Changes in the absolute numbers and proportions of open access articles from 2000 to 2021 based on the Web of Science Core Collection: a bibliometric study

Article information

Sci Ed. 2023;10(1):45-56
Publication date (electronic) : 2023 February 16
doi : https://doi.org/10.6087/kcse.296
1Clinical Research Institute, Incheon Sejong Hospital, Incheon, Korea
2Seoul National University, Seoul, Korea
Correspondence to Jeong-Wook Seo jwseo@snu.ac.kr
Received 2023 January 15; Accepted 2023 January 27.

Abstract

Purpose

The ultimate goal of current open access (OA) initiatives is for library services to use OA resources. This study aimed to assess the infrastructure for OA scholarly information services by tabulating the number and proportion of OA articles in a literature database.

Methods

We measured the absolute numbers and proportions of OA articles at different time points across various disciplines based on the Web of Science (WoS) database.

Results

The number (proportion) of available OA articles between 2000 and 2021 in the WoS database was 12 million (32.4%). The number (proportion) of indexed OA articles in 1 year was 0.15 million (14.6%) in 2000 and 1.5 million (48.0%) in 2021. The proportion of OA by subject categories in the cumulative data was the highest in the multidisciplinary category (2000–2021, 79%; 2021, 89%), high in natural sciences (2000–2021, 21%–46%; 2021, 41%–62%) and health and medicine (2000–2021, 37%–40%; 2021, 52%–60%), and low in social sciences and others (2000–2021, 23%–32%; 2021, 36%–44%), engineering (2000–2021, 17%–33%; 2021, 31%–39%) and humanities and arts (2000–2021, 11%–22%; 2021, 28%–38%).

Conclusion

Our study confirmed that increasingly many OA research papers have been published in the last 20 years, and the recent data show considerable promise for better services in the future. The proportions of OA articles differed among scholarly disciplines, and designing library services necessitates several considerations with regard to the customers’ demands, available OA resources, and strategic approaches to encourage the use of scholarly OA articles.

Introduction

Background

The current status of open access (OA) initiatives has been criticized by a number of stakeholders, one of whom has pointed out that “the current journal market is failing to operate optimally—particularly in relation to journal access and the cost of Gold OA” [1]. Slow progression, increasing costs, and resistance from researchers and publishers are issues related to OA compliance [2,3]. These negative views on OA are based on its production side, where the increased production of OA scholarly articles has been counteracted by the simultaneous increase in subscription-based publications. The increase in scholarly publications, including both subscription-based and OA journals, has resulted in growing subscription costs for libraries and article processing costs for authors aiming to publish OA articles [4].

OA for scholarly information has been a key agenda for knowledge production and exchange since the 2002 Budapest initiative [5]. The international community has made efforts to spread the OA campaign so that anyone can freely use research results without any economic, legal, and technical barriers [68]. On November 23, 2021, the United Nations Educational, Scientific and Cultural Organization (UNESCO) Recommendation on Open Science [9] was adopted by 193 member countries, including South Korea. The US governmental policy has been enhanced by removing the current 12-month embargo period on making federally funded research publications publicly accessible earlier [10].

To present a professional advisory opinion on establishing a national OA policy in Korea, the authors recently published a report analyzing the current OA status of scholarly publishing and its general principles, focusing on future practical tasks and the roles of various officials by synthesizing discussions about achieving OA [11]. As an extension of our report, this research will demonstrate some practical ways to implement OA services and formulate the further development of the government’s current OA policies [12].

This study investigated aspects of public library services for scholarly information, such as why and for whom these services are necessary and what their requirements are. More specifically, this study focused on the amount of available OA articles, which is the starting point of public services providing scholarly OA information. The content quality, subject domains, document types, and users’ preferences have to be considered when assessing the quantity of OA resources. Moreover, different types of OA involve different routes of access, such as gold OA, hybrid gold OA, and access through institutional repositories of public platforms for OA articles, which must be incorporated into the public database search.

Objectives

This study aimed to identify changes in the number and proportion of OA articles from 2000 to 2021 based on the Web of Science (WoS) Core Collection [13]. Specifically, first, changes were analyzed according to document type. Second, changes in the six major categories of research were investigated; and third, annual trends in OA and non-OA documents were traced.

Methods

Ethics statement

Neither approval by the Institutional Review Board nor obtainment of informed consent was required since this was a literature-based study.

Study design

This was a descriptive study based on a bibliometric analysis of the literature database.

Outcomes

The analysis involved querying the database for the number of documents available as OA at the production level. The absolute number and proportion of OA articles were used as basic parameters. The time of document production was categorized into two data sets: the cumulative data from 2000 to 2021 and 1-year data from 2021. Furthermore, the document type selection and subject domains for analysis were considered as parameters. Different types of OA have different access routes, and production and availability were studied as basic infrastructure for designing search and service systems.

