Abstract
European Parliament elections are often described as second-order national elections, in which less is at stake and parties and voters invest less. Does this hold for parties' campaigns on social media and the response of their audiences? I compare the Facebook pages of national parties in the twelve weeks before the European Parliament elections of 2014 and 2019 (151 party-elections, 96 pages, 15 countries) with party pages before national parliamentary elections in the same countries (173 party-elections, 79 pages, 75 of them also in the European sample), using posts collected through CrowdTangle. In national campaigns, parties posted about 84% more in the final four weeks than in the weeks before and received more than three times the interactions. In European campaigns, they posted about 54% more and received about twice the interactions. The European multiplier was more than a third lower (0.47 log points), and the percentage rise less than half as large. The gap held for pages observed in both kinds of election, with standard errors clustered by country, and with national elections from 2013 onwards. A larger part of the gap appears in per-post response: in national campaigns the median post drew about 40% more than before the campaign, in European campaigns about 8% more, a rise that could not be distinguished from zero. Eurosceptic and populist parties, which might be expected to treat European elections as first-order, showed no detectably larger surge, though these estimates are imprecise. On Facebook, European Parliament campaigns were real but second-order, for parties and even more for their audiences.
Cite
Hotham, T. (2026). Second-Order Campaigns on Facebook? European Parliament and National Elections on European Party Pages, 2014 and 2019. Preprint, Zenodo. https://doi.org/10.5281/zenodo.23414622