A new study suggests that perceived political polarization in the USA is influenced by a focus on difference, rather than similarities.

In just over three months, US voters will be asked to choose their new President. For most, the question is binary: Democrat or Republican? At the time of writing only 20 percent of independent voters are reportedly still undecided about who they will vote for. It's a choice that has come to polarize the American population. In 2019, eight in ten US adults saw the divisions between the parties increasing and expressed concern over this trend. The lines have been drawn, or so it seems. 

This is a story about how different ways of presenting the same data can strongly influence people’s opinions.

“There’s a general perception that Republican and Democratic voters are deeply divided but our new study suggests these differences of opinion between the two groups are greatly exaggerated and that the groups are actually very similar in the values they hold as important,” said author Dr Lukas Wolf from the Department of Psychology at the University of Bath in a statement.

In their study the team set out to investigate how presentation of similarity in values impacts voters' perception of their differences. Dr Wolf believes that “Generally, the media and social scientists compare groups by describing differences, for example by using barplots, which ignore overlap and show exaggerated divisions in US society.”

The researchers used data from a previous study polling Democrats and Republicans on their values, and presented it to the participants in one of two ways. One group was shown the results as overlapping distributions, thereby highlighting the similarities. The other group was shown bar plots of the mean of the distributions, with the y-axis truncated to strongly emphasize the differences. 

The participants who were shown distributions felt that Democrats and Republicans were more similar, and had more similar values. Importantly, this “increased their sense of hope about the future and inspired trust in people they have seen as fundamentally different to them,” said Dr Wolf. Perceived similarities also increased their perceived potential for compromise.

Unfortunately, the group that was shown the truncated barplot performed similarly to participants who were shown no data. This shows that people’s baseline assumptions are in line with large differences between the two parties. 

Dr Wolf added: “This perceived polarisation has detrimental consequences because people expect the worst from interactions with the other group and actively avoid them. It also means people become pessimistic about the future because the division in society stands in the way of compromise and cooperation for a better future.”

There is great responsibility in how data is presented to the public, which can have a societal impact.

This study is published in Social Psychological and Personality Science.