Physics and the Bush Gore Election
Measurement Uncertainty, Significant Figures, & Presidential Races
Oct 28, 2008
US citizen old enough to be politically aware at the time remember the 2000 nail-biting presidential election between Al Gore and George W. Bush. Deciding the winner required recounting the Florida votes and a Supreme Court opinion. Gore won the popular vote, but Bush won the electoral college and the election.
People often expect measurements to be perfectly accurate. Physicists and experimental scientists however know that perfectly accurate measurements do not exist. All measurements will involve some uncertainty. A vote count measures the will of the people. Hence, the fundamental problem was that the vote was closer than the probable uncertainty in the vote count.
Measurement Uncertainty & Significant Figures
Any experimental measurement will have an associated uncertainty. The uncertainty can result from poor experimental technique, however it does not have to. Even with excellent experimental technique there will be some experimental uncertainty or error in the measurement. It often results from design limitations of the instruments. For example a ruler cannot accurately measure the thickness of a sheet of paper. However even with the most accurate instruments possible, experimental uncertainty will result from the Heisenberg Uncertainty Principle.
Physicists and other experimental scientists know that there is a limit to the accuracy of any measurements. Scientists therefore estimate the probable uncertainty in their measurements and report the uncertainty in their data. Results of pre-election polls show the probable uncertainty in the polling results.
Experimental scientists are trained to deal with uncertainty in experimental results. They round off data to the correct number of significant digits, so that the reported numbers do not claim unjustified accuracy. For example if the length of a table is measured with a ruler that is accurate to a tenth of a centimeter, the result is 137.3 centimeters rather than 137.3845 centimeters. To report the latter number, scientists would need a more accurate ruler. The rules for dealing with significant figures boil down to not claiming unjustified accuracy.
Scientists consider uncertainties in experimental data when evaluating results. For example physicists trying to determine which of two tables was longer would look at both the lengths and the uncertainty in the lengths. If the difference in the lengths of the two tables were less that the uncertainty in the lengths, physicists would say that the tables were the same length.
Election Results
In the 2000 presidential election it came down to the Florida results. The infamous hanging chads complicated election workers' ability to determine some voters' intentions. Some ballots may have confused voters into voting for an unintended candidate. These two effects caused an uncertainty in the vote count.
In any election there will always be a few confused voters who inadvertently manage to vote for a candidate other than the one they intended. In the Florida 2000 Bush vs. Gore presidential election, the results were so close that this uncertainty in counting the ballots was larger than the difference in the number of votes for each candidate.
A physicist would have said that to within the measurement errors each candidate got the same number of votes. Such ties are not allowed in elections, so the Supreme Court eventually declared Bush the winner.
From the perspective of a physicist the fundamental problem in the 2000 presidential election was the fact that in Florida the inevitable experimental uncertainty in the vote count was larger than the difference in the number of votes.
