The medium is not the message, according to a study of a political campaign in Canada by Professors Torun Dewan, Macartan Humphreys and Daniel Rubenson. Their research, published in the February 2014 issue of the Economic Journal, finds little evidence that politicians vary according to their persuasive ability or that who endorses a campaign or policy matters. Political campaigns make a difference to democratic outcomes – but it is the content of the message that is key not who delivers it.
Co-author Daniel Rubenson comments:
”Our findings are surprising from the perspective of both campaigns and scholars of political persuasion.
”Campaigns devote large amounts of resources to determining who should deliver their message to voters – and analysts and pundits spend much energy examining and debating which features of campaigns are effective.
”Observers of electoral politics should take comfort in these findings. Despite a sense that modern democratic politics can often be a superficial game, our study suggests that voters pay more attention to substance than style.”
Persuasion has long been studied by economists and political scientists who seek to understand how different strategies in advertising or political campaigning lead individuals to update their beliefs and ultimately change their behaviour.
Political campaigns often seek to persuade voters of the merits of a particular alternative, be it a party, a candidate or a policy position. They do so using multiple channels: specific messages are used to enhance the profile of a candidate or position; individual canvassers are used to get the message across in a compelling way; and endorsements by public figures are used to enhance appeal.
Do such campaigns matter and which, if any, of their core elements are effective at persuasion? Is persuasion due to the message a politician delivers or due to who it is that”s speaking?
Answering these questions has stymied scholars of electoral politics and political communication. The difficulty is that politicians are smart. They tend to tailor what they say to their audience. Moreover, some politicians are particularly associated with certain messages. And political campaigns are strategic: they will often send certain politicians to say particular things in different locations.
To overcome these challenges, the authors of this study worked with a political campaign in a referendum on electoral reform in British Columbia, Canada, and randomly assigned messages, the canvassers delivering the messages and the public figures endorsing the campaign.
The findings are unexpected. The authors show evidence for a strong overall campaign effect – that is, political campaigns do matter. Both message-based and endorsement-based campaigns are effective. Indeed, they had a similar effect. Employing a message-based campaign or an endorsement-based campaign leads to about a six percentage point increase in the intention to vote yes in the referendum, compared with no campaign.
But strikingly, there is little evidence that politicians vary according to their persuasive ability or that who endorses a campaign or policy matters. In other words, despite apparently stark differences in ability, background and style, there are no differences between campaigners in their power to persuade.
Similarly, even though the public figures who lent their names and images to the campaign are from the left and right, media personalities or politicians, there is little difference in the boost they give to the campaign. Overall, the results point to an unexpectedly small role for differences between individuals working in political campaigns.
”Elements of Political Persuasion: Content, Charisma and Cue” by Torun Dewan, Macartan Humphreys and Daniel Rubenson is published in the February 2014 issue of the Economic Journal. Torun Dewan is in the Department of Government at the London School of Economics. Macartan Humphreys is in the Department of Political Science at Columbia University. Daniel Rubenson is in the Department of Politics at Ryerson University.
+44 207 955 6406 | email@example.com
+1 212 854 7431 | firstname.lastname@example.org
+1 416 979 5000 x4052 | email@example.com