The effect of module averaging on the HVI short fiber index measurement was evaluated on one hundred and forty eight modules randomly selected from the 2001 U.S. cotton crop. In addition to collecting module data, HVI data on over 115,000 samples, also randomly selected from the 2001 U.S. cotton crop, were used to establish the “normal” relationship of HVI short fiber index to upper half mean length and uniformity index. The normal relationship between the three measurements is strong as evidenced by R² = 0.99. As a result of this strong relationship, in order to be of value, the short fiber index measurement must be capable of providing accurate information on whether or not a cotton sample exhibits an abnormal amount of short fiber content for its length and uniformity index. Otherwise, the more precise measurements of length and uniformity index are actually more effective in predicting short fiber content.

The normal relationship of short fiber index to length and uniformity index was quantified and used as the basis for normalizing the module averaged short fiber index measurements. Increased precision resulting from module averaging enabled effective normalization. Given a tolerance of 1.0%, short fiber index, reproducibility between HVI’s for single bale testing averaged 57% and for module averaging averaged 95%. Of the 148 modules in the study, the detected abnormality of short fiber contents ranged from 15% more short fiber than normal to 15% less short fiber than normal.