Objectives: The accuracy of references was assessed in Turkish journals of orthopedics and traumatology.
Methods: All references cited in four journals (Acta Orthopaedica et Traumatologica Turcica, Arthroplasty Arthroscopic Surgery, Hacettepe Journal of Orthopaedic Surgery, and The Journal of Turkish Spinal Surgery) throughout 1995 to 1999 were numbered consecutively. One percent of references from each journal was randomly selected to be evaluated in comparison with primary sources. Errors were classified in three categories. Another randomized selection was made to check 50 references from each journal. The results were assessed by the t-test.
Results: Of 18,193 references cited in four journals, 182 references were checked. Errors were found in 117 references (63%), seven of which were major errors that prevented location of the primary source. When 50 references were checked from each journal, the mean inaccuracy rate was 60%. No significant differences were found between the journals with respect to the frequency of erroneous references.
Conclusion: The high rate of inaccuracy emphasizes the need for greater care in the use of references on the part of authors, journal editors, and reviewers.