Abstract
One of the most significant topics in statistics is the issue of variance estimation. In the research literature, various variance estimators are constructed based on traditional moments that are particularly affected by the presence of extreme values. Therefore, the focus of this paper is on the adoption of L-Moments features to propose some new calibration estimators for a variance with some suitable calibration constraints under stratified random sampling. The empirical efficiency of proposed estimators is calculated through simulation based on Covid-19 pandemic data for the period January 22, 2020, up to August 23, 2020. The results indicate that the proposed estimators are superior and highly efficient compared to the existing traditional estimator when the data includes extreme values.