This paper presents the first results from draw-a-scientist tests (DASTs) over five years that were used to measure the effect of 8-10 week long astronomy clubs and week long summer camps on 3rd-5th grade elementary school students' perceptions of scientists. We facilitated these DASTs prior to these clubs or camps, which provide a baseline for a student's initial conception of scientists, and once at the end, to determine whether their conception changed, possibly as a result of their involvement. In total we analyze 89 pairs of DASTs using a numerical grading scheme designed to quantify the presence of various features in the drawn scientist and their activities. We find that there is a gender imbalance in both the pre- and postclub drawings, with only 32% and 35%, respectively, of students drawing female scientists. We also find that a third to a half of the scientists have a stereotypical appearance and/or are performing stereotypical activities. Although we find insignificant changes (<5%) in most categories, we do find an 8% increase in the number of scientists that have a stereotypical appearance, which is worth following up, but a significant 12% decrease in the number of scientists who are performing stereotypical activities. In addition, we present some possible improvements to implementing DASTs and discuss other possible assessments that could provide a more direct method of gauging the effect of these astronomy clubs or camps.