Statistically speaking

We have been investigating data requirements for the Obesity e-Lab project recently.  This required some analysis of data created by the stata stats package.  However, we had to read it using the R tool instead (firstly because we can then use Taverna workflows to execute R scripts and secondly because Stata is not free).  After some head scratching we figured out that you had to use the ‘foreign’ library and the read.dta function.  This mean turning on the foreign library in the Package Manager using the R gui first and then executing the following command:

read.dta(“filename”, convert.dates = TRUE, tz = NULL, convert.factors = TRUE, missing.type = FALSE, convert.underscore=TRUE, warn.missing.labels=TRUE)