 %

Coverage in WoS and Scopus of the HumanitiesScholarly book publishingBased on complete data from the Norwegian HE sector 2005-2012

Book publications

Subfield Publications in books BCI Scopus

Classical Studies 278 9 % 2 %

Theatre and Drama 163 23 % 6 %

Linguistics 129 33 % 19 %

Ethnology 175 45 % 9 %

Literature 441 23 % 8 %

Archaeology and Conservation 80 32 % 12 %

Slavonic Studies 799 13 % 3 %

Architecture and Design 146 5 % 1 %

Philosophy and History of Ideas 112 8 % 0 %

Art History 324 41 % 19 %

Musicology 266 7 % 1 %

Theology and Religion 538 16 % 3 %

History 163 21 % 2 %

Media and Communication 1015 4 % 3 %

Asian and African Studies 168 26 % 3 %

Germanic Studies 93 26 % 1 %

Romance Studies 44 0 % 0 %

Scandinavian Studies 915 16 % 5 %

English Studies 136 24 % 19 %

Total humanities 5977 17 % 5 %

Differences in coverageWith a data source as a «standard»: A source of tensions instead?

All publications Journal publications

Subfield Total % in journals % in int. language % in WoS % in Scopus

Library and Information Science 389 83 % 98 % 56 % 80 %

Psychology 1940 79 % 79 % 66 % 72 %

Geography 853 78 % 86 % 72 % 78 %

Economics 1081 75 % 83 % 73 % 77 %

Business & Administration 2904 63 % 76 % 39 % 57 %

Law 2108 61 % 31 % 6 % 13 %

Anthropology 597 53 % 65 % 32 % 82 %

Gender Studies 358 48 % 38 % 19 % 37 %

Sociology 1157 46 % 60 % 40 % 48 %

Political Science 1655 45 % 76 % 64 % 73 %

Education & Educational Research 4861 43 % 51 % 22 % 35 %

Total 17903 58 % 66 % 42 % 54 %

Overview

1. Introduction: Why?

2. Methods: WoS, BCI, and Scopus compared to National CRIS-data

3. Results: Large variations among subfields in the SSH

4. Discussion:

a. Major differences between Elsevier and Thomson Reuters?

b. What types of publications are not covered?

c. Same pattern across countries?

d. What will happen to book publishing and the use of native languages in the SSH?

e. Will it become possible to achieve comprehensive coverage of scholarly publishing in the SH?

Major differences?

Scopus: Wider coverage of international and regional journals in the SSH

(WoS more limited)

Book Citation Index: Wider coverage of international English language publishers in the SSH

(Scopus more limited)

Overview

1. Introduction: Why?

2. Methods: WoS, BCI, and Scopus compared to National CRIS-data

3. Results: Large variations among subfields in the SSH

4. Discussion:

a. Major differences between Elsevier and Thomson Reuters?

b. What types of publications are not covered?

c. Same pattern across countries?

d. What will happen to book publishing and the use of native languages in the SSH?

e. Will it become possible to achieve comprehensive coverage of scholarly publishing in the SH?

Indexedjournals

Journals and series

Books

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Health

Sciences

Natural

Sciences

Engineering Social Sciences Humanities

Coverage of 70,500 scholarly publications from thehigher education sector in Norway 2005-2012.

ScopusWeb of Science

Incomplete coverage of internationaljournals in the SSH.

Very limited coverage of books.

Random or no coverage of the nationallevel (books and journals)

Österreichische Zeitschrift für Politikwissenschaftis covered …

But Österreichische Zeitschrift für Soziologie is not covered

Swedish Journal of Sociologyis covered …

But Swedish Journal of Political Science is not covered

Overview

1. Introduction: Why?

2. Methods: WoS, BCI, and Scopus compared to National CRIS-data

3. Results: Large variations among subfields in the SSH

4. Discussion:

a. Major differences between Elsevier and Thomson Reuters?

b. What types of publications are not covered?

c. Same pattern across countries?

d. What will happen to book publishing and the use of native languages in the SSH?

e. Will it become possible to achieve comprehensive coverage of scholarly publishing in the SH?

CRIS systems for PBF

Both Elsevier and Thomson Reuters are nowintegrating thecitation databases with researchmanagement toolsand CRIS-systems

ERIH will become dynamic and include the social sciences

Overview

1. Introduction: Why?

2. Methods: WoS, BCI, and Scopus compared to National CRIS-data

3. Results: Large variations among subfields in the SSH

4. Discussion:

a. Major differences between Elsevier and Thomson Reuters?

b. What types of publications are not covered?

c. Same pattern across countries?

d. What will happen to book publishing and the use of native languages in the SSH?

e. Will it become possible to achieve comprehensive coverage of scholarly publishing in the SH?

Publication typesBased on 59,100 scholarly publications from the higher education sector in Norway 2005-2011.

