Linear correlation in experimental design models applied to seed germination
Correlations between variables in completely randomized design models after the standardization of the effect of treatment are proposed to evaluate the associations. As an example, the correlations between germinability, germination duration, and germination uniformity of seeds of Brassica chinensis L. var. parachinensis (Bailey) Sinskaja and Matayba guianensis Aubl. were calculated using original data and residuals of the model. Correlations using the residuals were lower than those made with original data, showing the degree of association between variables without the interference of the effect of treatment. When significant differences are detected by multisample analysis, the effect of the treatment is obvious, but also for non significant F, the effect of the treatment is included and may modify the values of the correlation coefficient considerably.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media
Document Type: Research Article
Publication date: April 1, 2006
More about this publication?
- Seed Science and Technology (SST) is one of the leading international journals featuring original papers and review articles on seed quality and physiology as related to seed production, harvest, processing, sampling, storage, distribution and testing. This widely recognised journal is designed to meet the needs of researchers, advisers and all those involved in the improvement and technical control of seed quality.
- Editorial Board
- Information for Authors
- Subscribe to this Title
- Membership Information
- Ingenta Connect is not responsible for the content or availability of external websites