Controversies concerning the role of frozen section (FS) have been a matter of debate. The aim of this study was to identify the role of FS analysis in intraoperative decision making and analyze the effect of the cost in detecting thyroid malignancies in Turkey. Out of 214 consecutive patients who had been operated on for thyroid cancer between January 1996 and August 2004, 178 patients were evaluated retrospectively. All 178 patients were subjected to FS. Intraoperative FS correctly identified the pathology as malignant in 58.4% of patients. A true-positive FS result changed the surgical strategy in 30 (27.6%) cases False negative FS lesions were defined histologically as papillary microcarcinoma in 54%, follicular variant of papillary cancer in 18% and follicular cancer in 8% of cases. The sensitivities of FNAB and intraoperative FS in thyroid cancer patients were 22.5% and 58.4%, respectively. False negative FS results increased the cost for each informative FS from (sic)25 to (sic)42.7. Despite limitations, results of this Study reject the idea that the role of FS is becoming limited. We recommend routine frozen section in the operative assessment of thyroid nodules. Omitting FS may be suggested only in cases with a FNAB revealing malignancy.