Occurrence of author keywords and keywords plus in social sciences and humanities research : A preliminary study

Abstract

Keywords are important words and phrases of a research paper, expressing its essence. The longitudinal study of occurrence of keywords in specific discipline expresses trends of research. The present paper attempts to highlight trends of research in Social Sciences and Humanities disciplines in India, published during the time period, 2005 to 2014, by studying the keywords and keywords plus, assigned by the authors and Web of Science respectively. The study highlights the research themes which have been pursued by the Indian researchers during the ten year period. It also attempts to spotlight the extent of similarity between authors’ assigned keywords and WOS assigned keyword plus in the SSH publication from India. Descriptive and inferential statistics techniques have been applied in the study. Web of science was used for collecting the data.

Full text

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=tsim20
COLLNET Journal of Scientometrics and Information
Management
ISSN: 0973-7766 (Print) 2168-930X (Online) Journal homepage: https://www.tandfonline.com/loi/tsim20
Occurrence of author keywords and keywords
plus in social sciences and humanities research : A
preliminary study
Manorama Tripathi, Sunil Kumar, S. K. Sonker & Parveen Babbar
To cite this article: Manorama Tripathi, Sunil Kumar, S. K. Sonker & Parveen Babbar (2018)
Occurrence of author keywords and keywords plus in social sciences and humanities research :
A preliminary study, COLLNET Journal of Scientometrics and Information Management, 12:2,
215-232, DOI: 10.1080/09737766.2018.1436951
To link to this article: https://doi.org/10.1080/09737766.2018.1436951
Published online: 28 Nov 2018.
Submit your article to this journal
Article views: 9
View Crossmark data

©
Manorama Tripathi
Jawaharlal Nehru University
New Delhi 110063
India
manoramatripathi2@yahoo.com
Sunil Kumar
Indira Gandhi National Open
University
New Delhi 110044
India
sunilkumar@ignou.ac.in
S. K. Sonker
Babasaheb Bhimrao
Ambedkar University
Lucknow 226025
Uttar Pradesh
India
sksonker@yahoo.co.in
Parveen Babbar
Jawaharlal Nehru University
New Delhi 110063
India
parveenbabbar@gmail.com
Occurrence of author keywords and
keywords plus in social sciences and
humanities research : A preliminary
study
Manorama Tripathi
Sunil Kumar
S. K. Sonker
Parveen Babbar
Keywords are important words and phrases of a research paper, express-
ing its essence. The longitudinal study of occurrence of keywords in
specic discipline expresses trends of research. The present paper at-
tempts to highlight trends of research in Social Sciences and Humani-
ties disciplines in India, published during the time period, 2005 to 2014,
by studying the keywords and keywords plus, assigned by the authors
and Web of Science respectively. The study highlights the research
themes which have been pursued by the Indian researchers during the
ten year period. It also attempts to spotlight the extent of similarity be-
tween authors’ assigned keywords and WOS assigned keyword plus in
the SSH publication from India. Descriptive and inferential statistics
techniques have been applied in the study. Web of science was used for
collecting the data.
Keywords: Author keywords, Keyword plus, Research trends, Web of Science.
1. Introduction
Keywords are essential terms and phrases which high-
light the themes and focus of the research content. These
also indicate central areas which the researchers pursue
and study to generate new data and information and thus
extend frontiers of knowledge. These words highlight
elds, subelds, topics, issues covered in the articles, there-
fore keywords help researchers in locating and retrieving
COLLNET Journal of Scientometrics and Infor mation Management
ISSN : 0973-7766 (Print) 2168-930X (Online)
Vol. 12(2) December 2018, pp. 215–232
DOI : 10.1080/09737766.2018.1436951

