Answers are grounded in this paper and cite their sources.
OWING to rather limited facilities for handling experimental stock it is often necessary to make use of the repetition method of experimentation to obtain records on sufficient numbers that the results may be statistically accurate. When such gross measures as body weight or annual egg production are employed, there is special need for large experimental units. Studies at the Massachusetts Station indicate that there are no less than thirty phenotypes with respect to characteristics vitally affecting egg production. In order to obtain a reasonable distribution of these phenotypes in the different pens it would be necessary to carry units of no less than 90 to 100 birds. Wood and Stratton (1908) suggest a useful method for determining the number of experimental animals necessary in each lot to secure satisfactory odds that the differences in means will be significant. Data reported by Martin in Kentucky Station Bulletin 260 (1925) on page . . .
No takes yet. Share an insight, caveat, or question.
F.A. Hays (1932) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: