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 stable over time, but its interventions in amplification logics change over time. This should be seen to have an impact.
The project works with the Facebook URL Shares dataset, focussing on some 130,000 URLs shared over six years in the United States. The dataset contains URLs shared at least 100 times; it filters for signal amidst a dataset which has had differential noise added; the quality of news domains was ascertained via NewsGuard ratings (which retained some 100,000 URLs); and all this was estimated per quarter. The dataset also provides information on the broad (US) political orientation of engaging users.
Shares produce additional views: one share generates some 56 further views per URL. But this is conditional on who shares, what is shared, and when. Sharing is a signal entering a distribution machine, not a direct g channel to audiences.
Conversely, there is a partisan penalty: URLs shared by a more intensive partisan audience reach fewer people. One standard deviation of audience partisan intensity reduces share counts by some 2.3 million.
Journalism quality is also important: there were some 28,700 additional views for each point on the NewsGuard scale.
But how does this evolve over time, given Facebook’s interventions in algorithmic amplification? Click rates are broadly flat at some 6-7.5 views per share over six years; but views per share vary considerably. Meta’s severe intervention to penalise partisan sharing from Q3 2020 shows a clear effect, while quality news sources benefit considerably during this time. As this is rolled back these patterns bounce back to previous patterns.
This shows that amplification is the major driver of such patterns, not political homophony or ‘echo chambers’. It also shows the importance of greater transparency about such platform measures, and their impact on democratic discourses. Such research needs to be repeated for Facebook and other platforms again to see how similar processes operate there.
