THE 3-YEAR CRAFT PROJECT, COMPARING READING APPROACHES IN FIRST-GRADE TEACHING, STUDIED THE SKILLS-CENTERED APPROACH AND THE LANGUAGE-EXPERIENCE APPROACH. ABOUT 1,372 DISADVANTAGED CHILDREN IN NEW YORK CITY PARTICIPATED. THE RESULTS OF THE FIRST-GRADE STUDY AND THE PRELIMINARY RESULTS OF THE SECOND-GRADE STUDY WERE REPORTED LAST YEAR. THE RESULTS OF THE CONTINUATION OF THE SECOND-GRADE STUDY AND THE FIRST-GRADE REPLICATION RESULTS ARE PRESENTED IN THIS THIRD PROGRESS REPORT. SECOND-GRADE RESULTS SHOWED THAT THE SKILLS-CENTERED APPROACH WAS SUPERIOR TO THE LANGUAGE-EXPERIENCE APPROACH ON THE READING TESTS AND IN SPELLING. THE AUDIOVISUAL VARIATION OF THE LANGUAGE-EXPERIENCE APPROACH WAS LOWEST ON WORD DISCRIMINATION AND READING SUBTESTS. GREATER DIFFERENCES AMONG CLASSES WITHIN EACH METHOD INDICATED THAT THE TEACHER WAS MORE IMPORTANT THAN THE INSTRUCTIONAL TECHNIQUE. AN INTERCORRELATION MATRIX COMPUTATION REVEALED SOME FACTORS WHICH CORRELATED NEGATIVELY WITH READING ACHIEVEMENT. THE FIRST-GRADE REPLICATION STUDY USED THE FOUR ORIGINAL METHODS AND A PILOT METHOD WHICH COMBINED THE PHONOVISUAL TECHNIQUES WITH THE AUDIOVISUAL VARIATION OF THE LANGUAGE-EXPERIENCE APPROACH. THE SKILLS-CENTERED APPROACH PRODUCED SUPERIOR RESULTS IN SPELLING, AND THE PILOT METHOD PRODUCED SUPERIOR RESULTS IN VOCABULARY AND WORD STUDY SKILLS. THE FINAL CONCLUSIONS OF THE WHOLE PROJECT WILL BE REPORTED IN SPRING 1968. (NS)