Enhancing the use of soil conductivity as a co-variate to improve data analyses in small-plot field trials

Improvements can be made in the accuracy of crop growth data collected from small-plot field trials when ground conductivity is used as a co-variate in analysis of variance (AOV) models. The value of ground conductivity data is best understood in field trials where normal blocking and randomization fail to eliminate the effect of soil variation across a trial site. Comparison of analysed crop-growth means are presented from a nitrogen (N) rate trial established on a silty loam soil. An EM38 was used to collect ground conductivity data when plot soil-moisture levels were low and later when soil-moisture levels were high. Collecting EM38 data when the soil moisture was high improved the coefficient of variation, reduced the p-value and enhanced the differentiation among treatment means. Key words Ground conductivity, co-variate, analysis of variance, differentiation, P-value