**Project Description** Project on longitudinal (10 year) data, that investigates how self-esteem and social relationships are related by using continuous time modeling techniques. ---------- **Project Abstract** Research on the longitudinal association between self-esteem and satisfaction with social relationships led to ambiguous conclusions regarding the temporal order and strength of this relation. Existing studies have examined this association across intervals ranging from days to years, leaving it unclear as to what extent differences in timing may explain differences across studies. In the present study, we used continuous time structural equation models to examine cross-lagged relations between the constructs (i.e., CT-SEM), and also distinguished between-person differences from within-person processes (i.e., RI-CT-SEM). We analyzed 10 years of annual data from the Longitudinal Internet Studies of the Social Sciences (LISS; *N* = 14,741). When using CT-SEM, we found a bidirectional positive relation between self-esteem and satisfaction with social relationships, with larger effects over longer intervals. When using RI-CT-SEM, we found the largest effects of self-esteem and satisfaction with social relationships across intervals of one year, with smaller effect sizes at both shorter and longer intervals. Additionally, the effect of fluctuations in people’s satisfaction with social relationships on fluctuations in their self-esteem was greater than the reverse effect. Our results highlight the importance of considering time when examining the relation between self-esteem and interpersonal outcomes, and likely psychological constructs in general. *Keywords:* self-esteem; relationship satisfaction; continuous time modeling; cross-lagged panel model; random-intercept cross-lagged panel model ---------- **Pre-Registration** The reseach questions, hypotheses, and initial analytic plan were pre-registered at: https://osf.io/ngv3y/