Didn't realise this thread was still going! :-)
Phone app SPL meters are notoriously inaccurate and unreliable. The iOS ones tend to be better just because there are fewer variations and so less for the developer to check. Android and other mobile OS apps will generally be all over the shop.
Some apps do have bespoke calibration modes for specific popular phones, but there's nothing in them to take account of normal production tolerances, so they are still not particularly accurate. And the dynamic range of the MEMS chips and associated circuitry used as phone microphones is not great, either, so high SPL readings are even more suspect...
But, if all you need is a means of making comparative measurements at moderate SPLs then they are absolutely fine. Don't think of the displayed numbers as precise SPL values, and instead just match the readings as appropriate so that you can ensure one system is the same as another -- if that's a function you require (such as when comparing speakers).
The important thing, though, in a project studio, is not the absolute reference SPL, but just the fact that you have a repeatable reference so you can always set your monitoring level to the same value whenever you start or pick up a mix.
So the technique I recommend for that kind of application is simply to juggle the interface output level / monitor controller / speaker (or amp input) level to set the volume at a comfortable analytical level when replaying a mix-in-progress with a sensible headroom margin. Mark the volume control positions so you can easily and quickly recqll that working monitor level.
Then -- and only then -- play some band-limited (500Hz - 2kHz) pink-noise at -20dBFS rms and check the SPL on your phone or other SPL meter, and make a note of the number (and the app/phone you used) in case you need to reset the system to the same level again later. There is very good pink-noise file available for free on the Blue speaker website.http://abluesky.com/support/blue-sky-calibration-test-files/