German Twitter Data on Occupations

Background

There are many valuable insights on jobs and professions in different sectors of society based on their imminent and ascribed characteristics. Studying such characteristics traditionally was done by action research, surveys, questionnaires, etc. which typically take much time and resources to be concluded. This dataset accompanies our studies where we examine vocational education and training data on Twitter. We presented a generic framework to retrieve, process and analyze tweets, and discussed  research questions from computational social science: For example, how can we make Twitter data interoperable to otheravailable resources, e.g. classifications of occupations, tools and skills? Second, do we have enough data to process job collocational prestige analysis on a geographical basis? This presents a novel approach towards labor market research, making novel data interoperable which has not been considered in previous literature. Our approach and pipeline is generic and could be easily extended to other languages. It also contributes to prestige research by widening the question of ascribed prestige to the question of how information on occupations is collocated and what these contextualization tell us about how occupations are seen.

Data

This dataset contains the metadata related to 3,407,481 tweets (2007-2023). However, according to Twitter/X rules of data protection, we do not publish any data retrieved by their API. This dataset contains a unique id (not the tweet id), a job_id related to KldB (Klassifikation der Berufe), a job_title linked to that tweet, a timestamp and a sentiment. 

Contact

Please contact us if you have questions or want to use this dataset.