Is there enough data on that to be statistically valid?

Are a truly random sample of children brought up by gay parents? I'd expect some large selection bias there.