The Scholastic Aptitude Test (SAT) has been designed to test developed verbal and mathematics reasoning abilities of college-bound students, primarily high school juniors and seniors. For almost a decade there has been a research and development process to evaluate and change the entire SAT program. These changes were implemented in the SAT I: Reasoning Test in content and administration changes in April 1994 and scale recentering in April 1995. As of October 1995, a technical manual for the SAT I had not been published, but extensive research on the SAT I has been reported. The SAT I is normed on the 1990 reference group of 1,052,000 scores from 35 editions from October 1988 through June 1990. Field trials of the new test involved 162,692 high school juniors. By modeling the new test using item response theory, it has been estimated that the reliability of the new test is comparable to the old one, if not better. Definitive studies have not been done on test validity but research has suggested that recentering the scale has improved its predictive validity. Pains were taken in the test improvement process to ensure that the new test was not easier than the old one, even though mean test scores "rose." It could be said that changes to the SAT were not only justified but sorely needed to serve the test taking population adequately. Research on the effects of the new test specifications is needed, especially for changes in the mathematics section, that occurred after the field studies. (Contains 18 references.) (SLD)