Aggregation, or grouping, is a statistical procedure through which all members of a study within a specified range of scores (usually observed scores) are assigned a common or "group" score (for example, the group mean). The various social science methodology literatures agree on the costs of grouping: not only does one always lose information in grouping, in a wide variety of situations grouping introduces systematic error (bias). For most educational research applications the existing guidelines are probably appropriate. There is however, a class of situations in which grouping (of a particular type) will tend to compensate for errors in the original specification. That is, there are certain situations in which grouping produces a gain. Two special cases in which grouping is beneficial are discussed. Both cases involve estimations of effects in structural equation models. One case concerns grouping that minimizes (grouped) variation in confounding variables. The second case concerns the effects of grouping on measurement error. The benefits of the second case are less clear than in the first. The mathematical framework for both cases is presented as are areas for further investigation. (JKS)