A Local and Multi-Institutional Study of Open Access Engineering Publishing

The emergence of open access (OA), content that is online and free of charge, is challenging traditional subscription publishing models. It is estimated that worldwide OA journal publishing is currently 12%. This study examines OA journal publishing within engineering to quantify and characterize sources. Using Web of Science, journal articles within the 14 Web of Science engineering categories were identified. The articles were categorized based on OA, subject, author affiliation, and source title. Approximately 4.2% of the engineering journal articles within those 14 categories are OA, but limiting to include at least one U.S. author or coauthor decreased the number of OA articles to 1.2%. Comparisons were made between overall OA journal article publishing at the top 25 U.S engineering universities and the percentage of OA engineering journal articles by researchers at those universities. Despite overall OA publishing at the 25 U.S. engineering universities approaching the overall worldwide OA journal publishing level, OA engineering journal publishing is considerably less. A list of the most frequent OA engineering journal venues for the 25 engineering universities is presented. The paper concludes with a local study involving OA journal publishing within three engineering departments (Chemical, Civil, and Electrical & Computer Engineering) at a large research university. The local study reveals publishing patterns that are not initially intuitive. This study increases our understanding of current OA publishing within engineering and supports scholarly communication initiatives and outreach efforts. Literature Review Suber 1 defines open access (OA) as “...digital, online, free of charge, and free of most copyright and licensing restrictions.” There are two main types of OA: Gold which typically refers to free, immediate access to the publisher’s final version; and Green OA which refers to archiving a preprint or postprint. Most Gold OA journals charge the author a fee (“author pays” model). In addition there is delayed OA, which are often traditional subscription journals that provide free access to articles after an initial delay; and hybrid journals, which are traditional subscription journals that allow authors to pay a fee to make their individual articles OA. There have been a number of surveys that have examined researcher attitudes toward OA. Xia 2 compiled an extensive bibliography of surveys involving OA attitudes and conducted a longitudinal time-series analysis using data from 26 of those studies. Xia concluded that there has been an increase in OA awareness and willingness to publish in OA journals, but reputation and perceived lack of peer review remained concerns. 2 In a survey of engineering faculty at the University of Illinois, Mischo & Schlembach 3 reported that faculty had concerns about the economics of “author pays” model, yet 33% indicated having published in such journals. While there are a number of surveys examining attitudes about OA publishing, there are a dearth of bibliometric studies quantifying and characterizing OA publishing. Laakso & Bjork 4 estimated that approximately 12% of all peer-reviewed journal articles worldwide were Gold OA. There have been some larger bibliometric studies conducted to quantify OA publishing in India, 5 Germany, 6 and the UK, 7 but essentially none in the United States. In a study of 22 “prolific” researchers at the University of Utah, Amos et al. 8 found only 1% of those researchers published in an OA journal. In a larger study of ARL libraries, Hubbard 9 reported that annual Gold OA publishing was between 5% and 9% for the institutions studied. As part of a larger study, Bjork et al. 10 randomly sampled 209 engineering articles from Scopus and found that 4.8% of the articles were published in Gold OA journals. Using 107,052 articles selected randomly from Web of Science across 14 disciplines, Gargouri et al. 7 found only 1.3% of the engineering articles to be Gold OA. The two studies used different data sources and methodologies. Bjork et al. sampled from Scopus, which indexes 2,800 OA journals, and manually checked individual journals (and their articles) to determine the OA percentage. 10,11 Gargouri et al. sampled from Web of Science, which indexes 1,500 OA journals, and used a software bot and algorithm to estimate the OA percentage. 