Filling in the blanks

Some worksheets contain cells that have been left blank, e.g., in order to make the headings and subheadings easier to read. However, if for example you want to sort or filter the list or you want to use the table for analyses, you need to fill in the blanks. You could manually copy the value from the first filled cell above the blanks. An easier and faster way to do this is described on this site. The technique they describe requires the Go To function, which unfortunately is not available in Openoffice’s Calc. However, there is an alternative way (which works also in Excel). Continue reading “Filling in the blanks”


QGIS 1.4.0 released

Quantum GIS, an open source GIS, has released a new version last week; QGIS 1.4.0 ‘Enceladus’. It promises some welcome new features, such as New symbology infrastructure, a field calculator, an improved map/print composer, user interface improvements, a render caching capability, and the option to define SVG search paths enabling you to specify your own directories containing svg images for use as symbols. See here for more details.

Which text editor to use with R?

Although some users would prefer a graphical user interface for R, the arguably best way to work with R is through the command line. This can be done directly in the R console, but often it is more convenient to write scripts in a separate file, instead of typing them directly in the console. This requires a text editor, and you can of course use the default one that comes with your OS. However, there are text editors that offer different levels of integration with R. Below are the ones I tried. Continue reading “Which text editor to use with R?”

Why a GUI for R isn’t always a bad idea

The arguably best way to work with R is through the command line. Some even argue that the use of a graphical user interface (GUI) should be completely discouraged. And that is where I disagree. The command line can be quit intimidating for the new or occasional users, who may therefore benefit from a GUI. There are a number of different GUI available, for example R commander, JGR, Deducer (which includes a very nice plot builder), RKward, Rattle (geared towards data mining) and BiodiversityR (focuses on biodiversity analysis). None of them offer the same sophisticated graphical interface as e.g., S plus or SPSS, but they are in general easy to use and quit suitable for the more common type of analyzes. What they have in common is that while providing you with an graphical interface, they also show you the commands generated via the point and click dialogs. This helps in getting to know the R syntax and thus simplifies the learning curve of R. Continue reading “Why a GUI for R isn’t always a bad idea”