Ah yes, sorry, the OpenStudio Radiance measure does not compute daylight factors (because DF is not a very good daylight metric). Sorry! You could definitely calculate the external horizontal non-obstructed illuminance from the weather input file, but the Radiance measure also reports that out in the output file. Take a look in your run/[radiance_measure]/radiance/output/ts directory; in there are directories for each space in your model that had an illuminance map. Open any one of those in a text editor and you'll see from the header that the file stores the direct normal illuminance and global horizontal illuminance for each timestep. The the format is:
month,day,time,directNormalIllumimance(external),diffuseHorizontalIlluminance(external),daylightSensorIlluminance,pointIlluminance [lux]
e.g., this data:
1,1,08:00:00,6798.722108670218,2557.6913763455414,22.1,27.0,87.0,27.6,9.7
Indicates that on January 1 8:00 a.m., there was 6,799 lux from the sun (direct normal component) and 2,558 lux from the sky (diffuse horizontal component) recorded by the weather station, and five points inside the space received between 10 and 87 lux.
Note you cannot simply add these two values together for illuminance on the horizontal plane, as the direct normal value is recorded by a sensor that is normal to the sun. You have two options: A rough approximation would be to multiply direct normal illuminance by the cosine of the solar angle, and then add that to the global horizontal value. The other option is to do a separate calculation using this input data (using any of a number of tools/formulae).
We really should add external unobstructed illuminance (and perhaps even daylight factor) to the Radiance measure output. It's just not a very good metric, IMO. At the risk of sounding like a smartass, is your thesis attempting to illustrate DF's shortcomings? Because any daylighting evaluation based solely on DF in anything but environments of 100% diffuse celestial illumination (which do not exist in real life) would be highly suspect. This is one reason why other daylight metrics have been developed, such as daylight autonomy (DA), continuous daylight autonomy (cDA), and useful daylight illuminance (UDI) -- metrics that OpenStudio's Radiance measure does calculate.
