This research is motivated by the low level of student learning independence in Islamic Religious Education (PAI) at High School 13 Bandar Lampung, where the self-consistency indicator only reached 19%. This phenomenon indicates the need for innovative learning models that can foster students' self-confidence and responsibility for learning. This study aims to test the effectiveness of the Think Pair Share (TPS) Cooperative Learning model in improving learning independence, especially in the aspect of student self-consistency. This type of research is a quasi-experiment with a pretest-posttest control group design. The sample was selected using a simple random sampling technique consisting of an experimental class (TPS model) and a control class (conventional method). Data was analyzed using the Independent Samples T-Test after fulfilling the prerequisite tests of Shapiro-Wilk normality (Sig. X = 0.154; Y = 0.173) and Levene's homogeneity (Sig. 0.330). The research findings show that the TPS model has a highly significant partial effect on the dependent variable with a t-value of 57.326 and a significance of 0.000. The mechanism of individual thinking, pair discussions, and sharing results has proven effective in building self-confidence, courage to express opinions, and student responsibility. Theoretically, this study provides novelty in the application of the TPS model in the context of Islamic Religious Education at the high school level, especially in proving that academic procrastination behavior is a reliable predictor of students' levels of self-efficacy and independence. Practically, this study recommends the use of the TPS model to strengthen character and create a participatory classroom atmosphere.