The aim of this study was to investigate the relationship between cognitive emotional regulation strategies, test anxiety, and academic self-efficacy among secondary school students, with a focus on gender differences. The study included 166 male students and 136 female students from grammar and technical vocational schools in northwest Croatia, aged 15 to 19 years (M=17.21; SD=1.02). Data were collected using the Cognitive Emotion Regulation Questionnaire (CERQ), the Test Anxiety Scale, and the Academic Self-Efficacy Scale. Students completed the questionnaires online during school hours. The results revealed statistically significant gender differences in grade point averages, self-blame, rumination, and putting into perspective strategies, levels of test anxiety, and self-efficacy in achieving desired learning outcomes. Grade point averages positively correlated with academic self-efficacy, with these correlations varying by gender. A low but significant association was observed between grade point averages and test anxiety. Strategies of self-blame, rumination, and catastrophising in both genders are positively correlated with anxiety, with which planning and positive reappraisal strategies are negatively correlated. Regarding academic self-efficacy, significant positive associations were found with positive refocusing, planning, and positive reappraisal strategies for both genders, while catastrophizing was negatively associated. Significant negative correlations were found between test anxiety and academic self-efficacy for both genders. In predicting academic self-efficacy, significant predictors include the average school grades and "adaptive" emotional regulation strategies such as positive reappraisal and putting into perspective, as well as potentially planning and positive refocusing in the case of predicting components of academic self-efficacy. The percentages of explained variance range from 20.9% to 48.1%. Gender and components of test anxiety did not prove to be significant predictors. Unexpected findings were obtained in predicting self-efficacy in the learning process among female students, as the grade average was not a significant predictor, indicating the need for further research.