In this paper, in order to establish a confidence interval (general and shortest) for quantiles of normal distribution in the case of one population, we present a pivotal quantity that has non-central t distribution. In the case of two independent normal populations, we construct a confidence interval for the difference quantiles based on the generalized pivotal quantity and introduce a simple method for extracting its percentiles, by which a shorter confidence interval can be constructed. We will also examine the performance of the proposed methods by using simulations and examples.
