Analysis of Factors Influencing Moisture Defects in New Residential Buildings in Bangkok and Surrounding Areas Using Negative Binomial Regression

Abstract

This study examines the relationship between moisture-related defects in newly constructed residential buildings and two key factors: contractor type and environmental conditions. Data were collected from pre-handover inspection reports of 276 houses located in Bangkok and surrounding areas. The main variables of interest were the contractor type and the shortest distance from each house to the nearest waterway, with distances calculated using the Google Map API. The data, which are count-based and exhibit overdispersion, were analyzed using Negative Binomial Regression, which is appropriate for such data characteristics. The analysis revealed that houses built by local contractors were significantly more likely to experience moisture problems than those constructed by public companies. Meanwhile, an increase in the distance from waterways was associated with a decrease in moisture-related defects. However, the model’s explanatory power remained low, which may be attributed to the sample size or the limited number of factors considered. Further research should involve a larger sample size or include additional relevant variables to enhance the understanding of moisture issues in housing projects.