This study tests whether teachers' emotional labor in classroom settings is optimally conceptualized according to the type of emotional labor strategy involved (genuinely expressing, hiding, and faking emotions), the specific type of emotion being performed in class (e.g., enjoyment vs. pride vs. anxiety), or both strategy type and emotion type. Multitrait-multimethod analyses of 1,086 Canadian teachers' survey responses showed teachers' responses to emotional labor items to be most reliably differentiated according to both specific types of emotional labor strategies as well as the valence of the emotion being performed. Findings were largely consistent with common "display rules" encouraging expression of positive emotions and hiding of negative emotions by teachers in classroom settings. Results further showed teachers' emotional labor strategies for negative emotions to be particularly contingent on the specific discrete emotion involved, highlighting the complexity of expressing negative emotions as a behavior management strategy.