Data sources/measurement

Database for the collection of scholarly information

We used WoS [13] to collect data on scholarly publications. The documents indexed in this database are considered to have met the selection criteria and been certified as having a representative level of quality. The classification system of the academic disciplines and document types used in the database was applied; no exclusions or additions were made.

Assessment of the types of documents

We checked the number of documents available based on the 43 types of documents in the WoS database. The four main types—articles, proceedings papers, review articles, and letters—were analyzed. We found that a reasonable set of documents consisted of the aforementioned four main types. Thereafter, we sorted these four types by research discipline. The other 39 types were analyzed in a similar way.

Discipline-based quantitation of OA documents

The classification system of academic domains was adopted from that of WoS. A total of 254 disciplines were grouped into 25 subcategories, which were further grouped into six categories: (1) multidisciplinary; (2) health and medicine; (3) natural sciences, including environmental science, agriculture, and geoscience; (4) social sciences and others, including education, law, economics, and management; (5) engineering; and (6) humanities and arts. The numbers and proportions of OA articles were displayed according to the long-term trends of 22 years (2000–2021) and recent statistics of 2021. An additional analysis of the OA status of multidisciplinary science journals was conducted.

Annual trends in OA and non-OA documents

Detailed data on the trends on a yearly basis were shown for selective representative disciplines. The yearly trends were analyzed and discussed.

Bias

There was no bias in selecting the target articles.

Study size

No sample size estimation was required since this study included all target journals in two databases.

Statistical methods

Descriptive statistics were used for the data analysis.

Results

Assessment of the types of documents and OA

Articles were the most common document type. In total, 10.6 million articles were available as OA, corresponding to 35.6% of all produced research articles between 2000 and 2021 (Table 1).

Document types and the numbers of total and OA documents between 2000 and 2021 and in 2021

The proportion of OA among articles in 2021 was higher, at 49.5%. Proceedings papers, which are often produced in the engineering discipline, were the second most common document type. Slightly fewer than 1 million proceedings papers were OA, and they accounted for 13.9% of all produced papers of this type; however, this percentage increased to 24.9% in 2021.

We selected four types of documents, articles, proceedings papers, review articles, and letters, as the key sets for our data retrieval. These document types encompassed 90.1% of all OA documents between 2000 and 2021 and 91.5% of all OA documents in 2021. OA documents of these four types accounted for 32.4% of all publications between 2000 and 2021 and 48.0% in 2021.

The annual trends in OA and non-OA documents among the four major document types (Fig. 1) and among all types of documents (Fig. S1) are shown. The numbers (proportions) of 1-year production of OA documents in 2000 and 2021 were 148,642 (14.6%) and 1,521,946 (48.0%), respectively. The total number of both non-OA and OA documents increased from 2000 to 2021; however, the increment rate was higher for OA documents (×10.2) than for the total number (×3.1) and non-OA documents (×1.9).

Fig. 1.

Four major document types (articles, proceedings papers, review articles, and letters) and the numbers of open access (OA) and non-OA documents between 2000 and 2021.

Discipline-based quantitation of OA documents

The numbers and proportions of OA articles among the total publications are shown for six categories, 25 subcategories, and 254 disciplines. The numbers and proportions of OA are displayed in the form of long-term trends between 2000 and 2021 and the recent statistics from 2021. The data and interpretation of the data are shown for six categories, and a total list of six categories and 25 subcategories are shown in Table 2.

List of six categories and 25 disciplinary subcategories and the numbers of OA and total documents in 2000–2021 and 2021

Multidisciplinary sciences: the numbers and proportions of OA documents

This category corresponds to a single discipline: multidisciplinary sciences. The proportions of OA articles were 79% between 2000 and 2021 and 88.6% in 2021 (Table 2). The annual trends in OA and non-OA articles in the multidisciplinary sciences discipline showed a rapid increase in OA articles from 2008 to 2011 (Fig. 2). To explain this increase, data regarding the top 20 journals (based on the number of published articles) in this discipline are displayed. These include the numbers of total and OA documents and their proportions between 2000 and 2021 and 2021, along with the year that the journal was first indexed in the database (Table S1). Among the top 20 journals, nine journal titles are recently founded OA journals. Four journals—PLoS One, Scientific Reports, Proceedings of the National Academy of Sciences of the United States of America, and Nature Communications—published 74.4% (2000–2021) and 63.4% (2021) of all published OA articles in this discipline. A total of 686,485 documents were available as OA articles, and they constituted useful OA resources for every subject domain. The proportion of OA documents in Nature in 2021 was high (62.3% among all types of documents). The proportion of OA documents limited to the “article” type was higher (Fig. S2).

