Learning the ABCs of Tracking APCs
- 1. LEARNINGTHE ABCS OF TRACKING APCS Erin Calhoun Electronic Resources Intern, University of Toronto Libraries
- 2. University ofToronto Libraries 80,000 students. Around 700 undergrad and 200 grad programs 44 libraries across three campuses More than 12 million volumes in 341 languages, 1,500,000 electronic resources in various formats TorontoAcademic Health Science Network -- 20 community affiliated hospitals and healthcare sites
- 3. PURPOSE Purpose: “TrackArticle Processing Charges associated with UofT affiliated authors publishingOpen Access articles in hybrid and open access journals by collection bibliographic metrics” ■ Overall concern for “double dipping” in collection development spending and financially supporting researchers to public open access ■ UofT does not have a centralized reporting system where open access spending and publication data is stored – instead we have to play catch-up Goal: Determine volume of open access research & calculate APC cost
- 4. BACKGROUND Methodology for harvesting open access journals publication data Use DOAJ and Scopus to identify UofT affiliated submissions in fully OA publications Export key data about these submissions to a CSV file Identify UofT author’s email addresses under the Scopus field “corresponding address” University ofToronto Pay-It-Forward Analysis Project Report •Publishing output, library expenditures, researcher funding data •Conclusions on how an APC funded business model would operate & developed models on predicted costs •SNIP method
- 5. How did we do it: Organizing Data Export data:Web of Science WoS method OR Clarivates Analytics API to extract data Search for UofT authors in OA publications Locate the corresponding author Identify email domains associated with UofT
- 6. How did we do it: Calculating costs ■ =IF(SNIP>=3.207,5425,SUM((1245. 45+(769.83*SNIP)))) ■ The original PIF report, published in 2017, states the average APC for a journal is 1147.68.We updated this for inflation.
- 7. Testing Data ■ APC data provided about BMJ (2016-2019) ■ 2018: UofT paid $59,565.61 in APCs in BMJ (our method) ■ 2018: BMJ reports that UofT paid $42,133.72 in APCs in BMJ (their data) ■ Why? BMJ encourages partial payment of limited funds available; UofT authors could not have been the ones paying? ■ Consistent and standardized calculation model is preferred
- 8. Lessons Learned Difficult to track researchers affiliated with the University in a non- centralized system Calculations to generate estimations > searching for the exact amounts Significant variations in APCs paid by institutions and the fees set out by hybrid and OA journals (Pinfield, Salter, and Bath, 2016).
- 9. Future directions ■ Re-evaluate the average APC for a journal given inflation & rising costs – Evidence that APCs are rising (Soloman and Bjork, 2016) ■ Researchers at research intensive universities publish in traditional subscription journals, paying higherAPCs that fully-OA journals. – Results may be unbalanced if we treat subscription and OA journals the same (Soloman and Bjork, 2016; Pinfield,Salter, and Bath, 2016) ■ Target specific fields of study: biomedicine fields have high numbers of high-impact journals with highAPCs (Soloman and Bjork, 2012) – APCs for journals in specific fields do not correlate strongly with influence/impact (physics and math) (Corbyn, 2013)
- 10. Works Cited Soloman, David and Bo-Christer Bjork. (2016)Article Processing Charges for Open Access Publication – the situation for research intensive universities in the USA and Canada. Peer J, July 2016. DOI: 10.7717/peerj.2264. ----- (2012) A Study of Open Access Journals UsingArticle Processing Charges. Journal of the American Society for Information Science andTechnology, 63(8): 1485- 1495. 2012. Pinfield, Stephen and Salter, J. and Bath. P.A. (2016)The “total cost of publication” in a hybrid open-acess environment: Institutional approaches to funding journal article-processing charges in combination with subscriptions. Journal of the Association for I nformation Science andTechnology 67(7): 1751-1766) Thank you! erin.calhoun@mail.utoronto.ca