I would like to know what are you trying to establish here. Because if the question is that, Driving the pin in non-cohesive soils make readings are wrong? I would ask you, Do you think that making a hole of about 6 inches in non cohesive materials will not produce the same effect or disturbance. In defense of using nuclear gage, you drive a pin of about 5/8 or something like that, to the test depth and perform the reading in direct transmition, now, do you know the distance from the tip of the gage's rod to the Geiger-Müller detector in the back of the device, for instance if you drive your pin 8 inches, the distance from source to detector is about 12 or 13 inches,now I wonder how much disturbance can be introduced by the pin (remmember a little bit grater that 5/8" in diameter) driving in that length. That length is the thick of the mean that is being measured in density and it goes from source (rod's tip) to device's base.
Something that commonly happend with non-cohessive materials is that surface and top portion dry quite fast, and gage only measures moisture in top 3 or 4 inches at the most, so your readings of dry density could seem to be wrong or higher than real and obiously your compaction is also higher, well the first choice to blame always is the pin driving is affecting the readings. Remmember that it is highly recommendable to establish a correlation between nuke densities and sand cone densities even in cohesive soils when a job is starting to demonstrate that gage readings are right.
Gage is an excellent tool, but it is only that a tool, I remmeber a technician that was testing a material in a work where I was participating some years ago and he was concerned because gage readings indicate that compaction was about 75 or 79 percent in a layer, obviously every body at the job site do not even consider he's opinion, I taked him and his device to a material stockpile and asked him, do you have an idea of the compaction of this material, he answered In not sure may be 10 or 20 percent, no sir then I told him this material is about 70 or 75 percent of its maximum dry density, then we proposed a density and take some readings and surprise 76 percent, then he understand why nobody believe him, so nuke is like a calculator, give a child a calculator and then try that this child make some mental operations, probably he or she will not be able to make it, well it is the same thing with technicians, give them a nuke and Proctor results without learning how to perform a field density determination by other mean or without sure if he or she understand what are they doing and bam, you have created a nuclear monster, It's alive and hes in a job site ahhhhh!. Let's back to the basics before give a technician or even an engineer a nuke.
haha