Objective: Researches have pointed out perfectionism is a risk factor on development, maintenance and course of psychopathology. In recent years, clinical implications perfectionism is delineated in cognitive behavioural models of interventional approaches. In this vein, development of the Frost Multidimensional Perfectionism Scale (FMPS) seems to be a cornerstone in defining dimensions of the concept and recognizing possible linkages of perfectionism to various psychological outcomes. Of various instruments, the FMPS is one of the well-known instruments, translated into several languages, and addressed referring to its psychometric properties in many empirical studies. In this study, our aim was to examine psychometric properties of the FMPS, particularly validity of original factor structure in Turkish sample, by using confirmatory factor analysis. Methods: Data were collected from 492 undergraduates (183 males, 37.2%). Participants were administered the FMPS, the Limitedness Schemas Questionnaire, the Penn State Worry Scale, and the Brief Symptom Inventory. Item statistics were calculated. Validity of six dimensions of the scale in Turkish sample was tested by using confirmatory factor analysis with normality correction. Correlations of composite and subscale scores with psychological variables were computed. Reliability of the scale was assessed by performing inner consistencies and intra-correlations between two applications 15 days apart. Results: Confirmatory factor analysis show that six factor structure was valid in Turkish sample. Pearson correlations of psychological variables with the instrument scores were adequate. Inner consistency for overall items was 0.91. Chronbachs alpha for the subscales ranged between 0.64-0.94. 15 day intra-correlation for the composite score was 0.82. Conclusion: The Turkish version of the FMPS with its original six factor-structure is a valid and reliable instrument. It seems that Organization subscale represents healthy perfectionism. It is recommended not to add Organization subscale scores in computing composite scores particularly in studies concerning linkages to psychopathology. [Anadolu Psikiyatri Derg 2011; 12(3.000): 192-197]