The predictive roles of self-compassion, perceived social support, and psychological flexibility in early maladaptive schemas among college students: an exploration based on latent profile analysis.

Purpose: This study used latent profile analysis (LPA) to identify subgroups of early maladaptive schemas (EMSs) among college students based on the five core schema domains, and then investigated how these profiles related to self-compassion, perceived social support, and psychological flexibility. Methods: A total of 1,184 college students from universities in Northwest China were selected using cluster sampling (47.3% male, 52.7% female; 58.1% freshmen, 21.4% sophomores, 12.8% juniors, and 7.7% seniors). Participants completed a cross-sectional survey including the Short Form of the Young Schema Questionnaire, the Self-Compassion Scale, the Perceived Social Support Scale, and the Avoidance and Fusion Questionnaire (Short Form). Results: LPA identified three different EMSs profiles: low, moderate, and high. The high group scored significantly higher in disconnection and rejection, impaired autonomy and performance, impaired limits, other-directedness and over-vigilance and inhibition compared with the other two groups, while the low group demonstrated the lowest scores across all domains. Additionally, degrees of self-compassion, perceived social support, and psychological flexibility differed considerably between profiles. Multinomial logistic regression indicated that gender, grade level, self-compassion, perceived social support, and psychological flexibility significantly predicted profile membership in the expected directions. Conclusion: The study revealed clear variations in EMSs among college students, yielding three distinct profiles. The findings support the hypothesis, and provide a theoretical basis for developing targeted psychological interventions aimed at enhancing self-compassion, strengthening social support, and improving psychological flexibility.