A lot of statistical analysis and R packages were pioneered for/by ecologists. In ecology we get some of the more complex and interesting data, and questions to be asked of that data. For example, the popular R package vegan was made for ecological analysis.
My sense is that ecology tends to have more on spatial covariance (along with epidemiology), psychology on building scales out of latent factor models, and economics on temporal covariance.
(although there is substantial overlap between the three)
It's also worth mentioning that a lot of fields that you wouldn't normally think of as ecology use ecological tools like the vegan package. A good example is studies of the human microbiome. Studying the microbiome you want to measure the biodiversity of various samples from diseased and healthy individuals -- which is basically the same problem as traditional ecologists face when looking at, say, a polluted versus pristine environment.
> I’m a bit of an R nerd. Ok, that’s a lie; I’m a major R nerd. But for good reason, because R is incredibly useful in streamlining the scientific process, and increasing the ability to replicate findings with less human error.
If only we could make the spreadsheet not get used by most scientist it would be a better place.
That's why programs like Data Carpentry and Software Carpentry, mentioned in the article, are so important. http://www.datacarpentry.org. Most universities actually don't have good opportunities for active researchers (graduate students, postdocs, professors, research staff) to learn programming or effective data skills. But, researchers are really looking to learn the skills. https://www.embl-abr.org.au/news/braembl-community-survey-re... DC and SWC help fill that gap by teaching two-day hands-on workshops that meet researchers where they are to give them the skills and perspectives to get out of spreadsheets and into R and Python.
I'm no big fan of excel, but the modern xlsx format is "open", in that it's documented (though maybe not perfectly). It certainly is possible to read data from it using a variety of open source packages, from LO Calc to libraries like pyxl.
It's not that it's open it is that all the steps are seen and can be repeated. Excel is just manual labor without seeing how the person took the raw data and got to their conclusion.
I don't find the title misleading, but I think it's worth noting that the message isn't very R- or ecology-specific. The message is something like "if you do science, you should use scriptable data processing tools."
edit: I guess the message I described is on top of the more specific message that R, specifically, works well for ecology, specifically.
