A number of commentators have argued in recent years that the media overemphasises negativity in its content. Is this true?

Answering this question is no trivial matter, as it requires a standard against which the media’s coverage can be compared. That is, it is challenging to establish how negative or positive media content should be.

What we can certainly determine instead is how the sentiment (positive or negative) and emotional undertones (such as fear, anger or joy) of news content compare with the same metrics at different points in time. This allows us to establish whether news media content is becoming more positive over time, more negative or pretty much staying the same.

For the past two years, we have carried out a research project to find out precisely this – the sentiment and emotional undertones of 23 million headlines from a representative sample of 47 news outlets popular in the United States over the 2000 to 2019 time frame.

The sheer volume of headlines we considered meant that we needed to use artificial intelligence in the form of machine learning models, whose performance is on par with human raters, for the automated classification of sentiment and emotion in headlines.

What we found is that the sentiment of mainstream news media headlines has indeed become gradually more negative since the year 2000. What this means is that headlines with negative connotations, such as “Brazil Prison Riot Leaves 9 Dead”, are becoming more prevalent. In contrast, headlines with positive undertones, such as “A New Lens Restores Vision and Brings Relief”, are becoming less frequent.

Interestingly, when we partitioned news media outlets according to the ideological views they are widely associated with, we found that headlines from right-leaning news media have been consistently more negative than those from their left-leaning counterparts. Post-2013, the negativity of headlines in left-leaning news media appears to increase substantially. These trends might be partially related to a sharp uptick in news media usage of terminology that depicts prejudice (such as racism, sexism and homophobia) and political extremism (such as far right or far left).

When we further analysed the specific emotional undertones of the headlines, we discovered that the proportion denoting anger and fear has almost doubled in frequency over this period. Headlines embedded with anger and fear such as “Giving Poor Kids Free Meals at School Should Not Be Controversial. Tell That to Congressional Republicans!” or “Is rape epidemic in Sweden tied to influx of Muslim immigrants?” are becoming more prevalent. Sadness and disgust are also increasingly reflected in headlines, albeit to a lesser extent. In contrast, the proportion of emotionally neutral headlines is decreasing.

We found that right-leaning outlets tend to use headlines conveying anger more often than left-leaning media. On the other hand, the rise of headlines denoting fear and the decrease of emotionally neutral headlines has been very similar across media regardless of their ideological leanings.

How should we interpret these results? It is obvious from our analysis that the negativity, anger, sadness and fear conveyed by news media headlines are increasing over time. But why is this happening? Does this reflect a wider societal mood or just the sentiments and emotions of American newsrooms?

We believe that financial pressures to maximise click-through ratios as a response to decreasing media-industry revenue could be at play for the increasing negativity and emotionality of headlines over time. The crafting of headlines to advance political agendas through a shift away from fact-based standards of objectivity could also be playing a role.

The higher prevalence of headlines denoting negativity and anger in right-leaning news media is noteworthy, but we can only speculate about its causes.

One possibility is that right-leaning news media simply uses more negative language than left-leaning news media to describe the same phenomena. Some authors have argued that there is a connection between right-leaning political orientation and a disposition or sensitivity to negative stimuli and events.

Alternatively, this trend could be driven by differences in topic coverage between both types of outlets. But to be clear, these are all only possibilities. Much more research is needed to answer these questions.

One thing is certain, however. If you feel your news diet is more depressing these days, you’re not crazy, and just how depressing depends on what you read. The next task for researchers: digging deeper into how this increasing negativity of news content affects readers as well as democratic institutions and processes.

The views expressed in this article are the authors’ own and do not necessarily reflect Al Jazeera’s editorial stance.

QOSHE - Bad news: Head­lines are in­deed get­ting more neg­a­tive and an­gri­er - David Rozado
We use cookies to provide some features and experiences in QOSHE

More information  .  Close
Aa Aa Aa
- A +

Bad news: Head­lines are in­deed get­ting more neg­a­tive and an­gri­er

22 20 1
27.11.2022

A number of commentators have argued in recent years that the media overemphasises negativity in its content. Is this true?

Answering this question is no trivial matter, as it requires a standard against which the media’s coverage can be compared. That is, it is challenging to establish how negative or positive media content should be.

What we can certainly determine instead is how the sentiment (positive or negative) and emotional undertones (such as fear, anger or joy) of news content compare with the same metrics at different points in time. This allows us to establish whether news media content is becoming more positive over time, more negative or pretty much staying the same.

For the past two years, we have carried out a research project to find out precisely this – the sentiment and emotional undertones of 23 million headlines from a representative sample of 47 news outlets popular in the United States over the 2000 to 2019 time frame.

The sheer volume of headlines we considered meant that we needed to use artificial intelligence in the form of machine learning models, whose performance is on par with human raters, for the automated classification of sentiment and emotion........

© Al Jazeera


Get it on Google Play