Well, it isn't. There is no demogrpahic weighting applied, hence it isn't a snapshot of Scottish public opinion - but then again, it isn't meant to be either. I understand that some people believe folk will say whatever they think a canvasser wants to here, but I've never been convinced by that argument - otherwise you'd see 90/10 splits all over the place, and maybe there are folk like that but the size of the sample - 8,000 people, should help filter that uncertainty to a lower level.
The RIC I think, are a hugely important part of the campaign, they are highlighting and engaging with those parts of Scotland that are disenfranchised and who ar eopen to ideas about a different Scotland. If you can get and keep this demographic engaged and registered, then it could be a big stength for Yes.