In propositional domains, using a separate test set via random sampling or cross validation is generally considered to be an unbiased estimator of true error. In multirelational domains, previ- ous work has already noted that linkage of ob- jects may cause these procedures to be biased, and has proposed corrected sampling procedures. However, as we show in this paper, the exist- ing procedures ... [Show full abstract] only address one particular case of bias introduced by linkage. We recall that in the propositional case cross validation measures off-training set (OTS) error and not true error and illustrate the difference with a small experiment. In the multirelational case, we show that the dis- tinction between training and test set needs to be carefully extended based on a graph of poten- tially linked objects, and on their assumed proba- bilities of reoccurrence. We demonstrate that the bias due to linkage to known objects varies with the chosen proportion of the training/test split and present an algorithm, generalized subgraph sampling, that is guaranteed to avoid bias in the test set for more generalized cases.