Using Residual Analysis to Validate Rice Sowing Dates Experiment Model

Abstract

Abstract. Picking a model for a problem is a major undertaking. If the model fits well then it can be used to increase understanding of the problem and/or for prediction. Several statistical procedures have been used by researchers for planting date studies to suit their objectives. The ever-proliferation of Statistical procedures available to researchers has given room for use of diverse statistical design for research into finding optimum planting time for crops. Considering the subtle differences, advantages and disadvantage that these statistical designs pose, the results of such analysis may lead to false conclusion or be less reliable at least for comparative purposes. There is need to look at date of planting trials again to see the possibility of proffering a statistical model that could be commonly used by researchers. The main purpose of this work is to apply the residual analysis to check the suitability of the series of similar experimental model to describe the effects of sowing dates on yield of upland rice with the view of predicting optimum sowing date. Results show that the series of similar experiment methodology is able to model the changes associated with different sowing dates. The questions associated with model adequacy were discussed. Keywords: Sowing dates, rice, statistical model, residual analysis IntroductionStudies on dates of planting have been extensively conducted for several crops across different ecological zones. A review of studies on rice planting dates indicates that these research were necessitated by weather pattern [6,38]; pest management [5,18,32,36,37]; disease management [2,20]; and grain quality [9].Several statistical procedures have been used by researchers for planting date studies to suit their objectives. Obi [23] proposes series of similar experiment models for time of planting experiments. Some researchers [3,25,27], used this series of similar experiment model in their study. A survey of other works on planting date show that [35] used a randomized block design with four complete blocks to study the effects of planting dates and residue rate effect on growth partitioning and yield of corn. Also [2,18] used randomised complete block design in their planting date trials, while [28] in their trial on response of Soybean lines with "juvenile" trait to day length and sowing date combined both greenhouse and field work. In both cases they used factorial arrangement. Acikgoz [1] in his studies on effect of sowing time and planting method on rice yield per day also used factorial experiment. Fakorede [12], in maize planting date trial, used a randomised complete block design (RCBD) with a split-plot arrangement. He assigned dates to main plot, plant density to sub plot and genotypes to