Building envelope airtightness is important for residential energy use, occupant health and comfort. We analyzed the air leakage measurements of 134,000 single-family detached homes in US, using normalized leakage (NL) as the metric. Weatherization assistance programs (WAPs) and residential energy efficiency programs contributed most of the data. We performed regression analyses to examine the relationship between NL and various house characteristics. Explanatory variables that are correlated with NL include year built, climate zone, floor area, house height, and whether homes participated in WAPs or if they are energy efficiency rated homes. Foundation type and whether ducts are located outside or inside the conditioned space are also found to be useful parameters for predicting NL. We developed a regression model that explains approximately 68% of the observed variability across US homes. Of these variables considered, year built and climate zone are the two that have the largest influence on NL. The regression model can be used to predict air leakage values for individual homes, and distributions for groups of homes, based on their characteristics. Using RECS 2009 data, the regression model predicts 90% of US houses have NL between 0.22 and 1.95, with a median of 0.67. (C) 2013 Elsevier B.V. All rights reserved.