Statistics classes in school education in Japan often fail to narrow down multiple conflicting conclusions into one more persuasive conclusion. This paper analyzes and characterizes a relativistic statistics class from the perspective of inferentialism, thereby examining approaches to avoid relativism. As a result, I pointed out the inclusion of anomalies and the failure to pursue warrants as possible factors in the formation of relativistic norms and discussed strategies to avoid them. Based on the discussion, several challenges in adopting inferentialism as a normative analytical framework for educational phenomena are identified.