The term horse-race coverage refers to one of the most prominent types of election coverage (e.g., Schmuck et al., 2017) that strongly focuses on winners and losers (who is ahead?). Typically, it is related to opinion polls and/or election outcomes. Quite often also “a language of war or games to describe the campaign” (Aalberg et al., 2012, p. 167) is involved in this kind of news stories, although – in a narrow sense – this aspect does not seem to be an essential part of the concept of horse-race coverage (e.g., Banducci & Hanretty, 2014). Regarding the conceptual definitions, a development in the terminology may be noticed: “The original horse race news became part of the game frame which was later discussed as part of the strategy frame.” (Aalberg et al., 2012, p. 166) In other words, the term ‘game frame’ is sometimes used synonymously with ‘horse-race’ coverage (although some scholars discuss whether these concepts can actually be used synonymously; e.g., Banducci & Hanretty, 2014; de Vreese 2005; Valentino et al., 2001). Field of application/theoretical foundation: Horse-race coverage is a very popular concept that is analyzed in research on the media coverage of politics, especially in times of elections and election campaigns. References/combination with other methods of data collection: The analysis of horse-race coverage may be combined or compared with opinion polls and election outcomes. Furthermore, experimental studies that analyze potential effects of the horse-race coverage on recipients (e.g., political cynicism) exist (e.g., Lavrakas et al., 1991; Valentino et al., 2001). Example: Although often analyzed, the operationalization of horse-race coverage in quantitative content analyses differs. Aalberg et al. (2012) review existing concepts and operationalizations and provide a set of coding instructions, which are cited below. Coding instructions (direct quotation) by Aalberg et al. (2012, p. 177): Game frame [respectively horse-race coverage] Does the story deal with opinion polls and politicians’ or parties’ standing in the polls? This variable has two codes: 0 = no, 1 = yes. Coders should type 1 if the news story at least once mentions opinion polls and the standing of political parties or individual candidates in these. Coders should also type 1 if the news story includes references to generic ‘polls’ or ‘the opinion’ and the standing of political parties or candidates according to ‘polls’ or ‘the opinion’. Otherwise coders should type 0. Does the story deal with politicians, parties or other actors in relation to potential election outcomes and/or coalitions/government formation? This variable has two codes: 0 = no, 1 = yes. Coders should type 1 if the news story reports or speculates about election results or government/coalition formations. Otherwise coders should type 0. Does the story deal with politicians, parties or other actors winning or losing (elections, debates or in general)? This variable has two codes: 0 = no, 1 = yes. Coders should type 1 if the news story at least once refers to whether politicians, parties or other actors are winning or losing with respect to elections, debates or in general. Otherwise coders should type 0. Does the story make use of a language of sports or war? This variable has two codes: 0 = no, 1 = yes. Coders should type 1 if the news story at least once makes use of a language of sports and war, such as battle, competition, winning, or fight. Only exempted expression is ‘campaign’. Otherwise coders should type 0.
Read full abstract