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018
216
M. Tripathi, S. Kumar, S. K. Sonker and P. Babbar
relevant articles. The scholarly databases index publications according to the keywords
which are included in them. When a researcher conducts a query, the portal responds by
matching the keywords with the publications indexed in the database.
The authors use keywords to describe the main issues which they have dwelt upon
in their full – text articles. They usually provide ve to six keywords in keeping with the
guidelines of the journals. Web of Science generates and provides keyword plus for re-
search publications and facilitates researchers in locating and retrieving more extensive
relevant information. Keyword plus are words and phrases, which appear in the titles of
references cited by the authors. Author assigned keywords and WOS assigned keywords
plus express knowledge structure of the subject and interlinking of dierent research ar-
eas. The density or frequency of occurrence of keywords in any subject area also highlights
the emerging trends. Keywords plus is more descriptive than author – assigned keywords.
Thus keywords plus can express contents of the articles more succinctly. The present paper
dwells upon the ability of author – assigned keywords and keyword plus to highlight cov-
erage of the topics and subject areas. This study also highlights trends in research articles,
published in Social Sciences and Humanities (SSH) discipline during 2005 – 14 as reected
through author keywords and keywords plus.
2. Related Literature
Keywords of research papers describe contents of papers in systematic and exact man-
ner (Duriau, 2007, [6]). The keywords in scholarly publications help in assessing themes
and subthemes which have been extensively focused, without the need of consulting the
full texts. The keyword analysis highlights important research areas and explains interlink-
ing in dierent research areas (Schodl,2017; Li et al,2015 [13,9]). They are commonly used
words in a discipline and identify prime research themes and trends in that discipline,
thus the analysis of keywords is an essential part of text mining in understanding trends
in research (Conway, 2009,[4]). Wen and Huang (2012, [16]) analyzed title and keywords
of articles on oxidative stress, published in 1991 to 2010 , indexed in Web of Science and
highlighted that antioxidants in human or rat cells, Alzheimer’s disease, and diabetes were
the main research themes pursued by the researchers. Chang et al (2015, [1]) used the
technique of keyword analysis with bibliographic coupling to identify prominent research
subjects in Library and Information Science. Other studies also analyzed frequency of
keywords of scholarly articles to identify major evolving and diminishing research areas
and themes across dierent subjects like ecology (Ne and Corley,2009, [11]), software
engineering(Coulter et al,1998, [5]), renewable energy(Romo – Fernandez,2013, [12), law
and psychiatry (Lv,2017, [10]). Yang et al. (2015, [17]) analyzed keywords of research pub-
lications in Information Science discipline published during 2005 – 16 by using Author
keyword Coupling Analysis (AKCA) and author bibliographic coupling analysis(ABCA)
methods. This study reported that research in Information Science discipline evolved con-
tinuously during the period, but information retrieval showed a decline. The topics of pat-
ent analysis, open access, mapping of science and bibliometric evaluation also experienced

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018 217
Occurrence of author keywords and keywords plus in Social Sciences and Humanities...
substantial growth in that period. Zhang et al (2015, [20]) employed techniques of K – core
of SNA, co – word analysis,UCINET6.0 to highlight main research areas from 4575 papers
on creativity research, indexed in Web of science. The study showed that there were ve
main areas in creativity research, “practice applications of creativity, pathology and physi-
ology of creativity, individual – level creativity, organizational – level creativity, and the
basic theories and methodologies of creativity, and the regular spatial structure”.
Cobo et al (2011, [3]) explained growth and evolution of research in specic discipline
with the help of keywords. This study also described and compared dierent softwares
used to map scientic literature with the help of keywords. Taram and Doulani (2015,
[15]) analysed keywords in biochemistry discipline using LexiURL search tool, which pro-
vides information how links are distributed and arranged among dierent websites. This
functionality helps information seekers to identify websites, which prominently publish
research papers in areas of their interests.
The keywords of research publications are used in indexing and retrieving of infor-
mation, so numbers of downloads and citations are inuenced by keywords assigned by
authors (Uddin & Khan, 2016, [16]). Murphy et al ( 2003, [11]) advocated the authors to
consult controlled vocabulary, while assigning keywords for easy discoverability and re-
trieval of contents.
The analysis of keyword explains interlinking in dierent research areas ( Li etal,2015,
[9]). Research studies have also explored alternatives of keywords to understand and as-
sess contents of papers for example Kipp (2006, [8]) compared user tags with author – as-
signed keywords and indexer’s assigned descriptors for 165 journal articles and concluded
that all these are valuable as additional access points in catalogue searching. Zhang et al
(2016,[19]) compared author assigned keywords and keyword plus of 2397 research pa-
pers in patient adherence discipline published between January 2000 to May 2011. The
study concluded that the keyword plus terms were more descriptive than author assigned
keywords, however both are equally eective in highlighting the knowledge structure of
scientic elds.
3. Objectives of the study
The present study has the following objectives:
• To assess extent of overlapping in author assigned keywords and keyword plus as-
signed by WOS to research publications.
• To assess chances of occurrence of Author assigned keywords and keyword plus in
titles and abstracts of articles.
• To highlight trends of research in Social Sciences and Humanities disciplines fol-
lowed by Indian scholars with frequency of keywords and keywords plus terms.

