Hi Alicia,
I will chime in here because I had a similar question awhile back using the same AX3 device. I reached out directly to the people at Newcastle who developed the AX3, and unfortunately they alerted me to the fact that the light sensor on the AX3 is not meant for functional use. This was a major disappointment to me too! Here is a copy of the response I received from Dan Jackson, one of the engineers on the project:
"The light sensor is perhaps only really suitable as a general, relative, indicator of light, for example to distinguish a varying/stable, or rising/falling level, or general pattern over a day. The sensor is not used with a light pipe to the surface, so light is filtered in through the case and potting material, so may have inter-device variability and a complicated response with angle where the light may be diffused, and is easily obscured by the wearer’s clothing etc. In addition, as a secondary sensor, the value has a rather complicated update cycle (updated once a second, but only stored once for every 80/120 unpacked/packed accelerometer samples).
Overall, I’m afraid it’s not suitable for quantitative measurements on its own."
Hope this helps,
Tom