This dataset is from two experiments using eye movement methodology to assess the processing of four-character collocations in Chinese reading. The data resource including data files, R code, and stimulus sets for two experiments.