Problem with circular stats?

Working with the trial download, I have tried to calculate a mean direction (using Descriptive Statistics) from a series of circular compass data.
The means produced by Statistixl seemed correct for results in the quadrants 0-90deg, 90-180deg and 270-360 deg, but actual means that lay in the 180-270 (-x, -y) quadrant were calculated incorrectly by Statisti.

I can send some sample data if this will help exemplify the problem.

Many thanks,



  • Hi Matt

    Unfortunately Philip who is the primary statistician for statistiXL is away at a conference until the end of next week. If you can send a sample dataset to me though ( I will have him check through it as soon as he gets back.

    Best Regards

    Alan Roberts
Sign In or Register to comment.