Music and Moods

This project mainly focused on data analysis. I worked with three teammates to investigate how important features of musical mood change as gradualness of change and target mood are. This research consisted of a within-person analysis of the musical characteristics of short playlists that participants create in Spotify to change their moods away from sadness. I did some literature research to find validated resources as references to formulate the survey that used to measure the mood change and musical sophistication. After data collection, I worked with another teammate to do the descriptive analysis and statistical analysis about the patterns in the songs and transition models in song sequences. The regression was conducted to investigate the relations between music characteristics (energy and valence) and target mood. Different models (i.e., linear model, exponential model, and step model) were also used to compare with the pattern of changes of music characteristics, to see which one fitted best when predicting the sequence of characteristics.

In this project I...

• Design the survey that used to measure the mood change and musical sophistication

• Performed statistic analysis in Stata to explore the differences in the fit precision based on the starting amount of sadness, demographic data, music sophistication, and satisfaction with the chosen song by the participants

• Tools: Stata, Google Forms (Survey)