One challenge in social science work with computer-assisted methods is to critically examine the possibilities of the methods for their valid application and to keep an eye on possible limitations. Challenges in manual content analyses are (due to the high personnel resources) a clever selection of the material to be examined as well as the time-consuming training of the coders. After all, the results will only be of high quality and meaningful if they classify the media content reliably (i.e. reliably) and validly (i.e. "correctly" in terms of content). Above all, analyses of social media metrics must take into account the media logic of the medium under investigation, as the interaction options provided, such as liking, sharing or commenting, are designed differently by the platform operators. The analysis of networks in social media poses a challenge, as this is an investigation of the networking between user profiles or content on social media, rather than social actors. Such profiles do not necessarily correspond to social actors that exist offline. These include, for example, online communities or chatbots.
Especially in their combination, text and media content analyses can describe patterns and trends in media content on the one hand across very large amounts of data and on the other hand for selected time periods or content in a more in-depth and interpretative way.
When interpreting the results of computer-aided methods, particular attention should be paid to the validation of the results. Since a strictly representative selection of media content is difficult to achieve even with a clever selection of material, attention should always be paid to generalizability when interpreting manual content analyses.
Analyses in digital platforms must always bear in mind that this is a socio-technical process that is not only controlled by users, but is also influenced by automated factors such as algorithmic content recommendations. The results show how content is shaped, distributed, exaggerated or downplayed within a specific online environment, how social actors can be networked and mobilized around this content and how different actors act for different reasons to popularize and (de)legitimize content.
