USING A CROSS-SECTIONAL SAMPLING PLAN, THE STABILITY OF WECHSLER INTELLIGENCE SCALE FOR CHILDREN (WISC) PROFILES, MEAN SCALE SCORES ON THE FOLLOWING SUBTESTS--INFORMATION, COMPREHENSION, ARITHMETIC, SIMILARITIES, VOCABULARY, AND DIGIT SPAN WAS EXAMINED AS A FUNCTION OF SEX, AGE, AND ABILITY LEVEL. FROM 167 WISC FORMS (OF BLIND BOYS AND GIRLS AGED EIGHT TO 12) 80 FORMS WERE USED FOR SEX AND AGE ANALYSIS. A SECOND GROUP OF 39 FORMS DRAWN FROM THE SAME POOL WERE USED FOR THE ABILITY ANALYSIS. SEX AND AGE EFFECTS WERE ANALYZED IN A THREE-FACTOR DESIGN WITH REPEATED MEASURES ON SUBTESTS, AND ABILITY EFFECTS WERE ANALYZED IN A TWO-FACTOR DESIGN WITH REPEATED MEASURES ON SUBTESTS. RESULTS SHOWED MAIN EFFECTS OF SEX AND AGE WERE NOT SIGNIFICANT. THE MAIN EFFECT OF SUBTESTS WAS SIGNIFICANT (.001 LEVEL). THESE INTERACTIONS WERE NOT SIGNIFICANT--SEX AND AGE, SEX AND SUBTESTS, AGE AND SUBTESTS, SEX AND AGE AND SUBTESTS. THIS RESULT ATTESTS TO THE ACCURACY OF GROUPING PROCEDURES. THE SECOND ANALYSIS SHOWED THAT MAIN EFFECTS AND INTERACTION EFFECTS WERE SIGNIFICANT--ABILITY GROUPS (.001 LEVEL), SUBTESTS (.001 LEVEL), ABILITY GROUP AND SUBTESTS (.05 LEVEL). THIS SIGNIFICANT MAIN EFFECT WAS PLANNED, SINCE GROUPS WERE FORMED BY LOW, AVERAGE, AND HIGH IQ. SEX AND AGE DIFFERENCES DID NOT PRODUCE PROFILE DIFFERENCES WHILE GROUPING BY LOW, AVERAGE, AND HIGH IQ DID YIELD DIFFERENT PROFILES. A LIST OF FIVE REFERENCES IS INCLUDED. (AA)