Studies on sample size and number of replicates for guayule investigations
1945
Federer, W.T.
Statistical studies on data involving seven guayule strains for such plant characters as percentage of rubber, dry weight of shrub, and weight of rubber were conducted to determine the effect of size and number of samples on the experimental error. From these studies the determinations of the variance from each source are considered quite accurate in view of the large number of replicates employed. It is concluded for the 1-year guayule grown in this experiment that: 1. Significant differences existed between variety means for percentage of rubber and grams of rubber per plant. 2. The variability within the varieties for dry weight of shrub was much higher than for percentage of rubber. 3. The variability in grams of rubber per plant was nearly equal to that found in dry weight of shrub. The relative uniformity of percentage of rubber had little effect on the variability of grams of rubber per plant. 4. Little increase in precision in the experiment could be gained by harvesting more than 12 plants per sample. 5. For percentage rubber, nine replicates with 12 plants per sample gave an estimate of the standard error of the variety mean which was 1.68% of the mean. 6. For dry weight of shrub and grams of rubber per plant, a large number of replicates, 27, and a 12-plant sample were required to give an estimate of the standard error of the variety mean which was 3.95% of the mean. 7. For these data it required 24 times more plants for dry weight of shrub than for percentage of rubber to obtain a standard error that was about 4% of the mean.
Afficher plus [+] Moins [-]Mots clés AGROVOC
Informations bibliographiques
Cette notice bibliographique a été fournie par National Agricultural Library
Découvrez la collection de ce fournisseur de données dans AGRIS