'Big Data'

The next speaker in this session at the 2026 International Communication Association conference in Cape Town is Lucas Paulo da Silva, who reminds us that media outlets tend to chase large audiences. But can they do this across two ideological dimensions: economic and cultural? This might include left conservatives or right progressives, for instance.

Politically invested media actors tend to have very strongly correlated positions across issues, and so do party systems; if outlets are responsible to both economic and cultural dimensions, then this might make them less correlated over time, and this might also happen dynamically in response to …

The next speaker in this session at the 2026 International Communication Association conference in Cape Town is Rupert Kiddle, whose interest is in encoder-produced news embeddings. This is an increasingly common technique, which helps analyse and categorise news articles both for internal journalistic purposes and for scholarly research. But they are not very sensitive to differences over time, and instead engage in a kind of temporal averaging of embeddings; this can be addressed, but remains difficult.

Most models also remain intransparent about their training data and weighting approaches, so there is a need to develop new approaches. This project draws …

The next speaker in this session at the 2026 International Communication Association conference in Cape Town is Magdalena Wojcieszak, who is presenting work towards a news content quality classifier. The consumption of online news is diverse: people consume traditional news article,s blog content, YouTube news videos, news podcasts, and many different formats.

But how do we assess the quality of all this content? There are various different measures for this, and many of them are problematic, not least for their domain- rather than article-level assessments and their conflation of quality with ideological bias; some also include factuality and other features …

I missed the first speaker in the next session at the 2026 International Communication Association conference in Cape Town, so we’re straight on to a paper by the brilliant Fabio Giglietto, whose focus is on partisan alignment, journalistic quality, and algorithmic amplification on Facebook. How do URLs that are shared the same number of times on Facebook reach audiences of vastly different sizes?

This study explores the impact of partisanship and quality on amplification and reach on Facebook, and also takes into account shifts in Facebook’s algorithmic governance design over the years. The structure of Facebook’s social networks is relatively …

And the final speaker in this session at the 2026 International Communication Association conference in Cape Town is Felicia Loecherbach, whose interest is in video-based data donation, in the specific context of TikTok. Data donation is now a growing trend in digital media studies, of course, and many workable approaches to this have been developed, but to work with this approach remains difficult in practice.

TikTok is a key target for this – both because of its current importance with certain user demographics, the limitations of its API, and the fact that API-based observations do not accurately represent the experience …

The next speaker in this session at the 2026 International Communication Association conference in Cape Town is Jiyoun Suk. Her interest is in whether engagement metrics on social media actually measure attention: the average engagement rate across the major platforms is below 2%; in the US, for instance, some 70% never post or comment about political or social issues. So do attention and engagement actually align, and under what conditions – and what does this mean for what we can draw from the data we have?

We can expect a gap between attention and engagement because engagement requires crossing an …

The next speaker in this panel at the 2026 International Communication Association conference in Cape Town is Kilian Bühling, whose focus is on reconstructing deleted Telegram messages to prevent potential biases in data analysis. Message deletion and data loss is a common issue in working with digital trace data; the volume of deleted messages in a data sample increases over time, so data collected a substantial time after an even will miss quite a lot of messages.

This challenges the validity of analyses of such data, as well as the reproducibility of data analyses by other researchers; indeed, it is …

The next speaker in this session at the 2026 International Communication Association conference in Cape Town is Ernesto de León, whose interest is in who donates digital trace data. Data donation studies have become increasingly popular in recent times, of course, but continue to be plagued by sample bias: some users are simply much more likely to be willing to donate data to researchers.

The focus of criticism has been on the willingness to debate, the lack of data on non-participants, and the limited efforts to scope the downstream effects from biased data donation samples. The present project addressed these …

The third day at the 2026 International Communication Association conference in Cape Town starts for me with a session on digital trace data, and a paper by Kyle van Gaeveren, whose interest is in digital trace data. He notes that digital log data have always been very smartphone-dependent, and comparisons of this with laptop-derived trace data are important.

His project provided laptops to some 100 knowledge workers, and used a tool called ActivityWatch to trace which laptop and smartphone apps people were using; this tracked duration, frequency, and fragmentation, but also switching between devices and user mood. Part of the …