The Scientometrics of Nature

Nature is a weekly journal of multidisciplinary sciences with a huge range of followers in the scientific community. The purpose of the study is to analyse the citations of the full research articles published in Nature journal from 2006 to 2015. During this period Nature has published 8335 research articles. In the study the weightage of the research articles published in Nature were measured through immediacy index comparison of the research articles to the overall journal. The authorship pattern of the articles were measured and a cited references study was also conducted for measuring the self citations generated by Nature each year,the references per articles present and identifying core cited journals by Nature and their respective subject areas, which libraries in turn may use for collection development policies. Ranking of contributing countries and contributing organizations were also made by % share of article contribution. To check if any kind of biasness is present in Nature publications, a comparative evaluation of top contributing countries has also been made based on Nature Index. Web of Science core collection was used as the data source and Bibexcel bibliometric tool was used for data analysis purpose.


INTRODUCTION
In today's scientific society Nature is one of the most followed weekly scientific journal of multidisciplinary sciences having a impact factor of 41.45 as per journal citation reports 2014.First published in the year 1869, currently in the year 2015 it had a print circulation of 50,200 with a readership of 365,456 excluding the online readership of 8.7 million per month. 1 The rapid increase in its article submission rate, (almost 200 article submission per week) 2 only suggests its immense popularity and acceptability in the scientific community.Citation analysis of such a journal which is being considered as a world leader of multidisciplinary sciences will help in understanding interactions among scientific disciplines.Basic data were collected for the study with a quantitative approach of no of "citations gained" at article level, which has been further analysed for studying the selfcitations, authorship pattern and identification of core cited journals.Bibexcel 3 software tool was used for conducting the cited references study.Bibexcel tool was developed by Olle Pearson, which is a powerful tool in terms of flexibility for data management and interpretation.Bibexcel can analyse plain text format downloaded data from citation databases like WoS or Scopus, and later those data can be transferred to excel or statistical analysis software or can be used for visualization also through various mapping tools like Pajek, VOS viewer etc.

Review of Literature
Citation analysis of journals is not a new phenomenon, and it has its roots back till 1927 when Gross and Gross 4 has made an analysis of the prestigious Journal of American Chemical Society for identifying the top cited journals in it, thus preparing a core list of journals which they can use for collection development in Pomona College library, in Southern California.But later with the starting of Science citation index more in depth citation analysis of journals were possible which later were used for establishing network of authors, citations across subjects and papers 5,6 and connecting informal scholarly communication process. 7Glanzel and Moed 8 have made a review of different contextual and time cluster approaches used for scientometric measures of journal evaluation.Nebelong-Bonnevie and Frandsen 9 proposed the journal citation identity (i.e.references per different referenced work) and journal citation image as two indicators for journal evaluation.They analyzed Journal of Documentation (JOD) by using the data of Journal of Information Science (JIS) and Journal of the American Society for Information Science and Technology (JASIST) as standard of reference and comparison.Tsay and Shu 10 have made another approach to Journal of Documentation , where they evaluated the references present in the articles of that journal, and made a bridge that connects the journal to different subject areas.Garg, Srivastava and Bebi 11 made a different approach for evaluating journal citations and studied the cited references in the published articles of Journal of intellectual property rights and did a mapping of related subjects based on the references.Other than citation analysis studies conducted for evaluation of journals citations, [12][13][14] there are studies [15][16][17] which are conducted on thesis and dissertations of specific subject areas also to identify their citation patterns.

Objectives of the study
The study was conducted with the following objectives: 1. Analysis of type of items published by Nature.

Methodologies and Properties of Collected Data:
The accumulated data for the study comprises of bibliographical record of published items It is an indicator that shows us the higher visibility and acceptance of the researches published in the journal in a shorter period of time. 18The citation window selected for this indicator is the publication year only, i.e. the immediacy index for a journal for a year is the ratio of the no of citations gained by the journal in a year to the total items published in the journal in that year.Here the immediacy of the Nature journal and immediacy index of the research articles published in the journal were calculated for each year from 2006 to 2015 and their variation was compared (if there is any) From Table 2 it can be seen that articles comprises a large amount of citations of the total citations gained by the journal.From 2006 -2015 articles only generated most of the citations for the journal at an average of 86.24% each year.The Immediacy index of the research articles is on higher side in each year from the overall journal immediacy index.During the study period the average journal immediacy index was 2.75 whereas the same for research articles was 7.49 and their difference was 4.74.Immediacy index of the journal ranged between 2.2-3.09while that of articles ranged between 6.04-8.52 and their difference ranged between 3.74-5.45during the study period.This clearly shows that the research articles influence in overall  The difference between AII-JII has seen an increasing trend, this only shows the growing impact of Nature's research articles.

Objective 3: Authorship Pattern of Articles:
There were all total of 107334 authors distributed in 8335 articles.The total authors in each year have seen a growing trend.The average authorship per paper is also increasing.The collaborative index in each year shows a very high degree of collaboration in the research articles published in Nature.For calculating the degree of collaboration formula given by Subramaniyam 19     research and these citations we can trace by studying the references of an article.These references establish a connection between the citing journal and the cited journal through which a type of relation can be established between the articles published between both the journals.For cited references study bibexcel bibliometric tool was used.Downloaded plain text format article data from web of science data were used for analysis through bibexcel.From the records, cited references were pulled out first, after that from the references cited journals were pulled out for each year.During this process de-duplication was not done, i.e. if a journal has multiple presence in the references of a single article then that journal was given a count for each of its article.Cited references study helps us to identify the type of references present in the

Objective 4: Cited References Study
Citations given by an author to other authors are the motivations taken by any author during conducting a articles, understanding the self-citations a journal makes listing out the top referred journal in a particular journal's publications.
Here in Self-citations generated by Nature in each year: A journal self-citation is that scenario whenever an article published in it refers to an item published in that same journal.These self-citations usually have a great value while calculating the impact factor of the journal.These self-citations can be traced out through the study of references in the articles published in the journal.When the cited references study of the research articles of Nature was conducted we also found out the no of references in the articles that refers to articles published in Nature.
In bibexcel software tool this can be carried out with two options, one removing the duplicates and one including duplicates.For example there is an article in Nature which carries reference of three articles already published in Nature.Now when we will extract the reference data including duplication, then bibexcel will count three self-citations to Nature, and if we do the same removing duplication then bibexcel will count only one self-citation to Nature in such a scenario.Here in the study, the references were pulled out including duplication as because all the self-citations carry equal value for calculating impact factor.One of interesting fact that also been found out is that Nature remain at top of cited journal list in each year's publications (Table 6).
The amount of self citations in each year has increased in Nature journal but the total no of references was also increasing, so while comparing the self citations to the overall references, it hasn't showed a lot of variation (Figure 4).From 2006 to 2015, Nature has generated at an average of 10.33% of self citations each year with 3.12 self citations in each article.Self citations/article ranged from 2.67 to 3.58, highest in the year 2014 and lowest in the year 2006.

Prestige transferred to others
Findings from table 4 suggests that research articles of Nature tends to cite multiple articles from a single journal (average 10.6 articles from a single journal), therefore in the study an attempt was made to year wise analysis of top 5 journals which got most citations from Nature.Bibexcel bibliomeric tool was used for this analysis also.So if we proceed through the philosophy of Pinski and Narin, 20 we can say that these are the journals to which Nature has transferred most of prestige.Table 6 shows that in all the years most of the outgoing citations from Nature were circulated among Science, Proceedings of National Academy of Science (PNAS), Cell, Journal of Biological Chemistry (JBC) and The Astrophysical journal (AstroPJ), respectively where Nature occupied the top position every year in the list.
In the ten year period of study, this list was not altered, except the year 2013, when the sixth position i.e.The Astrophysical Journal was replaced by Acta Crystallo graphica Section D (ACtD).But from next year onwards this position was again retained by The Astrophysical Journal.The total amount of outgoing citations consumed by only these six journals is shown in table 6.From 2006 to 2015, on an average these journals only have consumed 29.23% of outgoing citations from Nature.So we can clearly say that the researches published in these journals have a great influence in the published researches of Nature.

Objective 5: Top most cited journals by Nature from 2006-2015 and their subject area
For listing out the top cited journal by Nature from 2006-2015, it is necessary that the problem of journal title duplication is nullified.We need to calculate the no of citations received from Nature by its cited journals over the study period, therefore we have merged downloaded .txtdata of articles to a single file and analysed it through bibexcel tool.For pulling out the outgoing citations, de-duplication of journal title was not done.There were all total 7992 journals present in the references of 8335 articles with 251476 article references.For being included in the top cited list of journal by Nature the minimum threshold value of received citations by a journal was considered as 1000 during the study period.With this criterion 41 journals featured in the top cited list that shared all total of 130232 references among them (i.e.51.78% of all outgoing references) where the top position was occupied by Nature itself, which is followed by the Science journal.The subject areas of the journals were also collected from the SCImago journal rank website for identifying the subject areas whose research articles gets most citations from Nature.SCImago has its own index of subjects under which it lists different journals.Some journals were listed under two or three subject categories.Those subject categories are shown in Table 7 separated by "|" symbol.Like that it was found that the top 41 journals were representing 11 different subject categories (Figure 5).40% journals were listed under the subject area of "Biochemistry, genetics and molecular biology", while 13% belonged to "Medicine", 12% to "Immunology and microbiology" and 8% each to "Earth and Planetary science" and "Physics and astronomy".If we consider a macro level analysis of subjects, then we can say that the researches published in Nature are little bit inclined to the field of biological and medicinal sciences, as we can list "Biochemistry, genetics     and molecular biology", "Medicine", "Neuroscience", "immunology and Microbiology" under the broader subject category of Biological and medicinal science.

