Search

The National Institute for Occupational Safety and Health and the Environmental Protection Agency sponsored the completion of an interlaboratory study to compare two fitting protocols specified by ANSI S12.6-1997 (R2002) [(2002). American National Standard Methods for the Measuring Real-Ear Attenuation of Hearing Protectors, American National Standards Institute, New York]. Six hearing protection devices (two earmuffs, foam, premolded, custom-molded earplugs, and canal-caps) were tested in six laboratories using the experimenter-supervised, Method A, and (naïve) subject-fit, Method B, protocols with 24 subjects per laboratory. Within-subject, between-subject, and between-laboratory standard deviations were determined for individual frequencies and A-weighted attenuations. The differences for the within-subject standard deviations were not statistically significant between Methods A and B. Using between-subject standard deviations from Method A, 3–12 subjects would be required to identify 6-dB differences between attenuation distributions. Whereas using between-subject standard deviations from Method B, 5–19 subjects would be required to identify 6-dB differences in attenuation distributions of a product tested within the same laboratory. However, the between-laboratory standard deviations for Method B were -0.1 to 3.0 dB less than the Method A results. These differences resulted in considerably more subjects being required to identify statistically significant differences between laboratories for Method A (12–132 subjects) than for Method B (9–28 subjects). [ABSTRACT FROM AUTHOR]