Polls are more than just numbers; they\u2019re a political weather forecast. When results show a drop in voter support, it\u2019s a warning signal that you might be drifting out of step with public concerns. For a politician, that\u2019s dangerous territory. Losing touch with voters can mean losing your seat.<\/p>\n\n\n\n
But how do politicians decide what to talk about when the pressure is on? They don\u2019t have a direct line to every voter\u2019s mind, so they look for signals\u2014and one of the clearest signals is the news.<\/p>\n\n\n\n
Turning to the Media Agenda<\/strong><\/p>\n\n\n\n The media plays a powerful role in shaping public opinion. The issues dominating front pages and news broadcasts often mirror the public\u2019s main concerns. When the polls go south, politicians seem to pay closer attention to these \u201ctop of the news\u201d topics\u2014and bring them into their own Facebook updates.<\/p>\n\n\n\n In our study of over 27,000 Facebook posts<\/strong> by 146 Danish members of parliament, we tracked what politicians were talking about week by week during the year 2016-2017. We matched their posts to the five issues most covered in major Danish newspapers at the same time.<\/p>\n\n\n\n The pattern was clear: when a party\u2019s poll numbers dropped, its politicians were about 17% more likely<\/strong> to post about those top media issues in the following weeks.<\/p>\n\n\n\n Figure 1<\/strong> shows this relationship clearly:<\/p>\n\n\n\n Facebook: Fast, Flexible, and Public<\/strong><\/p>\n\n\n\n Why Facebook? Because it\u2019s a low-cost, high-visibility platform. Unlike TV appearances or official speeches, it doesn\u2019t require approval from party leadership or the media. Politicians can post instantly, directly addressing both their followers and, indirectly, the wider public.<\/p>\n\n\n\n That makes it the perfect tool for a quick image adjustment\u2014showing voters, \u201cI hear you, I\u2019m talking about what matters to you right now.\u201d<\/p>\n\n\n\n Not Just a Personal Panic<\/strong><\/p>\n\n\n\n We thought the most insecure MPs\u2014those with the slimmest margins at the last election\u2014might react more strongly to bad polls. Surprisingly, they didn\u2019t. The shift towards media-driven topics wasn\u2019t a solo survival strategy; it was a herd moving. Politicians from the same party seemed to move alike when the polls turned bad, regardless of how personally vulnerable they were.<\/p>\n\n\n\n Even more surprisingly, they didn\u2019t just focus on issues in the news where the party traditionally has a strong competence reputation in the eyes of the voters. Instead, they often talked about their rivals\u2019 signature issues in the news\u2014perhaps an attempt to compete directly for attention and credibility.<\/p>\n\n\n\n Why This Matters<\/strong><\/p>\n\n\n\n This behavior shows the fast-moving, tactical side of modern politics. Social media doesn\u2019t just let politicians set their own agenda\u2014it also gives them a quick way to follow the public\u2019s concerns, especially when they\u2019re under pressure.<\/p>\n\n\n\n For voters, this means the issues you see politicians talking about online may not always be long-term priorities\u2014they might be rapid responses to the latest opinion polls and headlines.<\/p>\n\n\n\n For researchers, it\u2019s a reminder that the line between \u201cleading\u201d and \u201cfollowing\u201d the public agenda is thin. When the polls drop, even confident leaders can become followers. Previous research suggests that politicians are either \u2018leaders\u2019 or \u2018followers\u2019; however, we find that the answers depend on the quality of the polls. Losers tune in on the voters, winners stay on their preferred issues even if they are not in the news.<\/p>\n\n\n\n The Bigger Picture<\/strong><\/p>\n\n\n\n We studied a year without an election campaign, so these shifts happened during \u201cnormal\u201d politics. If anything, the effect could be even stronger during an election, when every poll feels like a make-or-break moment.<\/p>\n\n\n\n Social media has changed the rules of political communication, but one rule still applies: in politics, when you\u2019re losing, you listen harder. And on Facebook, that listening shows up fast.<\/p>\n\n\n\n Helene Helboe Pedersen<\/strong>\u00a0is a professor of political science at Aarhus University.<\/span> Her main research interest is political representation, and she has studied political representation via interest groups and political parties, from the perspective of voters as well as elites, and using qualitative as well as quantitative data. Her recent work on political representation appears in the European Journal of Political Research<\/em>, Party Politics<\/em>, and Democratization<\/em>.<\/p>\n\n\n\n Henrik Bech Seeberg<\/strong> is a professor of political science at Aarhus University. His main research interests are politicians\u2019 information processing and filtering, and political parties\u2019 competition to decide which policy issues top the political and public agenda. He currently leads a research project on youth representation and political parties\u2019 youth wings.<\/p>\n\n\n\n![\"\"](\"https:\/\/ubwp.buffalo.edu\/jopblog\/wp-content\/uploads\/sites\/234\/2025\/08\/Figure-1-1024x683.jpg\")
<\/figure>\n\n\n\n
When polls fall \u2013 marked by the vertical line on the horizontal timeline \u2013 the share of Facebook posts covering top media issues rises immediately (measured on the vertical axis). The jump in the marker after the vertical line means that losing politicians are following the media agenda more closely<\/em> after bad poll results compared to winning politicians. (Technical note: the Figure shows the differences in trends, i.e., in the average number of matches between the issue content of Facebook posts and top five media stories for the politicians experiencing a poll drop in the treatment group vs politicians experiencing a poll improvement or an unchanged poll in each of the four weeks before and after the poll.)<\/p>\n\n\n\nAuthors<\/h2>\n\n\n\n