Citations, Visits, Downloads, Purchases in OA Environment


Download presentation

Introduction and aim: In OA publishing environment as well as in general digital publishing visits(views) and downloads (readers) are of increasing interest for institutions, researchers and publishers (1,2). In the scientific community, citations both of printed and digital publications are considered as the most influential and reliable information on scientific efficiency which is of interest to researchers, publishers and to their working environments (3). On the other hand, numbers of visits (views) and downloads are normally not considered seriously in the evaluation of scientific impact of publications and their authors, as yet. Nevertheless, visits and downloads are often discussed and compared by research colleagues in addition to citing. The number of visits and downloads may help researchers to steer their researches towards topics of increasing interest in the scientific categories and fields (4). They also may be useful for assessments of importance and influence of scientific categories and fields. The survey of literature in scientometrics indicates increasing interest in these issues. Some OA publishers and institutions used to periodically publish the ranking by visits (views) and downloads of digital journals and papers to meet the interests of the wider science and engineering community. In commercial publishing, the number of purchased publications is for sure very important measure for success of publishers and merit of authors, but it is commonly not considered with high interest by the scientific community. However, the scientific community lately expresses increasing frustrations with respect to high prices of publications, access to exclusive bibliographic data bases, relevant information as well as against extra profits of scientific publishing industry using contributions of authors and reviewers and supporting institutions free of charge. In this review, relations among citations on one hand and visits and downloads on the other hand, as well as the relation between visits and downloads will be investigated on examples of open access indexed journals in engineering category having appropriate comparable data on citations, visits and downloads. Relations to purchased publications are not considered since that information are normally not publicly available. The aim of this research is to find out the links and relations, if any, among citing and viewing or downloading of papers published in open access modes in scientific bibliographic data bases.The interest of this review is focused on Croatian journals related to engineering. The research will consider the visits and downloads of papers published by two Croatian engineering journals compared to the most cited papers as well as the most visited and downloaded papers and their citations in period of last five years. In conclusion, the research discusses the importance of views and downloads and their role in the assessment of scientific impact of papers andauthor’s efficiency. Some of the pros and cons of these measures are considered.

Engineering vs. scientific publishing

Publishing in engineering has to meet some specific conditions with respect to scientific publishing. Engineering involves mathematical methods and empirical evidence to the innovation, design, construction, operation and maintenance of structures, machines, materials, devices, systems, processes, and organizations. The discipline of engineering encompasses a broad range of more specialized fields, each with a more specific emphasis on particular areas of applied mathematics, applied science, and types of application (5). In contrast to pure scientists, engineers and technicians are primarily educated and prepared for creative application of scientific principles to design or develop structures, machines,apparatus, or manufacturing processes, or works utilizing them singly or in combination; or to construct or operate the same with full cognizance of their design; or to forecast their behaviour under specific operating conditions; all as respects an intended function, economics of operation and safety to life and property (6,7). Engineers and technicians have different research, publishing duties and interests than pure scientists. Engineers and technicians insist on creative application of science in technical domain not necessarily on theory and science itself. They may perform different functions, such as practising engineers, research engineers or both. Practicing engineers work mostly in the production and fabrication. With respect to this, practicing engineers are prevailingly oriented to use and publish professional and review papers. Research engineers often work at universities and research institutes or laboratories.Hence, the research engineers are more oriented towards using and publishing original scientific papers, review papers as well as preliminary communications. Therefore, the publishing of professional papers is less attractive or not attractive at all to the scientific community although is more important to engineering profession. Having these differences in mind, it becomes clear how the publishing in engineering is not necessarily same as the pure scientific publishing.

International categorization of scientific and engineering research presentations

International categorization of scientific and professional research presentations is recommended by Guide for preparation of scientific papers for publications (8): original scientific paper (OSP) reports on unpublished results of original research. It should be written in such way that a qualified researcher on the basis of given information is able to: (i) repeat the experiment and obtain presented results with the same accuracy or within experimental error boundaries, or (ii) repeat the author’s observations, calculations or theoretical reasoning,and reach similar conclusions. Preliminary communication (PC) contains new scientific research results which require urgent publication. It does not need necessarily to provide the basis for the repetition of the experiments. Review paper (RP) should be an original, concise,and critical presentation of one subject area or a part of it, in which the information already published is assembled, analysed and discussed. Professional paper (PP) presents high achievements in engineering practice.

