Digital news media is continuously battling for reader attention. With fierce competition for engagement and loyalty, it is crucial to identify what drives online news consumption. A newly released study, Negativity drives online news consumption, analyzes news headlines’ impact on news consumption. The findings confirm that negative emotional words in news headlines increase consumption rates. However, the dataset is from 2013 to 2015. So, while the results are certainly worth consideration, it is important to consider whether the same would hold true today.

Methodology and analysis

This study comprises 22,743 randomized control trials (RCTs) with over 105,000 different variations of news headlines from Upworthy. Each random control test compares different variations of news headlines, an average of 4.31 variations, that all belong to the same news story. The headline variations are then compared to the generated click-through rate (CTR), defined as the ratio of clicks per impression. 

The headlines generated approximately 5.7 million clicks and more than 370 million impressions. The research notes that the results are comparable to traditional news sites. However, Upworthy differs from traditional news sources due to their early ‘click-bait’ headline practices. Therefore, there might not be a direct parallel to the headline practices of other types of news sites.

Detailed findings

The research shows that a higher share of negative language in news headlines increases click-throughs, and a higher percentage of positive language decreases them. Headlines with an average length of approximately 15 words with a single negative word increase the click-through rates by 2.3%. In contrast, including a positive word in a news headline significantly decreases the likelihood of CTR by about 1.0%. 

Overall, positive words are more common than negative words in the headlines analyzed ‒ 2.83% were positive words versus 2.62% negative words. Click-throughs overall are low, with the average CTR across all trial experiments at 1.39%, and the median click rate at 1.07%. Only a small proportion of the news headlines tested were associated with a high CTR.

The research also tests headlines across various news topics – economics, entertainment, lifestyle, etc. Interestingly, they found no difference in click-throughs across content categories. Further, the report notes an analysis in which the researchers study the effects of four emotions (anger, fear, joy, and sadness) and their role in news consumption. Words like sadness increase click-through rates, while words about fear decrease them. 

• A statistically significant and positive coefficient for sadness increases the probability of a user clicking the headline by 0.7%
• A statistically significant negative effect for joy and fear decreases the probability of a user clicking on a headline by 0.9% and 0.7%, respectively.
• No statistical coefficient estimates for anger.

Again, it’s important to understand the emotional response to these words today compared to the original testing.

The study’s research integrity is not in question here. This research offers insight into how the presence of certain words links to behavior from 2013 to 2015. However, digital media, news delivery and consumption platforms, and the internet itself constantly evolve. Therefore, while the study provides interesting insights on language choices in headlines, it might not hold true today. Updated research is essential to understand if the emotional effects on readership remain the same.