This standard describes practical methods for measuring the transmission characteristics of a telephone handset or headset by means of using a test connection to obtain objective measurements. Although the various methods generally produce similar results, some deviations can be expected depending on handset or headset characteristics. Measurement conditions should be noted when reporting results. The obtained test results may be used as a means of evaluating or specifying the transmission performance of a handset or headset on a standardized basis. This standard is intended for measuring the transmission performance over the frequency range most useful for speech, 100-5000 Hz. It is to be used for both single-frequency and varying-frequency measurements (such as continuous slow sweeps or equivalent, e.g., closely-spaced discrete steps). Continuous spectrum signals with one- or two-channel spectrum analysis methods, to the extent that they are equivalent to or better than the methods described herein, may also be used. These two types of signals (sinusoidal and continuous spectrum) give essentially identical results for devices measured in the linear region of their operation, when the measurement is sufficiently noise-free. For nonlinear devices, however, the continuous spectrum is preferable. Nonlinear devices include carbon microphones, compressors, automatic gain control (AGC) circuits, etc. For guidance in using continuous spectrum signals, see IEEE Std 269-1992 . Drawings and test procedures, as well as other pertinent information, are included as an aid for use in setting up the tests. The basic accuracy and reproducibility of measurements taken in accordance with this standard depend primarily upon the accuracy of the test equipment used, the care with which the calibration and measurements are conducted, and the inherent stability of the handset or headset under test. Where a given set of measurements is repeated in the same laboratory and by the same operator, a relatively high degree of reproducibility normally should be obtained. However, when comparing measurements taken by different laboratories, allowances should be made for the influencing factors mentioned above.