A READING PROGNOSIS TEST TO MEASURE FUTURE ABILITY, BASED ON PRESENT SKILLS AND KNOWLEDGE, OF CHILDREN FROM DIFFERENT SOCIOECONOMIC LEVELS WAS CONDUCTED. IT SOUGHT TO AVOID TWO MAJOR SHORTCOMINGS OF STANDARDIZED READING READINESS TESTS (1) THAT CHILDREN OF LOW SOCIOECONOMIC LEVEL WERE INCLUDED IN THE NORMS AND THEREFORE SCORES OF THESE CHILDREN WERE NOT ONLY AT THE LOW END OF THE SCALE BUT WERE CLUSTERED AMONG A FEW NUMBERS, AND (2) THAT NO CLEAR DIFFERENTIATION AMONG UNDERLYING SKILLS WAS GIVEN. A SAMPLE OF CHILDREN WAS DRAWN, WITH EQUAL NUMBERS FROM LOWER AND MIDDLE SOCIOECONOMIC LEVELS. THE TEST WAS CONSTRUCTED TO YIELD SCORES IN THREE AREAS, ALL CONTAINING SUBAREAS. TWO OF THESE, PERCEPTUAL DISCRIMINATION AND LANGUAGE, WERE DESIGNED TO MEASURE SKILLS NECESSARY TO BOTH BEGINNING AND ADVANCED READING. THE THIRD AREA, BEGINNING READING SKILLS, WAS DESIGNED TO MEASURE THE CHILD'S PRESENT STATUS. THREE STUDIES WERE UNDERTAKEN. IN EACH CASE, LATER IN THE SCHOOL YEAR A READING ACHIEVEMENT TEST WAS USED AS A VALIDITY CRITERION. NORMAL VARIABILITY WAS FOUND IN ALL SOCIOECONOMIC GROUPS. IT WAS CONCLUDED THAT POTENTIALLY POOR READERS CAN BE IDENTIFIED BEFORE FORMAL TRAINING IN READING TAKES PLACE AND THAT THEIR SKILL DEFICIENCIES CAN BE ASCERTAINED. TABLES GIVING TEST SCORES AND INTERCORRELATIONS WERE INCLUDED.