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018
218
M. Tripathi, S. Kumar, S. K. Sonker and P. Babbar
3.1. Method of study
Gareld (1990,[7]) advocated statistical analysis of keywords in identifying focus ar-
eas of research and predicting trends of research in that discipline. The present study has
applied quantitative statistical techniques to explain trends in research with the help of
frequency of keywords and keywords plus. Descriptive statistical techniques have applied
to explain trends in author assigned keywords and keyword plus from cohort data of ten
years. Inferential statistics techniques have been used to test null hypotheses postulated
in the study. This study is based on secondary data. The data for the study was extracted
from Social Science Citation Index (SSCI) and Arts and Humanities Citation Index (AHCI),
which are integral components of the Web of Science, an online bibliographic and citation
indexing service maintained and published by Institute for Scientic Information, at pres-
ent it is maintained by Clarivate Analytics. According to Clarivate Analytics (2017, [2]);
SSCI and AHCI have indexed research publications of 372 and 1782 journals respectively.
The data was collected on the following parameters: title of research paper, year of
publication, Author keywords, keyword plus, research areas, name of journals. The cohort
analysis of these parameters explained trends in research publications in SSH disciplines
in the ten year period. The data for the study was collected in September 2016.
The authors got 9525 records of Indian Scholars in Social Sciences and Humanities
disciplines, published during the time period, 2005 to 2014, which were analyzed in this
study. These records were downloaded in batches of 500 records (the maximum 500 re-
cords can be downloaded at one time from Web of Sciences) and saved as le in MS – Excel,
a component of MS oice. The authors considered the following types of documents i.e. ar-
ticles, proceedings, book reviews, review, and poetry for the analysis in the present study.
The authors used Ms – Excel and SPSS package for further quantitative analysis.
4. Data Analysis and Discussion
4.1. Number of publications
There were 9525 research publications from Indian scholars in SSH disciplines during
2005 – 2014. The trend in growth of research publication is reected in gure 1. There was
consistent growth in number of research articles from India, during 2005 to 2014 with a
spike with remarkable high numbers of papers in 2011. The number of publications in 2014
was almost four times of publications in 2005 There was exceptionally high growth in 2011.
4.2. Prime research Areas in Social Sciences and Humanities
Areas of research in Social Science and Humanities are dynamic and sensitive to chang-
ing needs of society and disciplines. The Cohort analysis of comparative growth in num-
bers of research publications in dierent disciplines helps in understanding change in
priorities in research over a period of time. Table 1 is conned to areas of research which
had more than 100 papers in the period of investigation. The research publications in each
discipline are blocked into two with each block of ve years

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018 219
Occurrence of author keywords and keywords plus in Social Sciences and Humanities...
Figure 1
Annual growth of research publications by Indians in SSH disciplines
Table 1
Popularity of research areas in social science and humanities in India from 2005-2014
Research Areas 2005-09 2010-14 % increase in 2010-14
with base years 2005-09
Grand
Total
No %No %
Business & Economics 399 11.3 889 14.8 122.8 1288
Sociology 194 5.5 212 3.5 9.3 406
Anthropology 236 6.7 167 2.8 – 29.2 403
Government & Law 123 3.5 226 3.8 83.7 349
History 142 4.0 142 2.4 0.0 284
Public Administration 120 3.4 156 2.6 30.0 276
Women’s Studies 123 3.5 138 2.3 12.2 261
Social Work 150 4.3 91 1.5 – 39.3 241
Psychology 62 1.8 159 2.6 156.5 221
Education & Educational Research 63 1.8 101 1.7 60.3 164
Anthropology 38 1.1 120 2.0 215.8 158
Literature 68 1.9 87 1.4 27.9 155
Health Care Sciences & Services 44 1.2 89 1.5 102.3 133
Area Studies 41 1.2 86 1.4 109.8 127
Religion 12 0.3 95 1.6 691.7 107
Sub total 2352 66.7 3518 58.6 49.6 5870
Rest of research areas and
interdisciplinary 1710 48.5 3243 54.0 89.6 4952
Grand Total 3525 100.0 6001 100.0 70.2 9525

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018
220
M. Tripathi, S. Kumar, S. K. Sonker and P. Babbar
Table 1 indicates that “Business and Economy’, Sociology and Anthropology were top
three research areas with the highest number of research articles by Indian scholars from
2005 to 2014. The cumulative number of research publications in ‘Business and Economy’
holding top position, is over three times of number of publications in Sociology holding
second position. There was substantial decline of number of articles in anthropology and
social works in the second ve year block. The remarkable increase in number of research
publications in the second 5 year block was noted in Religion, Anthropology, and Psychol-
ogy.
Keyword plus is as eective as Author keywords in highlighting the structure of subject
domains. But Zhang (2016,[19]) criticized keyword plus for not describing and represent-
ing the authors’ content in an adequate manner. Frequency of occurrence of keywords in a
discipline depicts key research areas of that discipline. Table2 reects descriptive statistics
for authors’ assigned keywords and WOS assigned keyword plus in terms of numbers of
keywords, keyword plus, average of these parameters of research publications of Indian
scholars during 2005 – 14. The table 2 is conned to popular research areas having over one
hundred research papers in 2005 to 2014.
verage No of keywords assigned by author per paper in specific discipline
Total no of keywords assigned by authors in that
discipline
=total no of research publication in that disciplines -no of
A
research publication
without keywords in that disciplines
Exploratory analysis of data collected for this study highlights that a number of journals
in social sciences and humanities disciplines do not make it compulsory for the authors to
assign keywords to the articles. it results in remarkable numbers of research papers were
without keywords. Table 2 indicates that the authors assigned four to ve keywords in
every publication in most of the disciplines. The mean number of authors assigned key-
word per paper was the highest in ‘Religion’ disciplines with 6.5 keywords per paper and
6.4 keywords per paper in History discipline, the lowest numbers of keywords were in
disciplines ‘Business & Economics’ with 4.8 keywords per paper. The standard deviation
values of keyword plus and keyword, expressing scatteredness on mean value, are 1.8 and
0.5 respectively, it indicates values of average no of keywords per paper across discipline
don’t have great variance in comparison of scatteredness in keyword plus.
Search results will be generic in disciplines which have less number of author assigned
keywords. Scholar will get either no papers or almost all the research papers as there are
very few keywords which occur frequently. Search results in disciplines with high number
of keywords per research publication, will be narrow as there are very few research papers
which have common keywords.
There were 2545 research publications for which Web of Science did not assign any
keyword plus. Table 2 indicates there were remarkable gaps in average numbers of key-
word plus per research publication across the research areas. The average numbers of

