This study presents a fuzzy logic framework for modeling self-efficacy, a critical component of self-regulated learning (SRL), using Likert-scale data. Traditional methods often treat Likert responses as strictly numeric with equal intervals; fuzzy logic enables a more complex representation by correlating responses to linguistic terms like “Low” or “High” and employing reasoning with partial membership. This study utilised a Mamdani-type fuzzy inference system to evaluate student responses to a self-efficacy assessment statement. Data from 149 students in three academic modules and several year levels was analysed using triangle membership functions and expert-defined criteria. The framework includes four main stages: fuzzification of inputs, rule-based inference, aggregation, and centroid defuzzification. Results showed that 71.8% of students agreed they could achieve their goals, with predicted self-efficacy scores averaging 3.98 out of 5. In contrast to conventional Likert scoring, the fuzzy methodology yielded significantly more conservative estimates ( \(\textrm{p} < 0.001\) , d = 0.30) while maintaining the validity of the fundamental construct (r = 0.99), which shows improved psychological measuring attributes. This approach illustrates the ability of fuzzy logic to improve the understanding of psychological constructs while providing beter interpretability than basic statistical averages.