New Zealand Journal of Agricultural Research abstracts

---

Should sequential designs be used for comparative experiments on animals?

P. D. JOHNSTONE

Invermay Agricultural Centre
Private Bag 50034
Mosgiel, New Zealand

Abstract  There has been some recent interest in sequential experiments as a means of reducing the number of animals used in comparative experiments. This paper describes sequential methods and explores their properties. Comparisons are made with fixed sample-sized experimental designs which result from traditional power calculations. Attention is drawn to potential problems in interpreting and managing sequential experiments and it is concluded that fixed sample-size experimental designs are still justified on both ethical and financial grounds for comparative experiments, particularly when estimates of the size of the response and their standard errors are important.

Keywords  sequential experiments; Sequential Probability Ratio Tests; SPRT; comparative experiments; animal ethics

New Zealand Journal of Agricultural Research, 1998, Vol. 41: 561-566

0028-8233/98/4104-0561 $7.00/0 (c) The Royal Society of New Zealand 1998

PDF file of entire paper: medium quality (379K); (scanned from paper original: notes about this process)

---

This year's abstracts | Journal home page | All abstracts | Publishing home page