7,12 Baldwin 13 is the only known bibliometric study to quantify Gold OA publishing among engineering researchers in the U.S. Baldwin 13 examined publications from three engineering departments (Chemical, Civil, and Mechanical) at five universities (Georgia Tech, MIT, Ohio State, University of Nebraska, and University of Michigan) and found 3% of the articles freely available at the publisher’s websites. While the articles were freely available, not all were necessarily published in Gold OA journals. Delayed OA, hybrid OA, and selected free issues/articles could have contributed to the 3%. In addition to an overall percentage, Baldwin 13 also provided the number of articles freely available at the publisher’s websites for the three engineering departments studied. The percentages that follow were computed using the data provided in her study: chemical engineering (3.1%), civil engineering (4.4%), and mechanical engineering (1.8%). In contrast to Baldwin, 13 the current study only examines Gold OA and reports on publication venues (i.e., journals) and subject areas. Rationale/Background The purpose of this study is to quantify and characterize Gold OA publishing locally and at 25 U.S. engineering universities. The results of this study inform subject librarians regarding current Gold OA journal venues, levels of educational and financial support needed, and engineering scholarship. More specifically, this study addresses the following research questions for the top U.S. engineering universities and selected engineering departments at Texas A&M University: How much Gold OA publishing is occurring? How much publishing is occurring in Gold OA “engineering journals”? Where are researchers publishing their Gold OA articles? There is a strong research focus at Texas A&M University (TAMU). This is evidenced by the Carnegie Classification of Institutions of Higher Education categorizing TAMU as a “Research University (Very High Research Activity)” and by its membership in the Association of American Universities. TAMU opened in 1876 as a land grant university and was one of the first universities to become a land, sea, and space grant institution. 14 The College of Engineering is comprised of 13 departments, with approximately 350 faculty members, 12,000 students, and 22 programs, and is the largest college on the TAMU College Station campus. 15 A greater understanding of scholarship at TAMU, and other U.S. universities, will lead to improved engagement on scholarly communication issues. In late 2013 Thomson Reuters introduced an “Open Access” facet (i.e., a limiter) to Web of Science that allows for facile determination of Gold OA articles indexed in Web of Science. There are currently over 1,500 Gold OA journals indexed in Web of Science. 12 Thomson Reuters assigns one or more of the 252 Web of Science Categories to the journals indexed. There are 14 main engineering categories (e.g., Engineering, Aerospace; Engineering, Biomedical; etc.), though other categories could be also considered engineering. Of the Gold OA journals indexed in Web of Science, only 58 journals are assigned to one or more of the 14 main engineering categories. Table 1 shows the distribution of OA journals by category, but note that only 11 of 14 categories have associated OA journals. Throughout the remainder of this paper, OA will refer to Gold OA unless otherwise specified. Table 1. Distribution of the OA Engineering Journals by Web of Science Category Methods Engineering Articles Using Thomson Reuters’ Web of Science, the number of engineering journal articles was determined using 14 of the 252 Web of Science Categories (Table 2). The 14 Web of Science Table 2. Engineering Web of Science Categories Engineering, Aerospace Engineering, Biomedical Engineering, Chemical Engineering, Civil Engineering, Electrical & Electronic Engineering, Environmental Engineering, Geological Engineering, Industrial Engineering, Manufacturing Engineering, Marine Engineering, Mechanical Engineering, Multidisciplinary Engineering, Ocean Engineering, Petroleum categories were searched as a single search string using the Advanced Search and the Publication Years limited to 1990-2013. The results were further refined by Document Type to include only Article and Review. The number of engineering articles that are OA was then determined using the “Open Access” facet. To provide context for the level of U.S. journal publishing, the “Countries/Territories” facet was also used to determine the number of engineering articles published (overall and OA) by national affiliation. Top 25 Engineering Universities For comparison, the top 25 engineering universities listed in the 2014 U.S. News & World 0 2 4 6 8 10 12 14 16 Report’s Best Graduate Schools (Engineering) were selected (Table 3). 16 The number of articles published by the 25 universities was determined individually and as a group (i.e., combined). The combined 25 universities were searched as a single search string using the OrganizationEnhanced field in the Basic Search. The Publication Years were limited to 1990-2013 and Document Type to Article and Review using facets. The engineering articles were determined using the Analyze tool to select publications within categories listed in Table 2. The “Open Access” facet was used to determine the number of OA articles, overall and engineering, for the 25 universities combined. The top OA and OA engineering journals in terms of frequency for the combined 25 universities were determined using the Analyze tool in Web of Science and sorted by