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018 221
Occurrence of author keywords and keywords plus in Social Sciences and Humanities...
Table 2
Descriptive statistics of authors’ assigned keywords and WOS assigned keyword plus
Research area
Keywords plus of WOS Authors’ assigned keywords
No of
papers
without
keywords
plus
Total no of
research
papers
No of
keyword
plus
Average no
of Keyword
Plus per
research
paper
No of papers
without
author
keywords
Total no of
research
papers
Numbers
of authors
keywords
Average no of
keywords per
paper
Business & Economics 313 1288 5827 6 291 1288 4825 4.8
Sociology 376 406 93 3.1 319 406 493 5.7
Anthropology 229 403 982 5.6 193 403 1032 4.9
Government & Law 299 349 161 3.2 176 349 928 5.4
History 260 284 50 2.1 242 284 268 6.4
Public Administration 175 276 419 4.1 169 276 570 5.3
Women’s Studies 207 261 198 3.7 171 261 451 5
Social Work 189 241 125 2.4 223 241 95 5.3
Psychology 24 221 1436 7.3 64 221 785 5
Education &
Educational Research 94 164 227 3.2 86 164 414 5.3
Literature 150 155 13 2.6 133 155 111 5
Health Care Sciences
& Services 24 133 637 5.8 44 133 476 5.3
Area Studies 107 127 42 2.1 92 127 191 5.5
Religion 98 107 13 1.4 70 107 242 6.5
Subtotal 2545 4415 2273 4415 
Research papers in rest of research
areas 5110 5110 
Total 9525 9525

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018
222
M. Tripathi, S. Kumar, S. K. Sonker and P. Babbar
keyword plus per research paper in ‘Religion’ research area was 1.4, on the higher side of
scale ‘Psychology’ discipline had 7.3 keywords plus per research publication.
4.3. Association in number of author keywords and keyword plus
Table 2 compares numbers of keyword and keyword plus across disciplines. Research
papers in Business & Economics disciplines have good number of keywords as well as
keyword plus; Health care sciences discipline have almost same number of keyword and
keyword plus; Religion discipline has the highest gap in keyword and keyword plus in
favor of higher number of keywords; on the other side the highest gap in favor of keyword
plus was in Psychology.
Null hypothesis:
H0 : Research areas having high number of author assigned keywords need not have
high number of keyword plus
The null hypothesis which examines association in average number of keywords and
average number of keyword plus per paper has been tested with Spearman’s rho Correla-
tion Coeicient. The test is conned to research areas which had more than one hundred
papers during 2005 to 2014.
The value of Spearman’s rho Correlation Coeicient shows signicant association be-
tween author assigned keywords and keywords plus. The negative value of Spearman’s
rho correlation coeicient indicates disciplines ranked high when sorted on number of
author assigned keywords may be ranked low when sorted on numbers of keywords plus
per paper.
The null hypothesis is rejected thus research areas having high number of author as-
signed keywords have comparatively low number of keyword plus.
4.4. Unique keywords and keyword plus
Common keywords in research papers indicate that theses papers might cover similar
issues thus study of unique keyword and keyword plus in research papers reect extent of
diversity in coverage of issues. Research areas with high number of unique keyword and
keyword plus reects high diversity in research coverage. Table 4 reects unique keywords
and keyword plus.
The highest number of unique keywords was in Religion discipline; Psychology disci-
pline had the highest numbers of unique keyword plus as well as there was almost no gap
in number of unique keyword and keyword plus.
4.5. Frequency of keywords plus
Keyword plus of a paper are words which appear in titles of research papers referred
by authors while writing papers. Popularity of keyword plus reects of titles of research

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018 223
Occurrence of author keywords and keywords plus in Social Sciences and Humanities...
Table 3
Association in average number of keywords and average number of keyword plus per paper in
popular research areas
Parameter Value
Spearman’s rho Correlation Coeicient in Numbers of keyword and keywords plus
in each discipline
– .567
Sig. (2 – tailed) .006
No of values 22
Table 4
Descriptive Analysis of unique keywords and keywords plus
Research area
Unique
Author
keywords
Unique
keyword
plus
Average
Unique
keywords
Average
Unique
keyword plus
Business & Economics 3461 2525 3.5 2.6
Sociology 404 74 4.6 2.5
Anthropology 796 579 3.8 3.3
Government & Law 663 117 3.8 2.3
History 205 39 4.9 1.6
Public Administration 442 336 4.1 3.3
Women’s Studies 340 148 3.8 2.7
Social Work 84 105 4.7 2
Psychology 705 865 4.5 4.4
Education & Educational Research 357 190 4.6 2.7
Literature 107 9 4.9 1.8
Health Care Sciences & Services 365 398 4.1 3.7
Area Studies 169 36 4.8 1.8
Religion 218 12 5.9 1.3
Correlation is signicant at the 0.01 level (2 – tailed).
publication frequently read and referred by Indian researchers. Table 5 reects frequency
of popular keywords plus to identify key research areas being pursued by Indians in social
sciences and humanities disciplines from 2005 to 2014. This table is conned to popular
keywords plus repeated in more than 100 research papers from 2005 to 2014.
Table 5 indicates that annual spread (frequency) of popular keywords had been uni-
form across ten years period. Model, Performance, Impact and India were the most popu-
lar keywords plus assigned by Web of Sciences in research publications in SSH disciplines
from 2005 to 2015, ; thus each area of research in SSH discipline has few words which
frequently appeared in titles of research publications. The study expressed ailiation of
keyword plus with areas of research.

