Choosing the right hearing protection device (HPD) should be based on two factors: comfort and performance. We could have a great debate on which of these factors is more important. I would argue that comfort should be of primary importance for one simple reason: If the device is comfortable, it will be worn! After considering comfort, HPD performance becomes the most important issue.

HPD performance is characterized by a single number rating called the Noise Reduction Rating, or NRR. So just how is this crucial number derived?

Do the math
The Noise Reduction Rating is determined by testing the HPD at a laboratory and then performing some computations. The computations used to determine the numerical value of the NRR have been codified by the Environmental Protection Agency (EPA) in its regulation 40 CFR 211. Inputs to the computations use the HPD’s average measured attenuation and standard deviation for each frequency band that was tested (more on this below). A normalized “C” weighted unprotected sound level and a normalized “A” weighted protected sound level must then be computed (determined by subtracting from the unprotected sound level the attenuation value plus two times the standard deviation). The NRR value is the difference between the two overall sound levels less another 3 dB for spectral uncertainty.

This computation methodology and an example are given in subsection 211.207 of the Code of Federal Regulations (40 CFR 211.207). (See also the National Institute of Occupational Safety and Health (NIOSH) Website,

The EPA regulation specifies that the HPD sound attenuation be determined using what is referred to as the “real ear method” (see subsection 206) and that it be tested to the American National Standards Institute (ANSI) standard S3.19-1974. However, in an attempt to get the latest copy of this standard, I learned that this standard has been rescinded and replaced by a newer standard, ANSI S12.6-1997 (R 2002), “Methods for Measuring the Real-Ear Attenuation of Hearing Protectors.”

Attenuation testing
Methodology exists to perform HPD attenuation testing with both human subjects (the “real ear method”) and physical testing with text fixtures and instrumentation. Both aforementioned ANSI standards use Real Ear Attenuation at Threshold (REAT) methods. For the physical test method, a special fixture resembling an ear and ear canal are used. Inside the artificial ear canal is a microphone. Sound level is measured inside and outside the ear with the HPD in place. The difference between the two values is the attenuation, typically measured as a function of frequency in one-third octave bands.

Methodology for the “real ear method” is similar to an audiogram, but instead of wearing a headset and sitting in a small soundproof booth, the subject sits inside a relatively large test chamber called a reverberation test room. Test sounds are then played inside this room at various amplitudes and frequencies (again in one-third octave bands: 125, 250, 500, 1000, 2000, 4000 and 8000 Hertz).

The test is conducted with and without the HPD in the subject’s ear. Amplitude of the sound level at each of the seven one-third octave bands is reduced until the lowest discernable level is determined. When the HPD is not on the human subject, the value determined is called the “open threshold of hearing,” and when the HPD is used, the value determined is called the “occluded threshold of hearing.” The difference between the occluded and open values is known as the “Real Ear Attenuation at Threshold” or REAT. Refer to ANSI S12.6 for more details.

According to S12.6-1997, these tests are conducted on 10 test subjects for earmuffs and on 20 subjects for earplugs. Each subject has their own earplug or earmuff, so testing takes into account manufacturing and individual differences. Each subject and HPD is tested twice during one visit to the lab. The average of each of the two trials for all 10 or 20 subjects is the reported average real ear attenuation. Standard deviation is also computed from the sample set.

Fitting methods
According to the introduction of the S12.6-1997 standard, numerous studies have shown that actual HPD performance in real-world situations is significantly less than what was measured by previous laboratory methods (namely S3.19-1974 and S12.6-1984). For this reason, the 1997 revision of S12.6 included two HPD fitting methods in the standard: the experimenter-supervised fit and the subject-fit.

For the experimenter-supervised fit, the test subject is instructed on proper placement of the HPD. The experimenter may show the subject how to fit the HPD before the test is conducted. During the test, the subject inserts the HPD and may adjust it while a test noise is played to minimize the perceived sound. The experimenter may visually inspect for a correct fit and ask the subject to reinsert the HPD if necessary. The intent of this process is to achieve the best possible performance of the HPD.

In the new subject-fit method, the subject is provided the HPD in its original packaging. The experimenter provides no instructions on insertion of the HPD; the only instructions are what’s written on the packaging. Further, the subjects cannot have any prior knowledge or training on using and wearing HPDs.

A major study conducted by the Federal Government and industry showed that the subject-fit method had as much as a 15-decibel lower attenuation for various types of earplugs and that differences for earmuffs were relatively much lower (no more than a 5-decibel difference). It seems that NRR ratings — at least as they apply to earplugs — are greatly influenced by the fitting protocol used: experimenter-fit or subject-fit.

Testing to S12.6
According to William Murphy, vice chair of the ANSI S12 Standards Committee on Noise and a scientist with NIOSH, some HPDs are now being testing to S12.6, but the EPA regulation does not require the new methodology to be used. The subject-fit approach has been around for at least 10 years (developed in 1997) and was reaffirmed in 2002. Why aren’t these updated methods, which result in more accurate but lower NRR values, being used? The answer rests with the EPA, which must revise and update regulation 40 CFR 211. Alas, according to Murphy, “the EPA is doing just that.”


Royster, J.D., Berger, E. H., Merry, C. J., Nixon, C. W., Franks, J. R., Behar, A., Casalie, J. G., Dixon-Ernst, C., Kieper, R. W., Mozo, B. T., Ohlin, D., and Royster L. H. (1996). “Development of a New Standard Laboratory Protocol for Estimating the Field Effectiveness of Hearing Protection Devices, Part I: Research of Working Group 11, Accredited Standards Committee S12 Noise,” Journal Acoustical Society of America, Volume 99 Edition 3, pages 1506-1526.