The present study tested the assumption of a positive and linear relation between years of schooling and school learning in the Philippine setting. It replicated a 1976 study that had cast doubt on this assumption in the Philippine public educational system. It tested three competing hypotheses for that finding: common sense, the 1976 arrested development hypothesis, and the alternative accelerated development hypothesis. To test these competing hypotheses, two factors were systematically varied: the grade levels of Ss and the levels of the tests used. The competing hypotheses have different predicted outcomes. A total n of 7097 from 96 schools participated in the study. The results showed that on all tests Grade 5 showed more competencies than Grades 4 and 6, although Grade 6 continued to perform better than Grade 4. When sub-test level was held constant in multiple comparisons, Grade 5 was learning more Grade 6 competencies, whereas Grade 6 was losing not only Grade 6 but also Grade 5 competencies. It is noted that whereas Grade 6 enjoyed a slight superiority in achievement scores circa 1976, the present study shows that Grade 5 enjoys an impressive superiority over Grade 6 circa 2003. That in Grade 6 one knows more competencies than in Grade 5 seems to be a myth. The common sense hypothesis has been ruled out. The results are consistent with the accelerated development hypothesis. (Contains 1 table, 4 figures and 1 footnote.)