Double quartile ranked set sampling procedure (DQRSS) and its properties for estimating the population mean are introduced. The performance of DQRSS with respect to simple random sampling (SRS), ranked set sampling (RSS) and quartile ranked set samples (QRSS) for estimating the population mean, is considered. The DQRSS estimator is unbiased of the population mean for symmetric distributions about its mean. In addition, the DQRSS method is more efficient than the SRS, RSS, and QRSS for all symmetric and asymmetric distributions considered in this study. For asymmetric distributions considered in this study, DQRSS estimator has a smaller bias.