A Social Media Study on Demographic Differences in Perceived Job Satisfaction

Area of Science:

• Computational Social Science
• Organizational Psychology
• Data Science

Background:

• Traditional job satisfaction metrics face challenges with scale, accuracy, and timeliness.
• Existing methods often overlook subjective employee expectations and demographic variations impacting job perception.

Purpose of the Study:

• To propose and validate a novel approach for assessing job satisfaction using large-scale social media data.
• To investigate demographic disparities in job satisfaction, focusing on pay and supervision.
• To explore linguistic differences reflecting varied worker priorities.

Main Methods:

• Leveraged a dataset of 1.5 million Twitter posts.
• Developed machine learning classifiers to evaluate perceived job satisfaction (average AUC of 0.84).
• Analyzed demographic variations by geography, sex, and race in the U.S.

Main Results:

• Job satisfaction on Twitter correlates with macroeconomic indicators like financial wellbeing and unemployment.
• Females report higher pay satisfaction but lower supervision satisfaction than males.
• White individuals report the lowest pay and supervision satisfaction.
• Linguistic analysis reveals distinct priorities: under-represented groups focus on basic needs, while majority groups emphasize self-actualization.

Conclusions:

• Social media data offers a scalable method to assess job satisfaction and its demographic nuances.
• Observed differences can be explained by the frame of reference and the job satisfaction paradox.
• Findings have theoretical and sociotechnical implications for enhancing worker wellbeing.