Problem with Grove GSR Sensor

I am using a grove GSR (galvanic skin response / Electro-dermal-activity) sensor, and keep getting significantly different baseline values when not wearing the sensor. The top value is supposed to be 450 or so and the lowest is zero. But sometimes the top value (no-one is wearing the sensor) is 340 - a huge difference from 450. I know that the sensor is affected by temperature and humidity and wonder if this is causing the fluctuation. I do believe that the sensor works, and fiddling with the potentiometer on it does not make a difference in this "problem." I am wondering if this is a common problem or how others have addressed this. I also wonder if the best way of measuring true deflection is taking the absolute value of number of units moved from baseline to amplitude at the time measured. Alternatively, maybe it is best to take the relative change from the baseline to calculate the deflection under a test condition.