Fig. 2.

The multidisciplinary sciences discipline and the numbers of open access (OA) and non-OA documents between 2000 and 2021.

Health and medicine: the numbers and proportions of OA documents in four discipline subcategories

This category comprises many different health and medicine disciplines, classified into four subcategories (Table S2). The proportions of OA articles were 37% to 40% between 2000 and 2021 and 52% to 60% in 2021 (Table 2). Detailed data on four subcategories are shown in Tables S3S6. The annual trends in OA and non-OA articles in the discipline of infectious diseases showed a rapid increase in OA documents in the most recent 10 years, whereas the number of fee-based articles was stationary (Fig. 3). The recent increase in OA articles is seen in almost every discipline, including oncology (Fig. S3).

Fig. 3.

The discipline of infectious diseases and the numbers of open access (OA) and non-OA documents in 2000–2021.

Natural sciences: the number and proportion of OA documents in seven subcategories

Seven subcategories in the natural sciences category showed relatively high numbers of total and OA articles (Table S7). The most recent data from 2021 showed proportions of 41% to 62%, and the long-term trends between 2020 and 2021 presented proportions ranging from 21% to 46% (Table 2). Two subcategories, biosciences and mathematics, showed high proportions (44%–46%) of OA. Four subcategories—physics, environmental, geoscience, and agriculture/fishery—had OA proportions of 29% to 33%, and the chemistry subcategory showed the lowest proportion (21%). Details of the subcategories are shown in Tables S8S14.

Particle physics and astronomy/astrophysics are known as fields in which OA research predominates, and the proportions of OA articles in 2021 were 88% and 83%, respectively (Table S8). The annual trends in astronomy/astrophysics showed a rapid shift to OA publishing from 2005–2006, and OA publications have thereafter predominated, accounting for 80% or more in recent years (Fig. 4). The annual trends in applied physics show a definitive transformation of fee-based publishing to OA publishing (Fig. S4)

Fig. 4.

The discipline of astronomy/astrophysics and the numbers of open access (OA) and non-OA documents between 2000 and 2021.

Data on 23 disciplines within the natural sciences showed that the subcategory of biosciences had the highest proportions of OA among the seven subcategories at 45.6% between 2000 and 2021 and 62.3% in 2021 (Table S9). Two disciplines (microbiology and virology) within the biosciences subcategory of the natural sciences had high proportions of OA documents, exceeding 60%. The annual trends in the numbers of OA and non-OA documents in the microbiology subcategory (Fig. S5) showed a predominance of OA articles over non-OA articles, and the pattern was similar to that of infectious diseases (Fig. 3). The discipline of biochemistry/molecular biology also showed an increase in OA articles, although the baseline of fee-based documents remained constant (Fig. S6).

The subcategory of chemistry showed a relatively low, but increasing, OA proportion (Table S10). The subcategory of mathematics showed a high OA proportion, similar to that of biosciences (Table S11). The subcategory of environmental sciences also showed a high proportion, but considering the common demands related to environmental issues, the figures were lower than expected (Table S12). However, the analysis of annual trends showed a recent rise in OA articles in the environmental sciences discipline (Fig. S7). The subcategory of geoscience had 30% of resources available as OA, corresponding to the middle of the natural sciences category (Table S13). The agriculture, fishery, and forestry subcategory showed a moderate penetration of OA (Table S14). The annual trends in dairy animal science and food science/technology showed a recent rise in OA articles (Figs. S8, S9). A few subcategories within the same category showed lower proportions; however, the total numbers of articles were small in these subcategories, suggesting selection bias.

Social sciences and others, including education, law, economics, and management: the numbers and proportions of OA documents in four subcategories

This category comprises four subcategories: social sciences; psychology; education; and law, economics, and management (Table S15). These subcategories have the common features of basic and applied social sciences. The proportions of OA articles were 23% to 32% between 2000 and 2021 and 36% to 44% in 2021 (Table 2). Detailed results for the four subcategories are shown in Tables S16S19. Eight disciplines in the subcategory of law, economics, and management showed similar OA proportions and increasing trends over time (Table S19). The annual trends in OA and non-OA articles in the economics discipline are shown in Fig. 5. The annual trends in OA and non-OA articles in five disciplines—education/educational research, business finance, management, political science, and law—are shown in Figs. S10S14.

Fig. 5.

The discipline of economics and the numbers of open access (OA) and non-OA documents between 2000 and 2021.

