In order to establish a good measure of mixing rate, jet mixing is examined using the DNS (direct numerical simulation) data of controlled jets. In the computation, spatial discretization is performed using a hybrid scheme that adopts a sixth-order compact scheme in the streamwise direction and a Fourier series in the lateral directions. The Reynolds number is 1500. Two measures such as statistical entropy and a mixedness parameter are investigated using a passive scalar concentration. Compared with simple measures, i.e., jet width, centerline velocity, and turbulent kinetic energy, it is found that the statistical entropy is a good measure for representing the mixing state in different controlled jets, and it is possible to use the fluctuating entropy to detect a highly mixed region and determine the correlation of the mixed regions with the vortical structure. In addition to statistical entropy, the mixedness parameter is also found to have useful properties for the estimation of mixing efficiency. These findings suggest that these measures are expected to contribute to the optimization of jet mixing.