Sampling and sample size in the estimation of maize ear traits

Abstract

The objective of this study was to evaluate the effects of random and non-random sampling and sample size on estimates and variance analysis for number of grains/ear (NGE), cob length (CL) and cob weight (CW) of ten cultivars. In the random sampling the plants were selected with the aid of a random number table. In the non-random sampling, the sampler tried to be random, but no random mechanism was used. The sample sizes studied were 1, 3, 5, 7, 9, 11, 13 and 15 ears for 26 ears harvested in each plot. The experiment was arranged in a randomized block design with five replications. F-test results for both sampling methods were in agreement for NGE, but not for CL and CW. There were differences between sampling methods for NGE and CW. It was concluded that at least 11 ears must be selected for estimation of the mentioned traits. This number of plants can be easily handled and requires much less labor input than sample sizes.