Engineering: the numbers and proportions of OA documents in five subcategories

This category comprises several different domains of engineering, classified into five subcategories (Table S20). The proportions of OA articles were 17% to 33% between 2000 and 2021 and 31% to 39% in 2021 (Table 2). The details of the five subcategories are shown in Tables S21S25. At least 16 disciplines in the subcategory of major engineering showed similar figures of OA proportions and increasing trends over time (Table S24). The annual trends in OA and non-OA articles in the discipline of electrical/electronic engineering are shown in Fig. 6. The OA trends in nanotechnology/nanoscience are presented in Fig. S15.

Fig. 6.

The discipline of electrical/electronic engineering and the numbers of open access (OA) and non-OA documents between 2000 and 2021.

Humanities and arts: the numbers and proportions of OA documents in four subcategories

The category of humanities and arts consists of an arbitrary list of disciplines with low OA penetration (Table S26). The most recent data from 2021 showed 29% to 38% penetration, and the long-term trends between 2020 and 2021 ranged from 12% to 22% (Table 2). Detailed data are shown in Tables S27S30.

Among the 10 disciplines in the humanities and arts, the lowest OA was found for literature, with proportions of 11.6% between 2000 and 2021 and 28.8% in 2021 (Table S29). Three subcategories—general humanities, history, and arts—showed proportions of 19% to 22% between 2000 and 2021 and 34% to 38% in 2021. The annual trends in the literature disciplines in a 22-year period are shown in Fig. 7. The number of non-OA articles in literature remained stationary, while that of OA articles has been increasing gradually. The annual trends in linguistics have shown increases in both OA and fee-based articles (Fig. S16).

Fig. 7.

The discipline of literature and the number of open access (OA) and non-OA documents between 2000 and 2021.

Annual trends in OA and non-OA documents

Four patterns are recognized in the annual trends of OA and non-OA documents.

Transformative pattern

A rapid increase in OA documents coupled with a rapid decrease in non-OA documents means that fee-based journals have been transformed into OA journals or authors have chosen OA publishing. Examples of this pattern are furnished by astronomy/astrophysics (Fig. 4), electrical/electronic engineering (Fig. 6), applied physics, and education/educational research (Figs. S4, S10)

Rapid increase in OA and a plateau in non-OA documents

A rapid increase in OA publications with a stationary pattern in non-OA documents was the most common pattern in the categories of health and medicine and natural sciences. The expansion of these academic domains is evident, and most new articles are published in newly established OA journals. Multidisciplinary sciences, infectious diseases, and microbiology provide examples of this pattern (Figs. 2, 3, S5). The amount of non-OA documents published in traditional journals remained constant in those disciplines. Some disciplines have strong fee-based journals with a high number of publications, so the relative proportion of non-OA documents remains high. Oncology, biochemistry, molecular biology, and agriculture–dairy animal sciences are examples (Figs. S3, S6, S8).

Increases in both OA and non-OA documents

Increases in both OA and non-OA documents were observed in domains where the amount of publications has recently increased. Examples in the social sciences category include the subcategories of economics, business finance, and law, while some examples in the natural sciences are the subcategories of environmental science and food science/technology. Additional examples of this trend include the subcategory of nanoscience/nanotechnology in the engineering category and the subcategory of linguistics in the humanities.

Other patterns

The fourth pattern involved a minimal increase in OA documents, with non-OA documents remaining the predominant type of publishing. Examples are literature (Fig. 7) and management (Fig. S12). Other nonspecific patterns occurred, which were probably related to the small number of published documents.

Discussion

Key results

The growing trends in the numbers and proportions of OA documents during the study’s approximately 20-year period must be recognized. The number of OA articles available on WoS between 2000 and 2021 was 12 million, amounting to 32.4% of all articles. The 1-year data from 2021 was 1.5 million, amounting to 48%. The proportion of OA documents was the highest in the multidisciplinary, natural sciences, and health and medicine categories, in which OA documents comprised 50% or more of the total documents. The categories of social sciences and others, engineering, and humanities and arts had proportions of around 30% to 40%, but these proportions are increasing.

Interpretation

The availability of OA resources can be expressed as the percentage of all documents. Data from previous research have shown different figures depending on the types of resources and years of publication. The proportion stood at 45% among articles published in 2015 and 28% in the cumulative data at the same time [14]. Our equivalent data would be 49.5% in 2021 and 35.6% between 2000 and 2021. However, the time of assessment is an important factor. A dramatic change occurred due to advances in the search function, and we believe that the aggregation of metadata on hybrid gold OA and green OA could uncover those hidden OA documents. Other examples of differences are a reported proportion 14% in 2019 [15] versus 48.4% in 2021 (current study) in the physics and astronomy subcategory. This difference is partly explained by the inclusion of only gold OA in the study of Demeter et al. [15], whereas our data include other types of OA. The details of document types were not explained. Thus, we selected only the four types necessary for services. The absolute number and proportional data among all resources were analyzed. We found that an optimal denominator was critical for our assessment of the proportion. For example, the document type was one of the important criteria for selecting document types for the denominator. Four document types (articles, review articles, proceedings papers, and letters) were chosen for the data pool of our denominator [16]. The annual trends used for evaluation were simplified to 2000–2021 and 2021, and an annual comparison between OA and non-OA was added for individual subjects.

