Clear-air wind profiler Doppler radars offer the capability to estimate the intensity of small-scale turbulent kinetic energy per unit mass (TKE) over a broad range of altitudes in the troposphere and lower stratosphere from the width of the observed Doppler spectrum. The TKE is theoretically related to the eddy dissipation rate and the eddy diffusivity. However, the observed spectral widths contain contributions related to the interaction of the radar beam with the large-scale wind which must be removed. These corrections are usually significant, sometimes being larger than the turbulence contribution to the observed spectral width. Much of the theory used to estimate the corrections was developed decades ago. An experiment to test the corrections applied to the spectral width was conducted at the highly versatile MU radar in Japan. Essentially simultaneous observations were made using two different beamwidths and two different zenith angles with oblique beams directed at the four cardinal compass points during a two-day period of relatively strong winds (peak winds over 50 ms-1). Theoretical predictions of changes in spectral width with respect to wind speed, azimuth, and altitude are compared with the observations. For example, the results show that the anisotropy of spectral widths between zonal and meridional beams can be removed by appropriate application of the traditional corrections.