Researchers these days are getting opportunities to study and investigate people’s emotional state with 2.77 million social media users worldwide in 2019. In 2011 Scoot A. Golder and his team from Department of Sociology, Cornell University published a research paper on ‘diurnal and seasonal mood vary with work, sleep, and daylength across diverse cultures ’ which mentions that positive emotional words are often found more in morning tweets. This research work from Scoot helped us to analysis that word choice reflects our psychological states.
However Jessie Sun, a graduate student from the University of California, Davis states contrary to what Scoot published in her research paper ‘The language of well-being: tracking fluctuations in emotion experience through everyday speech’. Jessie and her team found that emotion-related words do not, in fact, provide a good indication of a person’s mood, there might be another set of words that actually do.
In this blog post, we will explain to you what Jessie’s finding was contrary to Scoot’s findings and happy words used by social media users really mean the same.
How Jessie Sun & the Research Team Came to the Conclusion
For her research, Jessie and her team gave a recording device to 185 American university students for a week to record their communication activity. They measured positive and negative emotions of students and ended up with 150,000 recordings which research assistants transcribed for over a course of two years. They estimated each recording with respect to positive and negative emotion words it contained by running the text through Linguistic Inquiry and Word Count analysis programme.
This helped Jessie and her team to come up with 1,579 language-based emotion measurements that helped her to compare directly to participants’ self-reported mood.
Findings by Jessie Sun #1: Contrary to the Past Studies Assumptions
Jessie and her team came to realize that, contrary to the assumptions of former studies, the number of emotional words (positive and negative) has no correlation with participants’ actual mood. She explained that her findings suggest that researchers assume the fluctuations in the ‘use of emotional words’ can be used as a proxy for subjective motion experience, at least for spoken language.
Findings by Jessie Sun #2: Unrelated words to Emotion can predict the Mood
Through the analysis, Jessie found that participants’ mood can be predicted with another set of words that are not related to emotion. Socializing words such as “you” or “we” put forward a positive emotion, and math words such as “minus” and “numbers” weren’t as much. However, she came to realize that these associations were actually week and cannot definitely be used to measure emotion.
Finds by Jessie Sun #3: Usage of Positive words in Negative Context
Jessie found a more possible explanation for the null results in the study. They found that participant using alternatively positive dictionary words into a negative context: for example, the word pretty appear positive in the dictionary but it could be used in negative context (e.g. “it was pretty terrible”).
Jessie Sun & her team put forward a wonderful inference, even though with though taking the accounts of limitations. The study establishes the significance of checking the validity of the tool in psychology research. They nearly prove that early psychological fact was contrary to their research analysis. The users who use more positive or negative emotional words in social media, they are not actually experiencing those emotions.