NIOSHTIC-2 Publications Search
Statistical analysis of the benzene soluble fraction in field blanks used in NIOSH Method 5042.
Olsen L; Shulman S
American Industrial Hygiene Conference and Exposition, June 1-6, 2002, San Diego, California. Fairfax, VA: American Industrial Hygiene Association, 2002 Jun; :20
During the development of NIOSH Method 5042, an anomaly was noticed in the field blank data. Three of 60 field blanks had a higher than expected benzene soluble fraction. Since this event may occur with field samples, these results were not excluded when the data were evaluated. Although these events were observed with syringe-type clarification units (A and B) and not the recommended clarification unit, the cause of this event was not determined. Statistical evaluation of all data including a three-year follow-up indicated not only was this event still occurring, but it also was occurring with the recommended clarification unit and lower than expected results were also occurring. Pooled data indicated that the % outliers were 10.3% (+/-8.7), 3.8% (+/-2.2) and 8.4% (+/-4.4) for the A, B and recommended clarification unit, respectively. Pooled data also indicated that the standard deviations were increased by 1.39, 1.39 and 1.75 times for the A, B and recommended clarification unit, respectively. For the recommended clarification unit, these data indicated if three or five field blanks were collected, the probability of no outliers was about 0.8 and 0.7, respectively. Likewise, the probability of one outlier was about 0.2 and 0.3, respectively. In summary, ten atypical results were observed, and the cause remains undetermined. These events have two undesirable consequences: (I) Because the average weight of the field blanks was increased or decreased, field samples were either over or under corrected, and (2) the standard deviation of the field blank weights was increased resulting in higher LOD and LOQ values. Therefore, it was recommended that as many field blanks as is reasonable (five per day) need to be collected; also, it was recommended that a monitoring program be established to track this occurrence and, if possible, to identify and eliminate the cause(s).
Statistical-analysis; Benzenes; Sampling; Monitoring-systems; Analytical-methods
American Industrial Hygiene Conference and Exposition, June 1-6, 2002, San Diego, California
Page last reviewed: September 2, 2020
Content source: National Institute for Occupational Safety and Health Education and Information Division