Abstract: In order to investigate the relationships between various competition indices and their effects on forest structure, 12 different tree competition indices were selected and divided into three categories: tree distance, tree crown and tree relative size indices. The analysis was based on data from a field plot of a spruce-fir mixed broadleaf-conifer forest in the Changbai Mountain area. Statistical analysis involving a structural equation model (SEM) was used to evaluate the reliability and validity of these three categories of competition indices. The results show that tree distance indices show greater variation, while dispersion degrees are moderate and consistent. Correlation effects in the structural equation model are significantly and consistently positive. Our reliability analysis and a validity analysis reveal that tree distance indices exhibit superior internal consistency and the greatest potential to reflect the real tree competition. Tree crown indices show a high degree of variability and the reliability and validity values are the lowest. The tree crown indices were also correlated with the tree distance indices. The performance of the tree relative size indices, regarding variability, degree of dispersion, correlation effects, reliability and validity are between the tree distance and tree crown indice performances. The tree relative size indices are related to distance and crown indices. The conclusion is that tree distance-dependent competition indices are more stable and reliable, tree size-related indices are in second place, while tree crown-related competition indices are rather volatile, which can be expected as tree crowns are usually measured with error.