Open science curriculum for students, June 2019
1 like2,316 views
Open science is a movement aimed at making scientific research accessible to all, promoting collaboration and transparency in knowledge dissemination. It includes principles like open access to research data and adherence to fair data practices to improve reproducibility and trust in scientific findings. This approach is increasingly important for students, as it requires new skills and offers opportunities for novel research and collaboration.
![WILL ANYBODY TRUST CLOSED
SCIENCE AGAIN?
● Recent studies indicates that p-hacking [1] is a significant
problem – sometimes even without the scientist even being
aware of doing so (Ioannidis 2005; Head et al. 2015)
● Pre-registered (open) data provides a good insurance
against suspicion of both data dredging (and plain data
falsification).
[1] “p-hacking,” (data dredging, data fishing, …) occurs when researchers collect or
select data or statistical analyses until nonsignificant results become significant.
Ioannidis (2005). "Why Most Published Research Findings Are False". PLoS Medicine. doi:10.1371/journal.pmed.0020124.
Head et al. (2015) The Extent and Consequences of P-Hacking in Science. PLoS Biol. doi:10.1371/journal.pbio.1002106](https://image.slidesharecdn.com/2019-06-12-open-science-190614111803/85/Open-science-curriculum-for-students-June-2019-18-320.jpg)
Open science curriculum for students, June 2019
- 1. What is Open Science and why is it important for students? 12 June 2019, Trondheim, Norway Living Atlas Seminar http://bit.ly/gbifno-openscience
- 2. OPEN SCIENCE
- 3. WHAT IS OPEN SCIENCE? Open science is the movement to make scientific research (including publications, data, physical samples, and software) and its dissemination accessible to all levels of an inquiring society, amateur or professional (Woelfle et al. 2011). cf. Wikipedia Woelfle, M.; Olliaro, P.; Todd, M. H. (2011). "Open science is a research accelerator". Nature Chemistry. 3 (10): 745–748. doi:10.1038/nchem.1149
- 4. WHAT IS OPEN SCIENCE? Open science is transparent and accessible knowledge that is shared and developed through collaborative networks (Vicente-Saez et al. 2018). cf. Wikipedia Vicente-Saez, Ruben; Martinez-Fuentes, Clara (2018). "Open Science now: A systematic literature review for an integrated definition". Journal of Business Research. 88: 428–436. doi:10.1016/j.jbusres.2017.12.043
- 5. WHAT IS OPEN SCIENCE? Open Science can be seen as a continuation of, rather than a revolution in, practices begun in the 17th century with the advent of the academic journal (David 2004). cf. Wikipedia David, P. A. (2004). "Understanding the emergence of 'open science' institutions: Functionalist economics in historical context". Industrial and Corporate Change. 13 (4): 571–589. doi:10.1093/icc/dth023
- 6. Open Access (OA): Research results distributed online and free of costs or other barriers – often meaning free access to research articles. Open Science: Researchers to share their methods, computer code and research data in central data repositories. Open Data: is freely available to everyone to use and re-publish as they wish, without restrictions from copyright, patents or other mechanisms of control. FAIR data principles: findable, accessible, interoperable and reusable.
- 7. FAIR data principles Wilkinson et al. 2016 doi:10.1038/sdata.2016.18 FAIRdataprinciples Promotes maximum (re) use of research data. Researchers need to do more than simply post their data on the web for it to be useful.
