We tested the effectiveness on learning of hands-on, night-time laboratories that challenged student misconceptions in a non-major introductory astronomy class at Rensselaer Polytechnic Institute. We present a new assessment examination used to assess learning in this study. We were able to increase learning, at the 8.0 sigma level, on one of the moon phase objectives that was addressed in a cloudy night activity. There is weak evidence of some improvement on a broader range of learning objectives. We show evidence that the overall achievement levels of the four sections of the class is correlated with the amount of clear whether the sections had for observing, even though the learning objectives were addressed primarily in activities that did not require clear skies. This last result should be confirmed with future studies. We describe our first attempt to cycle the students through different activity stations in an attempt to handle 18 students at a time in the laboratories, and lessons learned from this.