Scientists from the University of Montana and the University of Calgary reported in 2007 on their use of local citizens to monitor and record observations of wildlife crossings of Highway 3 in southwestern Alberta, Canada. The data is being used to better understand wildlife movement in the area and how a planned highway improvement project might impact the environment.
This is a wonderfully useful approach for data collection in order to provide a potentially more broader look at a long-term natural pattern. What is particularly interesting about this report, however, is that is does address what will ultimately be the most critical issue for citizen science programs to overcome: how to guarantee that data collected from unregulated and potentially biased and subjective observers can be filtered into a set of data that can be considered scientific.
Citizen science data may never be considered as “real science” unless biases and unintentional errors can be monitored or filtered out. As long as the data collection sets are sufficiently large, then statistical analysis against a known, accurate sub-set of data can be used. Of course, once statistics is brought into the picture, then some generalizations are typically required, which can lessen the viability of the data.
I do believe that this is “legitimacy issue” is fundamental, and must be addressed by program leaders in the crowdsourcing citizen science projects.