Bulk Mg2Si crystals were grown using the vertical Bridgman melt growth method. The n-type and p-type dopants, bismuth (Bi) and silver (Ag), respectively, were incorporated during the growth. X-ray powder diffraction analysis revealed clear peaks of Mg2Si with no peaks associated with the metallic Mg and Si phases. Residual impurities and process induced contaminants were investigated by using glow discharge mass spectrometry (GDMS). A comparison between the results of GDMS and Hall effect measurements indicated that electrical activation of the Bi doping in the Mg2Si was sufficient, while activation of the Ag doping was relatively smaller. It was shown that an undoped n-type specimen contained a certain amount of aluminum (Al), which was due either to residual impurities in the Mg source or the incorporation of process-induced impurities. Thermoelectric properties such as the Seebeck coefficient and the electrical and thermal conductivities were measured as a function of temperature up to 850 K. The dimensionless figures of merit for Bi-doped and Ag-doped samples were 0.65 at 840 K and 0.1 at 566 K, respectively. Temperature dependence of the observed Seebeck coefficient was fitted well by the two-carrier model. The first-principles calculations were carried out by using the all-electron band-structure calculation package (ABCAP) in which the full-potential linearized augmented-plane-wave method was employed. The ABCAP calculation adequately presents characteristics of the Seebeck coefficients for the undoped and heavily Bi-doped samples over the whole measured temperature range from room temperature to 850 K. The agreement between the theory and the experiment is poorer for the Ag-doped p-type samples.