Originally Posted by kemoe
........the sensor is mounted in the mirror......
Does the manual say anything about what the accuracy is?
I was curious where the sensor was mounted. If the inside of your car can climb to over 100 degrees F on a winter day in the sun, think what the temperature might do inside your mirror. Same with the source of your reference temps. And then with a moving car vs. stationary temp sources - it may take awhile for ambient air to permeate the mirror assembly.
If you are looking for a meaningful comparison - you'll need a stable and controlled environment and a calibrated reference source. Depending on accuracy, traceable to NIST. How well insulated is your garage?
Comparing it to anything else may be part of the problem - especially if the sun is involved and/or the temperature changes quickly over a short period of time.
How important it is to you and to what expense you want to go to to find the answer is up to you. .... Remove the mirror (keep it electrically connected) and put it in an "Ice Bath" (bag of water and lots of ice). See what it reads after soaking for say 30-60 minutes. You know ice is frozen at 32 degrees F. - That becomes 'one' point for your graph/experiment.
Do this when the ambient temp is above freezing so you don't end up with a "Popsicled Mirror" (the other point on your graph deals with another known scientific point - "boiling water", 212 degrees F. But I don't recommend this or absolute zero either).