THE PROBLEM OF PLOT SIZE IN LARGE-SCALE YIELD SURVEYS
成果类型：
Article
署名作者：
SUKHATME, PV
刊物名称：
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
ISSN/ISSBN：
0162-1459
DOI：
10.2307/2280658
发表日期：
1947
页码：
297-310
关键词：
摘要：
For estimating crop yield from the yield of sample plots, large plots of 0.01 acres or more are used in some regions and very small plots of 0.0003 acres or less are employed in other regions. Three empirical comparisons of large and small plots in India suggest that the results from small plots are biased and tend to overestimate average yield. The bias evidently arises from uneven sowing and the inability of the sample-collectors to decide which of the plants on the borders of plots should be excluded.