Evaluating engineering and scientific research results through publications

The interests of the scientific and engineering community and the importance of publications for research are often evaluated by following indicators of their impacts: number of citations,number of views (visits), number of downloads (readers), and number of purchased papers.Besides, there are some others more sophisticated and special purpose measures, e.g. WoSClarivate and SCIMAGO used by scientometrics for different purposes. It is appropriate to investigate number of views and downloads on primary journal repositories. However, some other institutions, such as Mendeley, ResearchGate and Google Scholar also provide views(readings) and citations reports.

Comparing citations, visits and downloads

The most influential and widely adopted measure for bibliometric evaluation of published results both in sciences and engineering is the number of citations. Some common properties of numbers of citations: long tradition in evaluation, the controllability through bibliographic data bases, the applicability in general and in OA environment, huge and expensive ICT infrastructure, conditional accessibility. In OA environment digital publishing visits (views) and downloads are important indicators of interests of the research communities for research activities. Visits are mostly curiosity driven and may stop on reading the key words and the abstract. With this respect, views only partially express the importance of publications.Downloads are motivated by genuine interests and needs and normally follow the views and more appropriately define the importance of publications. Some properties of visits and downloads: the applicability in digital publishing environment, particularly apt for OA publishing, lack of control, e.g. multiple visit and downloads in order to unethically increase their numbers, traceability of viewers and downloaders is difficult and not always applicable,easy, immediate and automatic registration. Number of purchased papers are out of focus of researchers and normally cannot be used for evaluation of scientific efficiency in contrast for example to literature (bestsellers).

Results and Discussion: This paper investigates the number of visits and downloads of papers deposited on the central portal of Croatian scientific journals HRČAK compared to the most cited papers according to WoS Citation report following the Clarivate analytics, as well as the most visited and downloaded papers and their citations in a period of 5 years. HRČAK provide on annual basis ranking of Croatian journals by full text visits and total visits. Two Croatian engineering journals published by the University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture (FAMENA) are considered: Brodogradnja/Shipbuilding, in HRČAK since 2005, Transactions of FAMENA, in HRČAK since 2012.

Journal Brodogradnja/Shipbuilding

Brodogradnja/Shipbuilding is an international scientific peer reviewed open-access on-line digital journal launched in 1950 published quarterly. It is indexed by Web of Sciences (WoS) SCIExpanded since 2008 in JCR category Engineering Marine and by SCOPUS since 1985 in categories Marine Engineering and Mechanical Engineering. Since 2005 to 4th September 2018 there were all together 409643 visits (views). The following observations may be outlined for journal Brodogradnja: all 135 papers (100 %) have downloads (max 6781, min 118) and views (max 7232, min 213); total number of downloads is equal to 79400 and number of views is equal to 112665. There are 72 (53 %) cited papers out of all 135 papers, cited at least once with total of 183 citations, 56821 (71 %) downloads and 76862 (68 %) views. There are 63 (47%) non-cited papers (with 0 citations) out of all 135 papers with 22579 (29 %) downloads and 35803 (32 %) views. There are 25 (19 %) highly cited papers having 120 citations (65 %) greater than 2 with 22672 (29 %) downloads and 30444 (27 %) views. There are 110 (81 %) low cited papers having 63 citations (35 %) less than 3 with 56728 (71 %) downloads and 82221 (72 %)views. 110 low cited papers have 56728-22672=34056 (43 %) more downloads and 82221-30444=51777 (46 %) more views than 25 highly cited papers. 25 papers (19 %) with the highest number 41335 of downloads (52 %) and 48965 (43 %) views have 72 citations (39 %). Average number of downloads and views per citation for all 135 papers is equal to 433 and 515 respectively. Average number of downloads and views per paper for all 135 papers is equal to 588 and 834 respectively. Average number of downloads and views per 25 papers cited at least 3 times is equal to 906 and 1218 respectively. Average number of downloads and views per 47 papers cited less than 3 times is equal to 726 and 988 respectively. Average number of downloads and views per 63 papers without citations is equal to 358 and 568 respectively.Total number of views is 112665/79400=1.42 times greater than total number of downloads.By paper category in all 135 (100 %) publications there are 85 (63 %) OSP, 16 (12 %) RP, 9 (7%) PC and 25 (18 %) PP. There is about 70 % OSP+PC and 30 % PP+RP. By paper category in all 79400 (100 %) downloads there are 42497 (54 %) OSP, 12219 (15 %) RP, 4580 (6 %) PC and 20104 (25 %) PP. There is about 60 % OSP+PC and 40 % PP+RP. By paper category in 25 (19 %)publications with high citings there are 11 (44 %) of OSP, 6 (24 %) of RP, 1 (4 %) PC and 7 (28%) of PP. There is about 50 % of OSP+PC and of PP+RP.

