Rubrics have attained considerable importance in the authentic and sustainable assessment paradigm; nevertheless, few studies have examined their contribution to validity, especially outside the domain of educational studies. This empirical study used a quantitative approach to analyse the validity of a rubrics-based performance assessment. Raters evaluated the performance of 84 first-year university students producing service-learning projects for the Conservation-Restoration and Design degrees. The study data comprised the 9240 scores given by two teachers and three student tutors, who assessed the students' projects on three occasions during the semester. Factor analyses confirmed that the students attained the expected learning outcomes and made significant learning progress. This learning progress was also corroborated by analyses of variance. The attainment of the learning goals and the evidence of learning progress demonstrated the validity of the inferences drawn from the assessment system. In addition, the results highlighted the need to consider rubrics as a first-order teaching resource and not only as a scoring tool.