Most of the soil water content sensors today are designed to exploit the relatively stable relationship between soil water content and the real component of the complex dielectric permittivity (K') to measure soil water content. It has been shown that this relationship is affected by temperature which can therefore confound calibration equations that typically ignore temperature. Predictions of temperature effects based purely on the dielectric properties of water indicate a slightly negative temperature effect. However, positive effects have also been observed. In dispersive media it is expected that both the imaginary (K”) component and K' will increase with temperature so that knowledge of K” may be a predictor of temperature effects. We measured the effect of temperature on K' and K” measured with a Hydra Probe soil water sensor on 20 soils collected from across the USA that exhibit a wide range of properties. We found examples of both positive and negative temperature effects on K'. The positive effect resulted in a change in K' of about 4 over a 40 degree temperature change. Negative effects were about half that. For all soils there was a strong, positive, linear correlation between K” and temperature. The effect was as much as 20 times greater than that for K'. The magnitude of the temperature effect on K' was related to the magnitude of K”. When K”/K' exceeded about 0.4, positive temperature effects on K' were observed. These data are consistent with previous studies in demonstrating the close link between K' and K” in soils. Our measurements were made at 50 MHz. Although the same linkages exist at higher measurement frequencies, such as those used with TDR, the magnitude of the effects should be much less because the magnitude of K” declines with measurement frequency in this range.