As I just bought this cheap electronic load from AliExpress and there is 
so much discussion on the accuracy of this device I set up a little 
environment to get an impression. The entire procedure certainly is not 
particularily scientific at all, but given the price of the device 
itself at around 16 € (US$ 18) one can't expet much anyways - and at 
least the measurements were performed using two EEVblog BM235 DMM ;-)

I connected the load to my KORAD 3305P power supply and then stepped 
through 0,2 - 0,5 - 0,75 - 1,0 - 1,5 -2,0 - 2,5 - 3,0 - 3,5 - 4,0 - 4,5 
- 5,0 A at 12V. What puzzled me is, that the voltage readings on the 
load differed rather substancively from as well the readings at the 
power supply as the DMM. It did so no matter whether I osed the separate 
voltage measurement connector or just let the device read the voltage 
from the load connector itself. The read voltage was always too low 
whereas as well the power supply as the DMM showed 12 V (11,99 V at some 
cases). The error went up to ~ -6% at worst and -2.9% at average.

The current readings onthe other hand were pretty accurate and 
consistent throughout as well the entire spectrum and the readings of 
the power supply and the DMM. The deviation here was at 2% at 200mA, 
0,15% at 5A and 0,49% at average, which I regard as pretty good results.

Find attached the measurement data as ODS file and some additional 
information in form of partly detailed pictures of the device and the 
readings under test.

Maybe that's of any help for someone :-)

Best regards

Axel