Using R in the Statistical Office: the experience of Statistics Netherlands and Statistics Austria

Alexander Kowarik
Statistics Austria
Mark van der Loo
Statistics Netherlands


Over the last decade the R language and statistical environment has received a surge in popularity. Indeed, R has become one of the central tools for modern statistics and data science. Slowly but certainly, statistical offices are introducing R as a valid tool for statistical production as well. The Austrian and Dutch offices were amongst the first national statistical institutes to approve R as a tool for production. The aim of this paper is to describe how R was introduced and currently used in our offices. On one hand, we focus on practical issues such as infrastructural considerations, the use of R-based software by non-R programmers, and update policies. On the other hand we describe the activities that were undertaken to educate new users in the use of R. We also discuss work that was contributed back to the R community in general and the official statistics community in particular. Finally, we discuss how collaboration and standardization take place in an open source environment.

Keywords: R Software, Official Statistics
JEL Classification: C80, C88

[Full Text]

Romanian Statistical Review 1/2018