A STUDY OF SELECTION AND RETENTION OF CANDIDATES FOR ELEMENTARY TEACHER EDUCATION PROGRAMS WAS DESIGNED TO TEST AND IMPROVE THE SELECTION OF TEACHER CANDIDATES. THE 8-YEAR, CROSS SECTIONAL, LONGITUDINAL STUDY INVOLVED THREE MATCHED GROUPS (N-802) OF MATRICULATED FRESHMEN ELEMENTARY EDUCATION MAJORS DURING THE 1957-60 PERIOD. SCORES ON 11 STANDARDIZED TESTS WERE DESIGNED TO MEASURE SUCH VARIABLES AS PERSONALITY, MOTIVATION, INTEREST, ATTITUDES, AND SCHOLASTIC APTITUDE. HIGH SCHOOL RECORDS AND COLLEGE GRADES MADE UP THE REMAINDER OF THE OBJECTIVE DATA. SUBJECTIVE DATA CONSISTED OF SEVEN RECORDED PROFESSIONAL JUDGMENTS, FOUR OBTAINED DURING PRESERVICE TRAINING, AND THREE DURING THE SUBJECT'S 1ST INSERVICE YEAR. OF THE OBJECTIVE DATA, ONLY SELECTED SCALES OF THE MINNESOTA MULTIPHASIC PERSONALITY INVENTORY AND THE DYNAMIC FACTOR OPINION SURVEY APPEARED TO BE EFFECTIVE AS SCREENING DEVICES AT THE PRESERVICE LEVEL. INTERVIEWS DURING THE FRESHMEN YEAR INDICATED A SIGNIFICANT RELATIONSHIP BETWEEN EMOTIONAL BALANCE AND MOTIVATION, AND PRESERVICE COMPETENCE. AT THE SOPHOMORE LEVEL, ONLY SOCIAL INTELLIGENCE WAS SIGNIFICANTLY CORRELATED WITH PRESERVICE TEACHING SUCCESS. BASED ON STUDIES USING EITHER PRE- OR IN-SERVICE CRITERIA OF COMPETENCE, A SIGNIFICANT RELATIONSHIP WAS FOUND BETWEEN PRE- AND IN-SERVICE ASSESSMENTS OF COMPETENCE. RECOMMENDATIONS FOR INSTITUTIONAL USE OF THE FINDINGS ARE INCLUDED. (HA)