This article explores the uses of sources in coverage of the COVID-19 pandemic in social media posts of mainstream news organizations in Brazil, Chile, Germany, Mexico, Spain, the U.K., and the U.S. Based on computational content analysis, our study analyzes the sources and actors present in more than 940,000 posts on COVID-19 published in the 227 Facebook, Instagram, and Twitter accounts of 78 sampled news outlets between January 1 and December 31 of 2020, comparing their relative importance across countries, across media platforms, and across time as the pandemic evolved in each country. The analysis shows the dominance of political sources across countries and platforms, particularly in Latin America, demonstrating a strong role of the state in constructing pandemic news and suggesting that mainstream news organizations' social media posts maintain a strong elite orientation. Health sources were also prominent — consistent with the defining role of biomedical authority in health coverage—, while significant diversity of sources, including citizen sources, emerged as the pandemic went on. Our results also revealed that the use of specific sources significantly varied over time. These variations tend to go hand in hand with specific global milestones of the pandemic.

This study proposes the interventionist and the detached orientations to watchdog journalism through the conceptual lens of journalistic role performance. Based on a content analysis of 33,640 news stories from sixty-four media outlets in eighteen countries, we measure and compare both orientations across different countries using three performative aspects of monitoring: intensity of scrutiny, voice of the scrutiny, and source of the event. Our findings show that the interventionist approach of watchdog journalism is more likely to be found in democracies with traditionally partisan and opinion-oriented journalistic cultures or experiencing sociopolitical crises. In turn, the detached orientation predominates in democracies with journalistic traditions associated to objectivity. Although both orientations have a lower presence in transitional democracies, the detached watchdog prevails, while in non-democratic countries the watchdog role is almost absent. Our results also reveal that structural contexts of undemocratic political regimes and restricted press freedom are key definers of watchdog role performance overall. However, the type of political regime is actually more important—and in fact the most important predictor— for detached than for interventionist reporting.

Comparative research across the world has shown that nation-level variables are strong predictors of professional roles in journalism. There is, however, still insufficient comparative research about three key issues: cross-national comparison of journalistic role performance, exploration of how – or whether – organizational variables account for variation in role performance across countries, and the performance of specific journalistic roles that prevail in regions with post-authoritarian political trajectories. This article tackles these three issues by comparatively measuring journalistic performance in five Latin American countries. Based on a content analysis of 9841 news items from 18 newspapers, this article reports findings from Brazil, Chile, Cuba, Ecuador and Mexico, by analyzing the presence of the ‘interventionist’, ‘watchdog’, ‘loyal’, ‘service’, ‘infotainment’, and ‘civic’ roles. Results show that the region is far from homogeneous and that while ‘country’ is a strong predictor for most of the roles, other variables such as ‘media type’, ‘political orientation’, and ‘news topic’ are also significant predictors to varying levels.

Democracies with sharp violence and public insecurity have proliferated in recent decades, with many also featuring extreme economic inequality. These conditions have not been explicitly considered in comparative research on journalists’ work environments, an omission that may obscure important realities of contemporary journalism. We address this gap through analysis of journalist surveys in 62 countries. We confirm the existence of insecure democracies as an empirical phenomenon and begin to unravel their meaning for journalists. We find democracies with uneven democratic performance tend to have more journalist assassinations, which is the most extreme form of influence on work, and that levels of democratic performance, violence, public insecurity and economic inequality significantly shape how journalists perceive various influences in their work environment. Case studies of insecure democracies in Africa and Latin America address why these conditions sometimes (but not always) lead to journalist assassinations and other anti-press violence. They suggest anti-press violence is higher when sub-national state actors intensify criminal violence and when insecurity is geographically and topically proximate to journalists. How journalists’ perceive influences on work are therefore more complex and multidimensional than previous research has suggested. The study concludes by identifying areas for improvement in data collection.

Cross-national research has identified crime, corruption, and human rights abuses as explanations for threats against journalists in democracies and authoritarian hybrids plagued by antipress violence. In-depth studies additionally suggest gender or occupational characteristics such as risky newsbeats increase the likelihood of being threatened. We overcome data limitations in many of these studies by analyzing work-related threats reported by journalists in Mexico, a territorially uneven democracy. Findings confirm that contexts of criminal insecurity are the strongest predictor of threats but only for journalists who are frequently harassed. For the infrequently threatened, democratic normative commitments are a stronger predictor. Subnational government corruption is another important predictor of threat but operates counter to expectations. We believe this is because clientelism sufficiently controls journalists without the need for threat. Neither occupational traits nor gender were individually important predictors. Findings suggest future research should compare threat and harassment across lower and higher risk contexts, and measure public insecurity and clientelism at the local level where journalists actually work. Measurement improvements might better reveal the gender dynamics of threat. More broadly, comparative research and policy-making in democracies and authoritarian hybrids should focus on how local authoritarians limit journalists’ democratic normative aspirations.