Developers who don’t use emojis in their Github posts are three times more likely to drop off the platform during the following year than those who include emojis. Xuan Lu of the University of Michigan, Ann Arbor, and colleagues analyzed more than 60 million posts on the online work platform, presenting these findings in the open-access journal PLOS ONE on January 26, 2022.
Emotional signals from employees can be key indicators of attitudes, future work outcomes, and other factors of interest to employers. However, as an increasing number of employees work remotely, employers may find it difficult to observe emotions. Lu and colleagues hypothesize that emoji usage in online work platforms could be tracked to monitor employees’ emotions and engagement and could predict future work outcomes.
To explore that hypothesis, the researchers conducted several analyses examining emoji usage and work activities among hundreds of thousands of developers who used GitHub in the year 2018, ultimately considering more than 60 million posts in total.
The research team found that GitHub developers commonly used emojis, likely both for practical communications and to express emotions. Emoji-usage patterns varied among developers, and different patterns were associated with different factors, including types of work, time management, which programming languages were used, and more. It was not possible to determine in this study whether or not the emotions conveyed by the emojis were in fact sincerely felt by the user.
The researchers were particularly interested in emoji usage as a potential predictor of work outcomes; they found that developers who did not use emojis in 2018 were three times more likely to be inactive on GitHub in 2019. Further analyses suggested that the emotions conveyed by emojis used by workers may play an important role in machine-learning predictions of whether they later dropped off the platform—whether or not those emotions were sincerely felt.
Overall, these findings suggest that emoji usage can help predict employee dropout, but further research is needed. Future research could establish links between emoji usage and actual feelings, test whether similar usage patterns occur on other online platforms, or explore whether emoji usage could be altered to improve work outcomes.
The authors add: “In a remote setting, you can see workers replying to emails, you can see that they’re chatting on Slack channels, but you don’t know how they feel – you cannot see smiles, you cannot see anxiety. However, you can make fairly accurate predictions of whether people will drop out just based on how they use emojis. In fact, those who don’t use emojis are three times more likely to drop out of remote work. You don’t even need to look at their work productivity or the actual words they say— just look at how they use emojis.”
