Salstat now does box plots
Okay, what are box plots? They are a chart type where a box with lines describes a variable. First, the box: this is a plain box with a line in the middle. The line denotes the median: the upper part of the box denotes the 3rd quartile and the lower part of the box denotes the 1st quartile. The upper line, or whisker
, denotes the maximum value and the lower whisker denotes the minimum. See the chart picture (from Salstat!) below to get an idea.
They are really useful and simple way of understand how a variable is distributed at a glance.
And the best bit is that after lots of intensive work, Salstat now does box plots!
While this needs testing (and if you find anything that doesn't work, head on over to GitHub issues
and let us know), it seems to work fine so far with multiple variables both independent and dependent. The example above was for two dependent variables grouped by a single, 2-level independent variable.
Salstat's box plots do not yet calculate and display outliers just yet, and we're hoping to find a way for the different denotations to be customised (so you could, for example, have a box plot denoting the median, the 3rd and 7th deciles as box limits and the 5th and 95th percentiles denoted by the whiskers).
The latest code is available at Salstat's GitHub repository
for download (installation instructions included in the Readme.md).