This investigation sought to determine the extent to which examiners' evaluations of writing performance were influenced by the learning disability label and to identify mediating variables which modify expectancy effects. Forty-four preservice regular education teachers were randomly assigned to one of two expectancy conditions (learning-disabled and normal) and were asked to score three 4th-grade essays which differed in terms of overall writing quality. Prior to scoring the essays, subjects were informed of the educational status (learning-disabled or normal) of the children who wrote the stories. In addition, half of the subjects received a brief introduction on how to use the scoring procedure, while the other half received intensive practice and training. Results indicated that the learning disability label did influence the scores that examiners assigned to children's essays. However, labeling effects were successfully attenuated by providing examiners with training and practice in scoring written products. Furthermore, the marks assigned to students' essays were influenced far more by the writing competence demonstrated by the child than by the disability label attached to the child. (Author/JDD)