Abstract: The spatial distribution of seismic landslides is influenced by a wide range of factors. Understanding the relative importance of these factor is crucial for accurately predicting seismic landslide risks. However, most recently studies on factor importance have focused on individual earthquake events, making it difficult to identify overarching patterns and differences across multiple cases. This study focuses on the eastern margin of the Qinghai-Xizang Plateau and analyzes six earthquake-induced landslide events. Using the decisiveness (DC), we quantitatively assessed the absolute importance of 14 influencing factors: seismic intensity, seismic fault distance, geologic age, non-seismic fault distance, elevation, slope, aspect, geomorphology type, average annual precipitation, river distance, soil type, vegetation type, land use type, and distance to roads. These factors are categorized into six groups: seismic, geo-tectonic, topographic-geomorphic, meteorological-hydrological, soil-vegetation, and human-activity. The results show that: overall, seismic, geo-tectonic, topographic-geomorphic, and meteorological-hydrological factors have a greater influence on seismic landslide occurence, whereas soil-vegetation and human-activity factors are less significant. Moreover, the importance of individual factors varies with earthquake magnitude: for larger-scale earthquakes seismic and geo-tectonic factors dominates; for smaller-scale earthquakes, topographic and hydrological conditions are more influential. By ensuring consistency in the landslide and influencing factor datasets and employing an absolute importance assessment approach across multiple cases, this study provides a systematic analysis of the key drivers of seismic landslides. The findings offer valuable insights for seismic landslide risk assessment and mitigation strategies.