Abstract: In order to study the goodness-of-fit tests of logistic distribution, the author discusses the maximum likelihood estimation based on type Ⅱ censored sample parameters. From this reasearch the following results are obtained: The maximum likelihood estimation of the logistic distribution parameter exists uniquely, has limits and is consistent to the real parameter if the sample size is large enough and its censored ratio of two sides is a constant; The maximum likelihood estimation of the distribution parameter possesses of asymptoptic normality and asymptotic unbias under the same condition.