Data Papers: A New Submission Category
Publication: Journal of Structural Engineering
Volume 145, Issue 12
Experimental research in the field of structural engineering typically involves large specimens that are loaded in unique and intricate ways. Loading setups are generally large and expensive. They may involve multiaxial loading machines, shake tables, wind tunnels, hurricane and tornado simulators, tsunami loading basins, impact equipment, or explosive loading facilities. Given their unique nature, structural tests are densely instrumented and produce copious amounts of data that are spatially and temporally distributed. Journal publications that describe structural tests only archive a sampling of the test results.
Field studies, reconnaissance efforts, and survey-based research also produce vast amounts of data. Like experimental studies, only a small, processed portion of it reaches publication in major journals and conferences. The rest is typically documented in reports, mostly in hard copy format, that could become lost or misplaced over time. Even if it survives over multiple decades, data in old reports could still be not useful if they are not adequately described so that current and future users could understand their meaning and extract information from it for their research purposes.
The growing realization that data are valuable and perishable has prompted a conscious efforts to preserve and organize it. In 2004, the National Science Foundation’s George E. Brown Network for Earthquake Engineering Simulation (NEES) became one of the earliest efforts in the field of structural engineering (and civil engineering, more generally) to focus on data archiving and curation. Since then, the National Science Foundation, as well as many other funding agencies in the US and abroad, have required proposers to include a data management plan that describes how research data will be handled and eventually archived. Major journals have followed suit and now require authors to include a data availability statement in their papers. The intent is to enable replication of the research, validation, and transparency, and ultimately expedite the advancement of science.
To address the need to curate and archive data from unique structural engineering research efforts, the editorial board of the Journal of Structural Engineering (JSE) decided during its 2018 annual board meeting in Fort Worth, Texas, to explore publishing data papers in addition to its complement of other paper types. The decision was that the focus of these data papers should be only on data and how they are collected, not interpretation of the data, which is addressed by other paper categories. The intention was to cater to data producers, collectors, managers, and curators. To promote the highest level of data preservation efforts, the goal of the editorial board was for data papers to be counted as a measure of productivity in the field of structural engineering similar to regular technical papers.
A committee of associate editors from within the editorial board was formed to define data papers from the perspective of structural engineering and facilitate timely introduction of the new paper category. The committee was co-chaired by Professors Dimitrios Lignos [Ecole polytechnique fédérale de Lausanne (EPFL)] and Jamie Padgett (Rice University) and included Professors Eleni Chatzi (ETH Zurich), Kurt Gurley (University of Florida), Michael Scott (Oregon State University), and Farzin Zareian (University of California, Irvine). The committee recommended introduction of data papers as a new paper category and the idea was endorsed by the editorial board during its annual 2019 meeting in Orlando, Florida. The committee was charged with producing an inaugural special collection of data papers to demonstrate the intended scope of the new paper category and review process for such paper types. The relevant portion of the committee’s report is reproduced next.
ASCE Journal of Structural Engineering Data Papers: Draft Formatting and Review Guidelines
Aims and Scope
An ASCE JSE data paper is a peer-reviewed article that concisely describes the data, methods, and instrumentation used to acquire the data, associated metadata, data validation, and potential opportunities for reuse. The article must include a link to the complete data set archived at a publicly accessible repository.
General Requirements for ASCE JSE Data Papers
Data papers shall describe data sets of interest to the JSE audience. These data sets may be from field observations and measurements, laboratory experiments, or other data types relevant to the structural engineering community (e.g., expert opinion surveys). At this stage, data sets from purely numerical simulations are outside the scope of an ASCE JSE data paper. However, a collection of data from multiple sources including numerical simulations may be considered. The described data set must be publicly available through a digital object identifier (DOI), cited in the paper, and a link provided within the paper to ensure online access to the published data set. Emphasis should be placed on adequately describing the data, methods of collection, processing procedures, relevant notation, and significance or value of the data set to the structural engineering community. Authors should describe their methods for ensuring data reliability including relevant data summary tables or plots that could aid the peer-review process and facilitate efficient data reuse by the engineering community without the need of contacting the corresponding authors. Finally, the data paper should include a discussion of the relevance of the data set to the structural engineering community and potential opportunities for data reuse. The authors must also ensure that they have the rights to share the data publicly. If a peer-reviewed journal paper is accepted, it does not necessarily mean that a data paper companion is automatically published.