Transactions of FAMENA

Transactions of FAMENA is an international scientific peer reviewed open-access online journal with digital and printed version published quarterly. It is indexed by Web of Sciences (WoS) SCIExpanded in JCR categories Engineering Mechanical and Materials Science Multidisciplinary since 2007 and by SCOPUS in category Engineering Mechanics of Materials since 2002. Since 2012 to 4th September 2018 there were 196681 visits (views). The following observations maybe outlined for journal Transactions of FAMENA: all 168 papers (100 %) have downloads (max 9259, min 88) and views (max 9429, min 166); total number of downloads is equal to 109169 and number of views is equal to 135242. There are 84 (50 %) cited papers out of all 168 papers,cited at least once with total of 216 citations, 60623 (55 %) downloads and 74592 (55 %) views.There are 84 (50 %) non-cited papers (with 0 citations) out of all 168 papers with 60650 (45 %)views and 48546 (45 %) downloads. There are 28 (16 %) highly cited papers having 136 citations (63 %) greater than 2 with 28389 (26 %) downloads and 33591 (27 %) views. There are 140 (83 %) low cited papers having 80 citations (59 %) less than 3 with 80780 (75 %)downloads and 101651 (75 %) views. 140 low cited papers have 80780-28389=52391 (48 %)more downloads and 101651-33591=68060 (50 %) more views than 28 highly cited papers. 28 papers (16 %) with the highest number 57619 of downloads (53 %) and 62942 (47 %) views have 53 citations (25 %). Average number of downloads and views per citation for all 168 papers is equal to 505 and 624 respectively. Average number of downloads and views per paper for all 168 papers is equal to 650 and 806 respectively. Average number of downloads and views per 28 papers cited at least 3 times is equal to 747 and 883 respectively. Average number of downloads and views per 46 papers cited less than 3 times is equal to 700 and 891 respectively. Average number of downloads and views per 84 papers without citations is equal to 570 and 722 respectively. Total number of views is 135242/109169=1.24 times greater than total number of downloads. By paper category in all 168 (100 %) publications there are 144 (85%) OSP, and 24 (15 %) PP. By paper category in all 109169 (100 %) downloads there are 96551(88 %) OSP, and 12618 (12 %) PP.

Conclusions: The following assertions are in favour of views and downloads: there are no papers without views and downloads but there are papers without citations, numbers of downloads and views highly exceed the numbers of citation, and non-cited papers may have significant number of views and downloads. Some observations indicate the relations among views and downloads against citations: both 2nd order polynomial trend lines of views and downloads are increasing with number of citations; highly cited papers have higher average number of views and downloads than low or non-cited papers, number of downloads are slightly lower than number of views and do not differ too much; among low cited papers occurout standingly viewed and downloaded papers. These papers are probably of interest for practising and research engineers in other professional and research areas; among low viewed and downloaded papers occur outstandingly cited papers. These papers are probably of interest for scientist and research engineers in other scientific areas and categories; the two elaborated journals show highly similar patterns in relations of views, downloads and citations.In general, it may be noticed that: (i) views and downloads are important indicators of research efficiency; (ii) practicing engineers may find important information for their work in professional (PP) and review papers (RP); (iii) on the other hand, research engineers are oriented towards original scientific papers (OSP) and sometimes to preliminary communications (PC); (iv) the categorization of papers and papers is almost abandoned in most of the indexed journals both in sciences and engineering. This is appropriate to sciences but not entirely to engineering; (v) it is appropriate to maintain the international categorization in engineering with respect to importance of professional papers (PP) for the engineering community; (vi) views and downloads are easily, faster and more frequently accessed in OAdigital publishing (each published paper is viewed and downloaded) but not each published paper is cited; and (vii) views and downloads are already deeply involved in OA digital publishing and will be even more in the future.