Investigating the indoor environmental quality of different workplaces through web-scraping and text-mining of Glassdoor reviews

Giorgia Chinazzo*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

The analysis of occupants’ perception can improve building indoor environmental quality (IEQ). Going beyond conventional surveys, this study presents an innovative analysis of occupants’ feedback about the IEQ of different workplaces based on web-scraping and text-mining of online job reviews. A total of 1,158,706 job reviews posted on Glassdoor about 257 large organizations (with more than 10,000 employees) are scraped and analyzed. Within these reviews, 10,593 include complaints about at least one IEQ aspect. The analysis of this large number of feedbacks referring to several workplaces is the first of its kind and leads to two main results: (1) IEQ complaints mostly arise in workplaces that are not office buildings, especially regarding poor thermal and indoor air quality conditions in warehouses, stores, kitchens, and trucks; (2) reviews containing IEQ complaints are more negative than reviews without IEQ complaints. The first result highlights the need for IEQ investigations beyond office buildings. The second result strengthens the potential detrimental effect that uncomfortable IEQ conditions can have on job satisfaction. This study demonstrates the potential of User-Generated Content and text-mining techniques to analyze the IEQ of workplaces as an alternative to conventional surveys, for scientific and practical purposes.

Original languageEnglish (US)
Pages (from-to)695-713
Number of pages19
JournalBuilding Research and Information
Volume49
Issue number6
DOIs
StatePublished - 2021

Keywords

  • Built environment
  • data mining
  • indoor environmental quality
  • occupant perceptions

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction

Fingerprint

Dive into the research topics of 'Investigating the indoor environmental quality of different workplaces through web-scraping and text-mining of Glassdoor reviews'. Together they form a unique fingerprint.

Cite this