Scientometrics and Altmetrics Trends and Issues Dr Bidyarthi

Scientometrics and Altmetrics: Trends and Issues Dr. Bidyarthi Dutta Vidyasagar University Midnapore, West Bengal, India Email: bduttavu@mail. vidyasagar. ac. in

In LIS………… • Librametrics • Bibliometrics • Scientometrics • Informetrics • Webometrics……….

Areas of applications of statistical tools and techniques to various disciplines have given new names such as Econometrics, Sociometris and Informetrics etc. Infor/Biblio/Sciento…metrics were formed to develop statistical and mathematical method in order to study and analyses the characteristics of all kinds of information.

• Any system of measurement invariably involves one crux subject from methodological viewpoint, which is nothing but statistics.

• Prof. P. C. Mahalanobis in the early fifties described statistics as a key technology, because it is required for all socio-economic development activities and since statistical techniques are used in all development and forecasting studies.

• As statistics demands systematization and organization of data, similarly LIS demands organization of knowledge.

The terms ‘Informetrics’ was coined by Blackert, Siegal and Otto Nacke in 1979 but gained popularity by launch of the International Informetrics Conferences in 1987. Definition: According to Tague – Sutclifle, “Informetrics is the study of the quantitative aspects of information in any form, not just records or bibliographies and any social group, not just scientists”.

The quantitative studies in Library and Information Systems and Services at different time period were popular by various names such as • • • Statistical bibliography (1920 s) Librametry (1940 s) Bibliometrics (late 1960 s) Scientometrics (1960 s) and Informetrics (1980 s)

Leo Egghe and Ronald Rousseau have identified a place for Informetrics among other fields

Different domains of informetrics: the umbrella term • • • Scientometrics Bibliometrics Patentometrics Blogometrics News Metrics Webometrics Cybermetrics Wikimetrics Open source metrics and so on………….

Bibliometrics The term was coined by Pritchard in 1969. According to him, ‘”Bibliometrics means the application of mathematics and statistical method to books and other communication media”. Bibliometrics studies are mainly employed to investigate the following areas. �Scattering of articles �Author productivity �Citation analysis �Measures of journal productivity

Scientometrics…. . The term was coined in 1969 by Vassily V. Nalimov & Z. M Mulchenko. According to Van Raan “Scientometrics is defined as the quantitative study of science and technology”. Scientometrics studies are mainly applied for the following topics. � History of science �Growth of Science and Scientific institutions �Behavior of science and scientists �Science policy and decision indicators

Almind and Ingwerson coined the term Webometrics. According to Bjorneborn & Ingwerson “Webometrics is defined as the study of quantitative aspects of construction and use of information resources, structures and technologies on the web drawing on bibliometrics and informetrics approaches”. There are four areas of webometrics research as follows. �Web page content analysis. �Web link structure analysis. �Web usage analysis (Including log files of users’ search & browsing behavior). �Web terminology analysis (Including search engine performance).

Relationship between Different Metrics Thelwall, Vaughan and Bjorneboren in 2005 attempted to explained relationship between different metrics through the following diagram.

�The field of informetrics embracing the overlapping filed of bibliometrics and scienctometrics. �Webometrics is seen as entirely encompassed by bibliometrics because web documents in their various forms such as text, multimedia are all recorded information stored on the web server. �Webometrics is partially covered by scientometrics because many scholarly activities today are webbased. �Cybermetrics completely covers webometrics but exceeds the boundaries of bibliometrics because some activities in the cyberspace are not normally recorded.

Ravichandra Rao I. K. stated that the research areas of following topics are called as librametry, or as scientometrics or as bibliometrics or as infometrics: Ø Quantitative aspects of library and information science, especially use and user studies. Ø Quantitative studies related to book usage, acquisition, age distribution of documents etc. Ø Circulation studies. Ø Citation studies / analyses (impact factors and other measures) Ø Journal productivity (by coverage, by use, by citation, etc. ) Ø Author productivity. Ø Obsolescence and growth studies. Ø Quantitative analysis of science (- science indicators, country-wise, language-wise, subject-wise etc. ). Ø Identifying relations among various disciplines Ø Structure of subjects / disciplines Ø Evaluation of scientific research (by institutions, by individuals, by countries)

Bibliometric Approach • Concept of metrics was originated from here • S. C. Bradford & A. J. Lotka are known as Fathers of Bibliometric Approach • Derek De Solla Price generalized different bibliometric laws • Eugene Garfield showed the applications of bibliometric laws

Derek de Solla Price (1922 -83) • Was a physicist and historian of science • He estimated the number of the scholarly periodical titles being published by the end of 20 th century would exceed one million

• Little Science, Big Science and Beyond. .