The academic discipline was an important parameter for our assessment of the proportion of OA documents, reflecting its importance for user services. The domains of research subjects can be defined in different manners, and the details of the scope, level, and other service-related factors should be critically reviewed.

The most significant result of this study was the high rate of OA documents in the multidisciplinary category. This category did not show OA predominance group until 2009, when a rapid surge in OA documents was observed (Fig. 2). Two OA journals, PLoS One and Scientific Reports, were the definitive leading cause of this change. We also found a major change in researchers’ publishing patterns. Domain-specific research was gradually replaced by interdisciplinary or multidisciplinary research, and multidisciplinary journals are more often selected by authors for publication. An increase in OA documents was also noted in premium scholarly journals, such as Nature (Fig. S10) and Science. We believe that these journals have been influenced by the OA mandate policy in developed countries in the West. The best research papers supported by the governmental research funds of top-ranked countries are published in these top journals as hybrid gold OA. The documents in this multidisciplinary category are a fundamental resource for new OA-based library services.

Significant trends for OA were also observed in the health and medicine category. The public demands for health and well-being reflect this trend. Authors willingly share their research, and health professionals feel satisfaction if they become well-known for their scientific excellence. The high rates of OA in articles on infectious diseases and general healthcare are manifestations of the authors’ willingness to share. The participation of medical and healthcare professionals as volunteers for ancillary services, in addition to OA scholarly library services, is also expected.

Among the seven subcategories in the natural sciences category, biosciences and astronomy/physics had high proportions of OA publishing. Biosciences have a few overlapping features with healthcare and medicine, and basic and applied health researchers including students will benefit from OA articles on biosciences. The discipline of astronomy/physics is a peculiar case. Researchers in high-energy physics and astrophysics initiated their own OA projects and since 2014, have made collective efforts on OA. The Sponsoring Consortium for Open Access Publishing in Particle Physics (SCOAP3) covers more than 11 journals, books, and repositories [17]. However, chemistry and industry-oriented sciences showed lower proportions of OA.

Documents in the engineering category had relatively low OA proportions, and authors in these disciplines often worry about intellectual property rights when their articles are freely available. Although improving trends were observed during the study period (extending through 2021), the numbers are still lower than those in the categories of natural sciences and health and medicine. Within engineering, the multidisciplinary category showed a relatively high proportion of OA documents (57% in 2021). High OA proportions were further observed in marine and ocean engineering, metallurgy, and biomedical engineering. Low OA proportions were noted in chemical, geological, and environmental engineering.

Academic documents in the categories of “social sciences and others” and “humanities and arts” are published more often as monographs than as journal articles. Users favor electronic or analog versions of books, and digital transformation in these subject domains has been slower. It is necessary, however, to keep old literature available online so that it can be searched by those who want to obtain the corresponding knowledge in detail. Wikipedia and other types of online resources are used for these subjects, and OA documents will add more value to these existing knowledge resources.

Library services using OA resources are the ultimate goal of current OA initiatives [18]. The production of OA documents is important, and the visibility of those open documents can be enhanced by the development of research technologies [19] and timely library services for various types of users not limited to academic scholars. This study assessed the fundamental infrastructure for OA scholarly information services.

Limitations

The availability of the full text of those OA documents may be different among individuals because different types of OA require different routes for access. This study does not guarantee that users all have access to these OA documents. The Unpaywall [20], Google Scholar, and Scholytics services will help access the full text of those OA documents.

The complexity of OA document production and use is not covered in this article. The value of the conceptual status of journal goods from the viewpoint of Ostrom [21] and library services are summarized in our previous research report [11], Table S31, and Fig. S17.

Notes

Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Funding

The author received no financial support for this article.

Acknowledgements

The author is grateful to Dr. Eun-Hee Hyun (National Assembly Library of Korea, Seoul, Korea), So-Hyeong Kim (National Research Foundation of Korea, Seoul, Korea), and Eunjung Shin (Science and Technology Policy Institute of Korea, Sejong, Korea). They participated in discussions on the national policy of open access of Korea, which is part of the subject of this manuscript.

Supplementary Material

Supplementary files are available from https://doi.org/10.7910/DVN/W4EVZI

Supplementary tables

Table S1.

