Data scientists and statisticians have essentially identical skill sets. Yet over the next decade, the U.S. Bureau of Labor Statistics predicts there will be nearly six new data science jobs for every new opening in statistics. In this talk, I will first argue that the preference for data scientists is due to a difference in mindsets. Data science emphasizes collaboration and communication, which appeal to employers despite being historically undervalued by statisticians. I will then argue that this mindset can be taught at scale through data science laboratories (like those in the physical sciences) in which students address important campus problems by collecting data and communicating the results to campus experts who can evaluate the effectiveness of the solution. I will conclude with three examples from my own teaching: working with the campus transportation office to estimate the number of traffic violations, working with the campus diversity office to identify barriers to inclusion, and working with the campus sustainability office to monitor the local ecosystem.