University of Limerick Flag Poles
Tuesday, February 22, 2022

A study by researchers at University of Limerick has revealed that the ideological gap between Republicans and Democrats has consistently widened over the last three election cycles in the United States.

The analysis of US election data shows a trend of increasing polarisation and a growing ideological distance between Republicans and Democrats from 2012 to 2020.

The data came from the American National Election Survey who recruit around 5,000 participants prior to each US presidential election.

The UL research found that ideological distance and polarisation increased with each election cycle since 2012, and this increase was particularly striking from 2016 to 2020.

Using the data, the UL researchers employed network analysis to reveal that there is now less common ground between the two major US parties on key election issues than there was in the previous eight years.

Lead author of the study, published in the journal Analyses of Social Issues and Public Policy, Alejandro Dinkelberg, a PhD student in the Department of Mathematics and Statistics at UL, said the researchers had analysed the US data using a network of people connected by shared agreement on key political identity issues such as abortion, immigration, gay marriage and other items.

The UL research also reveals that within the two parties over the three elections Republicans have diverged and Democrats converged. It shows that Republicans have less agreement on their responses to these items in 2020 than 2012, indicating they were less cohesive.

“It is interesting to see that the Republicans are spread out and move further away from each other,” UL researcher Alejandro Dinkelberg observed.

“The opposite is the case for Democrats; their attitudes align more when it comes to questions about the US government’s influence on welfare distribution or on how strong to regulate business’ influence on the environment.”

The researchers show that the trend to 2020 illustrates that Democrats and Republicans in this sample become more distinct from one another, but also that Democrats generally become ideologically more similar to each other, while Republicans become more scattered, possibly showing that Trump-era politics resulted in polarisation within the Republican identity as well as between Republicans and other parties.

“We might not see political attitudes drift off to extremes, however, we can observe missing common ground between Republicans and Democrats,” clarified Dr Pádraig MacCarron, a postdoctoral researcher at the Centre for Social Issues Research and Mathematics Applications Consortium for Science and Industry (MACSI) at UL and a fellow author on the study.

“Polarisation and the hardening of attitudes could put brakes on America’s ability to find popular solutions to pressing social issues,” Dr MacCarron added.

The authors believe that the novel network-based approach employed in the study provides a unique way of analysing opinion-based identification and polarisation from opinion surveys.

“It is especially useful for identifying polarisation without extremism and, in the future, could be applied to identify unifying dimensions of agreement that cross party lines,” the authors argue in the paper.

The authors further believe that the methods used to convert survey data on certain attitudes offers an insight into group formation. With that, we can deepen our understanding about how different group interests embed within society.

This interdisciplinary, data based study uncovers trends and make us aware of issues we might address now or in the future, the researchers explained.

“Displaying attitudes and people in one network helps us to catch the multidimensional pattern of people’s identity,” explained Dr Mike Quayle of UL’s Department of Psychology, who was also an author on the study.

“We can see how closely people are linked together and with whom they are likely to engage with. Nowadays, people are more likely to connect with people with similar opinions due to the omnipresence of opinions, comments and identifications in online social networks like Twitter.

“We want to understand how this happens and what the important influential factors are,” Dr Quayle added.