Publication titles (publication years) and the numbers of total and OA documents of top 20 journals in the multidisciplinary science discipline between 2000 and 2021 and 2021

kcse-296-Supplementary-Table-1.pdf
Table S2.

Health and medicine category

kcse-296-Supplementary-Table-2.pdf
Table S3.

Eight disciplines in the general health subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-3.pdf
Table S4.

Twelve disciplines in the basic health subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-4.pdf
Table S5.

Thirteen disciplines in the health, clinical medicine subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-5.pdf
Table S6.

Twenty-one disciplines in the health by organ systems subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-6.pdf
Table S7.

Natural science category

kcse-296-Supplementary-Table-7.pdf
Table S8.

Twelve disciplines in the physics and astronomy subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-8.pdf
Table S9.

Twenty-three disciplines in the biosciences subcategory of the natural sciences and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-9.pdf
Table S10.

Nine disciplines in the chemistry subcategory of the natural sciences and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-10.pdf
Table S11.

Five disciplines in the mathematics subcategory of the natural sciences and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-11.pdf
Table S12.

Four disciplines in the environmental sciences subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-12.pdf
Table S13.

Six disciplines in the geosciences subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-13.pdf
Table S14.

Eleven disciplines in the agriculture, fishery, and forestry subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-14.pdf
Table S15.

Social sciences and others category

kcse-296-Supplementary-Table-15.pdf
Table S16.

Eighteen disciplines in the general social sciences subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-16.pdf
Table S17.

Eleven disciplines in the psychology subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-17.pdf
Table S18.

Five disciplines in the education and ethics subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-18.pdf
Table S19.

Eight disciplines in the law, economics, and management subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-19.pdf
Table S20.

Engineering category

kcse-296-Supplementary-Table-20.pdf
Table S21.

Twelve disciplines in the general engineering subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-21.pdf
Table S22.

Six disciplines in the architecture, urban planning, and construction subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-22.pdf
Table S23.

Nine disciplines in the computer engineering subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-23.pdf
Table S24.

Sixteen disciplines in the major engineering subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-24.pdf
Table S25.

Ten disciplines in the materials engineering subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-25.pdf
Table S26.

Humanities and arts category

kcse-296-Supplementary-Table-26.pdf
Table S27.

Seven disciplines in the general humanitiessubcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-27.pdf
Table S28.

Six disciplines in the history subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-28.pdf
Table S29.

Ten disciplines in the literature subcategory of the humanities and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-29.pdf
Table S30.

Eight disciplines in the arts subcategory and the numbers of total and OA documents in 2000–2021 and 2021

kcse-296-Supplementary-Table-30.pdf
Table S31.

The four types of economic status of journal goods according to Ostrom

kcse-296-Supplementary-Table-31.pdf

Supplementary figures

Fig. S1.

Total numbers of open access (OA) and non-OA documents with all 43 types of documents in years 2000–2021.

kcse-296-Supplementary-Fig-1.pdf
Fig. S2.

Yearly numbers of open access (OA) and non-OA documents in Nature.

kcse-296-Supplementary-Fig-2.pdf
Fig. S3.

The discipline of oncology and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-3.pdf
Fig. S4.

The discipline of applied physics and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-4.pdf
Fig. S5.

The discipline of microbiology and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-5.pdf
Fig. S6.

The discipline of biochemistry/molecular biology and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-6.pdf
Fig. S7.

The discipline of environmental science and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-7.pdf
Fig. S8.

The discipline of agriculture–dairy animal science and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-8.pdf
Fig. S9.

The discipline of food science technology, and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-9.pdf
Fig. S10.

The discipline of education/educational research and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-10.pdf
Fig. S11.

The discipline of business finance and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-11.pdf
Fig. S12.

The discipline of management and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-12.pdf
Fig. S13.

The discipline of political science and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-13.pdf
Fig. S14.

The discipline of law and the numbers of open access (OA) and nonOA documents in 2000–2021.

kcse-296-Supplementary-Fig-14.pdf
Fig. S15.

The discipline of nanoscience/nanotechnology and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-15.pdf
Fig. S16.

The discipline of linguistics and the numbers of open access (OA) and non-OA documents in 2000–2021.

kcse-296-Supplementary-Fig-16.pdf
Fig. S17.

