Full Length Research Paper

A multivariate analysis of variance (MANOVA) of the performance of sorghum lines in different agro-ecological regions of Zimbabwe

D. Maposa^1*, E. Mudimu² and O. Ngwenya³

1Department of Statistics and Operations Research, University of Limpopo, Turfloop Campus, Private Bag X1106, Sovenga, 0727, South Africa.

²Department of Decision Sciences, University of South Africa, P. O. Box 392, UNISA, Pretoria, 0003, South Africa.

³Department of Applied Mathematics, National University of Science and Technology, Zimbabwe.

*Corresponding author. E-mail: danmaposa@gmail.com. 

Tel: +27(0)152682792/2435.

Accepted 7 December, 2009

   Abstract

This study presents a multi-environment trial to evaluate the performance of sorghum lines in a target population of environments in Zimbabwe. The study attempts to determine whether genotypic variation and/or genotype-environment interaction have a significant effect on sorghum performance. It also attempts to determine whether the promising sorghum lines perform better than the established varieties with the aim of selecting promising sorghum lines of superior performance. To analyse the data, multivariate analysis of variance is used in this study. Results show that both environment and genotypic variation contribute to differences in sorghum line performance. Results in two of the selected sites (Kadoma and Matopos) suggest that sorghum lines significantly differ in their performance due to genotypic make-up. It is concluded that environment is the major contributor to differences in sorghum performance though genotypic make-up also play a part. Provisionally, promising lines of superior performance are NL9411 and NL9907.

Key words: Multi-environment, MANOVA, resettled farmers, sorghum production, genotype-environment.