# Add the dataset of DOAJ Seal journals

doaj_seal <- read_csv('DOAJ_Seal.csv')

1 Administrative Data

1.1 ID

Not Applicable

1.2 Funder

This research is being submitted for funding to the J. Bohannon Foundation.

1.3 Grant Reference Number

Not Applicable (Proposal in preparation)

1.4 Project Name

Do Reputable Open Access Journals Require Open Data Sharing?

1.5 Project Description

This study analyzes the submission requirements of the most reputable open access journals to determine the prevalence and characteristics of data sharing policies. This question is an important one for 21st century authors and readers because open data sharing is seen as a key component of open and more trusted scientific record.

According to the Research Data Alliance (RDA) Data Policy Standardisation and Implementation group:

“the prevalence of research data policies from institutions and research funders is increasing, so publishers and editors are paying more attention to standardisation and the wider adoption of data sharing policies.”

This study investigates whether the most reputable Open Access journals have data sharing polices and the characteristics of those policies. These policies require authors, in some fashion, to openly disseminate the data and software underlying their published articles.

Our investigation builds on the recent work of Castro et al. (2017) who assessed the prevalence and characteristics of data sharing policies from randomly-selected, English-language, open access journals. Their findings reveal that only a small minority of these journals have data sharing policies. These findings – which are consistent with those of other studies [see for example, (“Reproducible and Reusable Research: Are Journal Data Sharing Policies Meeting the Mark?” n.d.)) – may be skewed because of the authors’ rules of inclusion and exclusion (in particular, the choice to include open access journals merely because of their use of the Open Journal Systems (OJS) hosting platform; the choice to exclude non_English language journals).

In this study, we will include only the most reputable open access journals in our assessment of journal sharing policies, reglardless of language. We will analyze all journals that have attained the Seal of Approval from the Directory of Open Access Journals, DOAJ (shown below). We will apply the same coding framework devised by Castro et al. (2017) to the DOAJ Seal journals. We contend that a more rigorously screened population of open access journals, regardles of language, will yield a more accurate and reproducible set of findings than those published from the previous study.

DOAJ Seal of Approval

DOAJ Seal of Approval

DOAJ Seal journals are considered the most reputable because they:

“achieve a high level of openness, adhere to Best Practice and high publishing standards.The Seal is awarded to a journal that fulfills a set of criteria related to accessibility, openness, discoverability, reuse and author rights. It acts as a signal to readers and authors that the journal has generous use and reuse terms, author rights and adheres to the highest level of ‘openness’.”1

Moreover, the DOAJ Seal journals do include over 200 non-English language journals that merit analysis in this study. Excluding these from the analysis represents cultural bias that undermines reliable research.

According to Xie, Allaire, and Grolemund (2018) :

Inline LaTeX equations can be written in a pair of dollar signs using the LaTeX syntax, e.g., \(f(k) = {n \choose k} p^{k} (1-p)^{n-k}\)

Math expressions of the display style can be written in a pair of double dollar signs, e.g., \[f(k) = {n \choose k} p^{k} (1-p)^{n-k}\]

1.6 Researcher

Firstname Lastname, Principal Investigator, Institution

1.7 Researcher ID

1.8 ORCID: your ORCID

1.9 Date of First Version

1.10 Date of Last Update

2 Data Collection

2.1 Existing Data Being Reused

This study relies on the DOAJ Journal Metadata available as a csv file download from the DOAJ website. The csv file is updated every 30 minutes.

This data will be read into the open source tool OpenRefine to filter it for those journals awarded the DOAJ Seal, and to remove unneeded columns containing the web addresses for journal policies around plagiarism, submission fees, and other urls not related to this study.

The filtered version of the data set will be exported as a new file named doaj_seal.csv for importing into analytical software.

A sample of the doaj_seal.csv data set is shown below.

knitr::kable(head(doaj_seal, 4), caption = 'A Table of the first 4 rows of the DOAJ Seal data.')
A Table of the first 4 rows of the DOAJ Seal data.
JnlTitle Publisher PubCountry Fee WaiverPolicy Identifiers FirstYear Language ReviewProcess Plagiarism Sub2Pub JnlLicense AuthorCopyright DOAJ_Seal
Archives Animal Breeding Copernicus Publications Germany No Yes DOI 1999 English Peer review Yes 13 CC BY TRUE Yes
Bothalia: African Biodiversity & Conservation AOSIS South Africa No NA DOI 2014 English Double blind peer review Yes 12 CC BY TRUE Yes
Geographica Helvetica Copernicus Publications Germany No Yes DOI 1946 English, French, German, Italian Double blind peer review Yes 53 CC BY TRUE Yes
Hereditas BioMed Central United Kingdom Yes Yes DOI 2005 English Blind peer review Yes 6 CC BY TRUE Yes

2.2 Data being collected

The doaj_seal.csv data set will be copied and enhanced with additional columns, resulting in the processed dataset, doaj_seal_enhanced.csv. The following columns will be added to doaj_seal.csv, in conformance with the Coding Framework of Castro et al. (2017).

