Low student confidence in mathematics persists. The strength to forecast student confidence in mathematics using mathematics persistence and consensus classroom climate as predictors was examined. Predictive research design was used. Data gathered from 226 Senior High School students selected through stratified random sampling were analyzed using multiple linear regression analysis. The results showed that the criterion was significantly predicted. The strength of the model was strong (65.8%).   The Social Cognitive Theory was partially supported. Educational leaders may implement professional development programs relevant to improving mathematics persistence and consensus classroom climate. Qualitative research may extract themes considered as potential variables to account the remaining (34.2%) of the variance in the strength of the model.