Audience Participation in TikTok Metadata

Abstract

With the expansion of digital culture, an in-depth reflection on how to research audiences is necessary. If, formerly, the individual was placed in a social category that defined cultural tastes, now technology identifies patterns of behavior from the direct record of their actions. This text explores the type of knowledge that can be obtained on audience participation on TikTok. We propose a methodology that consists of the analysis of usage metadata. The fieldwork focuses on “Ac2ality”, an information account with 4.4 million followers in Spain. We analysed all videos shared over six weeks of the first quarter of 2023 (n=173). The purpose was to find (a) the degree of the linear correlation between the metadata for the same video and (b) the existence of correlations between metadata and type of video/ content. For each metadatum available with open access (comments, likes, saves, shares and views), four activity levels have been established (low, intermediate, high and very high). The majority trend indicates that the levels obtained by the metadata of the same content are not coincident, that is, a video will have more or less scope according to the observed metadata. The homogeneity of the videos means that only clear correlations between topic and metadata are detected. Topics with less presence can reach high levels of activity.