Climatological data records have a critical role in the design of renewable power systems. Various data sampling approaches were employed by literature. However, to the authors' knowledge, there is no work comparing them. This paper compares the results of four commonly used data records sampling scenarios (scenario 1:"15th July", scenario 2: "a single yearly-averaged day", scenario 3: "12 monthly-averaged days", and scenario 4: "full yearly data") for the multi-objective optimal design of a standalone hybrid PV-hydroelectric system by using NSGA-II. Investment cost and Loss of Power Supply Probability (LPSP) constitute the objective functions of the problem. The results will reveal that scenario 4 is the only scenario that fully satisfies the demanded power profile over the year, whereas scenarios 1 to 3 are not able to satisfy more than 95.3%, 72.3%, and 92.4% of the yearly demanded power profile, respectively. The currently employed optimization procedure, which combines NSGAII with a simple operational strategy, allows for the implementation of all the mentioned scenarios with tractable computation time (35-4150 min on an Intel (R) Core (TM) i7 4 GHz processor). This study can help designers to select a suitable data records through making a trade-off among data availability, computation time, and power supply probability. (C) 2019 Elsevier Ltd. All rights reserved.