A suggested statistical approach for dealing with the non-significant interactions between treatments

Abstract

A field experiment was conducted to study the effect of nitrogen (N) fertilizer and foliar appli-cation of potassium (K) and Mepiquat Chloride (MC) on yield of cotton. Seed cotton yield per plant and seed cotton and lint yield per hectare; have been increased due to the higher N rate and use of foliar application of K and MC. No significant interactions were found among the variables in the present study (N, K and MC) with respect to characters under investigation. Generally, interactions indicated that, the fa-vorable effects ascribed to the application of N; spraying cotton plants with K combined with MC on cotton productivity, were more obvious by applying N at 143 kg per hectare, and combined with spraying cotton plants with K at 957 g per hectare and also with MC at 48 + 24 g active ingredient per hectare. Sensible increases were found in seed cotton yield per hectare (about 40%) as a result of applying the same combination. However, this interaction did not reach the level of significance, so, statistical approach for dealing with the non-significant interactions between treatments, depending on the Least Significant Difference values has been suggested, to provide an opportunity to disclosure of the interaction effects regardless of their insignificance. As a matter of fact the original formula used in calculating the significance of interactions suffers a possible shortage, which can be eliminated through applying the new suggested formula.