Eugene Garfield (1925 -2017) • Father of Citation Index, Founder of ISI in 1964 (Science Citation Index, Social Science Citation Index and Arts & Humanities Citation Index) • That is Web of Knowledge today

Classical Indicator Approach • Impact Factor: I = (C 1+C 2)/(X 1+X 2) • Immediacy Index: I(i) = C/X • Cited Half Life • Citing Half Life

Modern Indicator Approach

Modern Indicator Approach h-index: an incomplete indicator h h e 2 = ∑(Cj – h) = ∑Cj - h 2 …. (1) (e-Index) j=1 Where cj are the citations received by the jth paper and e 2 denotes the excess citations within the h-core. Assuming, d 2 = ∑Cj …. (2) j=1 It is obtained, d 2 = e 2 + h 2; …. . (3) Here e ≥ 0 and e is a real number. Or e = √( d 2 - h 2)……. (4)

e-Index • Larger the e, the larger the net excess citations, and higher the loss of citation • The smaller the e, the more reliable the h-index

• Now, replacing the summation of equation (1), by integration, assuming the citation function Cj as a continuous dependent function of time (t), equation (1) may be rewritten as: • h • e 2 = ʃ(C(t) – h)dt (5) • 0 • Or, • h • e 2 = ʃC(t)dt – h 2 (6) • 0

• Citation Loss: associated with excess citation, i. e. eindex • Citation Fall: associated with low citation

• Now, an arbitrary point in h-e plane represents the overall information of citations received by all papers in the h-core. It is interesting to point out that the Euclidean distance between the origin and the point P(h, e) is equal to • R = (h 2 + e 2)1/2 = d………. (7)

• Thus, R-index (Equation 6) indicates state of citations of all papers in h-core. Now, rewriting Equation (3), it is found, • d 2 = h 2 + e 2 = h*(h + e 2/h) = R 2 ………………. (7 A) • Now, (h + e 2/h) = a, say………. (8) • Substituting equation (8) in Equation (7 A), it is found, R 2 = a*h…. . (9) • Another index in defined in Equation (9), i. e. a-index, which is the ratio between Rindex and h-index.

• Therefore, 4 indices are obtained here, i. e. h, e, a and R index. • These four indices can be divided into two types, fundamental and derived indices. A fundamental index is an independent variable and can be used to derive other indices. Here h and e are fundamental indices, since they are independent of each other. Whereas, a and R are derived indices that are calculated from h and e indices.

• h-index ignores excess citation causing loss of citation information. • To overcome this limitation of h index, Leo Egghe proposed another index in 2006, the g-index that is considered as an alternative for the h -index.

• The g-index does not average the numbers of citations. • It allows citations from highercited papers to be used to reinforce lower-cited papers in meeting this threshold.

g-index i. e. g 2≤∑Ci (i<g), ………. . (10), where Ci indicates total number of citations received by top cited i number of articles.

h-index may be holistically interpreted by e-index, also g-index is significant, the picture of h-index and e-index becomes complete after inclusion of R-index and a-index.

• Journal Level Metrics: IF, Immediacy Index, Half Life (Cited & Citing), SJR, SNIP • Author Level Metrics: Citations Per Paper (CPP), h-index, gindex, i 10 index, e-index, R-index, a-index • Article-Level Metrics: Citation mention, View, Share, Downloads, Bookmark, Impactstory, Blogs, Code etc

Alt + Metrics = Altmetrics

Altmetrics • Viewed - HTML views and PDF downloads • Discussed - journal comments, science blogs, Wikipedia, Twitter, Facebook and other social media • Saved - Mendeley, Cite. ULike and other social bookmarks • Cited - citations in the scholarly literature, tracked by Web of Science, Scopus, Cross. Ref and others • Recommended - for example used by F 1000 Prime

www. altmetrics. org

www. altmetric. com

altmetrics: a manifesto • Released On 26 October 2010. • NO ONE CAN READ EVERYTHING. We rely on filters to make sense of the scholarly literature, but the narrow, traditional filters are being swamped. However, the growth of new, online scholarly tools allows us to make new filters; these altmetrics reflect the broad, rapid impact of scholarship in this burgeoning ecosystem. We call for more tools and research based on altmetrics.

Major Altmetric Tools: Limitations • It works only when DOI is available, and in several databases such as Pub. Med and ar. Xiv. • It works when publishers embed Google Scholar friendly citation metadata. • Twitter mentions are only available for articles published since July 2011.

Major Altmetric tools

Impact. Story is a Web-based application that makes it easy to track the impact of a wide range of research artifacts (such as papers, datasets, slides, research code). The system aggregates impact data from many sources, from Mendeley to Git. Hub to Twitter and more, and displays it in a single, permalinked report.

Reader. Meter is a mashup visualizing author-level and article-level statistics based on the consumption of scientific content by a large population of readers. Readership data is obtained via the Mendeley API. Reports are available both as HTML and in a machinereadable version as JSON and are released under a CC -BY-SA 3. 0 license.

Science. Card is a website that automatically collects metrics (citations, download counts, altmetrics) for a particular researcher. All the researcher has to do is provide a unique author identifier such as Author. Claim or Microsoft Academic Search ID.

The PLo. S Impact Explorer allows you to browse the conversations collected by altmetric. com for papers published by the Public Library of Science (PLo. S).

Paper. Critic offers researchers a way of monitoring all types of feedback about their scientific work, as well as allows everyone to easily review the work of others, in a fully open and transparent environment.

• Crowdo. Meter is a web service that displays tweets linking to scientific articles, and allows users to add semantic information. Crowdo. Meter uses a subset of the Citation Typing Ontology (Ci. TO), an ontology for the characterization of citations, both factually and rhetorically. The results of this crowdsourcing effort are displayed in real-time.

The links below will allow to connect with both altmetrics users and individuals interested in subject: • #altmetrics on Twitter. A great deal of discussion about altmetrics takes place on Twitter through the use of the #altmetrics hashtag. • altmetric. org: This website serves as a hub for much of the activity surrounding altmetrics. Though their lists of Tools and Communities do not seem to have been updated recently, their listing of events is updated consistently. • Mendeley Altmetrics Group: From the group's description: The aim of this group is to discuss new approaches to the assessment of scholarly impact based on new metrics.

Thank You