A scholarly information system bridges the supply side to the customers’ side.

kcse-296-Supplementary-Fig-17.pdf

References

1. Tickell A. Open access to research publications: independent advice Government of the United Kingdom; 2016. https://apo.org.au/node/308486.
2. Wikipedia: the free encyclopedia [Internet]. Wikimedia Foundation; 2022. [cited 2022 Dec 15]. Available from: https://en.wikipedia.org.
3. Schmitt J. Can’t disrupt this: Elsevier and the 25.2 billion dollar a year academic publishing business [Internet]. Medium; 2015. [cited 2022 Dec 15]. Available from: https://medium.com/@jasonschmitt/can-t-disrupt-this-elsevier-and-the-25-2-billion-dollar-a-year-academic-publishing-business-aa3b9618d40a#.1d6l1bkia.
4. Frazier K. The librarians’ dilemma: contemplating the costs of the. D-Lib Mag 2001;7https://librarytechnology.org/document/8950/.
5. Budapest Open Access Initiative (BOAI). Budapest Open Access Initiative 2002 [Internet]. BOAI; [cited 2022 Dec 15]. Available from: http://www.budapestopenaccessinitiative.org.
6. OECD. Declaration on access to research data from public funding OECD Legal Instruments; 2004.
7. Global Research Council. Action plan towards open access to publications Global Research Council; 2013. Available from: https://globalresearchcouncil.org/fileadmin//documents/GRC_Publications/grc_action_plan_open_access_FINAL.pdf.
8. Max Planck Digital Library. Open Access 2020 Initiative (OA2020) executive summary Max Planck Digital Library; 2022.
9. UNESCO. UNESCO recommendation on open science UNESCO; 2021.
10. Office of Science and Technology Policy. Economic landscape of federal public access policy United States Government; 2022.
11. Seo JW, Hyun EH, Park K. National policy on open access for knowledge information S-Space; 2022. https://s-space.snu.ac.kr/handle/10371/176938.
12. Ministry of Science and ICT. Promotion of the Linkage and Utilization of National Knowledge Information Act [Internet]. Korean Law Information Center; 2021. [cited 2022 Dec 15]. Available from: https://www.law.go.kr/법령/국가지식정보연계및활용촉진에관한법률.
13. Clarivate. Web of Science [Internet]. Clarivate; c2023. [cited 2023 Jan 20]. Available from: https://clarivate.com/webofsciencegroup/solutions/web-of-science.
14. Piwowar H, Priem J, Larivière V, et al. The state of OA: a large-scale analysis of the prevalence and impact of Open Access articles. PeerJ 2018;6e4375. https://doi.org/10.7717/peerj.4375.
15. Demeter M, Jele A, Major ZB. The international development of open access publishing: a comparative empirical analysis over seven world regions and nine academic disciplines. Publ Res Q 2021;37:364–83. https://doi.org/10.1007/s12109-021-09814-9.
16. Seo JW, Chung H, Yun J, Park JY, Park E, Ahn Y. Usage trends of open access and local journals: a Korean case study. PLoS One 2016;11e0155843. https://doi.org/10.1371/journal.pone.0155843.
17. Wikipedia. Sponsoring Consortium for Open Access Publishing in Particle Physics [Internet]. Wikimedia Foundation; 2021. [cited 2022 Dec 15]. Available from: https://en.wikipedia.org/wiki/Sponsoring_Consortium_for_Open_Access_Publishing_in_Particle_Physics.
18. Seo JW, Chung H, Seo TS, et al. Equality, equity, and reality of open access on scholarly information. Sci Ed 2017;4:58–69. https://doi.org/10.6087/kcse.97.
19. Lanzolla G, Pesce D, Tucci CL. The digital transformation of search and recombination in the innovation function: tensions and an integrative framework. J Prod Innov Manage 2021;38:90–113. https://doi.org/10.1111/jpim.12546.
20. Chawla DS. Unpaywall finds free versions of paywalled papers Nature; 2017. https://doi.org/10.1038/nature.2017.21765.
21. Ostrom E. Beyond markets and states: polycentric governance of complex economic systems. Am Econ Rev 2010;100:641–72. https://doi.org/10.1257/aer.100.3.641.

Article information Continued

Fig. 1.

Four major document types (articles, proceedings papers, review articles, and letters) and the numbers of open access (OA) and non-OA documents between 2000 and 2021.

Fig. 2.

The multidisciplinary sciences discipline and the numbers of open access (OA) and non-OA documents between 2000 and 2021.

Fig. 3.

The discipline of infectious diseases and the numbers of open access (OA) and non-OA documents in 2000–2021.

Fig. 4.

The discipline of astronomy/astrophysics and the numbers of open access (OA) and non-OA documents between 2000 and 2021.

Fig. 5.

The discipline of economics and the numbers of open access (OA) and non-OA documents between 2000 and 2021.

Fig. 6.

The discipline of electrical/electronic engineering and the numbers of open access (OA) and non-OA documents between 2000 and 2021.

Fig. 7.

The discipline of literature and the number of open access (OA) and non-OA documents between 2000 and 2021.

Table 1.

