SARITA RANI*, MANOJ KUMAR AND NAVISH KAMBOJ
Department of Mathematics and Statistics, Chaudhary Charan Singh Haryana Agricultural University, Hisar –
125 004 (Haryana), India
*(e-mail : saritamalik@hau.ac.in; Mobile : 94677 72688)
(Received : February 25, 2022; Accepted : March 29, 2022)
ABSTRACT
To study the joint relationships of intercorrelated characters multivariate analysis technique is
appropriate whenever several responses are measured on each object or experimental unit. This paper
describes a general procedure of performing Bivariate Analysis of Variance technique for the secondary
data collected, on grain yield and straw yield for wheat crop from Department of Agronomy at Crop
Physiology Area, Chaudhary Charan Singh Haryana Agricultural University, Hisar in plot size of 5.0 x 3.6
m using randomized block design with three replications in the rabi season (2018) using 10 treatments.
Analysis of variance (ANOVA) and Multivariate analysis of variance technique (MANOVA) were performed
on secondary data for wheat to test the significance for the inter-correlated i. e. grain yield and straw
yield characters, respectively. It was observed that in case of MANOVA technique there was a significant
effect for treatment effects and replication effects for both characters, whereas ANOVA showed significant
effect for treatments of grain yield only and replication effects for straw yield only. So, this study
interpreted MANOVA technique should be applied when more than one inter-correlated characters are
being used for testing the significance in spite of a series of ANOVAs. Further, the use of several
univariate analyses leads to a greatly inflated overall Type I error rate.
Key words : Randomized block design, inter-correlated variables, analysis of variance, multivariate analysis of variance