A new and better way to detect media censorship
IMAGE:
A VISUALISATION OF THE AGENDA INDEPENDENCE SCORE, WHICH SHOWS HOW THE THREE NOMINALLY INDEPENDENT OUTLETS IN NICARAGUA DECLINE IN INDEPENDENCE OVER TIME. THE VISUALISATION ALSO SHOWS HOW ONE OUTLET (CANAL14) DECLINES MORE STEEPLY THAN THE OTHER TWO.
view moreCREDIT: BARNEHL EN SCHUMACHER
Worldwide news media are facing increasing pressure from autocrats to report favourably about their leaders and party politics. Political scientists launch a new computational method that can detect such media censorship by states while it is happening. This method provides valuable insights for communicating regime-driven media capture to the public. It is now described in detail in the scientific journal ‘Democratization’.
One of the first steps of would-be autocrats is to control the media, as seen in recent cases in Russia, Hungary, and Turkey. Most autocratic regimes today execute control not by monopolising media production, but by using legal, economic, and physical measures to sway the editorial choices of both public and private media in their favour. All to portray themselves and their party politics favourably to the public. Prominent examples of such measures are loosely defined laws, like anti-terrorism and anti-fake-news laws, used to silence opposition voices.
So far, expert surveys were employed to detect instances of such media capture by states. While these surveys are valuable, the authors argue they are inadequate. Their new computational method has the precision to uncover the influence of media capture on editorial decisions, assess immediate impacts from new censorship laws, and differentiate between various media outlets.
Comparing agenda and tone
Very simply explained, the computational method compares media agenda and tone of independent versus regime-owned outlets and through time. ‘Agenda and tone are two crucial components that autocrats aim to influence to emphasize topics favourable for their legitimacy and force a positive tone when mentioning the regime, especially its leader’, explain the authors.
Through this comparison the loss of editorial independence at the level of individual outlets are brought to the surface on a monthly or even weekly basis. ‘This takes into account that the process of media capture often does not progress uniformly across the media landscape. While some outlets remain independent from the regime, others, such as those owned by the state, the ruling party, or the leader's family or allies, are already fully captured.’
Tested as a valid method
To test the validity of this method, the authors applied it to the case of Nicaragua, where the regime has intensively cracked down on the media sector in the last years. ‘The country made for an ideal candidate, because it has experienced a steep deterioration of its media freedom in the last decade and the regime’s attacks on opposition media are well documented’, state the authors. Their method proved able to show how outlets responded differently to regime pressure. ‘We found clear differences in agenda and tone between the regime-owned outlets and the opposition outlets. And during and after the regime’s media crackdown, independent outlets shifted from a more critical stance to approximating regime preferences in tone and agenda.’
The defence of media from state interference
According to the authors, future research should evaluate the validity of their method across different media systems and stages of autocratization or democratization. ‘But the methodology could potentially revolutionize the future study of media capture,' they conclude. ‘It provides valuable insights for both researchers and for communicating regime-driven media capture to the public. It has the potential to reveal differences in outlets' resilience to repressive laws and allows for the timely assessment of the impact of seemingly innocuous fake-news or libel laws.’
Publication details
Hennes-Michel Barnehl and Gijs Schumacher, ‘Media capture, captured a new computational methodology to measure deteriorating media freedom’, In: Democratization
JOURNAL
Democratization
METHOD OF RESEARCH
Computational simulation/modeling
SUBJECT OF RESEARCH
Not applicable
ARTICLE TITLE
Media capture, captured a new computational methodology to measure deteriorating media freedom
ARTICLE PUBLICATION DATE
17-May-2024
Risk perception influenced less by media than previously thought
People don’t overestimate the frequency of dramatic causes of death
TECHNICAL UNIVERSITY OF MUNICH (TUM)
For decades, researchers have assumed that people overestimate the risk of dramatic causes of death, such as road traffic accidents. The reason given for this was that such deaths are the subject of far greater media attention than more significant but less spectacular mortality risks. However, a study at the Technical University of Munich (TUM) has now debunked this assumption. Although dramatic causes of death receive disproportionate media attention, deaths in the personal environment are more important for the risk perception.
From road traffic accidents to fires and murders, dramatic causes of death receive considerable media attention. The scientific consensus to date has been that, because the media report dramatic causes of death more frequently than “silent” causes such as heart attacks and diabetes, many people misjudge their prevalence. In specialist literature, this was considered one of the most settled findings in research into risk perception.
However, this assumption can ultimately be traced back to just a single study. In 1978, a team led by Sarah Lichtenstein at the Oregon Research Institute asked respondents to estimate the number of annual deaths for around 40 different causes. The study compared these assumptions with real figures and also examined the extent to which these causes of death were reported in the media and how this reporting was perceived by the respondents.
In the course of his research into risk perception, Thorsten Pachur, Professor of Behavioral Research Methods at TUM, discovered something surprising: the study’s conclusions were not underpinned by statistical analyses and had not been confirmed in subsequent studies. With this in mind, Pachur reevaluated the data from the original study. In addition, he incorporated the two subsequent studies (in which he had participated) that had also examined risk perception and conventional media reporting with similar lists of causes of death, reevaluating their data using the same method.
Results of influential study not replicable
Thorsten Pachur’s study has confirmed that dramatic causes of death have indeed been covered with disproportionate frequency in news reports given their actual prevalence, while unspectacular causes of death have been underrepresented.
However, Pachur’s analysis called the prevailing assumptions about people’s risk perception into question. His evaluation of the data confirmed that the respondents in the 1978 study inaccurately estimated the prevalence of seemingly spectacular risks. However, it was not possible to replicate this result with the data from the more recent studies. The results of a research experiment are only considered verified when they are replicable.
Instead, Pachur’s evaluation of the two more recent studies showed that overestimation or underestimation of a risk is not dependent on whether the respective cause of death is dramatic or non-dramatic. This conclusion remains valid even when evaluating the aggregated data of all three studies. And, when Pachur integrated further studies examining perceptions of mortality risks (but not media reporting), his conclusion was confirmed again.
“These insights do not call into question the fundamental notion that the media can influence people’s perception of risks,” emphasizes Pachur. “However, we should stop believing that a distortion in the level of reporting necessarily leads to a distortion in risk perception.”
Deaths in social environment are a more significant factor
Pachur also found a different explanation for people’s perceptions. Some of the studies he analyzed had also asked participants about their social environment. The new evaluation of this data has shown that the number of deaths of people known to an individual has a far more significant influence on the risk perception of the respective cause of death than the media.
“An important finding is that we are not helpless in the face of distortions in reporting,” says Pachur. “People are evidently quite capable of engaging consciously with news reports and incorporating other sources into their judgment.”
JOURNAL
Cognition
METHOD OF RESEARCH
Data/statistical analysis
SUBJECT OF RESEARCH
People
ARTICLE TITLE
The perception of dramatic risks: Biased media, but unbiased minds
No comments:
Post a Comment