The language of vocational interests on social media

There is a burgeoning interest in using natural language to study vocational interests. However, little to no research has used social media language to predict users’ interests. The present study investigated how accurately language used on Facebook predicts individuals’ self-ratings on eight basic interests: Agriculture, Engineering, Human Resource, Life Science, Management/Administration, Mechanics/Electronics, Media, and Social Science. This study employed closed-vocabulary (Linguistic Inquiry and Word Count 2015) and open-vocabulary approaches (Latent Dirichlet Allocation topic modeling) to analyze 3.2 million Facebook posts from 2,834 participants, who completed a 32-item basic interest measure (adapted from the Comprehensive Assessment of Basic Interests; CABIN; Su et al., 2019). Findings showed that the predictive accuracies of the linguistic models (mean r = .24; LDA topics) in assessing vocational interests are comparable to previous language research predicting personality traits (r = 0.27; Tay et al., 2020). Further, the present study revealed the unique language markers which characterize different basic interests. Together, the findings represent a novel advancement in vocational interest assessment. The results further suggest that automated language-based assessments can complement traditional self-report interest inventories to match people’s interests to their ideal occupations. Implications for research and applied settings were discussed.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Psychology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2023

Author dc:creator

• Du, Yan Yi Lance

Contributors dc:contributor

• Drasgow, Fritz
• Roberts, Brent W

• Vocational Interests
• Interest Assessment
• Language Analysis
• Social Media
• Facebook

Statement dc:rights

• Copyright 2023 Yan Yi Lance Du

Language dc:language

en, eng

Handle dc:identifier

https://hdl.handle.net/2142/120547

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F120547