Based on the fact that the uniaxial compressive strengths (UCS) of rock are normally distributed random variables, we first used mathematical statistics to analyze the scattering and the interval parameters of UCS of rock samples, then deduced a theoretical calculation formula about interval parameters of UCS of rock spamles and analyzed the relationship between the sample size and the mean of uniaxial compressive strength, the sample standard deviation, the confidence interval for population mean, and the confidence interval for population standard deviation of UCS of rock samples, and finally employed the previous test data to verify and analyze the interval parameters. The results show that the present single value method doesn't contain enough information and cannot guarantee the reliability of the comparison between tests data obtained under different test conditions. However, the interval estimation method can overcome the deficiency and can be used even if the sample size is small, but its confidence level will be lower (confidence interval width will be greater). When the sample size is small, the volatility of the sample standard deviation is one order of magnitude higher than that of the sample mean, the randomness and the discreteness of the sample standard deviation is greater, and the sample size for achieving stable convergence is larger. The confidence interval and the confidence level of UCS of rock samples are related to sample size, heterogeneity of rock sample, and randomness of rock strength distribution, etc.