Population-based reference intervals (RIs) are vital tools used to characterize health and disease based on laboratory values. The science and statistical basis for RI generation have evolved over the past 50 yr. Current veterinary-specific guidelines by the American Society of Veterinary Clinical Pathology exist for establishing RIs from nondomestic and wild animals. A list of 35 items that should be included during generation and publication of reference data was distilled from the currently available RI guidelines. The archives of five peer-reviewed journals were searched and 106 articles presenting laboratory reference data from nondomestic or wildlife species were identified and each reviewed by two authors to determine compliance with the list of 35 items. A compliance score was calculated as the number of articles that fulfilled the item out of the number where it would have been appropriate to fulfill the item. Most articles reported the number of reference individuals (compliance score 0.98), their partitioning demographics (compliance score 0.95), and sample collection and handling practices (compliance scores 0.97 and 0.96, respectively). Common deficiencies included omitting discussion of the validation status of the analytical methods for the species being evaluated (compliance score 0.12), documentation of use of exclusion criteria (compliance score 0.51), outlier detection (compliance score 0.43), appropriate statistical methods for the reference population (compliance score 0.34), and calculation and presentation of confidence intervals around the reference limits (compliance score 0.35). Compliance scores were not statistically different when stratified on the number of individuals in the largest and smallest evaluated group or the format of the article (full vs short format). Articles that cited RI generation guidelines fulfilled more of the required steps and provided a more complete description of their data (compliance score 0.74) than those that did not cite guidelines (compliance score 0.58). Additional attention to the science of and recommendations for RI generation is recommended to strengthen the utility of published data.