ABSTRACT This study aims at investigating the validity and reliability studies of the “Computer Anxiety Scale” (Ceyhan & Gurcan Namlu, 2000) on educational administrators. The data gathered from 143 educational administrators of state schools located in Eskişehir show that the scale consists of 2 factors. The first of these factors, affective anxiety regarding computers independently explains 37.99 % of the variance; the second factor, cognitive anxiety regarding computers forms 12.49 % of the variance. These two factors, which consist of 20 items altogether, explain 50.48 % of the variance. It was found that the scale was able to discriminate the computer anxiety level of educational administrators with respect to computer efficacy. With the scale, the correlation between Spielberger State and Trait Anxiety Inventory (Öner and Le Compte,1983) was calculated as 0.34 and 0.25. In addition, the correlation coefficient between the scale and Happner and Petersen’s Problem-Solving Inventory (Şahin, Şahin, Happner,1993) was found to be 0.40. The internal consistency coefficient was calculated as (a) 0.87. Item total score reliability coefficients were calculated between 0.23 and 0.71. As a result, the “Computer Anxiety Scale of Educational Administrators” (CAS-EA) was developed.