Medium temperature effects on capacitance soil water content sensors were reported in the literature. Multi-sensor capacitance probes (MCP) use calibration equations that link scaled frequency of the sensor readings to actual water content. The scaled frequency defined as (Air Count – Field Count)/(Air Count – Water Count) where air, water and field refer to the frequency count of each sensor placed in air (inside a PVC tube) at room temperature (22^o C), in a water bath at room temperature (22^o C), and in the field at variable field temperatures. The objectives of this study were: i) to test the effect of medium temperature on the accuracy of MCP at different water contents ranging from dry to full saturation; ii) to evaluate the effectiveness of temperature depend scaled frequency on improving the accuracy of MCP. Four MCPs were installed in four cylindrical soil columns, 16 cm diameter by 45 cm height. These soil columns were packed with quartz sandy at different water contents (0, 0.02, 0.04, 0.06, 0.08, 0.12 and 0.38 cm³ cm^-3). Cupper constantan thermocouples were inserted in the middle of the columns near the MCP to monitor soil temperature. Columns were placed in a water bath where temperature was varied between 0 and 45^o C. Sensors and thermocouples were read via data loggers at a 1 minute interval. Results show that the apparent water content estimated by the MCP increased with increase medium temperature. It was also found that temperature effect decrease with increase water content. Temperature depend scaled frequency improved the accuracy of the MCP substantially.