Inadequate and (or) inconsistent soil sample preparation techniques (SPT) contribute to excessive variance, difficulties in soil test interpretation, and incorrect lime and fertilizer recommendations. The objective of this study was to evaluate the effect of SPT of five laboratories in Quebec (Canada) on chemical parameters measurement reliability. Samples of fine (G1), medium (G2), and coarse (G3) textured soils were collected from the surface layer. Three 500 g portions of each soil were sent to each laboratory for preparation. In addition, all samples were analyzed by the same laboratory for routine analyses. Nested ANOVA in a hierarchical model were performed with components of SPT interlaboratory reproducibility, SPT intralaboratory replicability, and intralaboratory soil analysis repeatability. Before samples were analyzed, we observed an important interlaboratory heterogeneity of particle size distributions for the same samples; due to sample preparation techniques, this can affect results of the analyses. Of all variables analyzed, the only significant, outside acceptable variations due to SPT were (1) pH_water in G1; (2) P_M-III, Al_M-III, and (P/Al)_M-III in G1 and G2; (3) K_M-III, Ca_M-III, Mg_M-III and organic matter in G3; and (4) Mn_M-III and Cu_M-III in G1, G2, and G3. The steps in SPT, mostly drying and crushing, require standardization to reduce the variance of the entire soil testing process.