The calculation of R(2) and the test of significance for the least squ
ares regression solution through the origin are not well documented in
textbooks. Many of the statistical packages compute these statistics
in such a way that the researcher may easily, though inadvertently, be
misled. This paper contains a detailed discussion, supplemented by an
example, of the formulae required to properly calculate these statist
ics. Also, recommendations for correcting misleading computer results
are presented.