0 % 20 % 40 % 60 % 80 % 100 %

Eng Sci

Nat Sci

Health Sci

Social Sci

Humanities

Articles in journals/series Articles in books Books

Publication types and research methodologies

Articles in journals/series:

Individual, often empirical studies based on a limiteddataset or with a particular focus

Author name(s) on title page of each article

Edited volumes (articles in books):

A collaborative effort represented in several individualstudies that are coordinated and edited from the outset

Author name(s) on title page of each article; names of theeditors on the title page of the book

Books:

A coherent presentation of in-depth studies with a widerfocus

Author name(s) on the title of the book

Articles in journals/series as share of all publicationsBased on 59,100 scholarly publications from the higher education sector in Norway 2005-2011.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2005 2006 2007 2008 2009 2010 2011

Natural Sci

Health Sci

Engineering

Soc Sci

Humanities

0% 20% 40% 60% 80% 100%

Economics

Sociology

Linguistics

History

Articles in journals/series as share of all publicationsBased on 4,600 scholarly publications from the higher education sector in Norway 2005-2011.

Will English become the single scholarly language in the SSH?

No, probably not, because:

The so-called societal impact or relevance of the SSH depends on the scholars’ communication with their ownsociety and culture

Some findings may be less relevant to publish on theinternational level, but can be of high interest if publishedon the national level

The now widespread use of English among SSH scholars in some non-English speaking countries is so far accompaniedby the use of native language in some of their publications

Percentage publications in the national languageBased on 59,100 scholarly publications from the higher education sector in Norway 2005-2011.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2005 2006 2007 2008 2009 2010 2011

Humanities

Social Sci

Health Sci

Nat Sci

Engineering

Percentage publications in the national languageBased on 4,600 scholarly publications from the higher education sector in Norway 2005-2011.

0% 20% 40% 60% 80% 100%

Economics

Sociology

Linguistics

History

Overview

1. Introduction: Why?

2. Methods: WoS, BCI, and Scopus compared to National CRIS-data

3. Results: Large variations among subfields in the SSH

4. Discussion:

a. Major differences between Elsevier and Thomson Reuters?

b. What types of publications are not covered?

c. Same pattern across countries?

d. What will happen to book publishing and the use of native languages in the SSH?

e. Will it become possible to achieve comprehensive coverage of scholarly publishing in the SH?

Scholarly journal articles on the national level will be concentrated in a few journals. Example: SociologyBased on 435 scholarly journal articles from the higher education sector in Norway 2005-2011.

0 20 40 60 80

53 other journals

Dansk Sociologi

Work, Employment and Society

European Societies: The Official Journal of the European…

Pakistan Journal of Criminology

Sociologia Ruralis

Comparative Sociology

Qualitative Inquiry

Journal of Classical Sociology

International Journal of Sociology

International Journal of Social Research Methodology

Procedia - Social and Behavioral Sciences

European Journal of Social Theory

Asian Journal of Social Science

Qualitative Sociology Review

Leisure Studies

Sociologisk forskning

Sociology

Ethnic and Racial Studies

Journal of Youth Studies

Journal of Critical Realism

Sociological Review

Social Indicators Research

European Sociological Review

Acta Sociologica

Journal of Risk Research

Sosiologisk Årbok

Sosiologi i dag

Sosiologisk Tidsskrift

Typically, Scopus will have a wider coverage of these journalsExample: SociologyBased on 435 scholarly journal articles from the higher education sector in Norway 2005-2011.

0 20 40 60 80

53 other journals

Dansk Sociologi

Work, Employment and Society

European Societies: The Official Journal of the European…

Pakistan Journal of Criminology

Sociologia Ruralis

Comparative Sociology

Qualitative Inquiry

Journal of Classical Sociology

International Journal of Sociology

International Journal of Social Research Methodology

Procedia - Social and Behavioral Sciences

European Journal of Social Theory

Asian Journal of Social Science

Qualitative Sociology Review

Leisure Studies

Sociologisk forskning

Sociology

Ethnic and Racial Studies

Journal of Youth Studies

Journal of Critical Realism

Sociological Review

Social Indicators Research

European Sociological Review

Acta Sociologica

Journal of Risk Research

Sosiologisk Årbok

Sosiologi i dag

Sosiologisk Tidsskrift

Scholarly book publishing on the national level is evenmore concentrated in a few channels in each country

What is beyond coverage? Is comprehensive coveragefeasible?

Beyond coverage:

• Incomplete coverage of internationaljournals in the SSH.

• Very limited coverage of books.

• Random or no coverage of the nationallevel (books and journals)

• Feasible?

• Yes, if the coverage is restricted to scholarly publication channels (journals, series, book publishers) that publish with• Peer review• Authors from several institutions• Abstracts in English and standard

information about authors and affiliations

• Standards of referencing