Toward a Cross-Platform Framework: Assessing the Comprehensiveness of Online Rental Listings

Area of Science:

• Urban Studies
• Real Estate Economics
• Data Science

Background:

• Traditional housing surveys lack the temporal and spatial specificity needed for short-term rental market analysis.
• Digital transition of rental housing ads enables web scraping for studying listing practices and market dynamics.
• Existing research predominantly uses Craigslist data, potentially offering an incomplete view of rental markets.

Purpose of the Study:

• To investigate variations in rental unit types and features across multiple online platforms.
• To compare rental market characteristics using data from diverse online listing sites.
• To demonstrate how platform choice influences estimates of rental costs and unit size.

Main Methods:

• Employed a mixed-methods approach analyzing rental listings from six online platforms in five metropolitan areas.
• Utilized web scraping to collect data on listing criteria, search parameters, and result prioritization.
• Integrated scraped data with 2020 Fair Market Rents and 2019 American Community Survey data for comprehensive analysis.

Main Results:

• Significant variations observed in advertised rental unit types and features across different online platforms.
• Online platforms target distinct audiences, presenting unique market segment information.
• Platform-specific data yields markedly different estimates for local rental costs and unit size distribution.

Conclusions:

• Relying on a single online platform for rental market analysis can lead to biased or incomplete conclusions.
• A multi-platform data aggregation approach is crucial for a comprehensive understanding of rental housing dynamics.
• Future research should consider the heterogeneity of online rental listing platforms to improve market analysis accuracy.