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018
224
M. Tripathi, S. Kumar, S. K. Sonker and P. Babbar
Table 5
Annual density of popular keywords plus
Keywords plus 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Total
Model 8 19 14 30 37 51 73 54 60 64 410
Performance 4 8 7 11 21 37 41 32 35 51 247
Impact 10 3 8 8 17 25 34 22 28 34 189
Management 6 7 5 15 17 20 48 25 22 24 189
India 9 4 10 14 20 16 37 17 19 42 188
Behavior 8 6 4 12 16 18 37 21 21 21 164
Health 5 3 10 7 8 16 25 28 19 33 154
Women 7 4 8 6 10 16 23 16 10 17 117
Growth 8 5 6 12 10 13 18 13 19 10 114
Risk 6 5 5 6 8 11 17 16 13 23 110
Countries 8 4 1 2 13 10 11 20 15 18 102
Systems 2 3 2 5 11 18 27 12 15 7 102
Total 81 71 80 128 188 251 391 276 276 344 2086
Table 6
Annual density of keywords assigned by authors
Author assigned popular
keywords 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Total
India 36 54 59 81 99 124 151 132 160 142 1038
gender 3 5 4 1 4 12 13 13 8 11 74
HIV 2 4 2 6 6 12 6 18 12 6 74
Poverty 1 2 5 4 7 5 7 9 13 12 65
culture 2 1 2 2 3 8 11 7 5 13 54
innovation 4 2 3 4 2 7 8 5 6 2 43
development 1 3 2 2 5 4 6 7 5 7 42
Inventory 1 2 3 5 6 7 6 10 10 4 54
developing countries 5 3 5 6 3 3 8 1 3 4 41
Inequality 1 2 3 3 3 5 5 8 10 40
Total 55 77 87 114 138 185 221 207 230 211 1525
4.6. Frequency of Author assigned keywords
Table 6 reects trends in popular keywords assigned by authors. This table is conned
to popular keywords assigned by authors in over 40 research papers.
Table 6 indicates that the most popular keywords during 2005 – 14 were India, Gender,
HIV and Poverty. Table also indicates that ranking in popular keywords is uniform across
ten years period, thus there was no signicant change in research themes in SSH discipline
in India during 2005 – 14. Tables 5 and 6 indicates that ‘India’ is common word; equally
popular as keyword plus as well as keywords.

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018 225
Occurrence of author keywords and keywords plus in Social Sciences and Humanities...
Table 7 express descriptive statistics of number of keywords and keyword plus in indi-
vidual research publication by Indian scholars in social science and humanities disciplines
from 2005 – 2014.
There were 4041 research publications without keywords. 71.3 % of research articles
had 4 to 6 keywords in each paper. The maximum number of keywords in individual re-
search publication was eight; 8.2 % of research publications had eight author assigned
keywords.
Web of Science assigned keyword plus to 4993 research papers out of 9525.The number
of keyword plus in individual paper ranged from 1to 10; with mean value of 5.7 keywords
plus per paper. 22.3 % of research publications had 2 to 3 keyword plus.
4.7. Occurrence of keywords and keyword plus in titles of Research Articles
Titles of research articles always highlight theme of the full text. Titles may be consid-
ered as one line surrogates of full text articles. Search portals allow searching and retrieval
of research papers with specic words in titles or abstract of research paper. Searching of
specic word in title and abstract of research paper is fast in comparison of looking the
same word in entire texts of the papers. Searching of specic words in entire text may re-
sult in retrieval of research papers having little relevance to criteria specied by users, how-
ever searching by specic words of titles and abstracts of papers ensures greater relevant
Table 7
Number of keywords and keyword plus in research articles
Number of keyword/
keyword plus in each
paper
Keyword plus Keyword
Numbers
of research
papers
Percent Cumulative
Percent
Number of
papers Percent Cumulative
percent
1 697 14 14 4 0.07 0.07
2 604 12.1 26.1 66 1.2 1.28
3 461 9.2 35.3 678 12.4 13.64
4 426 8.5 43.8 1255 22.9 36.52
5 373 7.5 51.3 1610 29.4 65.88
6 293 5.9 57.2 1044 19 84.92
7 305 6.1 63.3 379 6.9 91.83
8 232 4.6 67.9 448 8.2 100
9 219 4.4 72.3 – – – – – –
10 1383 27.7 100 – – – – – –
Total 4993   5484
Research papers without keyword plus 4532 4041
Maximum numbers of keyword plus 10 8
Average number of keyword 5.69 5.05

