Campaign Polls Influence Election Results
March 25, 2019
The Jerusalem Post – A team of researchers from Ben-Gurion University of the Negev and the Technion-Israel Institute of Technology attempted to answer some questions about voting with the help of artificial intelligence (AI).
The study, conducted by Prof. Ya’akov “Kobi” Gal, head of Ben-Gurion University’s Human-Computer Decision Making Lab in the Department of Software and Information Systems Engineering, BGU’s master’s students Roy Fairstein and Adam Laoz, and the Technion’s Dr. Reshef Meir, presented thousands of participants with a range of polling data about fictional characters and then asked them to vote based on the information presented.
Each candidate was arbitrarily awarded a value equal to a monetary amount. The candidate who was equal to zero consistently received less than three percent of the vote. When this same candidate was ranked first in the survey, more than 17 percent of the respondents voted for him.
Then, when the 10-cent candidate led in the polls, more than half of the participants chose him instead of the 20-cent candidate. This showed that people tended to vote for the candidate who was leading in the polls regardless of his monetary worth.
Given this information, the researchers built a theoretical model that would allow them to predict how each participant would vote. The model proved to be very accurate. It also gave insight into what people take into consideration when they vote.
Most people tried to gear their personal preferences in a way that would align with the candidate who was most likely to win, illustrating the “bandwagon effect,” a psychological phenomenon in which people do something primarily because other people are doing it, even to the extent that they may ignore or override their own personal beliefs or preferences, explains Prof. Gal.
“When the respondents were asked to describe their decision-making process during actual election campaigns, they claimed not to have done this, in direct contrast to their behavior during the research project,” says Prof. Gal. “It is obviously not conscious.”
Most often individuals assume that the “majority” has information or knows something that he or she does not, but which might be beneficial to know, Prof. Gal explains.
Prof. Gal says that survey conductors should take this into account because the way they present information for the April 9 Israeli election could affect people’s decision-making.
Prof. Gal also notes that AI could play a role in determining election results in the near future, but first, there are other challenges to address in this arena, such as why poll results are becoming increasingly inaccurate. On this, he says, “People are becoming more and more strategic about how they report whom they vote for.”