Abstract
In micro-level data concerning the housing market, the apartment quality may be signaled via textual statements or the attached descriptions/photos. It may be done using hard information related to the easier-to-measure structural characteristics or soft information related to the apartment condition and design - soft quality. This paper checks whether the choice of the approach to handling the issue of soft quality of apartments influences the properties of hedonic models and the course of hedonic rent indices. The study shows that hedonic models that account for soft quality have better statistical properties than those without soft-quality-related variables. Among them, the models that include the information on quality extracted from descriptions of apartments prove to be the best. Still, considerable differences in the indicated course of hedonic rent indices have not been detected. However, the paper concludes that utilizing information on apartments’ soft quality may be crucial to understanding the market adjustment process to economic shocks. It has been proven that the price reaction of the Poznań (Poland) rental market to the COVID-19 pandemic and the Russian aggression on Ukraine has been diversified in the quality-related market segments.