Using a normalizing transformation to monitor process dispersion has received a great deal of attention. The exponentially weighted moving average (EWMA) control chart based on the logarithmic transformation of sample variance for monitoring increases in process variability is one such dispersion chart. The traditional EWMA-type dispersion chart resets the EWMA statistic to zero whenever it falls below zero. This paper proposes a new EWMA dispersion chart (NEWMA) by truncating negative normalized observations to zero in the traditional EWMA statistic. The comparison result shows that the NEWMA chart outperforms the traditional EWMA chart for detecting dispersion changes, especially at small changes.