- 8. What is FAIR Data? FINDABLE • Data and supplementary materials have sufficiently rich metadata and a unique and persistent identifier. ACCESSIBLE • Metadata and data are understandable to humans and machines. Data is deposited in a trusted repository. INTEROPERABLE • Metadata use a formal, accessible, shared, and broadly applicable language for knowledge representation. REUSABLE • Data and collections have a clear usage licenses and provide accurate information on provenance. https://libereurope.eu/wp-content/uploads/2017/12/LIBER-FAIR-Data.pdf FAIRData
- 9. SCIENCE CURRENCIES (CITATION) ● Peer-reviewed scholarly papers in high impact journals (still) maintain considerable weight for scientific careers. ● A movement is under way to build similar status for open data, open metadata, and other open science products…
- 10. Data Citation Principles 1. Data to be legitimate citable products of research. 2. Data citations giving scholarly credit and attribution. 3. In scholarly literature, whenever claims are based on data, data should always be cited. 4. Persistent method for identification of data, that is machine actionable, globally unique, universal. 5. Data citation facilitate access to data or at least to metadata. 6. Unique identifiers that persist even beyond the lifespan of the data. 7. Data citation identify and access the specific data that support verification of the claim (provenance, time-slice, version). 8. Flexible, but attention to interoperability of practices across communities. Data Citation Synthesis Group: Joint Declaration of Data Citation Principles. Martone M. (ed.) San Diego CA: FORCE11; 2014
- 11. Open research data policies The scientific journals (at Springer Nature) practices different guidelines and requirements for availability to the underlying research data for published research papers. Springer Nature has made a comprehensive report on practical incentives and appropriate norms to promote open data. http://www.springernature.com/gp/group/data-policy/policy-types
- 12. OPEN SCIENCE Kunnskapsdepartementet (2016) EU (2016) Competitiveness Council, 26-27/05/2016 EU (2007) INSPIRE Directive Norway is to be a careful pioneer in open access to research results. Norway to follow the ambition of EU on full open access to publicly funded research by 2020. Results of research supported by public and public-private funds freely available to and reusable by anyone.
- 13. OPEN RESEARCH DATA Forskningsrådet (2014). ISBN: 978-82-12-03361-0 The Research Council of Norway expects all research data from projects funded by the Research Council to be made freely available as open data. In some situations there can be valid and justified reasons for exceptions. (2014)
- 14. WHY TEACH STUDENTS OPEN SCIENCE ? ● We are in the middle of an ongoing paradigm shift in scientific practice (and impact metrics). ● The open science wave is moving fast! ● Young scientists will (already today) need different skills, than was needed previously – to succeed in academia.
- 15. Expanding possibilities… (for novel curiosity-driven research) Open science Traditional science Your student
- 16. REPRODUCIBILITY CRISIS "Scientific irreproducibility — the inability to repeat others' experiments and reach the same conclusion” (Nature 2016) Baker (2016) 1,500 scientists lift the lid on reproducibility. Nature. doi:10.1038/533452a
- 17. "Scientific irreproducibility — the inability to repeat others' experiments and reach the same conclusion — is a growing concern”. Baker (2016) Nature doi:10.1038/533452a Open Science solution: researchers to share their methods, data, computer code and results in central data repositories. Note that we also need herbarium specimen and bio-repositories (eg. museums).
- 18. WILL ANYBODY TRUST CLOSED SCIENCE AGAIN? ● Recent studies indicates that p-hacking [1] is a significant problem – sometimes even without the scientist even being aware of doing so (Ioannidis 2005; Head et al. 2015) ● Pre-registered (open) data provides a good insurance against suspicion of both data dredging (and plain data falsification). [1] “p-hacking,” (data dredging, data fishing, …) occurs when researchers collect or select data or statistical analyses until nonsignificant results become significant. Ioannidis (2005). "Why Most Published Research Findings Are False". PLoS Medicine. doi:10.1371/journal.pmed.0020124. Head et al. (2015) The Extent and Consequences of P-Hacking in Science. PLoS Biol. doi:10.1371/journal.pbio.1002106
- 19. Reuse of teaching curriculum
- 20. Why publish open data? ● Data produced using public funds should be regarded as a common good, and should be made available for inspection, interpretation and re-use by third parties. ● Needless duplication of data-collecting efforts and costs will be reduced. ● Open data increases transparency and overall quality of research. ● Published data can be re-analysed, verified, and improved by others. ● Data publication increase recognition and opportunities for collaboration. ● Published data can be cited and re-used, either alone or in combination with other data. ● Data owners and collection managers can trace data use and citation. ● Data creators, their institutions and funding agencies can be credited. ● Data can be integrated with other datasets across space and time. ● Open data increases potential for interdisciplinary research and re-use in new contexts not envisioned by the data creator. Penev et al. (2017) https://doi.org/10.3897/rio.3.e12431 20
