It is an interesting idea, to study SCA participants this way. But you're right about needing more porrf-reeding. Besides that, there's the classic sociological study flaw of using self-selected participants. The author himself mentioned this as a reason for the low response rate, but completely missed that it might skew any results.