Precipitation is one of the most important climatic variables, and due to its direct role in water resources, agriculture, and human livelihoods, its accurate assessment is of great significance. In recent years, gridded precipitation datasets, including station, satellite, and reanalysis data, have found widespread application due to their easy accessibility, lower cost, suitable spatial coverage, absence of missing data, and long temporal duration; however, examining their accuracy and validity is essential for scientific use. In this study, five precipitation datasets, namely APHRODITE and CRU (raingauge-based), PERSIANN-CDR (a combination of satellite and raingauge), and NCEP CFSR and ERA5 (reanalysis-based), were evaluated using the indicators R², NSE, NRMSE, BIAS, POD, FAR, and CSI, in comparison with data from 9 synoptic stations in the Khuzestan province. Since the number of indicators was large, the TOPSIS multi-criteria decision-making method was used for the final ranking of the datasets at each station. The results indicated that APHRODITE, CRU, and NCEP/CFSR provided the best performance, with APHRODITE exhibiting the highest agreement with observations across all stations; its average indicators were POD=0.756, FAR=0.290, and CSI=0.692. It was also found that some datasets may have lower accuracy in estimating precipitation amounts but demonstrate suitable performance in identifying rainy days. Therefore, the selection of a precipitation dataset should be based on the intended application.