edit: problem solved. Thankyou berkeman.
The I V curve of that circuit should not be linear, at least not for the first few volts of bias.
You will have to make some assumptions, but if you know the power of the light that comes out of the LED, you can work back to h.
Do you have some ideas about how you can use this equation?
[tex]E = h \nu[/tex]
I'm not sure what you should do about the assumption of how much of the power delivered to the LED (excluding the external dropping resistor of course) is converted to light, versus heat. Do you have the datasheet for the LED? That may give you that part of the calculation.
Separate names with a comma.