voting machine manufacturers have suggested that standard quality assurance testing of the voting machines be done by independent auditors. Therefore, governments must decide whether or not to test the machines, and then whether or not to use them. While the independent auditors are very good, they have limited experience with this generation voting machine technology. Thus, an accurate voting machine will pass the test 98% of the time. A machine that is not accurate will fail the test with probability .85. As there is no valid experience with the voting machines with reliable assessments of their accuracy, and the manufacturer's own testing results are suspect, the best estimate of the probability that the machines are accurate is 0.5. The audit (test) of the electronic voting machines is expensive and cumbersome, reducing social payoff by 25. If the government chooses to do an audit of the machines, this is done prior to the decision to use the machines for the election or not.

Using accurate electronic voting machines for an election will increase confidence in the legitimacy of the vote among most citizens and officials, reduce paperwork and time and effort in compiling results, and make people more confident in using electronic voting machines in the future, providing social payoff of 100. Using electronic voting machines that turn out to be inaccurate will compromise legitimacy of the election results, outrage citizens, cause international criticism, and eliminate the possibility of using improved electronic voting machines in the future, providing social payoff of 20. Using paper voting methods are more expensive and time consuming, but highly reliable, providing social payoff of 70.

(a) Create a decision tree for this government. Include all payoffs and probabilities.