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018
226
M. Tripathi, S. Kumar, S. K. Sonker and P. Babbar
results. Table 8 shows occurrence of keywords and keyword plus in the titles of research
articles.
There were 3157 articles which had at least one of keyword plus in their titles; while
1836 article titles did not have any word which matched with keyword plus. 80 % of re-
search articles had 1 or 2 keyword plus in their titles.
34.5 % of research papers had at least one of keywords in their titles; 26.6% of research
papers had two author – assigned keywords in their titles. 61.1 % of research papers had
one or two keywords in their titles. Table 8 indicates that readers can rely on article titles,
when keywords are not given, to understand what is covered in full texts. High occurrence
of keywords in article titles also ensures that reader get relevant research papers when
searching with keyword plus is conned to article titles. Table 9 reects occurrence of au-
thor assigned keywords in the abstracts of research articles.
There were 42 % of research articles which had 1 to 2 keywords in their abstracts. There
were 448 research articles with eight author keywords; out of these, 12 articles had all the
eight author keywords in their abstracts. There were 7.5 % of research articles which had
none of their keywords in their abstracts. Though abstracts are concise, still there are good
chances of getting keywords in them. Abstracts appropriately dene and describe contents
of the publications. Readers can rely on abstracts of research papers for getting glimpses of
contents in case keywords are not given.
Table 8
Number of keywords and keyword plus in title of research articles
No of keyword/
keyword plus in
title of research
paper
Keyword plus Keyword
Numbers
of papers Percent Cumulative
percent
Number of
papers Percent Cumulative
Percent
1 1685 53.4 53.4 1894 34.5 34.5
2 847 26.8 80.2 1457 26.6 61.1
3 451 14.3 94.5 667 12.2 73.3
4 130 4.1 98.6 194 3.5 76.8
5 30 1 99.6 46 0.8 77.6
6 9 0.3 99.8 7 0.1 77.8
7 5 0.2 100 1 0.02 77.8
Subtotal 3157 4266
Research papers without keyword plus in
title 1835 1217
Research papers without keyword plus 4533 4042
Total 9525 9525

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018 227
Occurrence of author keywords and keywords plus in Social Sciences and Humanities...
4.8. Overlapping and redundancy in Author assigned keywords and Web of Sciences Key words
plus
Overlapping in keywords and keyword plus in research articles is dened as presence
of same words as keyword and keyword plus. Overlapping index quanties extent of over-
lapping of keywords and keyword plus in research articles. High value of overlapping
index indicates that searching of research papers either on keyword or keyword plus will
lead to retrieval of same research articles, so readers can rely on keyword or keyword plus.
Research areas with low overlapping index will result in entirely dierent sets of results
when searching is done with keywords and another done on keyword plus. Disciplines
with high overlapping in keyword and keyword plus will always result in fast results, ir-
respective of indexing, by author or keyword plus. It would be prudent to minimize over-
lapping in keyword and keyword plus before conducting search. The redundancy index
indicates extent of avoiding of overlapping words as keyword and keyword plus. Research
areas with high level of overlapping in keywords and keywords plus, are likely to have
high value of redundancy index. Formulas applied for computing of overlapping index
and redundancy factor are as follows:
A(i) numbers of author assigned keywords in ith research paper
P(i) numbers of keywords plus in ith research paper
A(i) υ P(i) Total of numbers of keyword plus and keywords in ith Research paper –
Numbers of common words in keyword plus and keywords in that Research paper
A(i) « P(i) Numbers of common words in keyword plus and keywords in ith Research
paper.
Table 9
Numbers of keywords in the abstract of research articles
No of keywords No of papers Percent Cumulative Percent
1 875 17.3 17.3
2 1242 24.5 41.8
3 1338 26.4 68.2
4 897 17.7 85.9
5 486 9.6 95.5
6 178 3.5 99.0
7 40 0.8 99.8
8 12 0.2 100.0
No of research articles with at least one keyword in abstract 5068
No of research articles without any author keyword in abstracts 416
Research papers without keywords 4041
Total 9525

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018
228
M. Tripathi, S. Kumar, S. K. Sonker and P. Babbar
A(i) P(i)
A(i) P(
Overlapping Index =
Redundancy Index =
i)
A(i) P(i)
A(i)
υ
∩
∩
The Table 10 shows Overlapping Index and Redundancy Index of keywords and key-
word plus in popular research areas in social sciences and humanities disciplines in India
in 2005 to 2014.
The high value of overlapping index of a discipline indicates occurrence of same
word as author keywords and keyword plus in that discipline. The overlapping index for
Sociology was the highest,5.5, followed by Information Science & Library Science with
overlapping index of 2.9. The values of overlapping index and redundancy index for
Government & Law, History, Public Administration, Social Work, Literature, Health care
Table 10
Overlapping Index and Redundancy Index of keywords and keyword plus in
popular Research Areas
Research Area
Papers without
authors
keywords
Total numbers
of research
publications
Overlapping
Index
Redundancy
Index
Business & Economics 291 1288 0.1 0.23
Sociology 319 406 5.5 9.25
Anthropology 193 403 0.06 0.31
Government & Law 176 349 0 0
History 242 284 0 0
Public Administration 169 276 0 0
Women’s Studies 171 261 1.17 2.17
Social Work 223 241 0 0
Psychology 64 221 1.58 3.13
Information Science &
Library Science 49 193 2.92 8.33
Education & Educational
Research 86 164 1 1.14
Literature 133 155 0 0
Health Care Sciences &
Services 44 133 0 0
Area Studies 92 127 1.5 13
Religion 70 107 0 0

