Overlapping clustering allows a data point to be a member of multiple clusters, which is more appropriate for modeling many real data semantics. However, much of the existing work on overlapping clustering simply assume that a data point can be assigned to any number of clusters without any constraint. This assumption is not supported by many real contexts. In an attempt to reveal true data cluster structure, we propose sparsity constrained overlapping clustering by incorporating sparseness constraints into an overlapping clustering process. To solve the derived sparsity constrained overlapping clustering problems, efficient and effective algorithms are proposed. Experiments demonstrate the advantages of our overlapping clustering model.