A comparative evaluation was made of four analytical methods for the determination of formaldehyde (50000) - NIOSH method P+CAM 354, NlOSH P+CAM 318, NIOSH P+CAM 125, and a 2,4-dinitrophenylhydrazine (DNP) coated silica gel sorbent tube method. The evaluation was done on samples collected at a formaldehyde production facility (SIC- 2869) (site-1), a hospital dialysis unit (site-2), and a fiberglass insulation production facility (SIC-3296) (site-3). In the NIOSH P+CAM 354, the formaldehyde was collected on a reagent coated sorbent, in P+CAM 318 the formaldehyde was oxidized to formate by a charcoal sorbent, and in P+CAM 125 an impinger filled with a 1 percent sodium-bisulfite solution was used. Samples were collected (12 at a time) over 4 to 6 hours. At site-1, the DNP and P+CAM 354 methods gave comparable values while the P+CAM 318 method gave results that were considerably lower. Further studies showed that samples to be measured by the P+CAM 318 method should not be stored for more than 1 week due to instability. At site-2, the formaldehyde levels were at the limit of detection (slightly greater than 3 micrograms per milliliter) and both P+CAM 354 and the DNP method showed a high variability. At the last site, samples were collected with and without prefilters (P+CAM 354 and 125) and good agreement was observed between the determinations (approximately 0.1 to 0.2 parts per million concentration determined by the latter method). Phenol may have interfered with the determination by P+CAM 125, but no instability was found in samples from this method that were stored for 10 days. It is recommended that a prefilter be used with P+CAM 125 impinger sample collection when airborne particulate is present that may release formaldehyde.