4:22 AM The desire to increase the reproducibility of gold in hydrological research | |
In 6 influential journals on hydrology and Aqua resources, posted in 2017, the estimated fraction of studies, the results of which have all chances to be fully reproduced, was only 0.06 - 6.8 percent. This low degree of reproducibility is not considered a rarity in hydrological studies—a fact that almost all scientists are happy to confirm. Not least, the team of researchers from the Institute of Utah have discovered a conclusion, in order to make these studies more reproducible. In his own note "assessment of data availability and reproducibility of studies in hydrology and Aqua resources", posted in February. 26 in Nature's Scientific Data, David Rosenberg's co-authors came up with an online survey tool to assess the reproducibility of hosted studies. The group saw 360 notes from 6 magazines on Aqua resources posted in 2017. Of the 360 notes, they were able to fully reproduce only the results of 4 notes. "Our research tool breaks down the theory of scientific reproducibility into certain components of data availability, reproducibility of results and reproducibility of results," said Rosenberg, associate Professor of civil and environmental engineering at USU. "After that, we suggest, as the creators, magazines, sponsors and companies have a chance to apply the tool of sample survey for increasing low reproducibility of the characteristics." The creators are talking, in fact that the reproducibility has the ability to be divided into 3 components:
The group's online sample survey tool is made up of 15 questions and has a checklist of key components important for relic accessibility and reproducibility. Artifacts are a General term for all data, software, models, code, directions, and other materials that are important for playing back the results of a study. The panel found that, in fact, within 70 per cent of the selected notes, some material was available, but only within 48 per cent of the material was available online. Only within the limits of only 6% of the selected notes created artifacts publicly available, and only once the percentage of selected notes arranged artifacts cheap and had the opportunity to be fully reproduced. The creators said, in fact, that in many notes there are no instructions for obtaining the results. If the creators give instructions, they talk, the number of notes that could be found out on the reproducibility will increase twice. Notes that created all the artifacts cheap, had 6 out of 10 chances to play some or all of their results. 2 magazine interviewed by the panel urged taking notes, in order to show how it is possible to gain access to relics, and 4 magazine encouraged statements. Almost no magazines are not strongly asked from the creators of providing all the relics. The survey tool has the ability to help distinguish and encourage creators to achieve specific reproducibility values. For example, the creators have a good chance to use the sampling tool for self-conceit reproducibility of their own results. Rosenberg and his team also advise a system of medals to determine all possible values of reproducibility:
Rosenberg advises to position the icons of medals near online articles, in order to make the creators for their reproducibility and to simplify the readers intelligence the best practices reproducibility. Rosenberg and his team won 4 silver and 6 bronze medals from the 360 notes they discussed. The awarding of gold medals for the replicability of finds remains a necessary direction of future work. "We trust that the sampling tool will undoubtedly help to direct creators, magazines, sponsors and institutions to arrange scientific work more reproducible," Rosenberg said. "We welcome the discussion on improving the study tool and increasing the reproducibility of our science." | |
|
| |
| Total comments: 0 | |