Representación composicional (coda) de la agenda setting de los opinantes sobre política en la prensa escrita española durante las elecciones generales de 2015

Abstract

Resumen. Este articulo presenta y desarrolla un análisis composicional con representación de biplots como propuesta innovadora para un abordaje metodológico de representación de la teoría de la agenda-setting para la opinión sobre política en la prensa escrita. Para ello, se desarrolla un análisis de contenido de los artículos de opinión sobre política de los periódicos El País, El Mundo, ABC, La Razón y La Vanguardia, en el período de las Elecciones Generales españolas de 2015, para posteriormente aplicar el método composicional como procedimiento que permite representar la teoría de la agenda-setting, teniendo en cuenta la proporcionalidad de la presencia de los temas y superando los tradicionales métodos correlacionales. Los resultados prueban la viabilidad metodológica, a la vez que aportan una visión hasta el momento inexplorada sobre la agenda-setting en el caso en estudio. Palabras clave: Agenda-setting; análisis composicional; biplot; opinión política; prensa escrita. _____________ 1 Investigación financiada: los resultados de este trabajo forman parte del proyecto de investigación de doctorado del primer firmante, titulado El análisis composicional (CoDa) como herramienta para el mapeo

Mapping the agenda-setting theory, priming and the spiral of silence in Twitter accounts of political parties

“…The very nature of the AS theory makes analysis of content in the agenda typically focus on the relative or comparative importance of some issues over others (Blasco-Duatis et al 2018a, 2018b, 2018c. In other words, it is assumed that some media or social actors are busier than others, leading to an overall increase in communication volume, which is void of content-related interest.…”

“…The focus on relative importance found in research questions is usually not matched by a similar focus when selecting the statistical analysis method. One common way to assess similarity among senders of political information according to the content categories they send and to assess which of these categories contribute to their differences, is to compare the frequencies or rankings of the top contents (e.g., Blasco-Duatis et al 2017;Cui and Wu 2017;Guo et al 2015;Jungherr et al 2016;Lim 2011), with a general lack of focus on how content categories contribute to generating differences among senders, be they relative or absolute. This especially relevant when dealing with the spiral of silence, which deals with issues that constitute a minority of content, whose relative differences are likely high.…”

“…Nowadays, CoDa spans almost all the hard sciences and has started to be used in several fields of social science, such as education (Batista-Foguet et al 2015), economics (Fry 2011), marketing (Vives-mestres et al 2016), accounting (Linares-Mustarós et al 2018), tourism (Ferrer-Rosell and Coenders 2018), values (van Eijnatten et al 2015, social networks (Kogovšek et al 2013), time use (Martín-Fernández et al 2015) and election studies (Egozcue and Pawlowsky-Glahn 2011). Its application to content analysis in political communication (Blasco-Duatis et al 2018a, 2018b, 2018c and in other fields of communication (Mariné-Roig and Ferrer Rosell 2018;Wang et al 2018) is very recent, even though, as argued previously, such analysis poses the same problems in all respects as those faced by chemical and geological analyses. Absolute data are irrelevant and mostly tell about the length of the text, the word count, the popularity of the sender, and so on.…”

“…If the z vector contains zeros, they must be replaced beforehand. Among the zero replacement methods discussed in Martín-Fernández et al (2011) whose discussion is beyond the scope of this article, we use the same method proposed by Blasco-Duatis et al (2018a, 2018b, 2018c in the AS research context.…”

“…Existing differences between senders and content categories in AS have traditionally been studied either descriptively, by counting the occurrence of issues, or by means of Spearman correlations, taking only rank order into consideration and leaving out any other information. Recently, compositional analysis (CoDa), the standard statistical methodology when the researcher is interested in the relative importance of the parts of a whole (Aitchison 1986), has started to be applied in AS research (Blasco-Duatis et al 2018a, 2018b, 2018c. We especially focus on a data visualization tool, the compositional biplot, which allows us to see which issues are prioritized, who their senders are and the relationships between them, while taking into consideration the fact that the relevant information is the relative volume of each issue for each sender.…”