A data paper shall follow a format consistent with a general peer-reviewed ASCE journal paper. The length of a data paper is flexible and should offer as concise as possible a description of the aforementioned aspects.
Data Storage
Data sets described within the paper should appear in a research data repository that will provide authors with a DOI. The choice of the data-sharing repository is up to the author(s), considering that particular data-sharing policies may apply depending on the funding organization or other pertinent restrictions.
One relevant repository authors may consider is the DesignSafe-CI of the Natural Hazards Engineering Research Infrastructure (NHERI) supported by the National Science Foundation: https://www.designsafe-ci.org/. Through DesignSafe-CI, authors can store, curate, archive, and formally publish data on the Data Depot and receive a persistent DOI.
Other example general data repositories include Zenodo, Dryad, Figshare, Open Science Framework, Harvard Dataverse, and Hydroshare.
General Guidelines for Reviewers
Reviewers of an ASCE JSE data paper manuscript will ensure the paper meets the formatting, data and metadata, validation, and data access criteria. In particular, the review process shall comply with the all following:
•
Are the data subject appropriate for the Journal of Structural Engineering audience?
•
Does the writing meet the quality standards of an ASCE journal article?
•
Does the manuscript adequately describe the data and metadata including the relevant procedures for data postprocessing, if applicable?
•
Is data validation appropriately addressed in the manuscript?
•
Are the data collected, organized, and processed based on rigorous methods and standards?
•
Are the data archived in a publicly accessible data repository?
•
Do the archived data have an assigned DOI that is cited in the manuscript?
•
Are the archived data easily accessible via a provided link?
•
Is the full data set available as described in the manuscript?
•
Is there a summary presentation of the data sufficient to allow the domain expert reviewer to spot-check quality?
•
Does the paper offer examples and discussion of how the data may be useful to the structural engineering community and ideas on potential reuse scenarios?
Reviewers are not responsible for validating the quality of the data. Reviewers are only responsible for determining whether the authors present a compelling description of the data collection, validation, and reuse potential.
Example with Three Scenarios
A data paper is submitted to JSE on the subject of field measurements of ground-level wind velocities measured during a hurricane. The authors provide a DOI and data archive link as well as the metadata such as instrumentation, location and methods, and summary plots of the data. The reviewer does not analyze the data, but verifies that the data are accessible and complete as described. The reviewer confirms that the metadata are sufficient for a reader to properly interpret and utilize the data without need to interface with the authors. The paper is of sufficient quality with respect to formatting, grammar, and citations.
•
Scenario 1: The authors adequately describe the methods used to verify the reliability of the data, and the summary plots in the data paper show nothing unexpected in the wind velocities.
•
Outcome 1: The reviewer is justified in recommending publication because the authors have met their burdens.
•
Scenario 2: The authors fail to describe any methods they applied to verify the reliability of the data, but the summary plots in the data paper show nothing unexpected in the wind velocities.
•
Outcome 2: The reviewer is not justified in recommending publication, and should request a data validation section be included in a revision.
•
Scenario 3: The authors describe the methods they applied to verify the reliability of the data, but the summary plots in the data paper show an obvious anomaly in the data. For example, the wind speeds are not reasonably consistent with the known intensity of the hurricane at the measurement location, or the wind direction is not consistent with what is expected given the measurement location and path of the hurricane.
•
Outcome 3: The reviewer is not justified in recommending publication. The reviewer is not obligated to analyze the data and investigate cause. The reviewer need only note the anomalies and that the authors did not address them in their data validation section. The paper should either be rejected or returned for revision with specific requirements to address the anomalies.
Information & Authors
Information
Published In
Journal of Structural Engineering
Volume 145 • Issue 12 • December 2019
History
Received: Aug 11, 2019
Accepted: Sep 4, 2019
Published online: Sep 17, 2019
Published in print: Dec 1, 2019
Discussion open until: Feb 17, 2020
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.