A personalized, miniaturized air sampling system was evaluated to estimate the daily exposure of pediatric asthmatics to nitrogen dioxide (NO2). The lightweight device (170 g) uses a sampling pump connected to a solid sorbent tube containing triethanolamine (TEA)-impregnated molecular sieve. The pump is powered by a 9 V battery and samples air over a 24 h period at a collection rate of 0.100 L/min. After exposure, the solid sorbent is removed from the tubes for spectrophotometric analysis (Griess Assay). The lower detection limit of the overall method for NO2 is 11 microg/m3. The linearity, precision and accuracy of the sampler was evaluated. Different NO2 concentrations generated in the laboratory (range: 50 to 340 microg/m3) were simultaneously measured by the TEA tube samplers and colocated continuous chemiluminescent NO(x) analyzers (reference method). The coefficient of determination for the laboratory test derived from ordinary linear regression (OLR) was r2 = 0.99 (y(OLR) = 0.94 x -4.58) and the precision 3.6%. Further, ambient NO2 concentrations in the field (range: 10-120 microg/m3) were verified with continuous chemiluminescent monitors next to the active samplers. Re-weighted least squares analysis (RLS) based on the least median squares procedure (LMS) resulted in a correlation of r2 = 0.68 for a field comparison in Riverside, CA (y(RLS) = 1.01 x -0.94) and r2 = 0.92 in Los Angeles, CA (y(RLS) = 1.31 x -7.12). The precision of the TEA tube devices was 7.4% (at 20-60 microg/m3 NO2) under outdoor conditions. Data show that the performance of this small active sampling system was satisfactory for measuring environmental concentrations of NO2 under laboratory and field conditions. It is useful for personal monitoring of NO2 in environmental epidemiology studies where daily measurements are desired.