The impact of metaphorical language on social media engagement : evidence from the presidential run-off and major parliamentary parties during the 2023 Turkish general election

Abstract

This paper challenges the generally implied view in the literature on political language that metaphors attract citizens’ attention. We examined the influence of metaphorical expressions in social media posts on user engagement during the 2023 Turkish general elections, focusing on tweets by the two main presidential candidates who contested the run-off election and f ive major political parties that gained parliamentary representation (N = 4131). A data-driven framework measured the frequency and originality of metaphors in each post to assess their impact on likes, reposts, and replies. The results showed that metaphor frequency modestly increased engagement by 1–2%, while metaphor originality had a stronger effect, boosting likes and reposts by 6–7%. Replies, however, were less influenced, reflecting the ‘engagement ladder’ concept – the idea that as more mental effort is required, people tend to interact less. In addition, metaphor use intensified as critical election days approached, indicating strategic efforts to engage the audience. Overall, while metaphors enhance relatability and memorability, our findings showed that their actual impact on user engagement is modest and depends on contextual factors.