Berkeley Professor Leads Nation’s First Statewide Risk-Limiting Election Audit

Philip Stark, associate dean of mathematical and physical sciences and professor in the department of statistics at the University of California, Berkeley, led the country’s first statewide risk-limiting election audit in Colorado this November. At a press conference announcing the implementation of the post-election audit, Stark met with Colorado’s Secretary of State Wayne Williams, who explained Stark’s newly developed auditing process with a bag of dice and hat full of names to emphasize randomness in selection.

"When you enter into a risk-limiting audit, you assume that the answer is wrong ... that the reported winner did not really win," said Stark. "After that, you start collecting evidence, and if the evidence is convincing enough that the winner did actually win, the process stops."

With many in and out of the government questioning the validity and security of election results, Stark hopes other states will follow suit. Helping to make that happen is software company Free and Fair, which developed an open-source technology program operating with Stark’s new method that can be used by all states.

Stark is considered an authority figure in the field of election auditing, having served as an expert witness and providing testimony for high-profile legal cases. His research preferences include uncertainty quantification and inference, inverse problems, nonparametrics, risk assessment, earthquake prediction, election auditing, geomagnetism, cosmology, litigation, and food/nutrition.

In 2010, the ASA Board of Directors endorsed risk-limiting audits, recommending they be routinely conducted and reported in all federal, most state-wide, and at least a sampling of other governmental election contests. In the same letter, the ASA urged state and local officials to seek statistical advice about how to sample and analyze data to efficiently attain the desired level of risk control.