  • ‘Data Policy’ (Boolean)
    • Yes
    • No
  • ‘Data Sharing Policy’ (Factor, 6 levels)
    1. No mention 2 Implied
    2. Mentioned
    3. Explicitly encouraged
    4. Required, but not explictly tied to editorial decisions
    5. Required as a condition of publication
  • ‘Data Citation Policy’ (Facto, 3 levels)
    1. No mention
    2. Implied
    3. Explicitly encouraged

Investigators will examine the websites of each journal listed in the doaj_seal_enhanced.csv file to determine whether the data sharing policy is included in the Instructions to Authors. The Coding Framework published by Castro et al. (2017) will be applied.

2.2.1 Metadata

The journal metadata contained in the project’s data sets comes directly from the Directory of Open Access Journals.

The final outputs from the project will be documented in metadata files according to the DataCite DOI registration agency – see the [DataCite Metadata Schema 4.1] (https://schema.datacite.org/) for specific details. By following this standard metadata format, other researchers (and computers) will be able to find, access, and reuse the outputs from this project by searching the DataCite metadatabase.


4 Data collection and standards

All data retrieved for the DOAJ Seal of Approval are downloaded and stored in the open common separated value (csv) file format.

All data policies culled from the web sites of the DOAJ seal journals will be saved as .txt files. The data generated by applying Castro et al. (2017) Coding Framework will be stored in csv file format.

Analysis, visualization, and summarization of the study’s findings will be performed in the open source software R and RStudio using the tidyverse package (Wickham 2017). Reports produced from the study will be also be created in RStudio using the open source text format Rmarkdown and output to HTML documents, slides, and MS Word documents for submission to funders or publishers.

All files associated with the project will be maintained under the Git version control system and made openly available for download from the Principal Investigator’s GitHub repository.


5 Storage and Backup

During the active phase of the project data will be stored on and backed up to the Research Data Storage Facility (RDSF) at ? Institution. This facility represents 2 million pounds of digital resilient storage, with ongoing capital investment. The RDSF is overseen by a steering group of senior research and support staff, which includes the PVC Research. Backup procedures are robust (overnight backup, copies held remotely on tape) and secured access is in place

6 Deposit and long-term preservation

After the project conclusion, all research outputs will be assigned Digital Object Identifiers by DataCite to serve as persistent identifiers and to always resolve to the landing page for the project files (metadata, data, code, and reports).

Additionally, the Research Data Storage Facility (RDSF) at ? Institution will undertake to preserve each of the digital outputs mentioned above, for three years beyond the end of the project. Refreshment of storage media will be scheduled as required


7 Selection and Preservation

The final complete data set doaj_seal_enhanced and all published outputs produced based on this data set will be issued Digital Object Identifiers and stored on GitHub and Zenodo. All draft versions of project outputs will also be publicly available in Git Hub with unique hashes distinguishing each version that predates the final.


8 Data Sharing

The data and metadata will be made openly accessible under a CC-BY SA license in the CERN-maintained open access repository Zenodo, an open dependable home for the long-tail of science, enabling researchers to share and preserve any research outputs in any size, any format and from any science.3


9 Responsibilities and Resources

The Principal Investigator is responsible for implementing the Data management Plan and ensuring it is reviewed and revised as necessary. (S)he will be responsible for all data collection and recording; for data analysis and visualization; and for maintain all files under version control using git and GitHub.

The Data Management Specialist assigned to the project as an in-kind contribution from the ? Institution Library will be responsible for creating the DataCite metadata documentation for all outputs and ensuring timely DOI registration of each final output. (S)he will also deposit all final outputs to the Zenodo repository and update metadata associated with the DOI as necessary.


9.1 Principal Investigator’s BioSketch

About me…

References

Castro, Eleni, Merc‘ Crosas, Alex Garnett, Kasey Sheridan, and Micah Altman. 2017. “Evaluating and Promoting Open Data Practices in Open Access Journals.” Journal of Scholarly Publishing 49 (1). University of Toronto Press Inc. (UTPress): 66–88. doi:10.3138/jsp.49.1.66.

“Reproducible and Reusable Research: Are Journal Data Sharing Policies Meeting the Mark?” n.d. 5. doi:10.7717/peerj.3208.

Wickham, Hadley. 2017. Tidyverse: Easily Install and Load the ’Tidyverse’. https://CRAN.R-project.org/package=tidyverse.

Xie, Yihui, J. J. Allaire, and Garrett Grolemund. 2018. R Markdown - the Definitive Guide. Chapman & Hall/CRC. http://dx.doi.org/10.1038/nmat3283.