An evaluation is made of the NASS multiple frame (MF) survey estimator and a simple ratio estimator for various agricultural items using operational estimates from various quarterly agricultural surveys between June 1992 and May 1997. The relative mean deviation (R-MD) from the official statistics and the relative root mean square deviation (R-RMSD) from the official statistics were used as the performance criteria for each estimator. More specifically, the effect of the list frame coverage (which at the time of this analysis did not incorporate agricultural census records) on the R-RMSD and R-MD of these estimators is investigated. The empirical evaluations show that the R-RMSD tends to increase for both estimators when the list coverage is below eighty percent. The performance of the MF estimator is similar for the base and non-base survey periods for most items, and the ratio estimator performs similarly to the MF estimator when the list coverage exceeds eighty-five percent.