Statistical records have been created since the beginning of recorded history, about 3000 B.C.; nevertheless, apparently no specific guidelines for the appraisal of such records have been produced. This article, based on several decades of reflection about archivists' concern for the disposition of materials, examines the burdensome task archival appraisers face in dealing with the typically massive files created by major censuses and surveys. These records may include planning documentation, questionnaires, tabulations, studies, and publications. The article concludes with a suggested method of evaluating the records.