AbstractThis study is part of a growing wave of interest in foreign language (FL) learners’ emotions, their sources, and their effects. Previous studies have confirmed that there is a clear relationship between the emotions of foreign language enjoyment (FLE), foreign language classroom anxiety (FLCA), foreign language boredom (FLB), and foreign language performance. However, the relative importance of each emotion as a predictor of FL performance has yet to be examined, and as different teaching and learning strategies can elicit different emotions, it is difficult to determine whether FL teachers and learners should prioritize a specific emotion in course design and study. We, therefore, utilized structural equation modeling and latent dominance analysis on a sample of 502 Moroccan EFL learners in order to examine the relative importance of each emotion in predicting FL performance. We argue that it is crucial to use sophisticated statistical analyses and to collect samples from outside Western, educated, industrialized, rich, and democratic (WEIRD) countries. The latent dominance analysis revealed that FLCA had the strongest (negative) effect on English test scores. FLB had a significant—but slightly weaker—negative effect and FLE had a significant—but weaker still—positive effect. As such, it is vital that FL teachers and learners not underestimate the impact of anxiety on language learning.