I don't think they examined true randomness, but they did look at inter-astrologer consistency, and that was also no better than chance, which strongly suggests

I don't think they examined true randomness, but they did look at inter-astrologer consistency, and that was also no better than chance, which strongly suggests their answers were uniformly distributed (rather than being wrong, but everyone being wrong in the same way).