One of the main stages of agricultural testing is experimental planning. This study aimed to propose an experimental plan, with estimates of plot size, sample size and number of repetitions in the radish crop. The uniformity sowing tests were carried out in three plots of 6.5 m in length, arranged in four rows with spacing of 0.20 m between them and 0.10 m between plants, totaling 64 plants per row. The cultivars used were Redondo Vermelho Gigante and Gigante Siculo. Each plant was considered a basic experimental unit, and the number of leaves, shoot length, root circumference, plant height, root mass and shoot mass were evaluated. For each line, the plot size was estimated by the method of maximum curvature of the coefficient of variation. Sample size estimation followed Cochran's methodology. To estimate the number of repetitions, the least significant difference from the Tukey test was used. It is recommended to use the largest plot size (13 plants) and sample size with twelve plants in the plot. For a minimum difference of 50% to be considered by Tukey's test, between six and twelve repetitions are necessary.