The prevailing tendency in modern university reforms is towards "how people learn," following a learner-centered approach in which the learner is the main actor of the teaching-learning process. As a consequence, one of the key indicators of the teaching-learning process is the measurement of learner satisfaction within the classroom. Learner satisfaction has traditionally been measured using survey responses to a standard learning survey. However, more scientific analysis should be performed to assess adequately not only learner satisfaction but also the main dimensions that have a positive impact on learner satisfaction. The purpose of this paper is to define a structural and measurement model in which causal relationships among these different dimensions are adequately established. The methodology is based on a multivariate regression model (Structural Equation Models) to establish scientifically a structural model for learner satisfaction within a classroom, measuring its validity and reliability. The proposed approach has been applied to model learner satisfaction in an electronic instrumentation course at the University of Seville, Spain. The results and implications of this study will contribute to improve student satisfaction with respect to the dimensions considered. (Contains 5 tables and 3 figures.)