The use of multivariate page test in dealing the violations of principal assumptions in completely randomized block design

Abstract

Abstract There are various variance analysis tests, one of which is the Friedman test, it is used as an alternative to two way variance analysis technique. If there are unmet assumptions in the Friedman test, that is, they are not normally distributed and do not have a homogeneous variant, then the multivariate Page test can be used to determine whether there is a median difference between k (k ⩾ 3), a median population. The Page test will be applied in the case of the growth rate of soybean plants that obtain 5 types of treatment that are conditioned, namely seedling, watering, giving 100% KL, 75%, and 50% KL Based on statistical test using multivariate page indicated that there is influence on 5 types treatment of seed conditioning i.e dry seed (control), distilled water, 100 g PEG L-1 of water, 200 g PEG L-1 of water, and 300 g PEG L-1 of water against 3 response measurements i.e germination, growth rate and growing Randomness Of Soybean.

Access to Document

Other files and links

• Link to publication in Scopus
• Open Access Version Available

Fingerprint