a b s t r a c t
In this article, two modified ratio estimators of the population mean are suggested provided
that the first or third quartiles of the auxiliary variable can be established when the mean
of the auxiliary variable is known. The double-sampling method is used to estimate the
mean of the auxiliary variable if it is unknown. The suggested estimators are investigated
under simple random sampling (SRS) and median ranked set sampling (MRSS) schemes.
The new estimators when using MRSS are compared to their counterparts under SRS. The
bias and the mean square error of the proposed estimators are derived. It turns out that
the estimators are approximately unbiased and the MRSS estimators are more efficient
than the SRS estimators, on the basis of the same sample size, correlation coefficient, and
quartile. A real data set is used for illustration