Table 1. Distribution of the publication sample in Facebook

Source: Created by the authors

The sample was collected using the Netvizz application. The politicians who were analysed disseminated a total of 601 publications during the campaign (Table 1), of which 62% came from party accounts (373 posts), and 38% from leaders (228 posts). The statistical processing of the results was carried out using SPSS (v.23). The reliability of the inter-codifier, calculated with the Scott Pi formula, reaches a level of 0.97. Table 2 shows the analysis protocol used in this research, which was also based on previous work (López-Meri, Marcos-García & Casero-Ripollés, 2017).

Given that the objective is to analyse the functions that politicians attribute to Facebook during an election campaign, a series of categories and subcategories associated with the “function” variable have been designed. These are exclusive categories, or in other words, only one category is assigned to each analysis unit, which is the one that best represents the essence of the publication.