ANOVA (General Linear Model) was used to test for differences among treatments, and pair-wise comparisons were made using the Tukey-Kramer multiple-comparison test. We initially tested for differences in plant metrics among treatment replicates (three replicates per filter treatment) and found none, but did find highly significant differences among individual plants within container racks due to the red/far-red ratio growth response of plants (tallest plants were in the center of the rack, smallest on the outer margin). Due to this, data were blocked by treatment and by position in the cone-tainer racks (1 - 25); highly significant position effects were found (p < 0.00001). Linear regression was used to test for significance of various trends. NCSS Statistical Software (NCSS, 2007; release Aug. 2007; NCSS, Orem, UT, USA) was used for all statistical analyses.
Was this article helpful?