INTRODUCTION: There are many reports assessing specificity (Sp) and sensitivity (Se) of diagnostic tests for carpal tunnel syndrome (CTS) such as Tinel sign, Phalen test and electrodiagnostic (EDX) tests in the literature. Most of these assessments are done with the premise that the EDX test is the gold standard in the diagnosis of CTS. This assumption is not entirely correct. When a gold standard is assumed where actually none exists, the Se and Se of the tests maybe erroneously low. AIM: To calculate the Se and Sp of the diagnostic tests of CTS when a gold standard is not available. The Latent Class Analysis (LCA) is a mathematical model which allows calculation of Se and Sp in the absence of a gold standard. This is available as a software which can be downloaded from the internet. We used the LCA to calculate the Se and Sp of Tinel sign, Phalen test and EDX test in the diagnosis of CTS. The prevalence rate in the sample population also can be calculated with the LCA. MATERIAL AND METHODS: Charts of patients presenting with symptoms of upper extremity nerve compression were scrutinized. One hundred such patients were selected whom had concurrent recording of the Tinel sign, Phalen test and value of EDX test expressed as a binary (i.e positive or negative) result. Of the 100 patients, 185 hands with these data were identified. The data was analyzed using the LCA software. Results Using the LCA software, the Se and Sp of Tinel sign was 0.97 and 0.92, that of Phalen test was 0.9 and 0.89 and of EDX tests was 0.92 and 0.86 respectively. The prevalence rate of CTS in the sample population was 0.62. CONCLUSIONS: The estimates we obtained using the LCA are higher than previous reports. We believe that these reflects the true state of these tests as this method proportions the error between the three tests based on the best estimate from the observed data.