Social scientists sometimes encounter circular variables but they may lack the knowledge of how to deal with such data. A circular scale implies a periodical measurement, e.g. the time of day. Also directions, as measured on a compass, provide periodical measurements. A final example that is common in social sciences is the so-called circumplex measurement scale.
The periodicity of the scale implies that on, for instance, a 24 hour clock, the times 23.55 and 00.05 are only 10 minutes apart. On the compass, directions at 359 degrees and 1 degree are 2 degrees apart. If such data would be treated as linear these values would seem to be almost 24 hours or 360 degrees apart, which is not correct.
In this talk we start by giving examples of studies from psychology and educational sciences that include circular variables. Then, without diving into all technical details, we will show analyses for these examples. We will focus on how to visualize the data and how to interpret results of ANOVA and regression models with circular outcome variables.