Usually I also think that if some module for the task at hand is available from CPAN it is the way to go...

But this time I find that calculating the mean, variance and max and min values is so simple that using a module doesn't worth the trouble:

• find the most suitable module on CPAN, sometimes really hard!
• download and install the module
• study its documentation
• model your application around it
• find bugs related to wrong module usage.
• etc.

Well, on my original post I was also really thinking on the modules used by the OP (Statistics::Basic::*) that were causing a lot of convolution on his code. Statistics::Descriptive::Sparse seems more appropiate and easy to use