We present a discussion of efforts to obtain external validation of the estimated atmospheric parameters (Teff, log g, [Fe/H]) obtained from medium-resolution (R = 2000) SDSS spectroscopy and ugriz photometry, which are being employed for both the completed SDSS-I and the ongoing SEGUE survey. The SDSS/SEGUE spectroscopic pipeline makes use of a number of methods for the estimation of each parameter, with estimated internal errors on the order of σ(Teff) = 150 K, σ(log g) = 0.4 dex, and σ([Fe/H]) = 0.3 dex. Over the course of the past two years, we have obtained over 100 high-resolution optical spectra of SDSS/SEGUE stars using the HET, KECK and SUBARU telescopes. For the KECK/HIRES spectra, which have R = 40000, we have performed standard high-resolution analyses to estimate the stellar parameters. For the HET and KECK-ESI data, which have R = 15000 and R = 5000, respectively, we have performed synthetic spectra matching in order to to estimate the stellar parameters. We find that the derived stellar parameters agree well with the SDSS/SEGUE pipeline estimates for the temperature range 5000 K < Teff < 6500K; the errors are of the order of the internal errors expected from the SDSS/SEGUE pipeline. For effective temperatures in the range 4000 K to 5000 K the estimated parameters from the high-resolution spectroscopy exhibit offsets relative to the SDSS/SEGUE pipeline values on the order of ΔTeff = 200 K, Δlogg = 0.8 dex, and Δ[Fe/H] = 0.4 dex. Similar offsets exist for stars with T > 6500 K. The main reason for these offsets appears to arise due to varying microturbulence, for which the medium-resolution SDSS spectra are not sensitive. We also have performed external checks on pipline radial velocities. We find that the errors in radial velocities are on the order of 7 km/s for stars, which is at the expected level.