OP it is a real problem the accuracy or resistive sensors is all over the board. It takes very very little to mess one up (bending the probe wires too sharply can do that.) Hydrofarms are not very accurate in the tests I have done on them I saw 3.5 degrees apart for 3 of the 'new and improved versions' I have been supplied with. They are the newer ones off the line from the manufacturer so I don't know how bad the older ones might be.
I have always found IR guns to read incorrectly off shiny objects and read something in the bounced line of sight. I'll post a quick set up demonstrating this.
Hosted on Fotki
I certainly don't trust accurites my experience with them was they were poor at best no better than thermometers at 1/8 the price. IR guns are great tools but if you are at an angle they really want a matte surface glossy ones seem to cause problems.
In the image there is under my hand a platinum tip thermo coupler that has beed calibrated by the manufacturer in the last 6 months it is quite accurate (better be at the cost my company paid for it and the service contract on it...) As you can see the IR gun is +12ºF (it is set to ºC but is 7 degrees off) and yet it has been checked recently (black body ir calibrater) and passed.
Personally I'd stick a bit of masking tape down and measure off that unless you can get exactly at 90º and do the ice test and see if all things jive although a resistive type sensor can be correct at freezing and off at 80ºF.
I like to use a few thermometers (3 or 5) and check them against each other and use the ones that read the closest. that might not be a great test but it is likely to be close.