Discusses the potential consequences of conducting research in which the sample truly represents only a fragment of some theoretical population. Argues that most field-based reading research violates the assumption of representativeness and that this violation can result in severe misrepresentations of research results. (FL)