Pitfall trapping is a standard sampling method to compare the abundance or community structure of ground beetles. However, effects of sampling duration on biodiversity estimation of ground beetles according to different trap sizes have not been experimentally evaluated in temperate forests in Korea. Therefore, the objective of this study was to determine the interaction between trap sizes (SB, small‐sized bottle; MB, medium‐sized bottle; PC, plastic cup; PJ, plastic jar; PT, perforated type trap; FT, funnel‐type trap) and sampling duration (T1, 2 weeks × 2 sessions, 28 days; T2, 2 weeks × 4 sessions, 56 days; T3, 4 weeks × 2 sessions, 56 days) on estimation of ground beetle assemblages in Naejangsan National Park, a temperate forest in Korea. Funnel type and larger pitfall traps collected higher numbers of individuals and species than other trap sizes. Species composition of ground beetles was different by size of traps (SB, MB, and PC vs. PJ, PT, and FT). In particular, ground beetle composition in larger traps (PJ, PT, and FT) appeared to be influenced by environmental characteristics according to localities (e.g., soil characters and dominant tree species). These findings from our study support that pitfall trapping of ground beetles can be influenced by trap sizes per se as well as sampling durations and environmental characteristics. Thus, biodiversity monitoring in temperate forests should be conducted with long sampling duration (at least 28 days) using large‐sized traps (> 7.5 cm in diameter of trap mouthpart) considering expenses and study aims.