This study introduces a series of language signals that indicate whether a person finds their work meaningful (or meaningless). These signals are then integrated into a new, natural language measure of work meaningfulness. This algorithm can analyze a worker’s written description of their work, and using features of their writing determine whether they find their work meaningful with an average classification accuracy of 85%. As an additional, theoretical contribution, this study tests the relationship between work meaningfulness and construal level theory. Results indicate that personal pronouns and action verbs are most related to creating an impression of meaningfulness, but that identity statements and positive sentiment are more related to actual, self-reported meaningfulness. Additionally, construal level showed a significant, positive relationship with several measures of work meaningfulness.
Read
