COTTON VARIETY PERFORMANCE TRIALS - 1999

COMPARING VARIETIES

The performance of a variety can not be measured with absolute 
precision. Uncontrollable variability influences the yield of each 
variety.  One source of variability often found is the soil not being 
uniform throughout the test site (location).  There are other factors 
that contribute to this variability.  Some of the variables are 
uncontrollable due to economic and technical reasons.  Statistical 
analysis is used to evaluate and make comparisons on what is being 
studied (yield), and at the same time measuring and evaluating the 
uncontrollable variables.  Duncan's multiple range test was used to 
compare all variety yields to each other.  Varieties that are followed 
by the same letter are not statistically different from one another at 
.05% level.  This means if two yields are followed by different 
letters, then the higher yielding variety has a 95% chance of being 
superior to the lower yielding variety.  A measurement of the 
uncontrollable variables is C.V. (Coefficient of Variation).  As the 
C.V. increases, the amount of uncontrollable variables are increasing 
and influencing the test results.

Site maintained by people at AgEBB
agebb@missouri.edu