Objective 6: Ranking of countries and Organizations
There were total of 107334 authors (table 3) distributed in 111 countries affiliated to 2369 organizations all over the world in the study period.We can make a ranking of the countries and organizations by the authors affiliation address.The ranking was made by % share of paper contribution from each countries and organizations.

CONCLUSION
The study throws light on some of the basic facts related to the Nature journal.The very high immediacy index of the research articles only means the widespread visibility and quick acceptability of the research articles which actually have great influence in the overall journals impact factor.Nature tends to cite multiple articles of a single journal and except Nature itself its most of the outgoing citations are absorbed by five other journals as shown in Table 6.Though Nature generates a very high amount of self-citation (10.33% of total journal references) but it may only mean the higher value of its own published articles.The average authorship of the research articles has seen a growing trend over the period of the study, with highest average authorship per paper seen in the year 2015 with 19.92 authors per paper and the average being 12.88 in the study period (Table 3).Research articles of Nature cites articles from the field of biological and medical sciences more, which in turn we can say that the published researches are also from this domain only (Figure 5) and Nature has a good strength in these domains.
To check the regional biasness of Nature articles, comparative analysis was carried out using Nature Index for top performing countries in the journal group "Nature and Science" and that in world's other top journals from 2013-2016.Presence of biasness would have resulted in significant difference in the country names of both the group, but the analysis didn't found any significant variation in the country names in both the group except relative change in rankings.

2 .
Analysing the citation impact of research articles of Nature to its overall citations and make a comparison of immediacy index of Nature articles to the overall journal's immediacy index.(i.e.understanding the weight age of Nature articles in overall impact of the journal) 3. To know about the authorship pattern of Nature's research articles.4. Conducting a cited references study in the references of Nature to know about reference pattern and the amount of self citations generated by Nature in each year. 5. Identifying the top citing journals in Nature articles and identifying their subject areas, and thus identifying the subjects whose articles get more cited in Nature.6. Listing out the top 15 countries and top 15 organizations that have made highest no of research article contribution to Nature from 2006 to 2015.
impact of the Nature journal.The highest AII (8.52) and highest value of AII-JII (5.45) was seen in the year 2011 while the lowest AII (6.04) was seen in the year 2007 and lowest value of AII-JII (3.74) was seen in the year 2006.
was used, C=N m / (N m + N s ), Here, C= degree of collaboration, C≈1 means high degree of collaboration N m = No of multi authored paper, N s = no of single authored paper 2 to 10 authored papers have highest share each year, while single authored papers have the least share each year.From 2006 to 2015, 67.4% of articles published in Nature have 2 to 10 authors, and 20.2% papers have authorship between 11 to 20 authors.There were papers in Nature with more than 50 authors also.Papers which have seen authorship between 41 or more is 3.7% of total 8335 papers.From 2006 to 2015, the average authorship was 12.88 authors per article with only 147 single authored articles out of total 8335 articles with a degree of collaboration of 0.98.Highest average authorship was seen in the year 2015 with 19.92 authors per article and lowest was in the year 2006 with 8.03 authors per article.In the year 2014 only two single author paper was published out of 828 papers.The degree of collaboration was always above 0.95 in the study period.This only shows that the researches published in Nature are of highly collaborative nature.

Figure 3 :
Figure 3: Authorship density of Nature research articles

Figure 4 :
Figure 4: Comparison of self references to overall references data USA and England are the top contributing countries with approx 70% and around 19% share of papers out of 8335 research articles.Japan and China are the only Asian countries to be featured in the in the top 15 countries list.Again in top 15 organizations list University of California gained the top position with 16.11 % share of papers and Harvard University got second position with 12.57% of papers.In top 15 organizations list there is 10 organizations from USA, 3 from UK, one from Germany and one from France is featured.The massive amount of research funding of these countries may be one of the reasons of their performance in research which is reflected in their massive article contribution to Nature in the last 10 years.USA and China alone invested almost 26.4% & 19.8% of global research funding in the year 2015 as per the report 21 of Industrial Research Institute.While the feature of top developed countries in the top contributing countries to Nature makes sense, but it also raises question about the non-feature of other developing countries from Asia or Europe in the list when the global research funding share of Asia (41.2%, excluding China) and Europe (21.3%)combined in the year 2015 were 62.5% according to the same report.Some controversial conclusions like presence of biasness in the published articles of Nature towards developed countries can also be inferred from findings presented in Figure.6 & 7.So therefore to have a more clear picture of the status of different countries in world's other the top performing publications the study was extended beyond the network of Web of Science to Nature Index22 , which is a combined index of articles published in worlds top 68 journals and makes a ranking of countries and institutions based on article contribution.Nature Index, gives the prevailage of creating yearly ranking tables (current year yearly table is based on the previous year performance from January-December data) on regional and institutions level which further can be ranked on the basis of two parameters, i.e. subject disciplines and in the journal group of "Nature &Science".For the current study, two global ranking lists of top 20 countries were pulled out for each year from 2013-2016 (currently yearly tables are available from 2013 only), one (i.e.

