The response time of a mobile application (app), especially a mobile stock trading app, is an important factor that affects customer satisfaction. However, it is considerably difficult to accurately evaluate the performance of mobile apps owing to numerous real-world settings such as operating systems, hardware, and test environments. This paper presents a novel method to evaluate the response time of mobile apps on different mobile phones through combining network protocol analysis and information fusion. To make the data collected from the mobile app more reliable and credible, we recruited some volunteers to collect data on their mobile phones. Then we used the network protocol analysis method to obtain the response time of the mobile app on a mobile phone. Next, we adopted information fusion technology using the rank-score characteristic function to evaluate the response time of mobile apps on different mobile phones. Experiments were conducted to evaluate our approach on three types of mobile apps. The results showed that the proposed method can effectively evaluate the response time of mobile apps with low cost.