Total radiated power (TRP) is commonly accepted as an important figure of merit (FoM) for evaluating the over-the-air (OTA) performance of wireless devices enabled by the emerging fifth generation (5G) mobile communication technology. The statistically homogeneous and isotropic electromagnetic (EM) environment created by a reverberation chamber (RC) makes it an accurate, efficient, and economic testing facility for TRP measurement. In this paper, an improved analytical uncertainty model which is based on the average Rician K-factor (K <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">avg</sub> ) and the number of independent samples is proposed for TRP measurement using an RC. It has the flexibility to allow different stirring configurations in the calibration stage and the measurement stage, and gives insight into the measurement uncertainty without the tedious and inefficient empirical estimation processes. Estimators of K <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">avg</sub> are modelled and analyzed. Specifically, the maximum likelihood estimator (MLE) of K <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">avg</sub> is validated by the Monte Carlo simulation, and its unbiased correction is derived accordingly for improved uncertainty model accuracy. Extensive 9-Point estimation measurements are also conducted in order to evaluate the performance of the proposed analytical model.