Document types and the numbers of total and OA documents between 2000 and 2021 and in 2021

Document type 2000–2021
2021
OA Total OA Total
Major document type 12,424,254 (32.4) 38,392,794 1,521,946 (48.0) 3,169,557
Article 10,606,776 (35.6) 29,763,437 1,251,966 (49.5) 2,530,247
Proceedings paper 993,209 (13.9) 7,170,262 35,689 (24.9) 143,224
Review article 798,786 (43.2) 1,849,038 133,613 (59.9) 223,222
Letter 279,927 (28.0) 999,665 30,763 (49.0) 62,797
Others 1,362,775 (12.9) 10,581,479 142,251 (27.0) 527,383
Meeting abstract 390,880 (7.0) 5,555,017 27,973 (12.3) 228,172
Editorial material 673,357 (31.2) 2,154,842 70,794 (45.5) 155,582
Book review 54,867 (3.4) 1,615,365 6,175 (9.3) 66,082
News item 44,609 (9.6) 462,845 2,043 (15.6) 13,073
Correction 170,345 (52.4) 324,825 26,209 (74.4) 35,219
Early access 41,312 (27.0) 153,211 35,965 (27.6) 130,091
Biographical item 19,258 (16.3) 118,063 1,715 (34.3) 4,994
Poetry 391 (0.3) 116,131 36 (0.9) 3,926
Book chapter 12,571 (18.6) 67,619 899 (19.2) 4,673
Art exhibit review 111 (0.2) 53,606 11 (0.6) 1,708
Other 9,120 (5.1) 180,279 2,270 (32.7) 6,938
Total 13,787,029 (28.2) 48,974,273 1,664,197 (45.0) 3,696,940

Values are presented as number (%) or number only.

OA, open access.

Table 2.

List of six categories and 25 disciplinary subcategories and the numbers of OA and total documents in 2000–2021 and 2021

Category No. of disciplines 2000–2021
2021
Data table
OA Total OA Total
Multidisciplinary (sciences) 1 686,445 (79.0) 868,859 78,565 (88.6) 88,683 Table S1
Health and medicine
General 8 651,253 (40.6) 1,603,834 96,328 (60.6) 158,997 Table S3
Basic 12 915,685 (39.9) 2,295,253 103,462 (56.6) 182,939 Table S4
Clinical 13 1,601,713 (40.5) 3,956,573 209,337 (60.8) 344,556 Table S5
Oran systems 21 1,750,733 (37.4) 4,675,775 177,316 (52.1) 340,186 Table S6
Natural sciences
Biosciences 23 2,362,519 (45.6) 5,184,878 234,480 (62.3) 376,198 Table S9
Chemistry 9 834,015 (21.4) 3,895,728 127,198 (41.0) 310,400 Table S10
Mathematics 5 637,140 (44.4) 1,434,463 59,396 (54.6) 108,706 Table S11
Physics and astronomy 12 1,564,141 (33.4) 4,685,703 131,878 (48.4) 272,246 Table S8
Environmental science 4 463,807 (31.3) 1,479,662 87,502 (49.6) 176,498 Table S12
Geoscience 6 299,490 (31.1) 963,123 35,759 (46.2) 77,389 Table S13
Agriculture, fishery, and forestry 11 427,452 (28.9) 1,477,346 55,098 (45.2) 121,890 Table S14
Social sciences and others
General 18 359,693 (26.3) 1,368,694 45,918 (41.6) 110,474 Table S16
Psychology 11 292,831 (31.8) 921,275 34,277 (44.1) 77,712 Table S17
Education and ethics 5 162,213 (24.9) 651,781 23,799 (41.6) 57,218 Table S18
Law, economics, management 8 342,027 (23.8) 1,440,011 43,776 (36.2) 120,813 Table S19
Engineering
General 12 579,970 (33.4) 1,735,150 83,271 (39.0) 213,787 Table S21
Architecture, urban planning, and construction 6 114,478 (19.7) 580,902 15,260 (31.4) 48,643 Table S22
Computer 9 712,959 (18.0) 3,954,395 84,288 (36.9) 228,652 Table S23
Major 16 1,206,647 (17.1) 7,058,382 171,621 (34.3) 500,285 Table S24
Materials 10 632,433 (20.2) 3,129,774 94,708 (36.4) 259,976 Table S25
Humanities and arts
General 7 103,800 (22.2) 468,470 14,978 (37.9) 39,530 Table S27
Literature 10 19,770 (11.6) 171,127 2,864 (28.8) 9,957 Table S29
History 6 66,306 (19.7) 336,673 8,994 (36.9) 24,362 Table S28
Arts 8 108,873 (19.3) 564,252 13,267 (34.7) 38,240 Table S30

Values are presented as number (%) or number only.

OA, open access.