The rapid changes in lifestyle have led to increasingly complex environments for Chinese residents, and various sleep problems have become more common, including difficultly in falling asleep and waking up easily. Sleep, as an important physiological process for individuals to integrate body functions, restore energy and consolidate memory, is closely related to physical and mental health. Sleep quality is an important indicator for physical and mental health, and quality of life. Recent studies have shown that self-compassion, as a healthy self-attitude, is closely related to individuals’ sleep quality. However, the current research on how self-compassion influences sleep quality is still limited, and its mechanism is still unclear. Based on current literatures, self-compassion is closely related to individuals’ positive affect, negative affect, and self-control. At the same time, positive and negative affect, and self-control are proved to have wide influences on individuals’ sleep quality. Therefore, in the current study, we aim to reveal the psychological mechanism about how self-compassion predicts sleep quality through positive and negative affect, and self-control, thus helping people who suffer from poor sleep quality. This research collected self-reported data from 477 adults through the Internet (175 males, 302 females, average 31.7 years). The level of self-compassion was measured with the Self-Compassion Scale. The Positive and Negative Affect Scale was used to gather participants’ levels of positive and negative affect. Self-control and sleep quality were assessed by the Self-Control Scale, and Self-Rating Scale of Sleep. We analyzed the mediation effects of positive and negative affect, and self-control with AMOS 23.0. The main results of this study are as follows: (1) The correlations between each pair of self-compassion, positive affect, self-control as well as sleep quality were all significantly positive. The above variables were all negatively correlated with negative affect. (2) The direct effect of self-compassion on sleep quality was not significant, but self-compassion could predict sleep quality through the complete mediating effects of positive and negative affect, and self-control. There were four indirect paths: through the mediating role of positive affect; through the mediating role of negative affect; through the mediating role of self-control; through the chain mediating role of negative affect and self-control. Our study enriches the relevant research on self-compassion and complements the mechanism about how self-compassion predicts sleep quality. Our findings also shed light on how to handle individual sleep problems. Self-compassion is a dynamic psychological variable that can be adjusted. Brief and easy self-compassion-related trainings in daily life can help individuals improve their self-compassion levels, which provides a great opportunity for future interventions. However, in our study, the cross-sectional design was adopted, which made it difficult to determine causal links between the variables. Secondly, this study only focused on the effects of limited demographic variables such as gender and age on sleep quality, but more variables( e.g., family economic status, education) may play their roles in the relationships among self-compassion, positive and negative affect, self-control, and sleep quality. More demographic variables should be included in future studies to increase the generalizability of the findings. In addition, separately measured positive and negative affect may not accurately reflect individuals’ current emotional states. Future research can employ more comprehensive affect indicators, such as Affect Balance, to re-verify the reliability of the conclusions.