THIS STUDY ATTEMPTED TO REDUCE THE NUMBER OF VARIABLES FROM THE EDUCATIONAL OPPORTUNITIES SURVEY QUESTIONNAIRE FOR NINTH GRADE STUDENTS IN AN EMPIRICALLY MEANINGFUL WAY IN ORDER TO REDUCE THE VOLUME OF DATA PROCESSING AND COMPLEXITY OF LATER ANALYSES. AN ACHIEVEMENT COMPOSITE WAS FORMED BY WEIGHTING AND SUMMING A STUDENT'S SCORE ON THE FOLLOWING FIVE TESTS--NONVERBAL ABILITY, VERBAL ABILITY, READING COMPREHENSION, MATHEMATICS ACHIEVEMENT, AND GENERAL INFORMATION. FIFTY-SEVEN OF THE QUESTIONNAIRE ITEMS WERE CODED TO MAXIMIZE THEIR CORRELATION WITH THE ACHIEVEMENT COMPOSITE. THESE ANALYSES WERE CONDUCTED ON THE RESPONSES AND SCORES OF 12,000 OF THE 133,136 STUDENTS WHO PARTICIPATED IN THE SURVEY. THE ANALYSES YIELDED SEVEN MEANINGFUL INDEXES--EXPECTATIONS FOR EXCELLENCE, SOCIOECONOMIC STATUS, SOCIAL CONFIDENCE, ATTITUDE TOWARD LIFE, FAMILY STRUCTURE AND STABILITY, EDUCATIONAL DESIRES AND PLANS, AND STUDY HABITS. ALL OF THEM WERE FOUND TO BE MODERATELY CORRELATED WITH ACHIEVEMENT, RACIAL-ETHNIC DIFFERENCES, AND SEX. FURTHER ANALYSIS SHOWED THAT SOCIOECONOMIC STATUS AND FAMILY STRUCTURE AND STABILITY WERE IMPORTANT VARIABLES IN PREDICTING ACHIEVEMENT AND THE ATTITUDINAL INDEXES. ALSO, SEX WAS AN IMPORTANT EXPLANATORY VARIABLE FOR STUDY HABITS, AND RACIAL-ETHNIC DIFFERENCES WAS IMPORTANT IN EXPLAINING ACHIEVEMENT AND ATTITUDE TOWARD LIFE. (DK)