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018 229
Occurrence of author keywords and keywords plus in Social Sciences and Humanities...
science and services disciplines were 0, there was no overlapping in keyword and keyword
plus in these disciplines.
4.9. Overlapping in author keywords and keywords plus
There are 453 research papers having overlapping in keyword and keyword plus.
Analysis of Overlapping in author keywords and keywords plus is explained in table
11. Columns of the tables reect numbers of keyword plus in individual paper and rows
expresses numbers of authors’ assigned keyword. The cross section of rows and columns
represent numbers of papers with overlapping keyword and keyword plus.
Table 11
Overlapping in keyword and keyword plus
Number of author
assigned keywords
in each paper
Numbers of WOS assigned keyword plus in each paper
1 2 3 4 5 6 7 8
Numbers of
papers with
overlapping
No
overlapping Total
1 0 0 1 2 2 0 0 5 692 697
2 0 0 4 7 5 1 2 19 585 604
3 0 1 4 12 14 2 3 36 425 461
4 0 2 6 8 8 2 2 28 398 426
5 0 2 7 11 11 3 3 37 336 373
6 1 2 6 12 10 3 4 38 255 293
7 0 4 6 5 7 3 4 29 276 305
8 0 3 4 8 5 4 4 28 204 232
9 0 3 3 5 4 2 4 21 198 219
10 0 7 33 68 66 19 19 212 1171 1383
No of papers with
overlapping 0 1 24 74 138 132 39 45
No overlapping 4 65 654 1181 1472 912 340 403
Total 4 66 678 1255 1610 1044 379 448
Table 12
Association in numbers of keyword and keyword plus in individual paper
Parameter Numbers of
research papers
Value of Pearson
Correlation Sig (2 tailed)
Association in numbers of authors
keyword and numbers of keyword plus 9525 0.36 0.00

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018
230
M. Tripathi, S. Kumar, S. K. Sonker and P. Babbar
This table may be read like this that there are 305 research papers with 7 keyword out
of which 29 are with overlapping keyword plus. Row corresponding to 7 keyword express
that out of 29 overlapping papers; 4 papers were 3 keyword plus; 6 paper with 4 keyword
plus; 5, 7, 3 and 4 research papers with 5,6,7 and 8 keyword plus. Others rows and columns
of the table may be read in similar fashion. The table reects that chances of overlapping
improve with increase in numbers of keyword and keyword plus. The association in num-
bers of keyword and numbers of keyword plus in individual paper is reected in table 12.
The signicant level of values of Pearson Correlation test has proved association in
number of keywords plus and number of author assigned keywords in individual research
publications. However, very low values of the Pearson correlation express very weak as-
sociation in parameters under examination. Thus authors can improve chances of overlap-
ping by assigning more keyword to paper.
5. Conclusion
Volume of research publications of Indian scholars had substantially improved in most
research areas of Social Sciences and Humanities (SSH) disciplines during 2005 – 14 with
remarkable increase in publications in religion, anthropology, and psychology disciplines,
however there was minor drop in publication in anthropology and social work. Frequency
of popular keyword and keyword plus in SSH disciplines among Indian scholar remain
consistent during 2005 – 14 so there was no major change in research landscape in SSH
in India. Majority of journals do not make it compulsory for the authors to assign key-
words to their articles. Authors usually assign 4 to 6 keywords in individual papers across
various research areas in SSH disciplines. Web of Science is not consistent in assigning
keyword plus across research areas for example average number of keyword plus per re-
search paper in ‘Religion’ was 1.4 whereas ‘Psychology’ had 7.3 keywords plus per re-
search publication. There was statistically negative association in number of keywords and
keyword plus across research areas; this association states that research areas ranked high
in terms of number of keywords are usually ranked low when sorted on numbers of key-
word plus. The study highlights that the overlapping in the Author assigned keywords and
keyword plus diers with areas of research. The overlapping Index was the highest for the
research articles in Sociology. Number of unique keywords and keyword plus also diers
with areas of research. Indexing and sorting of research publications either on keyword or
keyword plus will be almost same in case of high overlapping in keyword and keyword
plus. Uniqueness in keywords expresses extent of diversity in coverage of research areas
thus research areas with high number of unique keywords express high diversity in cov-
erage of research areas. Search results in disciplines with few unique keywords will be
generic; search portal will display most of paper in case search criteria match with one of
few unique keywords. Researchers need to adjust search criteria on the basis of extent of
uniqueness of keyword in that discipline. There are good chances of nding keywords and
keyword plus in titles and abstracts of research papers; so these keywords should be used
for locating and retrieving specic information.

