Reproducible Science
In times of "Faked News" and "Alternative Facts" especially the reproducibility of scientific results is targeted in this learning resource.
Introduction
[edit | edit source]Reproducibility[1] is the ability to get
- (Result) the same research results or inferences,
- (Data)
- based on the same raw data or
- with data collected with the same or very similar sampling and research procedures (replicability)
- and finally
- (Algorithms)the same computer programs (e.g. Open Source R or KnitR) provided by researchers.
- (Capacity Building) optional access to capacity building material allows other scientist re-use the provide approach for their experimental design (see Open Educational Resources)
A related concept to reproducibility is replicability, meaning the ability to independently achieve non identical conclusions that are at least similar, when differences in sampling, research procedures and data analysis methods may exist.[2] Reproducibility and replicability together are among the main beliefs of ‘the scientific method’ - with the concrete expressions of the ideal/idea of such a method varying considerably across research disciplines and fields of study [3].
Learning Tasks
[edit | edit source]- Explore the KnitR concept of integration of content, data and algorithms for data processing.
- Compare the following two different scenarios:
- A researcher collects data and applies a several methodologies of data analysis (statistical and/or numerical approaches) to the data until a significat publishable result was found.
- A researcher has to publish a KnitR document (R-Markdown) prior to sampling the data. After sampling is finished the data was processed with the KnitR document and results will be published as significant or not significant.
- Analyse the WHO Document about clinical trial[4]. Explain, why is it important to publish negative results (e.g. statistical bias)! Why does the publication of negative results contribute to concept Reproducibility?
See also
[edit | edit source]External References
[edit | edit source]- Foundation for Open Access Statistics (FOAS) - Statistical Perspective on Reproducible Science,
References
[edit | edit source]- ↑ Reproducibility. (2018, January 10). In Wikipedia, The Free Encyclopedia. Retrieved 01:27, January 21, 2018, from https://en.wikipedia.org/w/index.php?title=Reproducibility&oldid=819591184
- ↑ Leek, Jeffrey T; Peng, Roger D (February 10, 2015). "Reproducible research can still be wrong: Adopting a prevention approach". Proceedings of the National Academy of Sciences of the United States of America 112 (6): 1645–1646. doi:10.1073/pnas.1421412111. http://www.pnas.org/content/112/6/1645.full.
- ↑ F., Repko, Allen. Interdisciplinary research : process and theory. Szostak, Rick, 1959- (Third ed.). Los Angeles. ISBN 9781506330488. OCLC 936687178. https://www.worldcat.org/oclc/936687178.
- ↑ WHO Statement on Public Disclosure of Clinical Trial Results, (2015) WHO-Web Portal - http://www.who.int/ictrp/results/reporting/en/ (accessed 2017/01/20)