Font Size: 

Model-assisted survey estimators estimate finite population parameters by using prediction methods to combine auxiliary information available at a population level with complex survey data. These methods assume that observations obtained for the sample can be matched without error to the auxiliary data.  We investigate properties of estimators that rely on matching algorithms that do not in general yield perfect matches.  We allow for multiple frames from which sampled elements can be selected and for multiple sources of auxiliary data. A class of multi-frame difference estimators with possibly imperfect matching is proposed.  The estimators are exactly unbiased under perfect matching but not under imperfect matching. The methods are investigated analytically and via simulation, using a study of recreational angling in South Carolina to build simulation populations with various structures.