Variability is an unavoidable reality. Physical phenomena such as loading conditions, material properties, and human behavior all exhibit variability. Engineers must deal with this variability when designing solutions. Unfortunately, an extensive body of human subjects research suggests that people---including engineers---consistently fail to understand variability. This deficit view of working with data is focused on statistical inference; identifying stable patterns in data.
However, engineering concerns are not identical to statistical concerns! Through a human subjects study using grounded theory methods, we have identified a novel cognitive resource that engineers use to analyze variability. Engineers can recognize the consequences of variability, and make data analysis choices that target those consequences. Targeting the consequences of variability is key to safe engineering design, but is not presently taught to engineering students.
In this poster I will describe the process of targeting variability, highlight factors that affect an engineer's proclivity to target (or not), and discuss implications for engineering education.
This work was funded by the NSF EEC under award number #2138463.
Are you a researcher? Would you like to cite this paper? Visit the ASEE document repository at peer.asee.org for more tools and easy citations.