- 21. Data Management Plan (DMP) A formal document that outlines HOW data are to be handled during a research project, and after the project is completed. The goal is to plan data management BEFORE the project begins. Including a plan for the COSTS of data management and archiving. This saves time in the long run, and promotes data fitness for reuse. Reduce duplication of existing scientific studies. Reduce the loss of data. https://en.wikipedia.org/wiki/Data_management_plan Illustration CC BY Jørgen Stamp
- 22. Why write Data Management Plans? A data management plan is a tool for making your research reproducible and thus trustworthy. Good data curation saves you research time, because you, your collaborators, and others, will find, understand, and get access to your (own) research data. Efficient data sharing provides broader distribution and impact for your research results. Open research data, available for reuse, strengthens open and curiosity- driven research, and scientific breakthrough not originally foreseen by the original data producer. https://en.wikipedia.org/wiki/Data_management_plan Illustration CC BY Jørgen Stamp
- 23. What is Metadata? Slide source CC BY EUDAT (2016) | Photo: CC-BY by Cea+ http://www.flickr.com/photos/centralasian/8071729256 Metadata, literally “data about data” are an essential component of a data management system, describing such aspects as the “what, where, when, who and how” pertaining to a resource. ‹#›
- 24. Why metadata? In general, metadata should allow a prospective end user of data to: 1. identify/discover its existence, 2. learn how to access or acquire the data, 3. understand its fitness-for-use, 4. learn how to transfer (obtain a copy of) the data, and 5. learn how the data should be used. Photo CC BY-SA Jennifer Fagan-Fry (NOAA) | GBIF Metadata Profile (2011) https://github.com/gbif/ipt/wiki/GMPHowToGuide ‹#›
- 25. Data entropy Illustration from: The Loss of Information about Data (Metadata) Over Time, Michener et al, 1997
- 26. What is a «data paper»? A data paper is a peer reviewed document describing a dataset, published in a peer reviewed journal. It takes effort to prepare, curate and describe data. Data papers provide recognition for this effort by means of a scholarly article. • Getting scholarly recognition for your datasets. • Promote and improve the fitness for reuse of research data. https://www.gbif.org/data-papers
- 27. Data papers explained A data paper is a searchable metadata document, describing a particular dataset or a group of datasets, published in the form of a peer-reviewed article in a scholarly journal. Unlike a conventional research article, the primary purpose of a data paper is to describe data and the circumstances of their collection, rather than to report hypotheses and conclusions. GBIF has been working with partners in academic publishing to promote the data paper as a means of bringing credit and recognition to all those involved in data publication; to alert the scientific community to the existence of biodiversity datasets and the value they can bring to particular research projects; and as a mechanism for quality assessment and control of data accessible through GBIF and other networks. https://www.gbif.org/data-papers
- 28. Why publish data papers? ● Improve the usability (fitness for use) of your published data! ● Receive credit through indexing and citation of the published paper. ● Increase the visibility and credibility of data resources you publish. ● Track more efficiently the use and citations of your data resources. ● Receive feedback and peer-review on your dataset. ● Improve the quality of your data resources. ● Increase your network of collaborators. ● Get more out of your data resources. ● Promote your openly published datasets.
- 29. Why publish data papers? Authoring clear, informative metadata is an essential step if biodiversity data are going to be discovered and used to inform research and decisions. This involves extra work, and data publishers need incentives to do it. In the absence of such incentives, too many datasets are published with poorly-documented metadata or, worse still, no metadata at all. Data papers help to overcome barriers to authoring of metadata by providing clear acknowledgement of all those involved in the collection, management, curation and publishing of biodiversity data. https://www.gbif.org/data-papers
- 30. By publishing a data paper, you will: Receive credit through indexing and citation of the published paper, in the same way as with any conventional scholarly publication, offering benefits to authors in terms of recognition and career building. Increase the visibility, usability and credibility of the data resources you publish. Track more effectively the usage and citations of the data you publish. https://www.gbif.org/data-papers
- 32. DATA CLEANING SKILLS Corrected in GBIF in April 2013
- 35. "We are increasingly relying on machines that derive conclusions from models that they themselves have created, models that are often beyond human comprehension, models that “think” about the world differently than we do" (David Weinberger 2017).
- 36. Scientist versus machine Singularity estimated to arrive in 2045 -- 26 year from now (Kurzweil 2005) ca 2045
- 37. The future is already here — it's just not very evenly distributed. William Gibson Will our data start watching us?
- 38. Who will our students compete with in the future job market?
- 40. What is Open Science and why is it important for students? 12 June 2019, Trondheim, Norway Living Atlas Seminar