This paper is based on experiments conducted during 91/92 and 92/93 cropping seasons. The objetive was to evaluate the sample size necessary to estimate yield components of beans. Sample size was calculated considering both finite and infinite populations. Results indicated that the number of required sampling units inpractice is only praticable for a precision equal or above 10% of the mean for number of seeds per pod and weight of 100 seeds and, equal or above 20% of the mean for number of pods per plant and seed weight per plant.
sample size
