Sample size for the estimation of Pearson’s linear correlation in crotalaria species

Tamanho de amostra para estimar coeficientes de correlação linear de Pearson em espécies de crotalária

Abstract:

The objective of this work was to determine the necessary sample size to estimate Pearson’s linear correlation coefficients of four species of crotalaria at precision levels. The experiment was carried out with Crotalaria juncea, Crotalaria spectabilis, Crotalaria breviflora, and Crotalaria ochroleuca, during the 2014/2015 crop year. Eight crotalaria traits were evaluated in 1,000 randomly collected pods per species. For each species, the correlation coefficients were estimated for the 28 pairs of traits, and the sample size necessary to estimate the correlation coefficients was determined at four precision levels [0.10, 0.20, 0.30, and 0.40 amplitudes of the 95% (CI_95%) confidence interval] by resampling with replacement. The sample size varies between crotalaria species and, especially, between pairs of traits, as a function of the magnitude of the correlation coefficient. At a certain precision level, the smallest sample size is required to estimate the correlation coefficients between highly correlated traits and vice-versa. To estimate the correlation coefficients with CI_95% of 0.20, 10 to 440 pods are required, depending on the species, pairs of traits, and magnitude of the correlation coefficient.

Index terms:
Crotalaria; linear relationships; resampling; sample precision