## Rigor and Reproducibility

The NIH (USA, National Institutes of Health) has opened a new web site on the subject, which although focused on Biomedical research, provides a good account of current trends and problems, how to overcome them and guidelines that could be easily adapted for the rest of the Biosciences including Plant Science.

# Rigor and Reproducibility

## Understanding statistics

Visualizations, specially dynamic ones can help in the understanding of statistical concepts. You can find some wonderful examples at a web site called R <- psychologist.

Do you understand hypothesis testing? and the controversy behind it?
Understanding Statistical Power and Significance Testing

Do you understand confidence intervals (CI)?
Interpreting Confidence Intervals

When you see a scatter plot are you able to guess a value for the correlation?
Interpreting Correlations

Have fun! and get some new insights about statistical tests.

## An article in Nature Tools on the “rise of R”

It is nice that R is so popular nowadays. I have been using it since 1999 or so, when Jaakko Heinonen introduced me to it, and been convinced since 2002 when I started teaching it to undergrads, that once one grasps the logic behind it, it is not difficult to use. At that time, I even developed a couple of simple packages for use in my courses. More recently, in the last three years I have been doing a lot of programming in R, as I am developing a suite of packages, while earlier I had been mostly writing simple scripts for analysing data.

R is as a programming language quite unusual and takes some time to learn to squeeze all the possibilities out of it, both in terms of performance and programming paradigms, but I have grown to like it a lot.

A very simple and a bit superficial note on R was published a couple of weeks ago in Nature. The citations record used, surely underestimates the use of R, especially early on, as it was frequent to not cite R as a publication, and least to have the “R project” as the ‘author’ but instead to mention it as a “product” in materials and methods, or to cite the paper “Ihaka R, Gentleman R. 1996. R: a language for data analysis and graphics. Journal of Computational and Graphical Statistics 5: 299–314.” In those times, even some editors refused to accept R itself as an entry in the list of references, something that did happen to me when I submitted a manuscript. The first article I published which cites R, or rather the 1996 paper on R, is from 2001, so although the first references to the “R project” may be from 2003 as mentioned in the Nature article, citations to books and articles describing R and R packages, appeared in the literature already in the late 1990’s.