Abstract
The Williams & Ratcliff (1980) and Zelen (1957) methods for the analysis of covariance in incomplete block designs with recovery of inter‐block information are compared using data from a balanced incomplete block design. The former is shown to be more efficient. Other advantages of the Williams & Ratcliff formulation are also presented.
Original language | English |
---|---|
Pages (from-to) | 337-341 |
Number of pages | 5 |
Journal | Australian Journal of Statistics |
Volume | 26 |
Issue number | 3 |
DOIs | |
Publication status | Published - Sept 1984 |
Externally published | Yes |