As the size and complexity of mission-critical application software continues to grow, the cost of Software Testing (ST) is also increasing. The numerous methods and processes used to evaluate ST dynamically and quantitatively to improve testing efficiency serve as practice problems in the ST field. Based on multi-dimensional test coverage models, this paper proposes a dynamic evaluation method for ST and discusses it from the viewpoint of testing monitoring information, dynamic analysis and evaluation models, and testing optimized strategy. Furthermore, a concept, Synthetic Test Coverage (STC), is defined in this paper, and its empirical formulas are also presented. Examples show that the methods are useful in helping ST evaluation groups to track and control the effects of ST and for improving the user’s ability to observe and control the ST process.