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018 231
Occurrence of author keywords and keywords plus in Social Sciences and Humanities...
References
[1] Chang, Y. W., Huang, M. H., and Lin, C. W. Evolution of research subjects in library and
information science based on keyword, bibliographical coupling, and co – citation analy-
ses.Scientometrics 105(3), 2015, 2071 – 2087.
[2] Clarivate Analytics. Master Journal List, Social Sciences Citation Index and Arts & Hu-
manities Citation Index, 2018. Retrieved from http://mjl.clarivate.com/cgi – bin/jrnlst/
jloptions.cgi?PC=SS and http://mjl.clarivate.com/cgi – bin/jrnlst/jloptions.cgi?PC=H
[3] Cobo, M. J., López–Herrera, A. G., Herrera–Viedma, E., and Herrera, F. Science mapping
software tools: Review, analysis, and cooperative study among tools.Journal of the Associa-
tion for Information Science and Technology,62(7), 2011,1382 – 1402.
[4] Conway, M. Mining a corpus of biographical texts using keywords.Literary and Linguistic
Computing,25(1), 2009, 23 – 35.
[5] Coulter, N., Monarch, I. and Konda, S. Software Engineering as seen through its Research
Literature: A Study in co – word Analysis.Journal of the Association for Information Science
and Technology 49(13), 1998, 1206 – 1223.
[6] Duriau, V. J., Reger, R. K., and Pfarrer, M. D. A content analysis of the content analysis
literature in organization studies: Research themes, data sources, and methodological re-
nements.Organizational research methods10(1),2007, 5 – 34.
[7] Gareld, E. KeyWords PIus: 1S1’s Breakthrough Retrieval Method. Part 1. Expanding Your
Searching Power on Current Contents on DisketteCurrent Contents,32, 1990,295 – 298. Re-
trieved from http://gareld.library.upenn.edu/essays/v13p295y1990.pdf
[8] Kipp, M. E. Complementary or discrete contexts in online indexing: A comparison of user,
creator and intermediary keywords. 2006 http://hdl.handle.net/10150/106315
[9] Li, H., An, H., Wang, Y., Huang, J., and Gao, X. Evolutionary features of academic articles
co – keyword network and keywords co – occurrence network: Based on two – mode aili-
ation network.Physica A: Statistical Mechanics and its Applications 450, 2016, 657 – 669.
[10] Lv H. Assessment of global law and psychiatry research in the period of 1993 – 2012:
Review and forecast on research emphases and research trends (Part – II). The Electronic
Library 35(6), 2017,1162 – 76.
[11] Murphy, L. S., Reinsch, S., Najm, W. I., Dickerson, V. M., Seinger, M. A., Adams, A., and
Mishra, S. I. (2003). Searching biomedical databases on complementary medicine: the use
of controlled vocabulary among authors, indexers and investigators.BMC Complementary
and Alternative Medicine,3(1), 2003 3 – 3.
[12] Ne, M. W., & Corley, E. A. 35 years and 160,000 articles: A bibliometric exploration of the
evolution of ecology.Scientometrics,80(3), 2009, 657 – 682.
[13] Romo – Fernández, L. M., Guerrero – Bote, V. P., & Moya – Anegón, F. Co – word based
thematic analysis of renewable energy (1990–2010).Scientometrics97(3), 2013, 743 – 765.

COLLNET JOURNAL OF SCIENTOMETRICS AND INFORMATION MANAGEMENT 12(2) DECEMBER 2018
232
M. Tripathi, S. Kumar, S. K. Sonker and P. Babbar
[14] Schodl, K., Klein, F., and Winckler, C. Mapping sustainability in pig farming research us-
ing keyword network analysis.Livestock Science196, 2017,28 – 35.
[15] Taram, K., & Doulani, A. A webometric analysis of major keywords and expressions in
biochemistry using LexiURL Searcher.The Electronic Library 33(6), 2015, 1163 – 1173.
[16] Uddin, S., & Khan, A. (2016). The impact of author – selected keywords on citation
counts.Journal of Informetrics 10(4), 2016, 1166 – 1177.
[17] Wen, H., & Huang, Y. Trends and performance of oxidative stress research from 1991 to
2010.Scientometrics 91(1), 2012, 51 – 63.
[18] Yang, S., Han, R., Wolfram, D., and Zhao, Y. Visualizing the intellectual structure of in-
formation science (2006–2015): Introducing author keyword coupling analysis.Journal of
Informetrics10(1), 2016, 132 – 150.
[19] Zhang, J., Yu, Q., Zheng, F., Long, C., Lu, Z., and Duan, Z. Comparing keywords plus of
WOS and author keywords: A case study of patient adherence research.Journal of the As-
sociation for Information Science and Technology 67(4), 2016, 967 – 972.
[20] Zhang, W., Zhang, Q., Yu, B., and Zhao, L. Knowledge map of creativity research based
on keywords network and co – word analysis, 1992–2011.Quality & Quantity 49(3), 2015,
1023 – 1038.