I think the same caveats exist that existed months ago.
I feel Intake air temperature is vital for this experiment. A so-called "MAF" really isn't. It's a guessmatic predicticator prognostitron (tm) of flow based on cooling of a hot wire, or hot film (better). This "voltage" or "frequency" rise may be non-linear, and further scaled by the ECU based on actual factory testing with all relevant plumbing in place.
I suggest you are making a number of assumptions about air temperature, humidity, density, as well as an assumed stoich at idle [maybe?] but I wonder if this is still true on GDI IC. I think there is a component of combustion that ultimately affects VE, albiet not directly. "Load" comes to mind.
Another issue is the associated plumbing before the MAF; it's a well-known caveat of installing aftermarket Cold Air Intake systems, which change the factory tested flow, and also change the response in the very narrow region of the hot wire or film, and can cause a vehicle to run lean and/or with high fuel trims.
For useful apples-apples comparisons, I think the only VE data that's useful is under signifigant load. Otherwise, who cares about VE at idle?
Besides the GDI issue, you might also want to restrict this to Normally Aspirated vehicles. VE goes out the window with any sort of supercharging. (In a good way, I admit.) There won't be any boost at idle, at least on turbosupercharged vehicles, including diesels.
There are simply too many variables, including ECU software implementations, calibrated intake plumbing, variable intake devices, and AFR strategies.
I don't want to discourage a noble effort, but I'm just not seeing the point. There's no way you can compare the VE of a Subaru turbo 2.5 EJ205 to a Honda VTEC 1.7 at Idle and have data that is useful. The BOTH engines have characteristics (turbo, VTEC) that change VE based on rpm and/or load.
The only useful data would be an additional plot on a dyno chart showing MAF values vs. RPM, along with the intake air temp, density, humidty, and other appropos corrections.
Still, I'll run my 1996 and newer vehicles and see if they correlate to any real meaning. Most of my fleet is pre-OBD2, which is a shame because I'd love to know the VE baseline of my '72 BMW 2002tii M10 racing engine.