doi: 10.7554/eLife.57067.
Evaluating the impact of open access policies on research institutions
Affiliations
- PMID:32924933
- PMCID: PMC7536542
- DOI: 10.7554/eLife.57067
Item in Clipboard
Evaluating the impact of open access policies on research institutions
Chun-Kai Karl Huang et al. Elife..
Display options
Format
Display options
Format
Abstract
The proportion of research outputs published in open access journals or made available on other freely-accessible platforms has increased over the past two decades, driven largely by funder mandates, institutional policies, grass-roots advocacy, and changing attitudes in the research community. However, the relative effectiveness of these different interventions has remained largely unexplored. Here we present a robust, transparent and updateable method for analysing how these interventions affect the open access performance of individual institutes. We studied 1,207 institutions from across the world, and found that, in 2017, the top-performing universities published around 80-90% of their research open access. The analysis also showed that publisher-mediated (gold) open access was popular in Latin American and African universities, whereas the growth of open access in Europe and North America has mostly been driven by repositories.
Keywords: meta-research; none; open access; repositories; research policy; scholarly publishing; universities.
© 2020, Huang et al.
Conflict of interest statement
CH, CN, RH, LM, KW, AO, CB No competing interests declared
Figures
Diagrammatic summary of how data is collected and mapped against open access definitions using information from Unpaywall metadata.
Percentages of institutional Total OA, Gold OA and Green OA (left to right) grouped by regions for 2017. Parallel figures for 2016 and 2018 are provided in Figure 2—figure supplement 1.
Percentages of institutional Total OA, Gold OA and Green OA (left to right) grouped by regions for 2016 and 2018, respectively.
The black lines represent the OA% as calculated from the data. The colour bars are 95% confidence intervals calculated with Šidák correction to compensate for the multiple comparisons effect. The colours represent the regions as per Figure 2. Evidently, universities topping this list achieve around 80–90% OA.
The colours represent the regions as in Figure 2. The overlaid boxplots exclude outliers as determined using the standard interquartile methods (i.e. values more than 1.5 interquartile range away from the first or third quartile are deemed outliers). Countries are ordered by their median Total OA% (left panel). This figure includes small universities (those with significantly fewer research outputs) which are left out of the main article.
Publisher-mediated open access (gold OA) vs repository-mediated open access (green OA) by institution for 2017. Each point plotted is a university, with size indicating the number of outputs analysed and colour showing the region. Articles can be open access through both publisher and repository routes so x and y values do not sum to give total open access. Animated version with figures for each year between 2007 to 2018 can be seen in Figure 3—animation 1.
(A) The annual change in percentage (rolling current year percentage minus the previous year percentage) of gold OA for six UK universities. The top three universities are those with the largest additional funding compared to the bottom three universities which received less additional funding. (B) The annual percentage of green OA through the home institutional repositories of four UK universities compared to high performing universities from elsewhere. (C) The annual percentages of hybrid OA at five universities in the Netherlands. (D) Three pairs of UK and US universities, selected based on having a similar size and level of green OA. The annual percentages of total green OA are depicted for each university. Additional figures are provided in Figure 4—figure supplement 1.
(A) The annual percentages of Hybrid OA publications for six UK universities. The top three are those that received the largest additional funding to support open access publishing compared to the bottom three that received less additional funding. (B) For the same set of UK universities, this panel shows the proportion of Hybrid OA publications within Total Gold OA publications. (C) The annual percentages of Total Gold OA for five Dutch universities, showing steady increase in Total Gold OA. (D) The annual percentages of Hybrid OA for the same five Dutch universities, showing a sharp increase in Hybrid OA around 2014 and 2015.
This figure compares the different open access routes taken by three groups of universities: three UK universities (University College London, University of Cambridge and University of Glasgow) that received substantial funding for open access publishing (combined gold and green OA increases), three UK universities (Loughborough University, University of St Andrews and Plymouth University) that received less funding (more green OA focused), and three Latin American universities (more gold OA focused). The dots represent the % of total gold OA publications and % of total green OA publications for the specified universities for each year from 2007 to 2017, where the arrow indicates the direction of time.
References
- Aguado-López E, Becerril-García A. Latin America’s longstanding open access ecosystem could be undermined by proposals from the Global North [LSE Latin America and Caribbean] 2019. [November 6, 2019].https://blogs.lse.ac.uk/latamcaribbean/2019/11/06/latin-americas-longsta...
- Barbour G, Nicholls S. Open access: should one model ever fit all? Australian Quarterly. 2019;90:3–9.
- Debat H, Babini D. Plan S in Latin America: a precautionary note. PeerJ. 2019;7:e27834. doi: 10.7287/peerj.preprints.27834v1. - DOI
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Miscellaneous