Figure 5 :Figure 6 :
Figure 5: Subject areas of top cited journals by Nature

Figure 7 :
Figure 7: Organizations ranking in terms of article contribution % of 8335

publishes under News section. Reviews, Correction, Biography, Book review and Editorial are other regular columns of Nature. From 2006-2015,
89 volumes data from volume 439 to 528.The source of data collection was Web of science core collection database.Basic search was made in the web of science core collection database typing "Nature" in the search box selecting "publication type"parameter within the time span of each year from 2006 to 2015.Thus Nature's publication for each year was obtained.Then the search result was refined to document type "Article" and then the full record of those refined result were downloaded in plain text format for evaluation through bibexcel software.The citation data of the refined result were also imported to MS Excel format for analysis.Within the period Nature has published all total of 26211 items that contains 8335 items under the tag "Article".Its different published items are shown in Table1.From 2006 to 2010 Nature used to combine its one year publica-Nature has published all total of 26210 items, highest being published in the year 2006 (2733 items) and lowest being in the year 2009 (2543 items).After going through the items published on record count (Table1), it is seen that Nature's publication under the editorial column has by Nature journal from 2006 to 2015 that includes all total of Nature's tion in 6 volumes (i.e. one volume containing two months publication) but from 2010 onwards it used to combine each year's publication in 12 volumes.Data analysis, Findings and Discussion Objective 1: Publication profile of Nature from 2006-2015 Nature publishes different items under its eight different sections as maintained in Web of Science database.Full research articles are published under the article section of Nature, which is the main concentration of this current article (the author will address these items as research articles from here on in this article).Short communications about ongoing researches are published under letters section.Scientific journalism is also one of the main scopes of Nature which it

Table 4 : Cited references of Nature articles
Table 4the column "Total journal references" means the references in the Nature articles that gives a backward citation to a journal article and the column "Total Journal titles in references" refers the journals to which those articles belong.From 2006-2015, there was all total of 251476 references present in 8335 articles, with average reference/article 27.46 and average journal title referred/article 2.85.Remember that here in column C of table4, the problem of journal title repetition was not nullified, i.e. there is great probability that the same journals which are cited in 2006, are also get cited in the article of 2007 and the same continues to other years also.Therefore the no 23785 showing total journals referred from 2006-2015 is actually not right, it's just a cumulative count of each year's record and it contains duplicate counts also and actual unique no of journal count is less than this.But here as we are considering each year's record separately so the data measurement in other columns and rows are accurate.Average ref/article ranged between 23.82-35.46whereas average journal title referred/article ranged between 2.47-3.3 in the study period.Existence of a significant difference between "average ref/article" and "average journal title referred/article"only suggests that the research articles in Nature usually tends to cite multiple articles from a single journal.Therefore to measure the average share of cited items coming from a single journal over the years, we also calculated "average referred article/ cited journal", which yields that in research articles of Nature, at an average 10.6 items from a single journalis being cited by authors in the study period.

Table 8
.1) based on article contribution in the journal group of Nature and Science and other (i.e.table 8.2) based on overhaul article contribution in all Nature index journals across all subject disciplines.Comparing both the table we can observe that there is not much of difference in the country names except slight changes in their rankings.But in table 8.2 entry of two new countries, India and Chile is observed.India maintained a very good ranking spot over these four year period, while Chile appeared only once in the 2016

Table 8 .2: Global Ranking of top 20 countries based on Article contribution to top 68 Nature Index journals 27,28,29,30
Country like Taiwan has only appeared once in table 8.1, but has appeared in all the four years in table 8.2.The interesting co appearance of similar countries in both the list may only suggest that what world's other top journals are publishing is also published by Nature & Science journal, so the question of biasness is nullified.The reason for non-appearance of India in table 8.1 besides having a good ranking spot in table 8.2 may only suggest that India maintained a good performance rate in other top journals but not in Nature & Science.