Ok,I got the Uni-T luxmeter in and did some testing on it. I compared it to my Mobilux DIN class A luxmeter which is about the best you can get in portable luxmeters, it has a very good optical filter.
The most expensive part of a good luxmeter is the optical filter in front of the sensor that makes sure that the relative sensitivities for all wavelengths are correct, following the standard defined luminosity curve. And this luxmeter being cheap, its filter does not do miracles, no surprises here. The longer wavelengths towards red do pretty ok, but like all other cheap luxmeters that I tested (see my sigline for tests on luxmeters) it over-reads the blues, up to 8.6 times too much for 470nm.
That said, as normally white light sources are measured instead of single colour light, the error is not nearly as bad. The calibration of my copy is not far from the Mobilux meter, and for sources from warm white to good quality cool white, readings from the Mobilux relative to the Uni-T meter varied but were within 15% of each other. An old very cool XR-E reads 23% too high on the Uni-T meter though.
This very cheap luxmeter seems pretty useful for flashlight lux-measurements, but maybe not so good for ambient light measurements (see below).
Test results :
I tested a number of different led light sources, measurements were from hotspots made by reflector flashlights, except the XR-E which was in a recoil thrower and the XM-L Color which was in an aspheric lens flashlight.
From left to right is the light source, by which factor the Uni-T luxmeter reads different from the Mobilux meter, and that same difference expressed in percentages.
Dedomed cool XP-E2, 0.95, –5%
3000K 90CRI XM-L, 0.90, –10%
Cool white XM-L2 1.03, +3%
5000K XM-L2 (Nitecore P12) 0.98, –2%
Very cool old-school XR-E 1.23, +23%
Royal blue XP-E2 470nm 4.00, +300%
XM-L Color red die 625nm 1.07, +7%
XM-L Color green die 532nm 0.93, –7%
XM-L Color blue die 450nm 8.66, +766%
_
Overcast daylight is measuring different too, the Uni-T measures 27% more. That could be due to the large fraction of 400-500nm light in cool white daylight (the Uni-T may even pick up some UV, although the plastic of the dome should absorb almost all of that), but there could be other reasons too, like different sensitivities to light that comes in at an angle (mind that throw measurements are always done with light hitting the sensor dome at 90 degrees, excluding one of the potential sources of error that luxmeters have)