The purpose of this study was to investigate the psychometric characteristics of the Math Anxiety Scale (MANX; Erol 1989, Unpublished master thesis, Bogazici University) with data collected from 952 middle school students in Turkey. The Rasch Rating Scale model was used to examine the MANX at the item level. The results revealed that although the MANX was sensitive to detect students with moderate levels of math anxiety and it was not targeted to identify those with very high and low math anxiety levels, it had high reliability and validity. Moreover, the majority of the MANX items were of good quality. The results of this study provide strong evidence for the validation of the MANX despite the need for deletion of eight misfit items and three items with the same item difficulties. Future research should consider possible revision or development of new items to capture gradations of challenges at the very high and low ends of the continuum.