The possibility of using translations of American reading tests for the evaluation of pupils belonging to different foreign groups was explored. Two parallel forms of a reading comprehension test geared to United States high school graduates and college entrants and the translations of these into Turkish and the relative retranslations back into English were administered to five groups of high school and college students in the United States and Turkey. Item difficulty and frequency of responses to item errors were highly stable in the two groups. There was great similarity in the total test scores of American and Turkish students at similar educational levels when the test was taken in their own language. This seems to indicate that translated reading tests remain culturally fair if total test scores, relative difficulty of reading passages, and indices of item difficulty are criteria for test fairness. (WL)