This study aimed to investigate the network structure and core features of depressive symptoms among older adults in China，and to compare the networks of depressive symptoms among older adults across different genders and severity levels of depression. Data from the 2020 China Health and Retirement Longitudinal Study （CHARLS） were used for the survey，which included 8 445 valid samples. Network analyses were conducted using R software. The results of the study showed that the core features of depressive symptoms in older adults were low mood and feeling unable to go on with their lives. There was a marginally significant difference in the network structure of depressive symptoms among older adults of different genders， while no significant differences in the strength of network connections. The centrality value of feeling unable to go on with their lives was significantly higher in men than in women. There were significant differences in certain edge weights between depressive symptom networks across genders. There were significant differences in the network structure and network strength of depressive symptoms among older adults with different severity levels of depression. The centrality values of feeling depressed and feeling unable to go on with their lives were higher in the group without depressive symptoms than in the group with depressive symptoms. These results suggest that the network structure of depressive symptoms among Chinese older adults is complex. It is recommended to prioritize core symptoms，such as feeling down，in the prevention and intervention of depressive symptoms among older adults，while developing personalized intervention strategies for gender differences to improve the mental health of older adults.