The method for the determination of heavy element contents in H II regions (the strong line method) is based on the assumption that some intensity combinations of strong emission lines in spectra of an H II region can serve as indicators of its metallicity and electron temperature in a nebula. Three strong line sets are considered, namely, the set A ( R₃, R₂, N₂, and S₂), set B (R₃, R₂, and N₂) and set C (R₃, N₂, and S₂). The strong line intensities are normalized to the H_β line intensity. The sets are tested for the availability of a unique correspondence between the intensities of the strong emission lines in the spectrum of an H II region and its chemical composition. A comprehensive set of H II region models is constructed for this purpose. Some unique correspondence is detected between the chemical composition of a nebula and the intensities of the strong emission lines from the set A and C. For the strong line set B, the unique correspondence disappears for 5 % of H II region models; some H II region models with significantly different chemical compositions can have similar intensities of the strong line set B. Because of this, the variants of the strong line method (calibrations) with the use of the strong line sets A and C are more reliable than that based on the strong line set B.