De. Powers et Sw. Leung, ANSWERING THE NEW SAT READING-COMPREHENSION QUESTIONS WITHOUT THE PASSAGES, Journal of educational measurement, 32(2), 1995, pp. 105-129
It has been reasonably well established that test takers can, to varyi
ng degrees, answer some reading comprehension questions without readin
g the passages on which the questions are based, even for carefully co
nstructed measures like the Scholastic Aptitude Test (SAT). The aim of
this study was to determine what test-taking strategies examinees use
, and which are related to test performance, when reading passages are
not available. The research focused on reading comprehension question
s similar to those that will be used in the revised SAT, to be introdu
ced in 1994. The most often cited strategies involved choosing answers
on the basis of consistency with other questions and reconstructing t
he main theme of a missing passage from all of the questions and answe
rs in a set. These strategies were more likely to result in successful
performance on individual test items than were any of many other poss
ible (and less construct-relevant) strategies.