Pfeffermann (1994) has proposed a solution to the long-standing problem of variance measures for time series seasonally adjusted by the X-11 method. The method uses X-11's linear filters for decomposing an observed series into trend, seasonal, and irregular. This paper pulls together refinements of the basic method and empirical evaluation in settings needed for practical applications, including ARIMA extrapolation and multiplicative adjustment. Also, comparisons are made to a method developed by Bell & Kramer (1999). Extensive use is made of estimates of sampling error variances and autocorrelations. This is especially important for household employment and unemployment series from the U.S. Current Population Survey, which has sizable autocorrelations at a 12-month lag and beyond. Employment series from the Current Employment Statistics establishment survey use a ratio estimator, similar in style to many index formulas, that requires special considerations. Evaluation is based on important series from both surveys and simulation experiments styled on actual series.